@article{TAC23_Ristic,
 author = {Marko Ristic and Benjamin Noack and Uwe D. Hanebeck},
 journal = {IEEE Transactions on Automatic Control},
 title = {Distributed Range-Only Localisation that Preserves Sensor and Navigator Privacies},
 year = {2023}
}

@inproceedings{IFAC23_Fennel,
 address = {Yokohama, Japan},
 author = {Michael Fennel and Lukas Driller and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd IFAC World Congress (IFAC 2023)},
 month = {July},
 pdf = {IFAC23_Fennel.pdf},
 title = {Observability-based Placement of Inertial Sensors on Robotic Manipulators for Kinematic State Estimation},
 year = {2023}
}

@inproceedings{IFAC23_Reith-Braun,
 address = {Yokohama, Japan},
 author = {Marcel Reith-Braun and Albert Bauer and Maximilian Staab and Florian Pfaff and Georg Maier and Robin Gruna and Thomas Längle and Jürgen Beyerer and Harald Kruggel-Emden and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd IFAC World Congress (IFAC 2023)},
 month = {July},
 title = {GridSort: Image-based Optical Bulk Material Sorting Using Convolutional LSTMs},
 year = {2023}
}

@inproceedings{FUSION23_Hanebeck,
 address = {Charleston, USA},
 author = {Uwe D. Hanebeck},
 booktitle = {Proceedings of the 26th International Conference on Information Fusion (Fusion 2023)},
 month = {June},
 title = {Progressive Bayesian Particle Flows Based on Optimal Transport Map Sequences},
 year = {2023}
}

@inproceedings{FUSION23_Baur,
 address = {Charleston, USA},
 author = {Tim Baur and Johannes Reuter and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 26th International Conference on Information Fusion (Fusion 2023)},
 month = {June},
 title = {Progressive Bayesian Particle Flows Based on Optimal Transport Map Sequences},
 year = {2023}
}

@inproceedings{FUSION23_Frisch,
 address = {Charleston, USA},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 26th International Conference on Information Fusion (Fusion 2023)},
 month = {June},
 pdf = {Fusion23_Frisch.pdf},
 title = {Deterministic Sampling of Arbitrary Densities Using Equal Sphere Packing of Volume under the Density (PoVuD)},
 year = {2023}
}

@inproceedings{FUSION23_Fennel,
 address = {Charleston, USA},
 author = {Michael Fennel and Serge Garbay and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 26th International Conference on Information Fusion (Fusion 2023)},
 month = {June},
 pdf = {Fusion23_Fennel.pdf},
 title = {Intention Estimation with Recurrent Neural Networks for Mixed Reality Environments},
 year = {2023}
}

@inproceedings{FUSION23_Reith-Braun,
 address = {Charleston, USA},
 author = {Marcel Reith-Braun and Jakob Thumm and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 26th International Conference on Information Fusion (Fusion 2023)},
 month = {June},
 pdf = {Fusion23_Reith-Braun.pdf},
 title = {Approximate First-Passage Time Distributions for Gaussian Motion and Transportation Models},
 year = {2023}
}

@inproceedings{FUSION23_Ernst,
 address = {Charleston, USA},
 author = {Eugen Ernst and Florian Pfaff and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 26th International Conference on Information Fusion (Fusion 2023)},
 month = {June},
 title = {Multitarget-Multidetection Tracking Using the Kernel SME Filter},
 year = {2023}
}

@article{TM23_Maier,
 author = {Georg Maier and Marcel Reith-Braun and Albert Bauer and Robin Gruna and Florian Pfaff and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Jürgen Beyerer},
 doi = {10.1515/teme-2023-0033},
 journal = {tm -- Technisches Messen, De Gruyter},
 month = {May},
 title = {Simulation study and experimental validation of a neural network-based predictive tracking system for sensor-based sorting},
 url = {https://doi.org/10.1515/teme-2023-0033},
 year = {2023}
}

@inproceedings{ACC23_Ernst,
 address = {San Diego, CA, USA},
 author = {Eugen Ernst and Florian Pfaff and Uwe D. Hanebeck and Marcus Baum},
 booktitle = {Proceedings of the 2023 American Control Conference (ACC 2023)},
 month = {May},
 title = {The Kernel-SME Filter with Adaptive Kernel Widths for Association-free Multi-target Tracking (to appear)},
 year = {2023}
}

@inproceedings{ACC23_Vieth,
 address = {San Diego, CA, USA},
 author = {Jonathan Vieth and Marcel Reith-Braun and Albert Bauer and Florian Pfaff and Georg Maier and Robin Gruna and Thomas Längle and Harald Kruggel-Emden and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2023 American Control Conference (ACC 2023)},
 month = {May},
 pdf = {ACC23_Vieth.pdf},
 title = {Improving Accuracy of Optical Sorters Using Closed-Loop Control of Material Recirculations (to appear)},
 year = {2023}
}

@article{arXiv23_Hanebeck,
 author = {Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2303.02412},
 month = {March},
 title = {Progressive Bayesian Particle Flows based on Optimal Transport Map Sequences},
 url = {https://arxiv.org/abs/2303.02412},
 year = {2023}
}

@article{arXiv23_Li,
 author = {Kailai Li and Ziyu Cao and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2301.09033},
 month = {January},
 title = {Continuous-Time Ultra-Wideband-Inertial Fusion},
 url = {https://arxiv.org/abs/2301.09033},
 year = {2023}
}

@article{Access22_Siebler,
 author = {Benjamin Siebler and Stephan Sand and Uwe D. Hanebeck},
 doi = {10.1109/ACCESS.2022.3223693},
 journal = {IEEE Access},
 pages = {123080-123093},
 pdf = {Access22_Siebler.pdf},
 title = {Bayesian Cramér--Rao Lower Bound for Magnetic Field-Based Localization},
 volume = {10},
 year = {2022}
}

@article{CIT22_Bauer,
 author = {Albert Bauer and Georg Maier and Marcel Reith-Braun and Harald Kruggel-Emden and Florian Pfaff and Robin Gruna and Uwe D. Hanebeck and Thomas Längle},
 journal = {Chemie Ingenieur Technik},
 pdf = {CIT22_Bauer.pdf},
 title = {Benchmarking a DEM--CFD Model of an Optical Belt Sorter by Experimental Comparison},
 url = {https://doi.org/10.1002/cite.202200124},
 year = {2022}
}

@article{TAC22_Ristic,
 author = {Marko Ristic and Benjamin Noack and Uwe D. Hanebeck},
 journal = {IEEE Transactions on Automatic Control},
 title = {Distributed Range-Only Localisation that Preserves Sensor and Navigator Privacies (to appear)},
 year = {2022}
}

@article{JAIF22_Zea,
 author = {Antonio Zea and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 title = {Modeling Spatial Uncertainty for the iPad Pro Depth Sensor},
 year = {2022}
}

@inproceedings{FB22_Maier,
 address = {Karlsruhe, Germany},
 author = {Georg Maier and Marcel Reith-Braun and Albert Bauer and Robin Gruna and Florian Pfaff and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Jürgen Beyerer},
 booktitle = {Proceedings of the Image Processing Forum 2022},
 month = {November},
 pdf = {FB22_Maier.pdf},
 title = {Machine Learning-Based Multiobject Tracking for Sensor-Based Sorting},
 year = {2022}
}

@article{AT22_Fennel,
 author = {Michael Fennel and Antonio Zea and Uwe D. Hanebeck},
 journal = {at -- Automatisierungstechnik},
 month = {October},
 title = {Intuitive and Immersive Teleoperation of Robot Manipulators for Remote Decontamination},
 url = {https://doi.org/10.1515/auto-2022-0057},
 year = {2022}
}

@inproceedings{MFI22_Baur,
 address = {Cranfield, United Kingdom},
 author = {Tim Baur and Johannes Reuter and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2022 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2022)},
 month = {September},
 pdf = {MFI22_Baur.pdf},
 title = {Harmonic Functions for Three-Dimensional Shape Estimation in Cylindrical Coordinates},
 year = {2022}
}

@inproceedings{MFI22_Frisch,
 address = {Cranfield, United Kingdom},
 author = {Daniel Frisch and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2022 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2022)},
 month = {September},
 pdf = {MFI22_Frisch.pdf},
 title = {Optimal Sensor Placement for Multilateration Using Alternating Greedy Removal and Placement},
 year = {2022}
}

@inproceedings{MFI22_Fennel,
 address = {Cranfield, United Kingdom},
 annote = {Winner Best Paper Award Certificate (PDF)},
 author = {Michael Fennel and Lukas Driller and Antonio Zea und Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2022 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2022)},
 month = {September},
 pdf = {MFI22_Fennel.pdf},
 title = {Calibration-free IMU-based Kinematic State Estimation for Robotic Manipulators},
 year = {2022}
}

@article{PowTec22_Bauer,
 author = {Albert Bauer and Georg Maier and Marcel Reith-Braun, Harald Kruggel-Emden and Florian Pfaff and Robin Gruna and Uwe D. Hanebeck and Thomas Längle and Jürgen Beyerer},
 journal = {Powder Technology},
 month = {September},
 title = {Numerical Modelling of an Optical Belt Sorter Using a DEM--CFD Approach Coupled with Particle Tracking Algorithm and Comparison with Experiments},
 year = {2022}
}

@inproceedings{Fusion22_Frisch-Rejection,
 address = {Linköping, Sweden},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Frisch-Rejection.pdf},
 title = {Rejection Sampling from Arbitrary Multivariate Distributions Using Generalized Fibonacci Lattices},
 year = {2022}
}

@inproceedings{Fusion22_Frisch-PCD,
 address = {Linköping, Sweden},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Frisch-PCD.pdf},
 title = {Deterministic Sampling on the Circle Using Projected Cumulative Distributions},
 year = {2022}
}

@inproceedings{Fusion22_Pfaff,
 address = {Linköping, Sweden},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Pfaff.pdf},
 title = {The State Space Subdivision Filter for SE(3)},
 year = {2022}
}

@inproceedings{Fusion22_Baur,
 author = {Tim Baur and Johannes Reuter and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 ddress = {Linköping, Sweden},
 month = {July},
 pdf = {Fusion22_Baur.pdf},
 title = {Extent Estimation of Sailing Boats Applying Elliptic Cones to 3D LiDAR Data},
 year = {2022}
}

@inproceedings{Fusion22_XianqingLi,
 address = {Linköping, Sweden},
 author = {Xianqing Li and Zhansheng Duan and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_XianqingLi.pdf},
 title = {Recursive Joint Cramér-Rao Lower Bound for Nonlinear Parametric Systems with Colored Noises},
 year = {2022}
}

@inproceedings{Fusion22_Li,
 address = {Linköping, Sweden},
 annote = {Winner Best Paper Award: First Runner Up Certificate (PDF)},
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Li.pdf},
 title = {Circular Discrete Reapproximation},
 year = {2022}
}

@inproceedings{Fusion22_Prossel,
 address = {Linköping, Sweden},
 author = {Dominik Prossel and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Prossel.pdf},
 title = {Dirac Mixture Reduction Using Wasserstein Distances on Projected Cumulative Distributions},
 year = {2022}
}

@inproceedings{Fusion22_Zea,
 address = {Linköping, Sweden},
 author = {Antonio Zea and Michael Fennel and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Zea.pdf},
 title = {Robot Joint Tracking With Mobile Depth Cameras for Augmented Reality Applications},
 year = {2022}
}

@inproceedings{Fusion22_Noack,
 address = {Linköping, Sweden},
 author = {Benjamin Noack and Clemens Öhl and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Noack.pdf},
 title = {Event-Based Kalman Filtering Exploiting Correlated Trigger Information},
 year = {2022}
}

@inproceedings{Fusion22_Zhang,
 address = {Linköping, Sweden},
 author = {Donglin Zhang and Zhansheng Duan and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 25th International Conference on
Information Fusion (Fusion 2022)},
 month = {July},
 pdf = {Fusion22_Zhang.pdf},
 title = {Asynchronous Multi-Radar Tracking Fusion with Converted Measurements},
 year = {2022}
}

@article{Haptics21_Fennel,
 author = {Michael Fennel and Antonio Zea and Uwe D. Hanebeck},
 journal = {IEEE Transactions on Haptics},
 pdf = {Haptics21_Fennel.pdf},
 title = {Optimization-Driven Design of a Kinesthetic Haptic Interface with Human-Like Capabilities},
 year = {2021}
}

@inproceedings{Fusion21_Frisch,
 address = {Sun City, South Africa},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 24th International Conference on Information Fusion (Fusion 2021)},
 month = {November},
 pdf = {Fusion21_Frisch.pdf},
 title = {Deterministic Gaussian Sampling With Generalized Fibonacci Grids},
 year = {2021}
}

@inproceedings{Fusion21_Pfaff,
 address = {Sun City, South Africa},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 24th International Conference on Information Fusion (Fusion 2021)},
 month = {November},
 pdf = {Fusion21_Pfaff.pdf},
 title = {Deep Likelihood Learning for 2-D Orientation Estimation Using a Fourier Filter},
 year = {2021}
}

@inproceedings{Fusion21_Dunik,
 address = {Sun City, South Africa},
 author = {Jindřich Duník and Ondřej Straka and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 24th International Conference on Information Fusion (Fusion 2021)},
 month = {November},
 pdf = {Fusion21_Dunik.pdf},
 title = {Cooperative Unscented Kalman Filter with Bank of Scaling Parameter Values},
 year = {2021}
}

@article{arXiv21_Li,
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2110.10411},
 month = {October},
 title = {Hyperspherical Dirac Mixture Reapproximation},
 url = {https://arxiv.org/abs/2110.10411},
 year = {2021}
}

@article{TII21_Thumm,
 author = {Jakob Thumm and Marcel Reith-Braun and Florian Pfaff and Uwe D. Hanebeck and Merle Flitter and Georg Maier and Robin Gruna and Thomas Längle and Albert Bauer and Harald Kruggel-Emden},
 journal = {IEEE Transactions on Industrial Informatics},
 month = {September},
 pdf = {TII21_Thumm.pdf},
 title = {Mixture of Experts of Neural Networks and Kalman Filters for Optical Belt Sorting},
 year = {2021}
}

@inproceedings{MFI21_Pfaff,
 address = {Karlsruhe, Germany},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2021 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2021)},
 month = {September},
 pdf = {MFI21_Pfaff.pdf},
 title = {Conditional Densities and Likelihoods for Hypertoroidal Densities Based on Trigonometric Polynomials},
 year = {2021}
}

@inproceedings{MFI21_Zhao,
 address = {Karlsruhe, Germany},
 author = {Haibin Zhao and Christopher Funk and Benjamin Noack and Uwe D. Hanebeck and Michael Beigl},
 booktitle = {Proceedings of the 2021 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2021)},
 month = {September},
 pdf = {MFI21_Zhao.pdf},
 title = {Kalman Filtered Compressive Sensing Using Pseudo-Measurements},
 year = {2021}
}

@inproceedings{MFI21_Radtke,
 address = {Karlsruhe, Germany},
 author = {Susanne Radtke and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2021 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2021)},
 month = {September},
 pdf = {MFI21_Radtke.pdf},
 title = {Learning and Exploiting Partial Knowledge in Distributed Estimation},
 year = {2021}
}

@inproceedings{MFI21_Frisch,
 address = {Karlsruhe, Germany},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2021 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2021)},
 month = {September},
 pdf = {MFI21_Frisch.pdf},
 title = {Gaussian Mixture Estimation from Weighted Samples},
 year = {2021}
}

@inproceedings{MFI21_Baur,
 address = {Karlsruhe, Germany},
 author = {Tim Baur and Johannes Reuter and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2021 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2021)},
 month = {September},
 pdf = {MFI21_Baur.pdf},
 title = {Shape Estimation and Tracking using Spherical Double Fourier Series for Three-Dimensional Range Sensors},
 year = {2021}
}

@article{Sensors21_Pfaff,
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 doi = {https://doi.org/10.3390/s21186314},
 journal = {Sensors},
 month = {September},
 title = {The State Space Subdivision Filter for Estimation on SE(2)},
 url = {https://www.mdpi.com/1424-8220/21/18/6314},
 year = {2021}
}

@article{SoftwareImpacts21_Fennel,
 author = {Michael Fennel and Stefan Geyer and Uwe D. Hanebeck},
 doi = {https://doi.org/10.1016/j.simpa.2021.100109},
 issn = {2665-9638},
 journal = {Software Impacts},
 month = {August},
 pdf = {SoftwareImpacts21_Fennel.pdf},
 title = {RTCF: A Framework for Seamless and Modular Real-Time Control with ROS},
 url = {https://www.sciencedirect.com/science/article/pii/S2665963821000427},
 year = {2021}
}

@article{RAL21_Li,
 author = {Kailai Li and Meng Li and Uwe D. Hanebeck},
 journal = {IEEE Robotics and Automation Letters},
 month = {July},
 number = {3},
 pdf = {RAL21_Li.pdf},
 title = {Towards High-Performance Solid-State-LiDAR-Inertial Odometry and Mapping},
 url = {https://ieeexplore.ieee.org/document/9392274},
 vol = {6},
 year = {2021}
}

@article{arXiv21_Mayer,
 author = {Jana Mayer and Johannes Westermann and Juan Pedro Gutiérrez H. Muriedas and Uwe Mettin and Alexander Lampe},
 journal = {arXiv preprint arXiv:2107.09647},
 month = {July},
 title = {Proximal Policy Optimization for Tracking Control Exploiting Future Reference Information},
 url = {https://arxiv.org/abs/2107.09647},
 year = {2021}
}

@article{JAIF21_Radtke,
 author = {Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {June},
 number = {1},
 pages = {3--16},
 title = {Fully Decentralized Estimation Using Square-Root Decompositions},
 url = {https://confcats_isif.s3.amazonaws.com/web-files/journals/entries/3-16.pdf},
 volume = {16},
 year = {2021}
}

@inproceedings{ECC21_Westermann,
 address = {Virtual},
 author = {Johannes Westermann and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2021 European Control Conference (ECC 2021)},
 month = {June},
 pdf = {ECC21_Westermann.pdf},
 title = {Adaptive Sampling for Global Meta Modeling Using a Gaussian Process Variance Measure},
 year = {2021}
}

@article{LCSS21_Fennel,
 abstract = {The programming of manipulators is a common task in robotics, for which numerous solutions exist. In this work, a new programming method related to the common master-slave approach is introduced, in which the master is replaced by a digital twin created through haptic and visual rendering. To achieve this, we present an algorithm that enables the haptic rendering of any programmed robot with a serial manipulator on a general-purpose haptic interface. The results show that the proposed haptic rendering reproduces the kinematic properties of the programmed robot and directly provides the desired joint space trajectories. In addition to a stand-alone usage, we demonstrate that the proposed algorithm can be easily paired with existing visual technology for virtual and augmented reality to facilitate a highly immersive programming experience.},
 author = {Michael Fennel and Antonio Zea and Johannes Mangler and Arne Roennau and Uwe D. Hanebeck},
 journal = {IEEE Control Systems Letters},
 month = {June},
 pdf = {LCSS21_Fennel.pdf},
 title = {Haptic Rendering of Arbitrary Serial Manipulators for Robot Programming},
 year = {2021}
}

@article{CAE21_Maier,
 abstract = {Automatic quality control has long been an integral part of the processing of food and agricultural products. Visual inspection offers solutions for many issues in this context and can be employed in the form of sensor-based sorting to automatically remove foreign and low quality entities from a product stream. However, these methods are limited to defects that can be made visible by the employed sensor, which usually restricts the system to defects appearing on the surface. An alternative non-visual solution lies in impact-acoustic methods, which do not suffer from this constraint. However, these are strongly limited in terms of material throughput and consequently not suitable for large scale industrial application. In this paper, we present a novel approach that performs inspection based on optically acquired motion data. A high-speed camera captures image sequences of test objects during a transportation process on a chute with a specific structured surface. The trajectory data is then used to classify test objects based on their motion behavior. The approach is evaluated experimentally on the example of distinguishing defect-free hazelnuts from ones that suffer from insect damage. Results show that by merely utilizing the motion data, a recognition rate of up to 80% for undamaged hazelnuts can be achieved. A major advantage of our approach is that it can be integrated in sensor-based sorting systems and is suitable for high throughput applications.},
 author = {Georg Maier and Anja Shevchyk and Merle Flitter and Robin Gruna and Thomas Längle and Uwe D. Hanebeck and Jürgen Beyerer},
 doi = {https://doi.org/10.1016/j.compag.2021.106147},
 issn = {0168-1699},
 journal = {Computers and Electronics in Agriculture},
 keywords = {Object trajectory, Motion classification, Sensor-based Sorting, Impact-acoustic},
 month = {June},
 pages = {106147},
 pdf = {CAE21_Maier.pdf},
 title = {Motion-Based Visual Inspection of Optically Indiscernible Defects on the Example of Hazelnuts},
 url = {https://www.sciencedirect.com/science/article/pii/S0168169921001654},
 volume = {185},
 year = {2021}
}

@article{arXiv21_Frisch-2,
 author = {Daniel Frisch and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2106.05109},
 month = {June},
 title = {Gaussian Mixture Estimation from Weighted Samples},
 url = {https://arxiv.org/abs/2106.05109},
 year = {2021}
}

@article{LCSS21_Ristic-2,
 author = {Marko Ristic and Benjamin Noack and Uwe D. Hanebeck},
 doi = {10.1109/LCSYS.2021.3084405},
 issn = {2475-1456},
 journal = {IEEE Control Systems Letters},
 month = {May},
 pdf = {LCSS21_Ristic-2.pdf},
 title = {Cryptographically Privileged State Estimation With Gaussian Keystreams},
 url = {https://doi.org/10.1109/LCSYS.2021.3084405},
 year = {2021}
}

@article{SIMP21_Zea,
 author = {Antonio Zea and Uwe D. Hanebeck},
 doi = {https://doi.org/10.1016/j.simpa.2021.100057},
 issn = {2665-9638},
 journal = {Software Impacts},
 keywords = {Robotics, Data visualization, Augmented reality},
 month = {May},
 pages = {100057},
 title = {iviz: A ROS visualization app for mobile devices},
 url = {https://www.sciencedirect.com/science/article/pii/S2665963821000051},
 volume = {8},
 year = {2021}
}

@article{Sensors21_Funk,
 author = {Christopher Funk and Benjamin Noack and Uwe D. Hanebeck},
 doi = {https://doi.org/10.3390/s21093059},
 journal = {Sensors},
 month = {April},
 pdf = {Sensors21_Funk.pdf},
 title = {Conservative Quantization of Covariance Matrices with Applications to Decentralized Information Fusion},
 url = {https://www.mdpi.com/1424-8220/21/9/3059},
 year = {2021}
}

@article{Sensors21_Li,
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 doi = {10.3390/s21092991},
 journal = {Sensors},
 month = {April},
 pdf = {Sensors21_Li.pdf},
 title = {Progressive von Mises--Fisher Filtering Using Isotropic Sample Sets for Nonlinear Hyperspherical Estimation},
 url = {https://www.mdpi.com/1424-8220/21/9/2991},
 year = {2021}
}

@article{RAL21_Fennel,
 abstract = {Despite significant advances in robot autonomy, manual intervention by a human operator is necessary in many situations. This usually requires qualified staff and some robot-specific input device even for the comparatively simple case of platform locomotion. For this reason, we propose a novel path generation method applicable to car-like vehicles. With this method, the operator “draws” a desired 2D path by walking in a large-scale haptic interface while a guiding force is exerted, which ensures that the generated path can later be accurately followed by a path tracking controller running offline on a remote robot. We present a local optimization-based path planner, a higher-level path generation algorithm utilizing the aforementioned planner, and a force feedback law. Experiments show improved feasibility of the generated paths without affecting the operator's ability to make decisions independently.},
 author = {Michael Fennel and Antonio Zea and Uwe D. Hanebeck},
 doi = {10.1109/LRA.2021.3067846},
 issn = {2377-3766},
 journal = {IEEE Robotics and Automation Letters},
 keywords = {Haptic interfaces;Robots;Kinematics;Turning;Force;Legged locomotion;Trajectory;Haptics and haptic interfaces;human performance augmentation;motion and path planning;telerobotics and teleoperation},
 month = {April},
 number = {2},
 pages = {4088-4095},
 pdf = {RAL21_Fennel.pdf},
 title = {Haptic-Guided Path Generation for Remote Car-Like Vehicles},
 volume = {6},
 year = {2021}
}

@article{LCSS21_Li,
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 doi = {10.1109/LCSYS.2020.3005066},
 issn = {2475-1456},
 journal = {IEEE Control Systems Letters},
 month = {April},
 number = {2},
 pages = {647--652},
 pdf = {LCSS21_Li.pdf},
 title = {Unscented Dual Quaternion Particle Filter for SE(3) Estimation},
 url = {https://doi.org/10.1109/LCSYS.2020.3005066},
 volume = {5},
 year = {2021}
}

@article{IEEESensors21_Siebler,
 author = {Benjamin Siebler and Stephan Sand and Uwe D. Hanebeck},
 doi = {10.1109/JSEN.2020.3024073},
 issn = {1530-437X, 1558-1748},
 journal = {IEEE Sensors Journal},
 language = {english},
 month = {February},
 number = {3},
 pages = {3388-3397},
 pdf = {IEEESensors21_Siebler.pdf},
 publisher = {Institute of Electrical and Electronics Engineers (IEEE)},
 title = {Localization with Magnetic Field Distortions and Simultaneous Magnetometer Calibration},
 volume = {21},
 year = {2021}
}

@article{arXiv21_Frisch,
 author = {Daniel Frisch and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2102.04528},
 month = {February},
 title = {Deterministic Sampling on the Circle using Projected Cumulative Distributions},
 url = {https://arxiv.org/abs/2102.04528},
 year = {2021}
}

@article{LCSS21_Ristic,
 abstract = {Fast covariance intersection is a widespread technique for state estimate fusion in sensor networks when cross-correlations are not known and fast computations are desired. The common requirement of sending estimates from one party to another during fusion forfeits local privacy. Current secure fusion algorithms rely on encryption schemes that do not provide sufficient flexibility. As a result, excess communication between estimate producers is required, which is often undesirable. We propose a novel method of homomorphically computing the fast covariance intersection algorithm on estimates encrypted with a combination of encryption schemes. Using order revealing encryption, we show how an approximate solution to the fast covariance intersection weights can be computed and combined with partially homomorphic encryptions of estimates, to calculate an encryption of the fused result. The described approach allows secure fusion of any number of private estimates, making third-party cloud processing a viable option when working with sensitive state estimates or when performing estimation over untrusted networks.},
 author = {Marko Ristic and Benjamin Noack and Uwe D. Hanebeck},
 doi = {10.1109/LCSYS.2020.3000649},
 issn = {2475-1456},
 journal = {IEEE Control Systems Letters},
 month = {January},
 number = {1},
 pages = {217--222},
 pdf = {LCSS21_Ristic.pdf},
 title = {Secure Fast Covariance Intersection Using Partially Homomorphic and Order Revealing Encryption Schemes},
 url = {https://doi.org/10.1109/LCSYS.2020.3000649},
 volume = {5},
 year = {2021}
}

@article{arXiv20_Kurz,
 author = {Gerhard Kurz and Florian Faion and Florian Pfaff and Antonio Zea and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2012.13580},
 month = {December},
 title = {Three-dimensional Simultaneous Shape and Pose Estimation for Extended Objects Using Spherical Harmonics},
 url = {https://arxiv.org/pdf/2012.13580},
 year = {2020}
}

@inproceedings{IROS20_Zea,
 address = {Las Vegas, USA},
 author = {Christoph Pohl and Kevin Hitzler and Raphael Grimm and Antonio Zea and Uwe D. Hanebeck and Tamim Asfour},
 booktitle = {Proceedings of the 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2020)},
 month = {October},
 pdf = {IROS20_Zea.pdf},
 title = {Affordance-Based Grasping and Manipulation in Real World Applications},
 year = {2020}
}

@article{arXiv20_Li,
 author = {Kailai Li and Meng Li and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:2010.13150},
 month = {October},
 title = {Towards High-Performance Solid-State-LiDAR-Inertial Odometry and Mapping},
 url = {https://arxiv.org/abs/2010.13150},
 year = {2020}
}

@inproceedings{MFI20_Li,
 address = {Virtual},
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2020)},
 month = {September},
 pdf = {MFI20_Li.pdf},
 title = {Nonlinear von Mises--Fisher Filtering Based on Isotropic Deterministic Sampling},
 year = {2020}
}

@inproceedings{MFI20_Frisch,
 address = {Virtual},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2020)},
 month = {September},
 pdf = {MFI20_Frisch.pdf},
 title = {Efficient Deterministic Conditional Sampling of Multivariate Gaussian Densities},
 year = {2020}
}

@inproceedings{MFI20_Funk,
 address = {Virtual},
 author = {Christopher Funk and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2020)},
 month = {September},
 pdf = {MFI20_Funk.pdf},
 title = {Conservative Quantization of Fast Covariance Intersection},
 year = {2020}
}

@inproceedings{MFI20_Pollithy,
 address = {Virtual},
 author = {Daniel Pollithy and Marcel Reith-Braun and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2020)},
 month = {September},
 pdf = {MFI20_Pollithy.pdf},
 title = {Estimating Uncertainties of Recurrent Neural Networks in Application to Multitarget Tracking},
 year = {2020}
}

@inproceedings{MFI20_Pfaff,
 address = {Virtual},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2020)},
 month = {September},
 pdf = {MFI20_Pfaff.pdf},
 title = {Estimating Correlated Angles Using the Hypertoroidal Grid Filter},
 year = {2020}
}

@article{arXiv20_Zea,
 archiveprefix = {arXiv},
 author = {Antonio Zea and Uwe D. Hanebeck},
 eprint = {2008.12725},
 journal = {arXiv preprint arXiv:2008.12725},
 month = {August},
 primaryclass = {cs.RO},
 title = {iviz: A ROS Visualization App for Mobile Devices},
 url = {https://arxiv.org/abs/2008.12725},
 year = {2020}
}

@inproceedings{IFAC20_Rosenthal,
 author = {Florian Rosenthal and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Rosenthal.pdf},
 title = {Stability Analysis of Polytopic Markov Jump Linear Systems with Applications to Sequence-Based Control over Networks},
 year = {2020}
}

@inproceedings{IFAC20_Frisch,
 author = {Daniel Frisch and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Frisch.pdf},
 title = {Optimal Reduction of Dirac Mixture Densities on the 2-Sphere},
 year = {2020}
}

@inproceedings{IFAC20_Pfaff,
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Pfaff.pdf},
 title = {The Spherical Grid Filter for Nonlinear Estimation on the Unit Sphere},
 year = {2020}
}

@inproceedings{IFAC20_Li-PoseGraph,
 author = {Kailai Li and Johannes Cox and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Li-PoseGraph.pdf},
 title = {Improved Pose Graph Optimization for Planar Motions Using Riemannian Geometry on the Manifold of Dual Quaternions},
 year = {2020}
}

@inproceedings{IFAC20_Li-UPF,
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Li-UPF.pdf},
 title = {Hyperspherical Unscented Particle Filter for Nonlinear Orientation Estimation},
 year = {2020}
}

@inproceedings{IFAC20_Noack,
 author = {Benjamin Noack and Christopher Funk and Susanne Radtke and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Noack.pdf},
 title = {State Estimation with Event-Based Inputs Using Stochastic Triggers},
 year = {2020}
}

@inproceedings{IFAC20_Radtke,
 author = {Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Radtke.pdf},
 title = {Reconstruction of Cross-Correlations between Heterogeneous Trackers Using Deterministic Samples},
 year = {2020}
}

@inproceedings{IFAC20_Mayer,
 author = {Jana Mayer and Ajit Basarur and Mariana Petrova and Fabian Sordon and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 1st Virtual IFAC World Congress (IFAC-V 2020)},
 month = {July},
 pdf = {IFAC20_Mayer.pdf},
 title = {Position and Speed Estimation of PMSMs Using Gaussian Processes},
 year = {2020}
}

@inproceedings{Fusion20_Basarur,
 address = {Virtual},
 author = {Ajit Basarur and Jana Mayer and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 23rd International Conference on Information Fusion (Fusion 2020)},
 month = {July},
 pdf = {FUSION20_Basarur.pdf},
 title = {Position and Speed Estimation for BLDC Motors Using Fourier-Series Regression},
 year = {2020}
}

@inproceedings{Fusion20_Radtke,
 address = {Virtual},
 annote = {Winner Best Paper Award Certificate (PDF)},
 author = {Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 23rd International Conference on Information Fusion (Fusion 2020)},
 month = {July},
 pdf = {FUSION20_Radtke.pdf},
 title = {Fully Decentralized Estimation Using Square-Root Decompositions},
 year = {2020}
}

@inproceedings{Fusion20_Frisch,
 address = {Virtual},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 23rd International Conference on Information Fusion (Fusion 2020)},
 month = {July},
 pdf = {FUSION20_Frisch.pdf},
 title = {Progressive Bayesian Filtering with Coupled Gaussian and Dirac Mixtures},
 year = {2020}
}

@inproceedings{Fusion20_Li,
 address = {Virtual},
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 23rd International Conference on Information Fusion (Fusion 2020)},
 month = {July},
 pdf = {FUSION20_Li.pdf},
 title = {Dual Quaternion Sample Reduction for SE(2) Estimation},
 year = {2020}
}

@inproceedings{Fusion20_Pfaff,
 address = {Virtual},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 23rd International Conference on Information Fusion (Fusion 2020)},
 month = {July},
 pdf = {FUSION20_Pfaff.pdf},
 title = {A Hyperhemispherical Grid Filter for Orientation Estimation},
 year = {2020}
}

@inproceedings{ICRA20_Frisch,
 address = {Virtual},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Robotics and Automation (ICRA 2020)},
 month = {May},
 pdf = {ICRA20_Frisch.pdf},
 title = {Association-Free Multilateration Based on Times of Arrival},
 year = {2020}
}

@inproceedings{ICRA20_Li,
 address = {Virtual},
 author = {Hannes Möls and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 IEEE International Conference on Robotics and Automation (ICRA 2020)},
 month = {May},
 pdf = {ICRA20_Li.pdf},
 title = {Highly Parallelizable Plane Extraction for Organized Point Clouds Using Spherical Convex Hulls},
 year = {2020}
}

@inproceedings{ECC20_Li,
 address = {Virtual},
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2020 European Control Conference (ECC 2020)},
 month = {May},
 pdf = {ECC20_Li.pdf},
 title = {Grid-Based Quaternion Filter for SO(3) Estimation},
 year = {2020}
}

@inproceedings{PLANS20_Siebler,
 abstract = {In this paper a train localization method is proposed that uses local variations of the earth magnetic field to determine the topological position of a train in a track network. The approach requires a magnetometer triad, an accelerometer, and a map of the magnetic field along the railway tracks. The estimated topological position comprises the along-track position that defines the position of the train within a certain track and the track ID that specifies the track the train is driving on. The along-track position is estimated by a a recursive Bayesian filter and the track ID is found from a hypothesis test. In particular the use of multiple particle filter, each estimating the position on different track hypothesis, is proposed. Whenever the estimated train position crosses a switch, a particle filter for each possible track is created. With the position estimates of the different filters, the likelihood for each track hypothesis is calculated from the measured magnetic field and the expected magnetic field in the map. A comparison of the likelihoods is subsequently used to decide which track is the most likely. After a decision for a track is made, the unnecessary filters are deleted. The feasibility of the proposed localization method is evaluated with measurement data recorded on a regional train. In the evaluation, the localization method was running in real time and overall an RMSE below five meter could be achieved and all tracks were correctly identified.},
 author = {Benjamin Siebler and Oliver Heirich and Stephan Sand and Uwe D. Hanebeck},
 booktitle = {2020 IEEE/ION Position, Location and Navigation Symposium (PLANS)},
 doi = {10.1109/PLANS46316.2020.9110149},
 issn = {2153-3598},
 keywords = {accelerometers;Bayes methods;distortion;magnetometers;particle filtering (numerical methods);position measurement;rail traffic control;railways;recursive filters;target tracking;recursive Bayesian filter;track ID;multiple particle filter;track hypothesis;position estimates;measured magnetic field;expected magnetic field;regional train;joint train localization;earth magnetic field distortions;train localization method;track network;railway tracks;estimated topological position;along-track position;track identification;magnetometer triad;accelerometer;hypothesis test;particle filter;measurement data},
 month = {April},
 number = {},
 pages = {941-948},
 pdf = {PLANS20_Siebler.pdf},
 title = {Joint Train Localization and Track Identification based on Earth Magnetic Field Distortions},
 volume = {},
 year = {2020}
}

@proceedings{AT20_Pfaff-Editorial,
 doi = {10.1515/auto-2020-0023},
 editor = {Florian Pfaff and Uwe D. Hanebeck},
 journal = {at -- Automatisierungstechnik},
 month = {April},
 title = {Guest Editorial of the Special Issue on Sensor-Based Sorting},
 url = {https://www.degruyter.com/view/journals/auto/68/4/article-p229.xml},
 year = {2020}
}

@article{AT20_Pfaff,
 author = {Florian Pfaff and Christoph Pieper and Georg Maier and Benjamin Noack and Robin Gruna and Harald Kruggel-Emden and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Thomas Längle and Jürgen Beyerer},
 doi = {10.1515/auto-2019-0134},
 journal = {at -- Automatisierungstechnik},
 month = {April},
 title = {Predictive Tracking with Improved Motion Models for Optical Belt Sorting},
 url = {https://www.degruyter.com/view/journals/auto/68/4/article-p239.xml},
 year = {2020}
}

@article{AT20_Maier,
 author = {Georg Maier and Florian Pfaff and Andrea Bittner and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Uwe D. Hanebeck and Thomas Längle and Jürgen Beyerer},
 doi = {10.1515/auto-2019-0128},
 journal = {at -- Automatisierungstechnik},
 month = {April},
 title = {Characterizing Material Flow in Sensor-Based Sorting Systems Using an Instrumented Particle},
 url = {https://www.degruyter.com/view/journals/auto/68/4/article-p256.xml},
 year = {2020}
}

@inproceedings{CISS20_Hanebeck,
 address = {Princeton, New Jersey, USA},
 author = {Uwe D. Hanebeck},
 booktitle = {Proceedings of the 54th Annual Conference on Information Sciences and Systems (CISS 2020)},
 month = {March},
 pdf = {CISS20_Hanebeck.pdf},
 title = {Deterministic Sampling of Multivariate Densities based on Projected Cumulative Distributions},
 year = {2020}
}

@article{EURASIP20_Frisch,
 author = {Daniel Frisch and Uwe D. Hanebeck},
 issn = {1687-1499},
 journal = {EURASIP Journal on Wireless Communications and Networking},
 month = {February},
 number = {1},
 pages = {43},
 title = {Commentary to: TDOA versus ATDOA for Wide Area Multilateration System},
 url = {https://doi.org/10.1186/s13638-020-1656-1},
 volume = {2020},
 year = {2020}
}

@article{TIE20_Maier,
 author = {Georg Maier and Florian Pfaff and Christoph Pieper and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Jürgen Beyerer},
 doi = {10.1109/TIE.2020.2970643},
 journal = {Transactions on Industrial Electronics},
 month = {February},
 pdf = {TIE20_Maier.pdf},
 title = {Experimental Evaluation of a Novel Sensor-Based Sorting Approach Featuring Predictive Real-Time Multiobject Tracking},
 url = {https://ieeexplore.ieee.org/document/8984697},
 year = {2020}
}

@inproceedings{CDC19_Rosenthal,
 address = {Nice, France},
 author = {Florian Rosenthal and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 58th IEEE Conference on Decision and Control (CDC 2019)},
 month = {December},
 pdf = {CDC19_Rosenthal.pdf},
 title = {Sequence-Based Stochastic Receding Horizon Control Using IMM Filtering and Value Function Approximation},
 year = {2019}
}

@article{arXiv19_Hanebeck,
 author = {Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:1912.12875},
 month = {December},
 title = {Deterministic Sampling of Multivariate Densities Based on Projected Cumulative Distributions},
 url = {https://arxiv.org/abs/1912.12875},
 year = {2019}
}

@inproceedings{ICCAIS19_Zhang,
 address = {Chengdu, China},
 author = {Qingpeng Zhang and Zhansheng Duan and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 International Conference on Control, Automation and Information Sciences (ICCAIS 2019)},
 month = {October},
 pdf = {ICCAIS2019_Zhang.pdf},
 title = {Recursive LMMSE Centralized Fusion with Compressed Multi-Radar Measurements},
 year = {2019}
}

@inproceedings{ACC19_Rosenthal,
 address = {Philadelphia, PA, USA},
 author = {Florian Rosenthal and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 American Control Conference (ACC 2019)},
 month = {July},
 pdf = {ACC19_Rosenthal.pdf},
 title = {Sequence-Based Receding Horizon Control Over Networks with Delays and Data Losses},
 year = {2019}
}

@inproceedings{ACC19_Mayer,
 address = {Philadelphia, PA, USA},
 author = {Jana Mayer and Maxim Dolgov and Tobias Stickling and Selim Özgen and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 American Control Conference (ACC 2019)},
 month = {July},
 pdf = {ACC19_Mayer.pdf},
 title = {Stochastic Optimal Control Using Gaussian Process Regression Over Probability Distributions},
 year = {2019}
}

@article{LCSS19_Schmitt,
 abstract = {With modern communication technology, sensors, estimators, and controllers can be pushed apart to distribute 
intelligence over wide distances. Instead of congesting channels by periodic data transmissions, smart sensors can decide 
on their own whether data are worth transmitting. This paper studies event-based transmissions from sensor to estimator. 
The sensor-side event trigger conveys usable information even if no transmission is triggered. In the absence of data, 
such implicit information can still be exploited by the remote Kalman filter. For this purpose, an easy-to-implement 
triggering mechanism is proposed based on a Finite Impulse Response prediction that is compared against a stochastic 
decision variable. By the aid of the stochastic event trigger, the implicit information retains a Gaussian representation 
and can easily be processed by the Kalman filter. The parameters for the stochastic trigger are retrieved from the 
Finite Impulse Response filter, which contributes to reducing the communication rate significantly, as shown in simulations.},
 author = {Eva Julia Schmitt and Benjamin Noack and Wolfgang Krippner and Uwe D. Hanebeck},
 doi = {10.1109/LCSYS.2019.2918024},
 issn = {2475-1456},
 journal = {IEEE Control Systems Letters},
 month = {July},
 number = {3},
 pages = {769--774},
 pdf = {LCSS19_Schmitt.pdf},
 title = {Gaussianity-Preserving Event-Based State Estimation with an FIR-Based Stochastic Trigger},
 url = {https://doi.org/10.1109/LCSYS.2019.2918024},
 volume = {3},
 year = {2019}
}

@inproceedings{Fusion19_Zea,
 address = {Ottawa, Canada},
 author = {Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Zea.pdf},
 title = {Refining Pose Estimation for Square Markers Using
Shape Fitting},
 year = {2019}
}

@inproceedings{Fusion19_Noack,
 abstract = {Decentralized data fusion is a challenging task even for 
linear estimation problems. Nonlinear estimation renders data fusion 
even more difficult as dependencies among the nonlinear estimates 
require complicated parameterizations. It is nearly impossible to 
reconstruct or keep track of dependencies. Therefore, conservative 
approaches have become a popular solution to nonlinear data fusion. As a 
generalization of Covariance Intersection, exponential mixture densities 
have been widely applied for nonlinear fusion. However, this approach 
inherits the conservativeness of Covariance Intersection. For this 
reason, the less conservative fusion rule Inverse Covariance 
Intersection is studied in this paper and also generalized to nonlinear 
data fusion. This generalization employs a conservative approximation of 
the common information shared by the estimates to be fused. This bound 
of the common information is subtracted from the fusion result. In doing 
so, less conservative fusion results can be attained as an empirical 
analysis demonstrates.},
 address = {Ottawa, Canada},
 author = {Benjamin Noack and Umut Orguner and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on 
Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Noack.pdf},
 title = {Nonlinear Decentralized Data Fusion with Generalized 
Inverse Covariance Intersection},
 year = {2019}
}

@inproceedings{Fusion19_Frisch,
 address = {Ottawa, Canada},
 author = {Daniel Frisch and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Frisch.pdf},
 title = {ROTA: Round Trip Times of Arrival for Localization with Unsynchronized Receivers},
 year = {2019}
}

@inproceedings{Fusion19_Shao,
 address = {Ottawa, Canada},
 author = {Teng Shao and Zhangsheng Duan and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Shao.pdf},
 title = {Multi-Rate Asynchronous Distributed Filtering Under Randomized Gossip Strategy},
 year = {2019}
}

@inproceedings{Fusion19_Bultmann,
 address = {Ottawa, Canada},
 author = {Simon Bultmann and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Bultmann.pdf},
 title = {Stereo Visual SLAM Based on Unscented Dual Quaternion Filtering},
 year = {2019}
}

@inproceedings{Fusion19_Li,
 address = {Ottawa, Canada},
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Li.pdf},
 title = {Hyperspherical Deterministic Sampling Based on Riemannian Geometry for Improved Nonlinear Bingham Filtering},
 year = {2019}
}

@inproceedings{Fusion19_Pfaff,
 address = {Ottawa, Canada},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Pfaff.pdf},
 title = {Fourier Filters, Grid Filters, and the Fourier-Interpreted Grid Filter},
 year = {2019}
}

@inproceedings{Fusion19_Kronauer,
 address = {Ottawa, Canada},
 author = {Tobias Kronauer and Florian Pfaff and Benjamin  Noack and Wei Tian and Georg Maier and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Kronauer.pdf},
 title = {Feature-Aided Multitarget Tracking for Optical Belt Sorters},
 year = {2019}
}

@inproceedings{Fusion19_Radtke,
 address = {Ottawa, Canada},
 author = {Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 22nd International Conference on Information Fusion (Fusion 2019)},
 month = {July},
 pdf = {Fusion19_Radtke.pdf},
 title = {Distributed Estimation using Square Root Decompositions of Dependent Information},
 year = {2019}
}

@article{arXiv19_Li,
 author = {Kailai Li and Johannes Cox and Benjamin Noack and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:1907.13566},
 month = {July},
 title = {Improved Pose Graph Optimization for Planar Motions Using Riemannian Geometry on the Manifold of Dual Quaternions},
 url = {https://arxiv.org/abs/1907.13566},
 year = {2019}
}

@inproceedings{ECC19_Radtke,
 address = {Naples, Italy},
 author = {Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 European Control Conference (ECC 2019)},
 month = {June},
 pdf = {ECC19_Radtke.pdf},
 title = {Distributed Estimation with Partially Overlapping States based on Deterministic Sample-based Fusion},
 year = {2019}
}

@inproceedings{ECC19_Li,
 address = {Naples, Italy},
 author = {Kailai Li and Daniel Frisch and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 European Control Conference (ECC 2019)},
 month = {June},
 pdf = {ECC19_Li.pdf},
 title = {Geometry-Driven Deterministic Sampling for Nonlinear Bingham Filtering},
 year = {2019}
}

@article{JSS19_Kurz,
 author = {Gerhard Kurz and Igor Gilitschenski and Florian Pfaff and Lukas Drude and Uwe D. Hanebeck and Reinhold Haeb-Umbach and Roland Y. Siegwart},
 doi = {10.18637/jss.v089.i04},
 journal = {Journal of Statistical Software},
 month = {May},
 title = {Directional Statistics and Filtering Using libDirectional},
 url = {https://dx.doi.org/10.18637/jss.v089.i04},
 year = {2019}
}

@inproceedings{MFI19_Oezgen,
 address = {Taipei, Republic of China},
 author = {Selim Özgen and Saskia Kohn and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2019)},
 month = {May},
 pdf = {MFI19_Oezgen.pdf},
 title = {State Estimation with Model-Mismatch-Based Secrecy against Eavesdroppers},
 year = {2019}
}

@inproceedings{ICPS19_Rosenthal,
 address = {Taipei, Republic of China},
 author = {Florian Rosenthal and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 IEEE International Conference on Industrial Cyber-Physical Systems (ICPS 2019)},
 month = {May},
 pdf = {ICPS19_Rosenthal.pdf},
 title = {A Control Approach for Cooperative Sharing of Network Resources in Cyber-Physical Systems},
 year = {2019}
}

@inproceedings{MFI19_Radtke,
 address = {Taipei, Republic of China},
 author = {Susanne Radtke and Kailai Li and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2019)},
 month = {May},
 pdf = {MFI19_Radtke.pdf},
 title = {Comparative Study of Track-to-Track Fusion Methods for Cooperative Tracking with Bearings-only Measurements},
 year = {2019}
}

@inproceedings{MFI19_Li,
 address = {Taipei, Republic of China},
 author = {Kailai Li and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2019)},
 month = {May},
 pdf = {MFI19_Li.pdf},
 title = {Geometry-Driven Stochastic Modeling of SE(3) States Based on Dual Quaternion Representation},
 year = {2019}
}

@inproceedings{MFI19_Pfaff,
 address = {Taipei, Republic of China},
 author = {Florian Pfaff and Kailai Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2019)},
 month = {May},
 pdf = {MFI19_Pfaff.pdf},
 title = {Association Likelihoods for Directional Estimation},
 year = {2019}
}

@inproceedings{CCNC19_Rosenthal,
 abstract = {This work is concerned with our ongoing research
project CoCPN: Cooperative Cyber Physical Networking that
has the goal to allow cooperation between control applications
and the communication system that constitute cyber-physical
systems. We describe the envisioned architecture of CoCPN and
outline how it improves the flexibility of cyber-physical systems
by cooperatively sharing a common network infrastructure. We
also present our simulation tool CoCPN-Sim that we developed
as a method to thoroughly investigate the interaction between
control applications and the communication system. By providing
the results of selected simulations using the well-known inverted
pendulum, we identify potential aspects that can be exploited
for cooperation and hence serve as starting points towards more
flexible and adaptive cyber-physical systems.},
 address = {Las Vegas, USA},
 author = {Florian Rosenthal and Markus Jung and Martina Zitterbart and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2019 16th IEEE Annual Consumer Communications & Networking Conference (CCNC 2019)},
 days = {11},
 keywords = {Cyber-Physical Systems; Cooperation; Simulation; Event-Based Control},
 month = {January},
 pdf = {CCNC19_Rosenthal.pdf},
 title = {CoCPN -- Towards Flexible and Adaptive Cyber-Physical Systems Through Cooperation},
 year = {2019}
}

@incollection{LNEE18_Rosenthal,
 abstract = {In this article, we are concerned with state estimation in Networked Control Systems where both control inputs and measurements are transmitted over networks which are lossy and introduce random transmission delays. We focus on the case where acknowledgment packets transmitted by the actuator upon reception of applicable control inputs are also subject to delays and losses, as opposed to the common notion of TCP-like communication where successful transmissions are acknowledged instantaneously and without losses. As a consequence, the state estimator in the considered setup has only partial and belated knowledge concerning the actually applied control inputs which results in additional uncertainty. We derive an estimator by extending an existing approach for the special case of UDP-like communication which maintains estimates of the applied control inputs that are incorporated into the estimation of the plant state. The presented estimator is compared to the original approach in terms of Monte Carlo simulations where its increased robustness towards imperfect knowledge of the underlying networks is indicated.},
 address = {Cham},
 author = {Florian Rosenthal and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System},
 doi = {https://doi.org/10.1007/978-3-319-90509-9_2},
 editor = {Sukhan Lee and Hanseok Ko and Songhwai Oh},
 isbn = {978-3-319-90509-9},
 pages = {22--38},
 publisher = {Springer International Publishing},
 title = {State Estimation in Networked Control Systems with Delayed and Lossy Acknowledgments},
 url = {https://link.springer.com/chapter/10.1007/978-3-319-90509-9_2},
 year = {2018}
}

@article{PowTec18_Pieper,
 abstract = {State-of-the-art optical sorting systems suffer from delays between the particle detection and separation stage, during which the material movement is not accounted for. Commonly line scan cameras, using simple assumptions to predict the future particle movement, are employed. In this study, a novel prediction approach is presented, where an area scan camera records the particle movement over multiple time steps and a tracking algorithm is used to reconstruct the corresponding paths to determine the time and position at which the material reaches the separation stage. In order to assess the benefit of such a model at different operating parameters, an automated optical belt sorter is numerically modelled and coupled with the tracking procedure. The Discrete Element Method (DEM) is used to describe the particle–particle as well as particle–wall interactions, while the air nozzles required for deflecting undesired material fractions are described with Computational Fluid Dynamics (CFD). The accuracy of the employed numerical approach is ensured by comparing the separation results of a predefined sorting task with experimental investigations. The quality of the aforementioned prediction models is compared when utilizing different belt lengths, nozzle activation durations, particle types, sampling frequencies and detection windows. Results show that the numerical model of the optical belt sorter is able to accurately describe the sorting system and is suitable for detailed investigation of various operational parameters. The proposed tracking prediction model was found to be superior to the common line scan camera method in all investigated scenarios. Its advantage is especially profound when difficult sorting conditions, e.g. short conveyor belt lengths or uncooperative moving bulk solids, apply.},
 author = {Christoph Pieper and Florian Pfaff and Georg Maier and Harald Kruggel-Emden and Siegmar Wirtz and Benjamin Noack and Robin Gruna and Viktor Scherer and Uwe D. Hanebeck and Thomas Längle and Jürgen Beyerer},
 doi = {10.1016/j.powtec.2018.09.003},
 journal = {Powder Technology},
 month = {December},
 title = {Numerical Modelling of an Optical Belt Sorter Using a DEM--CFD Approach Coupled with Particle Tracking Algorithm and Comparison with Experiments},
 url = {https://doi.org/10.1016/j.powtec.2018.09.003},
 year = {2018}
}

@inproceedings{SDF18_Li,
 abstract = {In this paper, we propose the Progressive Bingham
Filter (PBF), a novel stochastic filtering algorithm for nonlinear
spatial orientation estimation. As an extension of the orientation
filter previously proposed only for the identity measurement
model based on the Bingham distribution, our method is able to
handle arbitrary measurement models. Instead of the sampling-
approximation scheme used in the prediction step, a closed-
form solution is possible when the system equation is based on
the Hamilton product. Besides stochastic approaches, we also
introduce the Spherical Averaging Method (SAM), which is an
application of the Riemannian averaging technique. The two
approaches are then applied to a specific problem where the
wavefront orientation is estimated based on Time Differences of
Arrival (TDOA) and evaluated in simulations. The results show
theoretical competitiveness of the PBF.},
 author = {Kailai Li and Daniel Frisch and Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2018)},
 month = {October},
 pdf = {SDF18_Li.pdf},
 title = {Wavefront Orientation Estimation Based on Progressive Bingham Filtering},
 year = {2018}
}

@article{arXiv18_Dolgov,
 author = {Maxim Dolgov and Uwe D. Hanebeck},
 journal = {arXiv preprint arXiv:1809.09193},
 month = {September},
 title = {A Distance-based Framework for Gaussian Processes over Probability Distributions},
 url = {https://arxiv.org/abs/1809.09193},
 year = {2018}
}

@article{arXiv18_Hanebeck,
 abstract = {We propose a homotopy continuation method called FLUX for approximating
complicated probability density functions. It is based on progressive
processing for smoothly morphing a given density into the desired one.
Distributed ordinary differential equations (DODEs) with an artificial time
$γ ın [0,1]$ are derived for describing the evolution from the initial
density to the desired final density. For a finite-dimensional
parametrization, the DODEs are converted to a system of ordinary
differential equations (SODEs), which are solved for $γ ın [0,1]$ and
return the desired result for $γ=1$. This includes parametric
representations such as Gaussians or Gaussian mixtures and nonparametric
setups such as sample sets. In the latter case, we obtain a particle flow
between the two densities along the artificial time.
FLUX is applied to state estimation in stochastic nonlinear dynamic
systems by gradual inclusion of measurement information. The proposed
approximation method (1) is fast, (2) can be applied to arbitrary nonlinear
systems and is not limited to additive noise, (3) allows for target
densities that are only known at certain points, (4) does not require
optimization, (5) does not require the solution of partial differential
equations, and (6) works with standard procedures for solving SODEs. This
manuscript is limited to the one-dimensional case and a fixed number of
parameters during the progression.
Future extensions will include consideration of higher dimensions and on the
fly adaption of the number of parameters.},
 author = {Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {August},
 title = {FLUX: Progressive State Estimation Based on Zakai-type Distributed Ordinary Differential Equations},
 url = {https://arxiv.org/abs/1808.02825},
 year = {2018}
}

@inproceedings{Fusion18_Oezgen,
 address = {Cambridge, United Kingdom},
 author = {Selim Özgen and Marco F. Huber and Florian Rosenthal and Jana Mayer and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Oezgen.pdf},
 title = {Retrodiction of Data Association Probabilities via Convex Optimization},
 year = {2018}
}

@inproceedings{Fusion18_Radtke,
 abstract = {Optimal fusion of estimates that are computed in a distributed fashion is a challenging task. In general, the sensor nodes cannot keep track of the cross-correlations required to fuse estimates optimally. In this paper, a novel technique is presented that provides the means to reconstruct the required correlation structure. For this purpose, each node computes a set of deterministic samples that provides all the information required to reassemble the cross-covariance matrix for each pair of estimates. As the number of samples is increasing over time, a method to reduce the size of the sample set is presented and studied. In doing so, communication expenses can be reduced significantly, but approximation errors are possibly introduced by neglecting past correlation terms. In order to keep approximation errors at a minimum, an appropriate set size can be determined and a trade-off between communication expenses and estimation quality can be found.},
 address = {Cambridge, United Kingdom},
 author = {Susanne Radtke and Benjamin Noack and Uwe D. Hanebeck and Ondřej Straka},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Radtke.pdf},
 title = {Reconstruction of Cross-Correlations with Constant Number of Deterministic Samples},
 year = {2018}
}

@inproceedings{Fusion18_Li-SE2,
 address = {Cambridge, United Kingdom},
 author = {Kailai Li and Gerhard Kurz and Lukas Bernreiter and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Li-SE2.pdf},
 title = {Nonlinear Progressive Filtering for SE(2) Estimation},
 year = {2018}
}

@inproceedings{Fusion18_Li-SLAM,
 address = {Cambridge, United Kingdom},
 author = {Kailai Li and Gerhard Kurz and Lukas Bernreiter and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Li-SLAM.pdf},
 title = {Simultaneous Localization and Mapping Using a Novel Dual Quaternion Particle Filter},
 year = {2018}
}

@inproceedings{Fusion18_Aristov,
 abstract = {With the advent of cheap sensor technology, multisensor data fusion algorithms have been becoming a key enabler for efficient in-network processing of sensor data. The information filter, in particular, has proven useful due to its simple additive structure of the measurement update equations.
In order to exploit this structure for an efficient in-network processing, each node in the network is supposed to locally process and combine data from its neighboring nodes. The aspired in-network processing, at first glance, prohibits efficient privacy-preserving communication protocols, and encryption schemes that allow for algebraic manipulations are often computationally too expensive. Partially homomorphic encryption schemes constitute far more practical solutions but are restricted to a single algebraic operation on the corresponding ciphertexts. In this paper, an additive-homomorphic encryption scheme is used to derive a privacy-preserving implementation of the information filter where additive operations are sufficient to distribute the workload among the sensor nodes. However, the encryption scheme requires the floating-point data to be quantized, which impairs the estimation quality. The proposed filter and the implications of the necessary quantization are analyzed in a simulated multisensor tracking scenario.},
 address = {Cambridge, United Kingdom},
 author = {Mikhail Aristov and Benjamin Noack and Uwe D. Hanebeck and Jörn Müller-Quade},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Aristov.pdf},
 title = {Encrypted Multisensor Information Filtering},
 year = {2018}
}

@inproceedings{Fusion18_Sigges,
 address = {Cambridge, United Kingdom},
 author = {Fabian Sigges and Christoph Rauterberg and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Sigges.pdf},
 title = {An Ensemble Kalman Filter for Feature-Based SLAM with Unknown Associations},
 year = {2018}
}

@inproceedings{Fusion18_Havlik,
 address = {Cambridge, United Kingdom},
 author = {Jindřich Havlík and Ondřej Straka and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Havlik.pdf},
 title = {Stochastic Integration Filter: Theoretical and Implementation Aspects},
 year = {2018}
}

@inproceedings{Fusion18_Wang,
 address = {Cambridge, United Kingdom},
 author = {Qi Wang and Zhansheng Duan and Xiao-Rong Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (Fusion 2018)},
 month = {July},
 pdf = {Fusion18_Wang.pdf},
 title = {Convex Combination for Source Localization Using Received Signal Strength Measurements},
 year = {2018}
}

@article{IETSP2018_Dunik,
 abstract = {Transformation of a random variable is a common need in a design of many algorithms in signal processing, automatic
control, and fault detection. Typically, the design is tied to an assumption on a probability density function of the random variable,
often in the form of the Gaussian distribution. The assumption may be, however, difficult to be met in algorithms involving nonlinear
transformation of the random variable. This paper focuses on techniques capable to ensure validity of the Gaussian assumption of
the nonlinearly transformed Gaussian variable by approximating the to-be-transformed random variable distribution by a Gaussian
mixture distribution. The stress is laid on an analysis and selection of design parameters of the approximate Gaussian mixture
distribution to minimise the error imposed by the nonlinear transformation such as the location and number of the Gaussian mixture
terms. A special attention is devoted to the definition of the novel Gaussian mixture splitting directions based on the measures of
non-Gaussianity. The proposed splitting directions are analysed and illustrated in numerical simulations.},
 author = {Jindřich Duník and Ondřej Straka and Benjamin Noack and Jannik Steinbring and Uwe D. Hanebeck},
 doi = {10.1049/iet-spr.2017.0286},
 issn = {1751-9683},
 journal = {IET Signal Processing},
 month = {July},
 number = {9},
 pages = {1073-1081},
 title = {On Directional Splitting of Gaussian Density in Nonlinear Random Variable Transformation},
 url = {https://digital-library.theiet.org/content/journals/10.1049/iet-spr.2017.0286},
 volume = {12},
 year = {2018}
}

@inproceedings{ACC18_Dolgov,
 address = {Milwaukee, Wisconsin, USA},
 author = {Maxim Dolgov and Gerhard Kurz and Daniela Grimm and Florian Rosenthal and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2018 American Control Conference (ACC 2018)},
 month = {June},
 pdf = {ACC18_Dolgov.pdf},
 title = {Stochastic Optimal Control Using Local Sample-Based Value Function Approximation},
 year = {2018}
}

@inproceedings{ACC18_Kurz,
 address = {Milwaukee, Wisconsin, USA},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2018 American Control Conference (ACC 2018)},
 month = {June},
 pdf = {ACC18_Kurz-PGF.pdf},
 title = {Improved Progressive Gaussian Filtering Using LRKF Priors},
 year = {2018}
}

@inproceedings{ACC18_Rosenthal,
 address = {Milwaukee, Wisconsin, USA},
 author = {Florian Rosenthal and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2018 American Control Conference (ACC 2018)},
 month = {June},
 pdf = {ACC18_Rosenthal.pdf},
 title = {Scheduling of Measurement Transmission in Networked Control Systems Subject to Communication Constraints},
 year = {2018}
}

@article{Sensors2018_Dormann,
 abstract = {For multisensor data fusion, distributed state estimation techniques that enable a local processing of sensor data are the means of choice in order to minimize storage and communication costs. In particular, a distributed implementation of the optimal Kalman filter has recently been developed. A significant disadvantage of this algorithm is that the fusion center needs access to each node so as to compute a consistent state estimate, which requires full communication each time an estimate is requested. In this article, different extensions of the optimally distributed Kalman filter are proposed that employ data-driven transmission schemes in order to reduce communication expenses. As a first relaxation of the full-rate communication scheme, it can be shown that each node only has to transmit every second time step without endangering consistency of the fusion result. Also, two data-driven algorithms are introduced that even allow for lower transmission rates, and bounds are derived to guarantee consistent fusion results. Simulations demonstrate that the data-driven distributed filtering schemes can outperform a centralized Kalman filter that requires each measurement to be sent to the center node.},
 author = {Katharina Dormann and Benjamin Noack and Uwe D. Hanebeck},
 doi = {10.3390/s18041034},
 issn = {1424-8220},
 journal = {Sensors},
 month = {April},
 number = {4},
 pdf = {Sensors18_Dormann.pdf},
 title = {Optimally Distributed Kalman Filtering with Data-Driven Communication},
 url = {https://www.mdpi.com/1424-8220/18/4/1034},
 volume = {18},
 year = {2018}
}

@inproceedings{IoTDI18_Jung,
 address = {Orlando, FL, USA},
 annote = {Best Poster Award Runner-Up},
 author = {Markus Jung and Florian Rosenthal and Martina Zitterbart},
 booktitle = {Proceedings of the 2018 IEEE/ACM Third International Conference on Internet-of-Things Design and Implementation (IoTDI)},
 month = {April},
 title = {Poster Abstract: CoCPN-Sim: An Integrated Simulation Environment for Cyber-Physical Systems},
 year = {2018}
}

@proceedings{TII18_Hanebeck,
 editor = {Uwe D. Hanebeck and Marcus Baum and Marco F. Huber},
 journal = {IEEE Transactions on Industrial Informatics},
 month = {March},
 title = {Guest Editorial Special Section on Multisensor Fusion and Integration for Intelligent Systems},
 volume = {14},
 year = {2018}
}

@inproceedings{SBSC18_Maier,
 abstract = {In the field of machine vision, sensor-based sorting is an important real-time application that enables the separation of a material feed into different classes. While state-of-the-art systems utilize scanning sensors such as line-scan cameras, advances in sensor technology have made application of area scanning sensors feasible. Provided a sufficiently high frame rate, objects can be observed at multiple points in time. By applying multiobject tracking, information about the objects contained in the material stream can be fused over time. Based on this information, our approach further allows predicting the position of each object for future points in time. While conventional systems typically apply a global, rather simple motion model, our approach includes an individual motion model for each object, which in turn allows estimating the point in time as well as the position when reaching the separation stage. In this contribution, we present results from our collaborative research project and summarize the present advances by discussing the potential of the application of area-scan sensors for sensor-based sorting. Among others, we introduce our simulation-driven approach and present results for physical separation efficiency for simulation-generated data, demonstrate the potential of using motion-based features for material classification and discuss real-time related challenges.},
 address = {Aachen, Germany},
 author = {Georg Maier and Florian Pfaff and Christoph Pieper and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Jürgen Beyerer},
 booktitle = {Proceedings of the Eighth Conference on Sensor-Based Sorting & Control 2018 (SBSC 2018)},
 month = {March},
 pdf = {SBSC18_Maier.pdf},
 title = {Application of Area-Scan Sensors in Sensor-Based Sorting},
 year = {2018}
}

@article{TII18_Kurz,
 author = {Gerhard Kurz and Florian Pfaff and Uwe D. Hanebeck},
 doi = {10.1109/TII.2017.2757011},
 journal = {IEEE Transactions on Industrial Informatics},
 month = {March},
 number = {3},
 pages = {1197-1206},
 pdf = {TII18_Kurz-DiscreteFiltering.pdf},
 title = {Application of Discrete Recursive Bayesian Estimation on Intervals and the Unit Circle to Filtering on SE(2)},
 url = {https://doi.org/10.1109/TII.2017.2757011},
 volume = {14},
 year = {2018}
}

@article{TAES17_Zea,
 abstract = {This paper presents a novel approach to track a nonconvex shape approximation of an extended target based on noisy point measurements. For this purpose, a novel type of random hypersurface model (RHM) called Level-set RHM is introduced that models the interior of a shape with level-sets of an implicit function. Based on the Level-set RHM, a nonlinear measurement equation can be derived that allows to employ a standard Gaussian state estimator for tracking an extended object even in scenarios with moderate measurement noise. In this paper, shapes are described using polygons, and shape regularization is applied using ideas from active contour models.},
 author = {Antonio Zea and Florian Faion and Marcus Baum and Uwe D. Hanebeck},
 journal = {IEEE Transactions on Aerospace and Electronic Systems},
 title = {Level-Set Random Hypersurface Models for Tracking Nonconvex Extended Objects},
 url = {https://ieeexplore.ieee.org/document/7855600},
 year = {2017}
}

@inproceedings{CDC17_Kurz,
 address = {Melbourne, Australia},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 56th IEEE Conference on Decision and Control (CDC 2017)},
 month = {December},
 pdf = {CDC17_Kurz-HSKF.pdf},
 title = {Linear Regression Kalman Filtering Based on Hyperspherical Deterministic Sampling},
 year = {2017}
}

@article{arXiv17_Kurz,
 abstract = {In this paper, we present libDirectional, a MATLAB library for directional statistics and directional estimation. It supports a variety of commonly used distributions on the unit circle, such as the von Mises, wrapped normal, and wrapped Cauchy distributions. Furthermore, various distributions on higher-dimensional manifolds such as the unit hypersphere and the hypertorus are available. Based on these distributions, several recursive filtering algorithms in libDirectional allow estimation on these manifolds. The functionality is implemented in a clear, well-documented, and object-oriented structure that is both easy to use and easy to extend.},
 author = {Gerhard Kurz and Igor Gilitschenski and Florian Pfaff and Lukas Drude and Uwe D. Hanebeck and Reinhold Haeb-Umbach and Roland Y. Siegwart},
 journal = {arXiv preprint: Computation (stat.CO)},
 month = {December},
 title = {Directional Statistics and Filtering Using libDirectional},
 url = {https://arxiv.org/abs/1712.09718},
 year = {2017}
}

@article{RTIP17_Maier,
 abstract = {Utilizing parallel algorithms is an established way of increasing performance in systems that are bound to real-time restrictions. Sensor-based sorting is a machine vision application for which firm real-time requirements need to be respected in order to reliably remove potentially harmful entities from a material feed. Recently, employing a predictive tracking approach using multitarget tracking in order to decrease the error in the physical separation in optical sorting has been proposed. For implementations that use hard associations between measurements and tracks, a linear assignment problem has to be solved for each frame recorded by a camera. The auction algorithm can be utilized for this purpose, which also has the advantage of being well suited for parallel architectures. In this paper, an improved implementation of this algorithm for a graphics processing unit (GPU) is presented. The resulting algorithm is implemented in both an OpenCL and a CUDA based environment. By using an optimized data structure, the presented algorithm outperforms recently proposed implementations in terms of speed while retaining the quality of output of the algorithm. Furthermore, memory requirements are significantly decreased, which is important for embedded systems. Experimental results are provided for two different GPUs and six datasets. It is shown that the proposed approach is of particular interest for applications dealing with comparatively large problem sizes.},
 author = {Georg Maier and Florian Pfaff and Matthias Wagner and Christoph Pieper and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Jürgen Beyerer},
 doi = {10.1007/s11554-017-0735-y},
 journal = {Journal of Real-Time Image Processing},
 month = {November},
 title = {Real-Time Multitarget Tracking for Sensor-Based Sorting},
 url = {https://doi.org/10.1007/s11554-017-0735-y},
 year = {2017}
}

@inproceedings{MFI17_Dormann,
 abstract = {The centralized Kalman filter can be implemented
in such a way that the required calculations can be distributed
over multiple nodes in a network, each of which processes
only the locally acquired sensor data. The main downside of
this implementation is that it requires each distributed sensor
node to communicate with the fusion center in every time step
so as to compute the optimal state estimate. In this paper,
two distributed Kalman filtering algorithms are proposed to
overcome these limitations. The first algorithm merely requires
communication of each local sensor node with the fusion center
in every other time step. The second algorithm even allows
for a lower communicate rate. Both algorithms apply event-based 
communication to compute consistent estimates and to
reduce the estimation error for a fixed communication rate.
Simulations demonstrate that both algorithms perform better in
terms of the mean squared estimation error than the centralized
Kalman filter.},
 address = {Daegu, Republic of Korea},
 author = {Katharina Dormann and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2017)},
 month = {November},
 pdf = {MFI17_Dormann.pdf},
 title = {Distributed Kalman Filtering With Reduced Transmission Rate},
 year = {2017}
}

@inproceedings{MFI17_Pfaff,
 abstract = {Optical belt sorters can be used to sort a large
variety of bulk materials. By the use of sophisticated algo-
rithms, the performance of the complex machinery can be
further improved. Recently, we have proposed an extension
to industrial optical belt sorters that involves tracking the
individual particles on the belt using an area scan camera. If the
estimated behavior of the particles matches the true behavior,
the reliability of the separation process can be improved. The
approach relies on multitarget tracking using hard association
decisions between the tracks and the measurements. In this
paper, we propose to include the orientation in the assessment
of the compatibility of a track and a measurement. This allows
us to achieve more reliable associations, facilitating a higher
accuracy of the tracking results.},
 address = {Daegu, Republic of Korea},
 author = {Florian Pfaff and Gerhard Kurz and Christoph Pieper and Georg Maier and Benjamin Noack and Harald Kruggel-Emden and Robin Gruna and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Thomas Längle and Jürgen Beyerer},
 booktitle = {Proceedings of the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2017)},
 month = {November},
 pdf = {MFI17_Pfaff.pdf},
 title = {Improving Multitarget Tracking Using Orientation Estimates for Sorting Bulk Materials},
 year = {2017}
}

@inproceedings{MFI17_Pfaff-SphericalHarmonics,
 abstract = {For manifolds with topologies that strongly differ from the standard topology of R^n, using common filters created for linear domains can yield misleading results. While there is a lot of ongoing research on estimation on the unit circle, higher-dimensional problems particularly pose a challenge. One important generalization of the unit circle is the unit hypersphere. In this paper, we propose a recursive Bayesian estimator for the unit sphere S^2 based on spherical harmonics for arbitrary likelihood functions and rotationally symmetric system noises. In our evaluation, the proposed filter outperforms the particle filter in a target tracking scenario on the sphere.},
 address = {Daegu, Republic of Korea},
 author = {Florian Pfaff and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2017)},
 month = {November},
 pdf = {MFI17_Pfaff-SphericalHarmonics.pdf},
 title = {Filtering on the Unit Sphere Using Spherical Harmonics},
 year = {2017}
}

@inproceedings{MFI17_Kurz,
 abstract = {The group of rotations in three dimensions SO(3) plays a crucial role in applications ranging from robotics and aeronautics to computer graphics. Rotations have three degrees of freedom, but representing rotations is a nontrivial matter and different methods, such as Euler angles, quaternions, rotation matrices, and Rodrigues vectors are commonly used. Unfortunately, none of these representations allows easy discretization of orientations on evenly spaced grids. We present a novel discretization method that is based on a quaternion representation in conjunction with a recursive subdivision scheme of the four-dimensional hypercube, also known as the tesseract.},
 address = {Daegu, Republic of Korea},
 author = {Gerhard Kurz and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2017)},
 month = {November},
 pdf = {MFI17_Kurz-Tesseract.pdf},
 title = {Discretization of SO(3) Using Recursive Tesseract Subdivision},
 year = {2017}
}

@inproceedings{MFI17_Rosenthal,
 abstract = {In this paper, we consider state estimation in
Networked Control Systems where both control inputs and
measurements are transmitted via networks which are lossy
and introduce random transmission delays. In contrast to the
common notion of TCP-like communication, where successful
transmissions are acknowledged instantaneously and without
losses, we focus on the case where the acknowledgment packets
provided by the actuator upon reception of applicable control
inputs are also subject to delays and losses. Consequently,
the estimator has only partial and belated knowledge on the
actually applied control inputs, which results in additional
uncertainty. We derive an estimator for the considered setup by
generalizing an existing approach for UDP-like communication
which integrates estimates of the applied control inputs into the
overall state estimation. The presented estimator is assessed in
terms of Monte Carlo simulations.},
 address = {Daegu, Republic of Korea},
 author = {Florian Rosenthal and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2017)},
 month = {November},
 pdf = {MFI17_Rosenthal.pdf},
 title = {State Estimation in Networked Control Systems With Delayed And Lossy Acknowledgments},
 year = {2017}
}

@inproceedings{MFI17_Oezgen,
 abstract = {This study explores the non-parametric estimation of a shape boundary from noisy points in 2D when the sensor characteristics are known. As the underlying shape information is not known, the offered algorithm estimates points on the shape boundary by using the statistics of the subsets of point cloud data.

The novel approach proposed in this paper is able to find corner points in a local geometry by only using sample mean and covariance matrices of the subsets of the point cloud. While the proposed approach can be used for any class of boundary functions that demonstrates symmetry; for this paper, the analysis and experiments are performed on a connected line segment.},
 address = {Daegu, Republic of Korea},
 author = {Selim Özgen and Florian Faion and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2017)},
 month = {November},
 pdf = {MFI17_Oezgen.pdf},
 title = {A Non-Parametric Inference Technique for Shape Boundaries in Noisy Point Clouds},
 year = {2017}
}

@article{TAC17_Dolgov,
 abstract = {In this paper, we address infinite-horizon optimal control
of Markov Jump Linear Systems (MJLS) via static output feedback. Because
the jump parameter is assumed not to be observed, the optimal control
law is nonlinear and intractable. Therefore, we assume the regulator to
be linear. Under this assumption, we first present sufficient
feasibility conditions for static output-feedback stabilization of MJLS
with non-observed mode in the mean square sense in terms of linear
matrix inequalities (LMIs). However, these conditions depend on the
particular state-space representation, i.e., a coordinate transform can
make the LMIs feasible, while the original LMIs are infeasible. To avoid
the issues with the ambiguity of the state-space representation, we
therefore present an iterative algorithm for the computation of the
regulator gain. The algorithm is shown to converge if the MJLS is
stabilizable via mode-independent static output feedback. However,
convergence of the algorithm is not sufficient for stability of the
closed loop, which requires an additional stability check after the
regulator gains have been computed. A numerical example demonstrates the
application of the presented results.},
 author = {Maxim Dolgov and Uwe D. Hanebeck},
 doi = {10.1109/TAC.2017.2703924},
 issn = {0018-9286},
 journal = {IEEE Transactions on Automatic Control},
 month = {October},
 number = {10},
 pdf = {TAC17_Dolgov.pdf},
 title = {Static Output-Feedback Control of Markov Jump Linear Systems Without Mode Observation},
 url = {https://ieeexplore.ieee.org/document/7927447/},
 volume = {62},
 year = {2017}
}

@article{TM17_Maier,
 abstract = {Sensor-based sorting provides state-of-the-art solutions for sorting cohesive, granular materials. Typically, involved sensors, illumination, implementation of data analysis and other components are designed and chosen according to the sorting task at hand. A common property of conventional systems is the utilization of scanning sensors. However, the usage of area-scan cameras has recently been proposed. When observing objects at multiple time points, the corresponding paths can be reconstructed by using multiobject tracking. This in turn allows to accurately estimate the point in time and position at which any object will reach the separation stage of the optical sorter and hence contributes to decreasing the error in physical separation. In this paper, it is proposed to further exploit motion information for the purpose of material characterization. By deriving suitable features from the motion information, we show that high classification performance is obtained for an exemplary classification task. The approach therefore contributes towards decreasing the detection error of sorting systems.},
 author = {Georg Maier and Florian Pfaff and Florian Becker and Christoph Pieper and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Jürgen Beyerer},
 doi = {10.1515/teme-2017-0063},
 journal = {tm -- Technisches Messen, De Gruyter},
 month = {October},
 title = {Motion-Based Material Characterization in Sensor-Based Sorting},
 url = {https://doi.org/10.1515/teme-2017-0063},
 year = {2017}
}

@inproceedings{PARTICLES17_Pieper,
 abstract = {In the growing field of bulk solids handling, automated optical sorting systems are
of increasing importance. However, the initial sorter calibration is still very time consuming
and the precise optical sorting of many materials still remains challenging. In order to
investigate the impact of different operating parameters on the sorting quality, a numerical
model of an existing modular optical belt sorter is presented in this study. The sorter and particle
interaction is described with the Discrete Element Method (DEM) while the air nozzles required
for deflecting undesired material fractions are modelled with Computation Fluid Dynamics
(CFD). The correct representation of the resulting particle–fluid interaction is realized through
a one–way coupling of the DEM with CFD. Complex shaped particle clusters are employed to
model peppercorns also used in experimental investigations. To test the correct implementation
of the utilized models, the particle mass flow within the sorter is compared between experiment
and simulation. The particle separation results of the developed numerical model of the optical
sorting system are compared with matching experimental investigations. The findings show
that the numerical model is able to predict the sorting quality of the optical sorting system with
reasonable accuracy.},
 address = {Hannover, Germany},
 author = {Christoph Pieper and Georg Maier and Florian Pfaff and Harald Kruggel-Emden and Robin Gruna and Benjamin Noack and Siegmar Wirtz and Viktor Scherer and Thomas Längle and Uwe D. Hanebeck and Jürgen Beyerer},
 booktitle = {Proceedings of the 5th International Conference on Particle-Based Methods (PARTICLES 2017)},
 month = {September},
 pdf = {Particles17_Pieper.pdf},
 title = {Numerical Modelling of the Separation of Complex Shaped Particles in an Optical Belt Sorter Using a DEM--CFD Approach and Comparison with Experiments},
 year = {2017}
}

@inproceedings{Fusion17_Kurz,
 address = {Xi'an, China},
 author = {Gerhard Kurz and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 20th International Conference on Information Fusion (Fusion 2017)},
 month = {July},
 pdf = {Fusion17_Kurz-TorusFilter.pdf},
 title = {Nonlinear Toroidal Filtering Based on Bivariate Wrapped Normal Distributions},
 year = {2017}
}

@inproceedings{Fusion17_Li,
 address = {Xi'an, China},
 author = {Xianqing Li and Zhansheng Duan and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 20th International Conference on Information Fusion (Fusion 2017)},
 month = {July},
 title = {Performance Ranking of Multiple Nonlinear Filters Using Ranking Vector and Voting Fusion},
 year = {2017}
}

@inproceedings{Fusion17_Noack,
 abstract = {Decentralized data fusion is a challenging task.
Either it is too difficult to maintain and track the information
required to perform fusion optimally, or too much information
is discarded to obtain informative fusion results. A well-known
solution is Covariance Intersection, which may provide too
conservative fusion results. A less conservative alternative is
discussed in this paper, and generalizations are proposed in
order to apply it to a wide class of fusion problems. The Inverse
Covariance Intersection algorithm is about finding the maximum
possible common information shared by the estimates to be
fused. A bound on the possibly shared common information
is derived and removed from the fusion result in order to
guarantee consistency. It is shown that the conditions required
for consistency can be significantly relaxed, and also other causes
of correlations, such as common process noise, can be treated.},
 address = {Xi'an, China},
 author = {Benjamin Noack and Joris Sijs and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 20th International Conference on Information Fusion (Fusion 2017)},
 month = {July},
 pdf = {Fusion17_Noack.pdf},
 title = {Inverse Covariance Intersection: New Insights and Properties},
 year = {2017}
}

@inproceedings{Fusion17_Pfaff-Set,
 abstract = {For distributed estimation, algorithms have to be
specifically crafted to minimize communication between the
sensor nodes. As an adjusted version of the regular Kalman filter,
the distributed Kalman filter (DKF) allows for deriving optimal
results while not requiring regular communication. To achieve
this, the DKF requires that each node has full knowledge about
the system model and measurement models of all nodes. However,
the DKF is not sufficient if the characteristics of the errors in the
system and measurement models are not purely stochastic. In this
paper, we present a distributed version of a combined stochastic
and set-membership Kalman filter. The proposed filter optimizes
the approximations of the set-membership uncertainties and can
even yield better results than the regular centralized filter.},
 address = {Xi'an, China},
 author = {Florian Pfaff and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 20th International Conference on Information Fusion (Fusion 2017)},
 month = {July},
 pdf = {Fusion17_Pfaff-Set.pdf},
 title = {Optimal Distributed Combined Stochastic and Set-Membership State Estimation},
 year = {2017}
}

@inproceedings{Fusion17_Pfaff-IDKF,
 abstract = {With the ubiquity of information distributed in
networks, performing recursive Bayesian estimation using
distributed calculations is becoming more and more important.
There are a wide variety of algorithms catering to different
applications and requiring different degrees of knowledge about
the other nodes involved. One recently developed algorithm is
the distributed Kalman filter (DKF), which assumes that all
knowledge about the measurements, except the measurements
themselves, are known to all nodes. If this condition is met,
the DKF allows deriving the optimal estimate if all information
is combined in one node at an arbitrary time step. In this
paper, we present an information form of the distributed Kalman
filter (IDKF) that allows the use of explicit system inputs at
the individual nodes while still yielding the same results as a
centralized Kalman filter.},
 address = {Xi'an, China},
 author = {Florian Pfaff and Benjamin Noack and Uwe D. Hanebeck and Felix Govaers and Wolfgang Koch},
 booktitle = {Proceedings of the 20th International Conference on Information Fusion (Fusion 2017)},
 month = {July},
 pdf = {Fusion17_Pfaff-IDKF.pdf},
 title = {Information Form Distributed Kalman Filtering (IDKF) with Explicit Inputs},
 year = {2017}
}

@inproceedings{Fusion17_Sigges,
 abstract = {We present a particle filter for multi-object tracking
that is based on the ideas of the Approximate Bayesian
Computation (ABC) paradigm. The main idea is to avoid the explicit
computation of the likelihood function by means of simulation.
For this purpose, a large amount of particles in the state space is
simulated from the prior, transformed into measurement space,
and then compared to the real measurement by using an appropriate
distance function, i.e., the OSPA distance. By selecting the
closest simulated measurements and their corresponding particles
in state space, the posterior distribution is approximated. The
algorithm is evaluated in a multi-object scenario with and without
clutter and is compared to a global nearest neighbour Kalman
filter.},
 address = {Xi'an, China},
 author = {Fabian Sigges and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 20th International Conference on Information Fusion (Fusion 2017)},
 month = {July},
 pdf = {Fusion17_Sigges.pdf},
 title = {A Likelihood-Free Particle Filter for Multi-Object Tracking},
 year = {2017}
}

@inproceedings{IFAC17_Sijs,
 abstract = {Event-based sampling of sensor signals has become a mature alternative to time-periodic sampling completed with solutions for event-based estimation and control. Among those solutions there is a class of estimators exploiting the fact that an event was not triggered. Not receiving a new measurement is then interpreted as a sensor signal that has not violated the event criteria, which means that the signal is still within the triggering set defining the event. Such implied measurement information is exploited by the estimator, though it is only valid when no event occurred. The approach is thus sensitive to package loss and latency, as the estimator might be incorrect in assuming that no event took place. A solution to anticipate for package loss on the estimation error is studied in this article, and it is further turned into a first solution when the estimator is part of a feedback-control loop.},
 address = {Toulouse, France},
 author = {Joris Sijs and Benjamin Noack},
 booktitle = {Proceedings of the 20th IFAC World Congress (IFAC 2017)},
 month = {July},
 title = {Event-based State Estimation in a Feedback Loop with Imperfect Communication Links},
 year = {2017}
}

@article{AT17_Pfaff,
 abstract = {State-of-the-art optical belt sorters commonly employ line scan cameras and use simple assumptions to predict each particle's movement, which is required for the separation process. Previously, we have equipped an experimental optical belt sorter with an area scan camera and were able to show that tracking the particles of the bulk material results in an improvement of the predictions and thus also the sorting process. In this paper, we use the slight gap between the sensor lines of an RGB line scan camera to derive information about the particles' movements in real-time. This approach allows improving the predictions in optical belt sorters without necessitating any hardware modifications.},
 author = {Florian Pfaff and Georg Maier and Mikhail Aristov and Benjamin Noack and Robin Gruna and Uwe D. Hanebeck and Thomas Längle and Jürgen Beyerer and Christoph Pieper and Harald Kruggel-Emden and Siegmar Wirtz and Viktor Scherer},
 doi = {10.1515/auto-2017-0009},
 journal = {at -- Automatisierungstechnik, De Gruyter},
 month = {June},
 pdf = {AT17_Pfaff.pdf},
 title = {Real-Time Motion Prediction Using the Chromatic Offset of Line Scan Cameras},
 url = {https://doi.org/10.1515/auto-2017-0009},
 year = {2017}
}

@article{Automatica17_Noack,
 abstract = {In distributed and decentralized state estimation systems, fusion methods are employed to systematically combine multiple estimates of the state into a single, more accurate estimate. An often encountered problem in the fusion process relates to unknown common information that is shared by the estimates to be fused and is responsible for correlations. If the correlation structure is unknown to the fusion method, conservative strategies are typically pursued. As such, the parameterization introduced by the ellipsoidal intersection method has been a novel approach to describe unknown correlations, though suitable values for these parameters with proven consistency have not been identified yet. In this article, an extension of ellipsoidal intersection is proposed that guarantees consistent fusion results in the presence of unknown common information. The bound used by the novel approach corresponds to computing an outer ellipsoidal bound on the intersection of inverse covariance ellipsoids. As a major advantage of this   inverse covariance intersection method, fusion results prove to be more accurate than those provided by the well-known covariance intersection method.},
 author = {Benjamin Noack and Joris Sijs and Marc Reinhardt and Uwe D. Hanebeck},
 doi = {10.1016/j.automatica.2017.01.019},
 journal = {Automatica},
 month = {May},
 pages = {35--41},
 pdf = {Automatica17_Noack.pdf},
 title = {Decentralized Data Fusion with Inverse Covariance Intersection},
 url = {https://dx.doi.org/10.1016/j.automatica.2017.01.019},
 volume = {79},
 year = {2017}
}

@inproceedings{SPIE17_Hanebeck,
 abstract = {We are tracking multiple targets based on noisy measurements. The targets are labeled, the measurements are unlabeled, and the association of measurements to targets is unknown. Our goal is association-free tracking, so the associations will never be determined as this is costly and impractical in many scenarios. By employing a permutation-invariant and differentiable point set distance measure, we derive a modified association-free multi-target measurement equation. It maintains the target identities but is invariant to permutations in the unlabeled measurements. Based on this measurement equation, we derive an efficient sample-based association-free multi-target Kalman filter. The proposed new approach is straightforward to implement and scalable.},
 address = {Anaheim, California, USA},
 author = {Uwe D. Hanebeck and Marcus Baum and Peter Willett},
 booktitle = {SPIE - Signal Processing, Sensor/Information Fusion, and Target Recognition XXVI},
 month = {April},
 pdf = {SPIE17_Hanebeck.pdf},
 title = {Symmetrizing Measurement Equations for Association-free Multi-target Tracking via Point Set Distances},
 year = {2017}
}

@inproceedings{OCM17_Maier,
 abstract = {Sensor-based sorting provides state-of-the-art solutions for sorting of cohesive, granular materials. Systems tailored to a task at hand, for instance by means of sensors and implementations of data analysis. Conventional systems utilize scanning sensors which do not allow for extraction of motion-related information of objects contained in a material feed. Recently, usage of area-scan cameras to overcome this disadvantage has been proposed. Multitarget tracking can then be used in order to accurately estimate the point in time and position at which any object will reach the separation stage. In this paper, utilizing motion information of objects which can be retrieved from multitarget tracking for the purpose of classification is proposed. Results show that corresponding features can significantly increase classification performance and eventually decrease the detection error of a sorting system.},
 address = {Karlsruhe, Germany},
 annote = {Winner Best Paper Award Certificate (PDF)},
 author = {Georg Maier and Florian Pfaff and Florian Becker and Christoph Pieper and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Jürgen Beyerer},
 booktitle = {Proceedings of the 3rd Conference on Optical Characterization of Materials (OCM 2017)},
 month = {March},
 pdf = {OCM17_Maier.pdf},
 title = {Improving Material Characterization in Sensor-Based Sorting by Utilizing Motion Information},
 url = {https://www.ksp.kit.edu/9783731506126},
 year = {2017}
}

@article{TAES17_Kurz,
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 doi = {10.1109/TAES.2017.2650079},
 issn = {0018-9251},
 journal = {IEEE Transactions on Aerospace and Electronic Systems},
 month = {February},
 number = {1},
 pages = {530--534},
 pdf = {TAES17_Kurz.pdf},
 title = {Deterministic Sampling on the Torus for Bivariate Circular Estimation},
 volume = {53},
 year = {2017}
}

@incollection{MunozMorcillo2016,
 author = {Jesús Munoz Morcillo and Florian Faion and Antonio Zea and Uwe D. Hanebeck and Caroline Y. Robertson-von Trotha},
 booktitle = {Space and Time Visualisation},
 editor = {Boştenaru Dan, Maria and Crăciun, Cerasella},
 isbn = {978-3-319-24942-1},
 pages = {173--191},
 pdf = {STV16_Morcillo.pdf},
 publisher = {Springer International Publishing},
 title = {e-Installation: Synesthetic Documentation of Media Art via Telepresence Technologies},
 year = {2016}
}

@inproceedings{CDC16_Dolgov,
 address = {Las Vegas, Nevada, USA},
 author = {Maxim Dolgov and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 55th IEEE Conference on Decision and Control (CDC 2016)},
 month = {December},
 pdf = {CDC16_Dolgov.pdf},
 title = {Finite-horizon Dynamic Compensation of Markov Jump Linear Systems without Mode Observation},
 year = {2016}
}

@article{JAIF16_Kurz,
 author = {Gerhard Kurz and Igor Gilitschenski and Roland Y. Siegwart and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {December},
 number = {2},
 pages = {138--156},
 title = {Methods for Deterministic Approximation of Circular Densities},
 url = {https://confcats_isif.s3.amazonaws.com/web-files/journals/entries/JAIF_Vol11_2_3.pdf},
 volume = {11},
 year = {2016}
}

@article{JAIF16_Pfaff,
 author = {Florian Pfaff and Gerhard Kurz and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {December},
 number = {2},
 pages = {206--226},
 title = {Multivariate Angular Filtering Using Fourier Series},
 url = {https://confcats_isif.s3.amazonaws.com/web-files/journals/entries/JAIF_Vol11_2_7.pdf},
 volume = {11},
 year = {2016}
}

@inproceedings{MFI16_Kurz,
 abstract = {Various applications necessitate the estimation of
quantities defined on intervals or the unit circle, which can
also be parameterized as an interval. These applications include
estimation of joint angles that are either limited to a certain
range or that are 360-degree-periodic. For this purpose, we
consider two approaches based on discretizing the state space
that use fundamentally different density representations. We
show how prediction and measurement update for systems with
nonlinear dynamics and nonlinear measurement models can
be performed in each representation. In particular, we discuss
the choices that go into designing discrete filters, which are
sometimes taken for granted. A thorough comparison and a
numerical evaluation of both approaches show the advantages
and disadvantages of each method.},
 address = {Baden-Baden, Germany},
 author = {Gerhard Kurz and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Kurz.pdf},
 title = {Discrete Recursive Bayesian Filtering on Intervals and the Unit Circle},
 year = {2016}
}

@inproceedings{MFI16_Zea,
 abstract = {ABSTRACT},
 address = {Baden-Baden, Germany},
 author = {Antonio Zea and Florian Faion and Jannik Steinbring and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Zea.pdf},
 title = {Exploiting Negative Measurements for Tracking Star-Convex Extended Objects},
 year = {2016}
}

@inproceedings{MFI16_Faion,
 address = {Baden-Baden, Germany},
 author = {Florian Faion and Antonio Zea and Benjamin Noack and Jannik Steinbring and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Faion.pdf},
 title = {Camera- and IMU-based Pose Tracking for Augmented Reality},
 year = {2016}
}

@inproceedings{MFI16_Maier,
 abstract = {State-of-the-art sensor-based sorting systems provide
solutions to sort various products according to quality
aspects. Such systems face the challenge of an existing delay
between perception and separation of the material. To reliably
predict an object's position when reaching the separation
stage, information regarding its movement needs to be derived.
Multitarget tracking offers approaches through which this
can be achieved. However, processing time is typically limited
since the sorting decision for each object needs to be derived
sufficiently early before it reaches the separation stage. In this
paper, an approach for multitarget tracking in sensor-based
sorting is proposed which supports establishing an upper bound
regarding processing time required for solving the measurement
to track association problem. To demonstrate the success of
the proposed method, experiments are conducted for data-sets
obtained via simulation of a sorting system. This way, it
is possible to not only demonstrate the impact on required
runtime but also on the quality of the association.},
 address = {Baden-Baden, Germany},
 author = {Georg Maier and Florian Pfaff and Christoph Pieper and Robin Gruna and Benjamin Noack and Harald Kruggel-Emden and Thomas Längle and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Jürgen Beyerer},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Maier.pdf},
 title = {Fast Multitarget Tracking via Strategy Switching for Sensor-Based Sorting},
 year = {2016}
}

@inproceedings{MFI16_Noack,
 abstract = {For decentralized fusion problems, ellipsoidal intersection has been proposed as an efficient fusion rule
that provides less conservative results as compared to the well-know covariance intersection method. Ellipsoidal intersection
relies on the computation of a common estimate that is shared by the estimates to be fused. In this paper,
an algebraic reformulation of ellipsoidal intersection is discussed that circumvents the computation of the common estimate.
It is shown that ellipsoidal intersection corresponds to an internal ellipsoidal approximation of the intersection of covariance ellipsoids.
An interesting result is that ellipsoidal intersection can be computed with the aid of the Bar-Shalom/Campo fusion formulae.
This is achieved by assuming a specific correlation structure between the estimates to be fused.},
 address = {Baden-Baden, Germany},
 author = {Benjamin Noack and Joris Sijs and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Noack.pdf},
 title = {Algebraic Analysis of Data Fusion with Ellipsoidal Intersection},
 year = {2016}
}

@inproceedings{MFI16_Pfaff,
 abstract = {Multitarget tracking problems arise in many real-world
applications. The performance of the utilized algorithm
strongly depends both on how the data association problem is
handled and on the suitability of the motion models employed.
Especially the motion models can be hard to validate. Previously,
we have proposed to use multitarget tracking to improve optical
belt sorters. In this paper, we evaluate both the suitability of
our model and the tracking and then of our entire system
incorporating the image processing component via the use of
highly realistic numerical simulations. We first assess the model
using noise-free measurements generated by the simulation and
then evaluate the entire system by using synthetically generated
image data.},
 address = {Baden-Baden, Germany},
 author = {Florian Pfaff and Christoph Pieper and Georg Maier and Benjamin Noack and Harald Kruggel-Emden and Robin Gruna and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Thomas Längle and Jürgen Beyerer},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Pfaff.pdf},
 title = {Simulation-Based Evaluation of Predictive Tracking for Sorting Bulk Materials},
 year = {2016}
}

@inproceedings{MFI16_Steinbring,
 abstract = {As an alternative to Kalman filters and particle filters, recently the progressive Gaussian filter (PGF) was proposed for estimating the state of discrete-time stochastic nonlinear dynamic systems. Like Kalman filters, the estimate of the PGF is a Gaussian distribution, but like particle filters, its measurement update works directly with the likelihood function in order to avoid the inherent linearization of the Kalman filters. However, compared to particle filters, the PGF allows for much faster state estimation and circumvents the severe problem of particle degeneracy by gradually transforming its prior Gaussian distribution into a posterior one. In this paper, we further enhance the estimation quality and runtime of the PGF by proposing a semi-analytic measurement update applicable to likelihood functions that only depend on a subspace of the system state. In fact, the proposed semi-analytic measurement update is not limited to the PGF and can be used by any nonlinear state estimator as long as its state estimate is Gaussian, e.g., the Gaussian particle filter.},
 address = {Baden-Baden, Germany},
 author = {Jannik Steinbring and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Steinbring_PGF.pdf},
 title = {Semi-Analytic Progressive Gaussian Filtering},
 year = {2016}
}

@inproceedings{MFI16_Steinbring2,
 abstract = {In this paper, we present a novel online approach for tracking whole-body human motion based on unlabeled measurements of markers attached to the body. For that purpose, we employ a given kinematic model of the human body including the locations of the attached markers. Based on the model, we apply a combination of constrained sample-based Kalman filtering and multi-target tracking techniques: 1) joint constraints imposed by the human body are satisfied by introducing a parameter transformation based on periodic functions, 2) a global nearest neighbor (GNN) algorithm computes the most likely one-to-one association between markers and measurements, and 3) multiple hypotheses tracking (MHT) allows for a robust initialization that only requires an upright standing user. Evaluations clearly demonstrate that the proposed tracking provides highly accurate pose estimates in real-time, even for fast and complex motions. In addition, it provides robustness to partial occlusion of markers and also handles unavoidable clutter measurements.},
 address = {Baden-Baden, Germany},
 author = {Jannik Steinbring and Christian Mandery and Florian Pfaff and Florian Faion and Tamim Asfour and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2016)},
 month = {September},
 pdf = {MFI16_Steinbring_Tracking.pdf},
 title = {Real-Time Whole-Body Human Motion Tracking Based on Unlabeled Markers},
 year = {2016}
}

@inproceedings{DEM16_Pieper,
 abstract = {Automated optical sorting systems are important devices in the growing field of bulk solids handling. The initial sorter calibration and the precise optical sorting of many materials is still very time consuming and difficult. A numerical model of an automated optical belt sorter is presented in this study. The sorter and particle interaction is described with the Discrete Element Method (DEM) while
the separation phase is considered in a post processing step. Different operating parameters and their influence on sorting quality are investigated. In addition, two models for detecting and predicting the particle movement between the detection point and the separation step are presented and compared, namely a conventional line scan camera model and a new approach combining an area scan camera model with particle tracking.},
 address = {Dalian, China},
 author = {Christoph Pieper and Harald Kruggel-Emden and Siegmar Wirtz and Viktor Scherer and Florian Pfaff and Benjamin Noack and Uwe D. Hanebeck and Georg Maier and Robin Gruna and Thomas Längle and Jürgen Beyerer},
 booktitle = {Proceedings of the 7th International Conference on Discrete Element Methods (DEM7)},
 month = {August},
 pdf = {DEM7_Pieper.pdf},
 series = {Springer Proceedings in Physics},
 title = {Numerical Investigation of Optical Sorting Using the Discrete Element Method},
 year = {2016}
}

@inproceedings{Fusion16_Steinbring,
 abstract = {In this paper, we present a novel approach to
optimally fuse estimates in distributed state estimation for linear
and nonlinear systems. An optimal fusion requires the knowledge
of the correct correlations between locally obtained estimates.
The naive and intractable way of calculating the correct
correlations would be to exchange information about every processed
measurement between all nodes. Instead, we propose to obtain
the correct correlations by keeping and processing a small set
of deterministic samples on each node in parallel to the actual
local state estimation. Sending these samples in addition to the
local state estimate to the fusion center allows for correctly
reconstructing the desired correlations between all estimates.
In doing so, each node does not need any information about
measurements processed on other nodes. We show the optimality
of the proposed method by means of tracking an extended object
in a multi-camera network.},
 address = {Heidelberg, Germany},
 author = {Jannik Steinbring and Benjamin Noack and Marc Reinhardt and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Steinbring.pdf},
 title = {Optimal Sample-Based Fusion for Distributed State Estimation},
 year = {2016}
}

@inproceedings{Fusion16_Baum,
 abstract = {The recently proposed Kernel-SME filter for multiobject
tracking is a further development of the Symmetric
Measurement Equation (SME) idea introduced by Kamen in
the 1990s. The Kernel-SME constructs a symmetric, i.e., permutation
invariant, measurement equation by transforming the
measurements to a kernel mixture function. This transformation
is scalable to a large number of objects and allows for deriving an
efficient closed-form Gaussian filter based on the Kalman filter
formulas. This work shows how the Kernel-SME approach can
systematically incorporate false and missing measurements.},
 address = {Heidelberg, Germany},
 author = {Marcus Baum and Shishan Yang and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Kernel-SME_Filter.pdf},
 title = {The Kernel-SME Filter with False and Missing Measurements},
 year = {2016}
}

@inproceedings{Fusion16_Chlebek,
 abstract = {We propose a novel progressive Gaussian filter for nonlinear stochastic systems. A Gaussian approximation of the posterior is computed without an explicit assumption of a linear relation between the system state and the measurement. This allows for better quality of the estimation compared to Kalman filters for nonlinear problems like the EKF or UKF. In this work, we use the progressive filter framework, which gradually incorporates information of a measurement into the state estimate
by considering a flow of probability mass from the prior to the posterior state estimate. We propose a novel particle flow by utilizing a simple linear model. This model predicts the movement of single particles over the course of the filter progression. The predicted trajectory is corrected using importance sampling and moment matching. The proposed method is evaluated in comparison with other state-of-the-art nonlinear Bayesian filters.},
 address = {Heidelberg, Germany},
 author = {Christof Chlebek and Jannik Steinbring and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Chlebek.pdf},
 title = {Progressive Gaussian Filter Using Importance Sampling and Particle Flow},
 year = {2016}
}

@inproceedings{Fusion16_Duan,
 abstract = {Fingerprinting localization is to estimate a mobile
terminal's location using its online received signal strength (RSS)
measurement and offline RSS database originated from multiple
access points (APs). Kernel-based fingerprinting localization is
such a competitive algorithm. However, all training data need to
be considered in its offline model learning stage. This render high
risks for overfitting. To alleviate this, we suggest to apply clustering
to the localization region of interest first and then use kernalbased
fingerprinting localization for each cluster. A byproduct of
clustering is that the computational load for each cluster is also
significantly reduced. To further reduce the computational load
within each cluster, we also suggest to apply principal component
compression to the raw RSS measurements to reduce their
dimensionality. The rationale for applying principal component
compression is that the distributions of the RSS measurements
at all calibration points (CPs) within each cluster will be
more similar after clustering. Performance evaluation using both
simulated data and real data show that the extended kernelbased
fingerprinting localization using clustering and principal
component compression have better location estimation accuracy
and less computational load.},
 address = {Heidelberg, Germany},
 author = {Zhansheng Duan and Qi Zhou and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Duan_Fingerprinting.pdf},
 title = {Extended Kernel-Based Location Fingerprinting in Wireless Sensor Networks},
 year = {2016}
}

@inproceedings{Fusion16_Faion,
 abstract = {We look at the task of estimating the parameters
of a geometric constraint from noisy points in 2D. The classical
approach of minimizing the Euclidean distance error between
points and constraint generally yields biased estimates for
non-linear constraints and higher noise levels. To deal with this
issue, the expected distribution of the distance error can be
explicitly incorporated in the estimator. However, for piecewise
linear constraints, e.g., polygons, only computationally
demanding sampling-based approaches are available. We propose two
major contributions in order to resolve this issue. First, we
derive closed-form expressions for the probability density of the
signed distance between noisy points and a polygon angle. Second,
based on this result, we develop a bias reduction method for
polygons, which can be calculated in closed-form as well. We
demonstrate that the quality of our approach can compete with
its sampling-based alternatives, but only demands a fraction of
their computational cost.},
 address = {Heidelberg, Germany},
 author = {Florian Faion and Maxim Dolgov and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Faion.pdf},
 title = {Closed-form Bias Reduction for Shape Estimation with Polygon Models},
 year = {2016}
}

@inproceedings{Fusion16_Gilitschenski,
 abstract = {In this work, a new method for approximating
circular probability distributions by a mixture of weighted
discrete samples is proposed. Particularly, the wrapped normal
distribution, the von Mises distribution, and the Bingham distribution
are considered. The approximation approach is based on
formulating a quantizer and a global optimality measure, which
can be optimized directly. Furthermore, a relationship between
the Bingham distribution and the von Mises distribution are
formulated showing that it is sufficient to approximate a von
Mises distribution with suitably chosen parameters in order to
obtain an optimal approximation of the Bingham distribution.
The resulting approximation is of particular interest for filtering
applications, because the involved optimality measure gives rise to
a general error estimate in propagation of uncertainties through
nontrivial functions in the circular domain.},
 address = {Heidelberg, Germany},
 author = {Igor Gilitschenski and Gerhard Kurz and Uwe D. Hanebeck and Roland Siegwart},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Quantization.pdf},
 title = {Optimal Quantization of Circular Distributions},
 year = {2016}
}

@inproceedings{Fusion16_Hanebeck,
 abstract = {This paper introduces an enhanced
method for progressive Bayesian estimation based
on a set of deterministic samples. The information
of a given measurement is gradually introduced in
order to avoid particle degeneration, which is usually
encountered in standard particle filters. The main
contribution of this paper is to derive a new method
for exploiting smoothness assumptions about the
unknown underlying density function of the state.},
 address = {Heidelberg, Germany},
 author = {Uwe D. Hanebeck and Martin Pander},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Hanebeck.pdf},
 title = {Progressive Bayesian Estimation with Deterministic Particles},
 year = {2016}
}

@inproceedings{Fusion16_Kurz-Control,
 abstract = {Chance-constrained control is a difficult problem
even if the considered system dynamics are linear. The difficulty
stems from the facts that the chance constraints are difficult
to evaluate and that the control law is nonlinear due to the
constraints. In this paper, we present a novel approach to chanceconstrained
control, where we solve the unconstrained control
problem first and then use a progressive method to gradually
introduce the chance constraints. This has significant advantages
compared to traditional methods, because we do not require a
feasible initial solution for the numerical optimization algorithm.
Finally, we evaluate the novel method and compare it to an
approach from literature.},
 address = {Heidelberg, Germany},
 author = {Gerhard Kurz and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Kurz-ChanceConstraints.pdf},
 title = {Progressive Closed-Loop Chance-Constrained Control},
 year = {2016}
}

@inproceedings{Fusion16_Kurz,
 abstract = {When approximating one probability density with
another density, it is desirable to minimize the information loss
of the approximation as quantified by, e.g., the Kullback–Leibler
divergence (KLD). It has been known for some time that in the
case of the Gaussian distribution, matching the first two moments
of the original density yields the optimal approximation in terms
of minimizing the KLD. In this paper, we will show that a similar
property can be proven for certain hyperspherical probability
distributions, namely the von Mises–Fisher and the Watson
distribution. This result has profound implications for momentbased
filtering on the unit hypersphere as it shows that momentbased
approaches are optimal in the information-theoretic sense.},
 address = {Heidelberg, Germany},
 author = {Gerhard Kurz and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Kurz-KLD2.pdf},
 title = {Kullback-Leibler Divergence and Moment Matching for Hyperspherical Probability Distributions},
 year = {2016}
}

@inproceedings{Fusion16_Noack,
 abstract = {State estimation concepts like the Kalman filter heavily rely on potentially noisy sensor data.
In general, the estimation quality depends on the amount of sensor data that can be exploited.
However, missing observations do not necessarily impair the estimation quality but may also convey
exploitable information on the system state. This type of information - noted as negative
information - often requires specific measurement and noise models in order to take advantage of it.
In this paper, a hybrid Kalman filter concept is employed that allows using both stochastic and
set-membership representations of information. In particular, the latter representation is
intended to account for negative information, which can often be easily described as a bounded
set in the measurement space. Depending on the type of information, the filtering step of the
proposed estimator adaptively switches between Gaussian and ellipsoidal noise representations.
A target tracking scenario is studied to evaluate and discuss the proposed concept.},
 address = {Heidelberg, Germany},
 author = {Benjamin Noack and Florian Pfaff and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Noack.pdf},
 title = {State Estimation Considering Negative Information with Switching Kalman and Ellipsoidal Filtering},
 year = {2016}
}

@inproceedings{Fusion16_Pfaff,
 abstract = {While nonlinear filtering for circular quantities is
closely related to nonlinear filtering on linear domains, the
underlying manifold enables the development of novel filters that
take advantage of the boundedness of the domain. Previously,
we introduced Fourier filters that approximate the density or its
square root using Fourier series. For these filters, we proposed
filter steps for arbitrary likelihoods and prediction steps for
the identity system model with additive noise. This paper adds
the capability of handling arbitrary transition densities in the
prediction step, which facilitates, e.g., the use of the filters for
nonlinear systems with additive noise. In the evaluation, the new
prediction steps for the Fourier filters outperform an SIR particle
filter, a grid filter, and a nonlinear variant of the von Mises filter.},
 address = {Heidelberg, Germany},
 author = {Florian Pfaff and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_nonlinearPred.pdf},
 title = {Nonlinear Prediction for Circular Filtering Using Fourier Series},
 year = {2016}
}

@inproceedings{Fusion16_Zea,
 address = {Heidelberg, Germany},
 author = {Antonio Zea and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th International Conference on Information Fusion (Fusion 2016)},
 month = {July},
 pdf = {Fusion16_Zea.pdf},
 title = {Tracking Elongated Extended Objects Using Splines},
 year = {2016}
}

@article{PowTec16_Pieper,
 abstract = {Optical sorters are important devices in the processing and handling
of the globally growing material streams. The precise optical sorting of many
bulk solids is still difficult due to the great technical effort necessary
for transport and flow control. In this study, particle separation with an automated
optical belt sorter is modeled numerically. The Discrete Element Method (DEM) is used
to model the sorter and calculate the particle movement as well as
particle – particle and particle – wall interactions. The particle ejection stage with
air valves is described with the help of a MATLAB script utilizing particle movement information
obtained with the DEM. Two models for predicting the particle movement between
the detection and separation phase are implemented and compared. In the first model,
it is assumed that the particles are moving with belt velocity and without any cross movements
and a conventional line scan camera is used for particle detection. In the second model,
a more sophisticated approach is employed where the particle motion is predicted with
an area scan camera combined with a tracking algorithm. In addition, the influence of
different operating parameters like particle shape or conveyor belt length on the
separation quality of the system is investigated. Results show that numerical simulations
can offer detailed insight into the operation performance of optical sorters and help to
optimize operating parameters. The area scan camera approach was found to be superior to
the standard line scan camera model in almost all investigated categories.},
 author = {Christoph Pieper and Georg Maier and Florian Pfaff and Harald Kruggel-Emden and Siegmar Wirtz and Robin Gruna and Benjamin Noack and Viktor Scherer and Thomas Längle and Jürgen Beyerer and Uwe D. Hanebeck},
 doi = {10.1016/j.powtec.2016.07.018},
 journal = {Powder Technology},
 month = {July},
 title = {Numerical Modeling of an Automated Optical Belt Sorter Using the Discrete Element Method},
 url = {https://dx.doi.org/10.1016/j.powtec.2016.07.018},
 year = {2016}
}

@inproceedings{ECC16_Chlebek,
 abstract = {We consider stochastic nonlinear time-variant systems
with imperfect state information in the context of model
predictive control. The optimal control performance can only be
achieved by closed-loop policies that anticipate future behavior.
However, the computation of these policies is in general not
tractable due to the presence of the dual effect, i.e., the control
actions not only influence the state but also the uncertainty
of its estimate. Thus, we propose an approximation to closedloop
control. We use a forward calculation approach, which is
derived from an open-loop feedback (OLF) control setup, but
implements the fundamental property of closed-loop control
that future measurement feedback is considered in the optimization.
By using a finite set of representative measurements,
the feedback behavior is anticipated only on basis of current
information. An optimization based on a continuation method
implements an effective calculation to obtain a sequence of
control inputs. The effectiveness of the presented approach is
demonstrated by means of a miniature walking robot.},
 address = {Aalborg, Denmark},
 author = {Christof Chlebek and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 European Control Conference (ECC 2016)},
 month = {June},
 pdf = {ECC16_Chlebek.pdf},
 title = {Approximation of Stochastic Nonlinear Closed-Loop Feedback Control with Application to Miniature Walking Robots},
 year = {2016}
}

@inproceedings{ECC16_Dolgov,
 abstract = {In this paper, we address control of Markov Jump Linear Systems
via dynamic output feedback without mode observation. Because the optimal
nonlinear control law for this problem is intractable, we assume a linear
controller. Under this assumption, the control law computation can be
expressed in terms of an optimization problem that involves Bilinear Matrix
Inequalities. Unfortunately, this problem is NP-hard. Thus, we propose an
alternative iterative algorithm and demonstrate it in a numerical example.
To the best of our knowledge, the paper presents the first solution of
mode-independent dynamic compensation of Markov Jump Linear Systems.},
 address = {Aalborg, Denmark},
 author = {Maxim Dolgov and Christof Chlebek and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2016 European Control Conference (ECC 2016)},
 month = {June},
 pdf = {ECC16_Dolgov.pdf},
 title = {Dynamic Compensation of Markov Jump Linear Systems without Mode Observation},
 year = {2016}
}

@article{JAIF16_Symmetric_S2KF_Steinbring,
 abstract = {Nonlinear Kalman Filters (KFs) are powerful and widely-used techniques when trying to estimate the hidden state of a stochastic nonlinear dynamic system. A novel sample-based KF is the Smart Sampling Kalman Filter (S²KF). It is based on deterministic Gaussian samples which are obtained from an offline optimization procedure. Although this sampling technique is quite effective, it does not preserve the point symmetry of the Gaussian distribution. In this paper, we overcome this issue by extending the S²KF with a new point-symmetric Gaussian sampling scheme to improve its estimation quality. Moreover, we also improve the numerical stability of the sample computation. This allows us to accurately approximate thousand-dimensional Gaussian distributions using tens of thousands of optimally placed and equally weighted samples. We evaluate the new symmetric S²KF by computing higher-order moments of standard normal distributions and investigate the estimation quality of the S²KF when dealing with symmetric measurement equations. Additionally, extended object tracking based on many measurements per time step is considered. This high-dimensional estimation problem shows the advantage of the S²KF being able to use an arbitrary number of samples independent of the state dimension, in contrast to other state-of-the-art sample-based Kalman Filters. Finally, other estimators also relying on the S²KF’s Gaussian sampling technique, e.g., the Progressive Gaussian Filter (PGF), will benefit from the new point-symmetric sampling as well.},
 author = {Jannik Steinbring and Martin Pander and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {June},
 number = {1},
 pages = {71--90},
 pdf = {JAIF16_Steinbring.pdf},
 title = {The Smart Sampling Kalman Filter with Symmetric Samples},
 volume = {11},
 year = {2016}
}

@article{SPL16_Kurz,
 abstract = {We introduce the unscented von Mises–Fisher filter (UvMFF), a nonlinear filtering algorithm for dynamic state estimation on the n-dimensional unit hypersphere. Estimation problems on the unit hypersphere occur in computer vision, e.g., when using omnidirectional cameras, as well as in signal processing. As approaches in literature are limited to very simple system and measurement models, we propose a deterministic sampling scheme on the unit hypersphere, which allows us to handle nonlinear system and measurement models. The proposed approach can be seen as a hyperspherical variant of the unscented Kalman filter (UKF). The advantages of the novel method are shown by means of simulations.},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 doi = {10.1109/LSP.2016.2529854},
 journal = {IEEE Signal Processing Letters},
 month = {April},
 number = {4},
 pages = {463-467},
 pdf = {Kurz-UnscentedVMF.pdf},
 title = {Unscented von Mises--Fisher Filtering},
 volume = {23},
 year = {2016}
}

@article{AES16_Kurz,
 abstract = {To facilitate recursive state estimation in the circular domain based on circular statistics, we introduce a general framework for estimation of a circular state based on different circular distributions. Specifically, we consider the wrapped normal (WN) distribution and the von Mises distribution. We propose an estimation method for circular systems with nonlinear system and measurement functions. This is achieved by relying on efficient deterministic sampling techniques. Furthermore, we show how the calculations can be simplified in a variety of important special cases, such as systems with additive noise, as well as identity system or measurement functions, which are illustrated using an example from aeronautics. We introduce several novel key components, particularly a distribution-free prediction algorithm, a new and superior formula for the multiplication of WN densities, and the ability to deal with nonadditive system noise. All proposed methods are thoroughly evaluated and compared with several state-of-the-art approaches.},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 doi = {10.1109/MAES.2016.150083},
 journal = {IEEE Aerospace and Electronic Systems Magazine},
 month = {March},
 number = {3},
 pages = {70-87},
 pdf = {AES16_Kurz.pdf},
 title = {Recursive Bayesian Filtering in Circular State Spaces},
 volume = {31},
 year = {2016}
}

@article{TM16_Pfaff,
 abstract = {Visual properties are powerful features to
reliably classify bulk materials, thereby allowing to detect
defect or low quality particles. Optical belt sorters are
an established technology to sort based on these properties,
but they suffer from delays between the simultaneous
classification and localization step and the subsequent
separation step. Therefore, accurate models to predict the
particles’ motions are a necessity to bridge this gap. In
this paper, we explicate our concept to use sophisticated
simulations to derive accurate models and optimize the
flow of bulk solids via adjustments of the sorter design.
This allows us to improve overall sorting accuracy and
cost efficiency. Lastly, initial results are presented.},
 author = {Florian Pfaff and Christoph Pieper and Georg Maier and Benjamin Noack and Harald Kruggel-Emden and Robin Gruna and Uwe D. Hanebeck and Siegmar Wirtz and Viktor Scherer and Thomas Längle and Jürgen Beyerer},
 doi = {/10.1515/teme-2015-0108},
 journal = {tm -- Technisches Messen, De Gruyter},
 month = {February},
 number = {2},
 pages = {77--84},
 pdf = {TM16_Pfaff.pdf},
 title = {Improving Optical Sorting of Bulk Materials Using Sophisticated Motion Models},
 url = {https://www.degruyter.com/view/j/teme.2016.83.issue-2/teme-2015-0108/teme-2015-0108.xml},
 volume = {83},
 year = {2016}
}

@article{TAC16_Gilitschenski,
 abstract = {In this work, we develop a recursive filter to estimate
orientation in 3D, represented by quaternions, using directional distributions.
Many closed-form orientation estimation algorithms are based on
traditional nonlinear filtering techniques, such as the extended Kalman
filter (EKF) or the unscented Kalman filter (UKF). These approaches
assume the uncertainties in the system state and measurements to be
Gaussian-distributed. However, Gaussians cannot account for the periodic
nature of the manifold of orientations and thus small angular errors
have to be assumed and ad hoc fixes must be used. In this work,
we develop computationally efficient recursive estimators that use the
Bingham distribution. This distribution is defined on the hypersphere
and is inherently more suitable for periodic problems. As a result,
these algorithms are able to consistently estimate orientation even in the
presence of large angular errors. Furthermore, handling of nontrivial
system functions is performed using an entirely deterministic method
which avoids any random sampling. A scheme reminiscent of the UKF
is proposed for the nonlinear manifold of orientations. It is the first
deterministic sampling scheme that truly reflects the nonlinear manifold
of orientations.},
 author = {Igor Gilitschenski and Gerhard Kurz and Simon J. Julier and Uwe D. Hanebeck},
 doi = {10.1109/TAC.2015.2423831},
 journal = {IEEE Transactions on Automatic Control},
 month = {January},
 number = {1},
 pages = {172-177},
 pdf = {TAC16_Gilitschenski.pdf},
 title = {Unscented Orientation Estimation Based on the Bingham Distribution},
 volume = {61},
 year = {2016}
}

@inproceedings{CDC15_Kurz,
 abstract = {Correlated uncertain angular quantities can be
modeled using the bivariate wrapped normal distribution.
In this paper, we focus on the problem of estimating the
distribution's parameters from a given set of samples. For
this purpose, we propose several new parameter estimation
methods and compare them to estimation techniques found in
literature. All methods are thoroughly evaluated in simulations.
One of the novel methods is shown to combine the advantages
of maximum likelihood estimation and moment-based methods,
thus outperforming current state-of-the-art techniques.},
 address = {Osaka, Japan},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 54th IEEE Conference on Decision and Control (CDC 2015)},
 month = {December},
 pdf = {CDC15_Kurz.pdf},
 title = {Parameter Estimation for the Bivariate Wrapped Normal Distribution},
 year = {2015}
}

@inproceedings{CDC15_Dolgov,
 abstract = {In this paper, we consider finite-horizon predictive
control of linear stochastic systems with chance constraints
where the admissible region is a convex polytope. For this
problem, we present a novel solution approach based on box
approximations. The key notion of our approach consists of
two steps. First, we apply a linear operation to the joint
state probability density function such that its covariance is
transformed into an identity matrix. This operation also defines
the transformation of the state space and therefore, of the
admissible polytope. Second, we approximate the admissible
region from the inside using axis-aligned boxes. By doing so, we
obtain a conservative approximation of the constraint violation
probability virtually in closed form (the expression contains
Gaussian error functions). The presented control approach is
demonstrated in a numerical example.},
 address = {Osaka, Japan},
 author = {Maxim Dolgov and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 54th IEEE Conference on Decision and Control (CDC 2015)},
 month = {December},
 pdf = {CDC15_Dolgov.pdf},
 title = {Chance-constrained Model Predictive Control based on Box Approximations},
 year = {2015}
}

@incollection{CRC15_Sijs,
 abstract = {To reduce the amount of data transfers
in networked systems, measurements can be taken at
an event on the sensor value rather than periodically
in time. Yet, this could lead to a divergence of estimation results when only the received measurement
values are exploited in a state estimation procedure.
A solution to this issue has been found by developing
estimators that perform a state update at both the event
instants as well as periodically in time: when an event
occurs the estimated state is updated using the measurement received, while at periodic instants the update is
based on knowledge that the sensor value lies within a
bounded subset of the measurement space. Several solutions for event-based state estimation will be presented
in this chapter, either based on stochastic representations
of random vectors, on deterministic representations of
random vectors or on a mixture of the two. All solutions aim to limit the required computational resources
by deriving explicit solutions for computing estimation
results. Yet, the main achievement for each estimation
solution is that stability of the estimation results are (not directly) dependent on the employed event sampling
strategy. As such, changing the event sampling strategy does not imply to change the event-based estimator
as well. This aspect is also illustrated in a case study
of tracking the distribution of a chemical compound
effected by wind via a wireless sensor network.},
 author = {Joris Sijs and Benjamin Noack and Mircea Lazar and Uwe D. Hanebeck},
 chapter = {Time-Periodic State Estimation with Event-Based Measurement Updates},
 editor = {Marek Miskowicz},
 month = {November},
 pages = {261--279},
 publisher = {CRC Press},
 title = {Event-Based Control and Signal Processing},
 url = {https://doi.org/10.1201/b19013},
 year = {2015}
}

@article{TAES15_Kurz,
 abstract = {We show an important property of the von Mises
distribution on the unit circle. If we approximate an arbitrary
circular distribution using a von Mises distribution, the result
obtained by trigonometric moment matching also minimizes
the Kullback–Leibler divergence (Theorem 1). This result is a
justification for circular filtering algorithms based on trigonometric
moment matching as the loss of information is minimized.
Furthermore, we show that Theorem 1 does not hold for the
wrapped normal distribution.},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 doi = {10.1109/TAES.2015.150406},
 journal = {IEEE Transactions on Aerospace and Electronic Systems},
 month = {October},
 number = {1},
 pages = {3480-3484},
 pdf = {TAES15_Kurz.pdf},
 title = {Trigonometric Moment Matching and Minimization of the Kullback--Leibler Divergence},
 volume = {51},
 year = {2015}
}

@article{SPL15_Baum,
 author = {Marcus Baum and Peter Willett and Uwe D. Hanebeck},
 doi = {10.1109/LSP.2015.2410217},
 journal = {IEEE Signal Processing Letters},
 month = {October},
 number = {10},
 pages = {1511-1515},
 title = {On Wasserstein Barycenters and MMOSPA Estimation},
 url = {https://ieeexplore.ieee.org/document/7054438/},
 volume = {22},
 year = {2015}
}

@inproceedings{SDF15_Faion,
 abstract = {We consider the task of recursively estimating the
pose and shape parameters of 3D objects based on noisy point
cloud measurements from their surface. We focus on objects
whose surface can be constructed by transforming a plane curve,
such as a cylinder that is constructed by extruding a circle.
However, designing estimators for such objects is challenging,
as the straightforward distance-minimizing approach cannot
observe all parameters, and additionally is subject to bias in
the presence of noise. In this article, we first discuss these issues
and then develop probabilistic models for cylinder, torus, cone,
and an extruded curve by adapting related approaches including
Random Hypersurface Models, partial likelihood, and symmetric
shape models. In experiments with simulated data, we show that
these models yield unbiased estimators for all parameters even
in the presence of high noise.},
 address = {Bonn, Germany},
 author = {Florian Faion and Antonio Zea and Jannik Steinbring and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2015)},
 month = {October},
 pdf = {SDF15_Faion.pdf},
 title = {Recursive Bayesian Pose and Shape Estimation of 3D Objects Using Transformed Plane Curves},
 year = {2015}
}

@inproceedings{SDF15_Kurz,
 abstract = {We propose a novel sampling algorithm for the
von Mises–Fisher distribution on the unit hypersphere. Unlike
previous works, we show a solution for an arbitrary number of
dimensions without requiring rejection sampling. As a result, the
proposed algorithm has a deterministic runtime. The key idea
consists in applying the inversion method to a one-dimensional
subproblem and analytically calculating the integral occurring in
the distribution function. The proposed method is most efficient
for odd numbers of dimensions. We compare the algorithm to a
state-of-the-art rejection sampling method in simulations.},
 address = {Bonn, Germany},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2015)},
 month = {October},
 pdf = {SDF15_Kurz-VMFSampling.pdf},
 title = {Stochastic Sampling of the Hyperspherical von Mises--Fisher Distribution Without Rejection Methods},
 year = {2015}
}

@inproceedings{MFI15_Baum,
 abstract = {This work investigates a novel method for dealing
with unknown data associations in Kalman filter-based
Simultaneous Localization and Mapping (SLAM) problems. The key
idea is to employ the concept of Symmetric Measurement
Equations (SMEs) in order to remove the data association
uncertainty from the original measurement equation. Based
on the resulting modified measurement equation, standard
nonlinear Kalman filters can estimate the full joint state vector
of the robot and landmarks without explicitly calculating data
association hypotheses.},
 address = {San Diego, California, USA},
 author = {Marcus Baum and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Baum.pdf},
 title = {Kalman Filter-based SLAM with Unknown Data Association using Symmetric Measurement Equations},
 year = {2015}
}

@inproceedings{MFI15_Faion,
 abstract = {In this paper, we propose a novel algorithm for automatically
calibrating a network of depth sensors, based on a moving calibration object.
The sensors may have non-overlapping fields of view in order to avoid interference.
Two major challenges are discussed. First, depending on where the object is
located relative to the sensor, the number and quality of the measurements strongly varies.
Second, a single depth sensor observes the calibration object only from one side.
Dealing with these challenges requires a simple calibration object as
well as an algorithm that can deal with under-determined measurements of
varying quality. A recursive Bayesian estimator is developed that determines the
extrinsic parameters by measuring the surface of a moving cube with
known pose. Our approach does not restrict the configuration of the network and
requires no manual initialization or interaction.
Ambiguities that are induced by the rotational cube symmetries are
resolved by applying a multiple model approach. Besides synthetic evaluation we
perform real data experiments and compare to state-of-the-art calibration.},
 address = {San Diego, California, USA},
 author = {Florian Faion and Marcus Baum and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Faion.pdf},
 title = {Depth Sensor Calibration by Tracking an Extended Object},
 year = {2015}
}

@inproceedings{MFI15_Gilitschenski,
 abstract = {This paper presents a novel algorithm for the
estimation of planar rigid-body motions. It is based on using a
probability distribution that is inherently defined on the non-
linear manifold representing these motions and on proposing
a deterministic sampling scheme that makes consideration of
complicated system models possible. Furthermore, we show
that the measurement update for a manifold equivalent to
noisy direct measurements can be carried out in closed form.
Thus, the resulting method avoids errors made due to local
linearization and outperforms methods that wrongly assume
Gaussian distributions, which we show by comparing the
proposed filter to the UKF.},
 address = {San Diego, California, USA},
 author = {Igor Gilitschenski and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Gilitschenski.pdf},
 title = {A Stochastic Filter for Planar Rigid-Body Motions},
 year = {2015}
}

@inproceedings{MFI15_Hanebeck,
 abstract = {Fusing two random vectors is simple, when they are characterized by continuous probability functions.
According to Bayes' law, fusion then consists of multiplying the two densities. When only empirical distributions
are given and a resulting empirical distribution is desired, Bayes' law is no longer applicable. Obviously,
fusion could now be performed by reconstructing the underlying continuous densities, subsequent multiplication,
and sampling of the result. As this is overly complicated, our goal is to perform a direct Bayesian fusion of the
two given empirical distributions. We devise a generalized multiplication procedure that mutually reweights appropriate
points of one density by local density values of the other density. The density values are efficiently estimated locally
by nearest neighbor operations. The method is symmetric in the sense that it uses points from both densities.},
 address = {San Diego, California, USA},
 author = {Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Hanebeck.pdf},
 title = {Bayesian Fusion of Empirical Distributions Based on Local Density Reconstruction},
 year = {2015}
}

@inproceedings{MFI15_Kurz,
 abstract = {Fusion of toroidal information, such as correlated
angles, is a problem that arises in many fields ranging from
robotics and signal processing to meteorology and bioinfor-
matics. For this purpose, we propose a novel fusion method
based on the bivariate von Mises distribution. Unlike most
literature on the bivariate von Mises distribution, we consider
the full version with matrix-valued parameter rather than
a simplified version. By doing so. we are able to derive
the exact analytical computation of the fusion operation. We
also propose an efficient approximation of the normalization
constant including an error bound and present a parameter
estimation algorithm based on a maximum likelihood approach.
The presented algorithms are illustrated through examples.},
 address = {San Diego, California, USA},
 annote = {Nominee Best Paper Award Certificate (PDF)},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Kurz.pdf},
 title = {Toroidal Information Fusion Based on the Bivariate von Mises Distribution},
 year = {2015}
}

@inproceedings{MFI15_Noack,
 abstract = {In many dynamic systems, the evolution of the
state is subject to specific constraints. In general, constraints
cannot easily be integrated into the prediction-correction structure
of the Kalman filter algorithm. Linear equality constraints
are an exception to this rule and have been widely used and
studied as they allow for simple closed-form expressions. A
common approach is to reformulate equality constraints into
pseudo measurements of the state to be estimated. However,
equality constraints define deterministic relationships between
state components which is an undesirable property in Kalman
filtering as this leads to singular covariance matrices. A second
problem relates to the knowledge required to identify and
define precise constraints, which are met by the system state.
In this article, ellipsoidal constraints are introduced that can
be employed to model a bounded region, to which the system
state is constrained. This concept constitutes an easy-to-use
relaxation of equality constraints. In order to integrate ellipsoidal
constraints into the Kalman filter structure, a generalized filter
framework is utilized that relies on a combined stochastic and
set-membership uncertainty representation.},
 address = {San Diego, California, USA},
 author = {Benjamin Noack and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Noack.pdf},
 title = {State Estimation for Ellipsoidally Constrained Dynamic Systems with Set-Membership Pseudo Measurements},
 year = {2015}
}

@inproceedings{MFI15_Pfaff,
 abstract = {Optical belt sorters are a versatile, state-of-the-art technology to sort
bulk materials that are hard to sort based on only nonvisual properties. In
this paper, we propose an extension to current optical belt sorters that
involves replacing the line camera with an area camera to observe a wider
field of view, allowing us to observe each particle over multiple time
steps. By performing multitarget tracking, we are able to improve the
prediction of each particle's movement and thus enhance the performance of
the utilized separation mechanism. We show that our approach will allow
belt sorters to handle new classes of bulk materials while improving cost
efficiency. Furthermore, we lay out additional extensions that are made
possible by our new paradigm},
 address = {San Diego, California, USA},
 author = {Florian Pfaff and Marcus Baum and Benjamin Noack and Uwe D. Hanebeck and Robin Gruna and Thomas Längle and Jürgen Beyerer},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Pfaff.pdf},
 title = {TrackSort: Predictive Tracking for Sorting Uncooperative Bulk Materials},
 year = {2015}
}

@inproceedings{MFI15_Steinbring,
 abstract = {Modeling 2D extended targets with star-convex Random Hypersurface Models (RHMs) allows for
accurate object pose and shape estimation. A star-convex RHM models the shape of an object with the aid of
a radial function that describes the distance from the object center to any point on its boundary. However,
up to now only linear estimators, i.e., Kalman Filters, are used due to the lack of a explicit likelihood function.
In this paper, we propose a closed-form and easy to implement likelihood function for tracking extended targets
with star-convex RHMs. This makes it possible to apply nonlinear estimators such as Particle Filters to estimate
a detailed shape of a target. We compared the proposed likelihood against the usual Kalman filter approaches
with tracking pose and shape of an airplane in 2D. The evaluations showed that the combination of the
Progressive Gaussian Filter (PGF) and the new likelihood function delivers the best estimation performance
and can outperform the usually employed Kalman Filters.},
 address = {San Diego, California, USA},
 author = {Jannik Steinbring and Marcus Baum and Antonio Zea and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Steinbring.pdf},
 title = {A Closed-Form Likelihood for Particle Filters to Track Extended Objects with Star-Convex RHMs},
 year = {2015}
}

@inproceedings{MFI15_Zea,
 abstract = {One of the challenges in shape tracking is how
to deal with associating measurements to sources in the shape,
while also taking to account parameters such as shape curvature
and noise characteristics. Partial Information Models (PIMs)
introduce a new approach that addresses this issue. The idea
is to reparametrize each measurement into two components,
one which depends on the position of its source on the shape,
and another which depends on how well it fits in the shape.
This allows for the derivation of a partial likelihood which
combines the strengths of probabilistic approaches and distance
minimization techniques. We propose an implementation of
PIMs using level-sets, which allow for a close approximation
of the distribution of distances we expect for a given shape. In
turn, this can be used to develop estimators that are highly
robust against high noise and occlusions.},
 address = {San Diego, California, USA},
 author = {Antonio Zea and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2015)},
 month = {September},
 pdf = {MFI15_Zea.pdf},
 title = {Shape Tracking using Partial Information Models},
 year = {2015}
}

@article{SPL15_Reinhardt,
 abstract = {One of the key challenges in distributed linear estimation
is the systematic fusion of estimates. While the fusion gains
that minimize the mean squared error of the fused estimate for
known correlations have been established, no analogous statement
could be obtained so far for unknown correlations. In this contribution,
we derive the gains that minimize the bound on the true covariance
of the fused estimate and prove that Covariance Intersection (CI)
is the optimal bounding algorithm for two estimates under completely
unknown correlations. When combining three or more variables, the
CI equations are not necessarily optimal, as shown by a counterexample.},
 author = {Marc Reinhardt and Benjamin Noack and Pablo O. Arambel and Uwe D. Hanebeck},
 doi = {10.1109/LSP.2015.2390417},
 journal = {IEEE Signal Processing Letters},
 month = {September},
 number = {9},
 pages = {1210--1214},
 pdf = {SPL15_Reinhardt.pdf},
 title = {Minimum Covariance Bounds for the Fusion under Unknown Correlations},
 url = {https://dx.doi.org/10.1109/LSP.2015.2390417},
 volume = {22},
 year = {2015}
}

@incollection{CRC15_Noack,
 abstract = {Distributed and decentralized processing and fusion of sensor data are becoming increasingly important. In view of the Internet of Things and the vision of ubiquitous sensing, designing and implementing multisensor state estimation algorithm have already become a key issue. A network of interconnected sensor devices is usually characterized by the idea to process and collect data locally and independently on the sensor nodes. However, this does not imply that the data are independent of each other, and the state estimation algorithms have to address possible interdependencies so as to avoid erroneous data fusion results.
Dependencies among local estimates generally can be traced back to common sensor information and common process noise. A wide variety of Kalman filtering schemes allow for the treatment of dependent data in centralized, distributed, and decentralized networks of sensor nodes, but making the right choice is itself dependent upon analyzing and weighing up the different advantages and disadvantages. This chapter discusses different strategies to identify and treat dependencies among Kalman filter estimates while pointing out advantages and challenges.},
 author = {Benjamin Noack and Joris Sijs and Marc Reinhardt and Uwe D. Hanebeck},
 booktitle = {Multisensor Data Fusion: From Algorithms and Architectural Design to Applications},
 editor = {Hassen Fourati},
 month = {August},
 pages = {169--192},
 publisher = {CRC Press},
 title = {Treatment of Dependent Information in Multisensor Kalman Filtering and Data Fusion},
 url = {https://doi.org/10.1201/b18851},
 year = {2015}
}

@inproceedings{Fusion15_Baum,
 abstract = {We consider the problem of clustering set-valued observations, i.e., each
observation is a set that consists of a finite number of real vectors. For this
purpose, we develop a $k$-means algorithm that employs the OSPA distance for
measuring the distance between sets. In particular, we introduce a novel
alternating optimization algorithm for the OSPA barycenter of sets with
varying cardinalities that is required for calculating cluster centroids
efficiently. The benefits of clustering set-valued data with respect to the OSPA
distance are illustrated by means of simulated experiments in the context of
target tracking and recognition.},
 address = {Washington D.C., USA},
 author = {Marcus Baum and Balakumar Balasingam and Peter Willett and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Baum.pdf},
 title = {OSPA Barycenters for Clustering Set-Valued Data},
 year = {2015}
}

@inproceedings{Fusion15_Chlebek,
 abstract = {We consider closed-loop feedback (CLF) stochastic model predictive control of
nonlinear time-invariant systems with imperfect state information. In this class of
control problems, future information feedback is considered in the decision making process,
and thus, the effect of the control influencing the state uncertainty is taken into account.
The main challenge in the solution is to find a good approximation to the arising stochastic
dynamic programming problem, which is computationally not tractable. In this work, future
information is considered in the form of conditional state probability densities. Thus,
the objective is it to optimize the state and its uncertainty as a combined problem.
We propose to discretize the state space by a novel scenario generation approach based on
deterministic sampling. A distance based threshold determines the narrowness of the discretization.
The dynamic programming problem is formulated such that the approximate cumulative control
cost function can be explicitly evaluated offline. The online calculation consists of a
one-step prediction and the interpolation of the explicit cost function in order to calculate
the control input. The effectiveness of this novel method is presented by means of a simulation.},
 address = {Washington D.C., USA},
 author = {Christof Chlebek and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Chlebek.pdf},
 title = {Stochastic Nonlinear Model Predictive Control Based on Deterministic Scenario Generation},
 year = {2015}
}

@inproceedings{Fusion15_Faion,
 abstract = {An extended object gives rise to several measurements that originate
from unknown measurement sources on the object. In this paper, we
consider the tracking and parameter estimation of extended objects that
are modeled as a curve in 2D such as a circle or an ellipse. A standard
model for such extended objects is to assume that the unknown
measurement sources are uniformly distributed on the curve. We argue
that the uniform distribution may not be the best choice in scenarios
where the true distribution of the measurements significantly differs
from a uniform distribution. Based on results from curve fitting and
errors-in-variables models, we develop a partial likelihood that ignores
the distribution of measurement sources and can be shown to outperform
the likelihood for a uniform distribution in these scenarios. If the
true measurement sources are in fact uniformly distributed, our new
likelihood results in a slightly slower convergence but has the same
asymptotic behavior.},
 address = {Washington D.C., USA},
 author = {Florian Faion and Antonio Zea and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Faion.pdf},
 title = {Partial Likelihood for Unbiased Extended Object Tracking},
 year = {2015}
}

@inproceedings{Fusion15_Gilitschenski,
 abstract = {We propose a novel measurement update procedure for orientation
estimation algorithms that are based on directional statistics. This
involves consideration of two scenarios, orientation estimation in the
2D plane and orientation estimation in three-dimensional space. We make
use of the von Mises distribution and the Bingham distribution in these
scenarios. In the derivation, we discuss directional counterparts to the
extended Kalman filter and a statistical-linearization-based filter. The
newly proposed algorithm makes use of deterministic sampling and can be
thought of as a directional variant of the measurement update that is
used in well-known sample-based algorithms such as the unscented Kalman
filter.},
 address = {Washington D.C., USA},
 author = {Igor Gilitschenski and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Gilitschenski.pdf},
 title = {Non-Identity Measurement Models for Orientation Estimation Based on Directional Statistics},
 year = {2015}
}

@inproceedings{Fusion15_Hanebeck-AssociationFreeTracking,
 abstract = {We consider association-free tracking of multiple targets without identities.
The uncertain multi-target state and the uncertain measurements cannot be described
by a random vector as this would imply a certain order. Instead, they are described by
an unordered random finite set (RFS). Particle-based random finite set densities are used
for characterizing the RFS in a simple and natural way. For recursive Bayesian filtering,
optimal multi-target state estimates are calculated by systematically minimizing an appropriate
set distance measure while directly operating on the particles. Although methods for calculating
point estimates of random finite set densities based on appropriate distance measures are
available in literature, the proposed recursive filtering is a novel contribution.},
 address = {Washington D.C., USA},
 author = {Uwe D. Hanebeck and Marcus Baum},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Hanebeck-AssociationFreeTracking.pdf},
 title = {Association-Free Direct Filtering of Multi-Target Random Finite Sets with Set Distance Measures},
 year = {2015}
}

@inproceedings{Fusion15_Hanebeck-Polytopes,
 abstract = {In this paper, we address the problem of probability mass
computation of a multivariate Gaussian contained within a polytope. This
computation requires an evaluation of a multivariate definite integral
of the Gaussian, whose solution is not tractable for higher dimensions
in a reasonable amount of time. Thus, research concentrates on the
derivation of approximate but sufficiently fast computation methods. We
propose a novel approach that approximates the underlying integration
domain, namely the polytope, using disjoint sectors such that the
probability mass contained within the sectors is maximized. In order to
derive our main algorithm, we first propose an approach to approximate
volume computation of a polytope using disjoint sectors. This solution
is then extended to the computation of the probability mass of a
Gaussian contained within the polytope. The presented solution provides
a lower bound on the true probability mass contained within the
polytope. Because the initial optimization problem is highly nonlinear,
we propose a greedy algorithm that splits the sectors with the highest
probability mass.},
 address = {Washington D.C., USA},
 author = {Uwe D. Hanebeck and Maxim Dolgov},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Hanebeck-Polytopes.pdf},
 title = {Adaptive Lower Bounds for Gaussian Measures of Polytopes},
 year = {2015}
}

@inproceedings{Fusion15_Kurz,
 abstract = {Robotic beating heart surgery requires accurate information about the
current state of the heart. For this purpose, it is of great importance
to have a good estimate of the heart's current phase, which in essence
corresponds to the percentage of the current heart cycle that has
already passed. Estimation of the heart phase is a highly nontrivial
problem as the heart motion is not exactly periodic. On the contrary, it
varies slightly from beat to beat and changes in frequency over time. In
order to derive a robust phase estimation algorithm, we rely on
directional statistics, a subfield of statistics that deals with
quantities that are inherently periodic, such as the phase of the
beating heart. The proposed methods are evaluated on a real data set and
shown to be superior to the state of the art.},
 address = {Washington D.C., USA},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Kurz.pdf},
 title = {Heart Phase Estimation Using Directional Statistics for Robotic Beating Heart Surgery},
 year = {2015}
}

@inproceedings{Fusion15_Noack,
 abstract = {A common approach to attack the simultaneous localization and mapping problem (SLAM)
is to consider factor-graph formulations of the underlying filtering and estimation setup.
While Kalman filter-based methods provide an estimate for the current pose of a robot and
all landmark positions, graph-based approaches take not only the current pose into account
but also the entire trajectory of the robot and have to solve a nonlinear least-squares
optimization problem. Using graph-based representations has proven to be highly scalable
and very accurate as compared with traditional filter-based approaches. However, biased
measurements as well as unmodeled correlations can lead to a sharp deterioration in the
estimation quality and hence require careful consideration. In this paper, a method to
incorporate biased or dependent measurement information is proposed that can easily be
integrated into existing optimization algorithms for graph-based SLAM. For biased sensor
data, techniques from ellipsoidal calculus are employed to compute the corresponding
information matrices. Dependencies among noise terms are treated by a generalization
of the covariance intersection concept. The treatment of both biased and correlated
sensor data rest upon the inflation of the involved error matrices. Simulations
are used to discuss and evaluate the proposed method.},
 address = {Washington D.C., USA},
 author = {Benjamin Noack and Simon J. Julier and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Noack.pdf},
 title = {Treatment of Biased and Dependent Sensor Data in Graph-based SLAM},
 year = {2015}
}

@inproceedings{Fusion15_Pfaff,
 abstract = {Recursive filtering with multimodal likelihoods and transition
densities on periodic manifolds is, despite the compact domain, still an
open problem. We propose a novel filter for the circular case that performs
well compared to other state-of-the-art filters adopted from linear
domains. The filter uses a limited number of Fourier coefficients of the
square root of the density. This representation is preserved throughout
filter and prediction steps and allows obtaining a valid density at any
point in time. Additionally, analytic formulae for calculating Fourier
coefficients of the square root of some common circular densities are
provided. In our evaluation, we show that this new filter performs well in
both unimodal and multimodal scenarios while requiring only a reasonable
number of coefficients.},
 address = {Washington D.C., USA},
 author = {Florian Pfaff and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Pfaff.pdf},
 title = {Multimodal Circular Filtering Using Fourier Series},
 year = {2015}
}

@inproceedings{Fusion15_Steinbring,
 abstract = {Since the last years, Graphics Processing Units (GPUs) have massive parallel execution capabilities even
for non-graphic related applications. The field of nonlinear state estimation is no exception here.
Particle Filters have already been successfully ported to GPUs. In this paper,
we propose a GPU-accelerated variant of the Progressive Gaussian Filter (PGF). This allows
us to combine the advantages of the particle flow with the ability to process thousands of
measurements at once in order to improve state estimation quality. To get a meaningful
comparison between its CPU and GPU variants, we additionally propose a likelihood for tracking
a sphere and its extent in 3D based on noisy point measurements. The likelihood considers the
physical relationship between sensor, measurement, and sphere to best exploit the information of the received measurements.
We evaluate the GPU implementation of the PGF using the proposed likelihood in combination with tens of thousands of measurements.
Although the CPU implementation fully exploits parallelization techniques such as SSE and OpenMP, the GPU-accelerated PGF
reaches speedups over 20 and real-time tracking can nearly be achieved.},
 address = {Washington D.C., USA},
 author = {Jannik Steinbring and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Steinbring.pdf},
 title = {GPU-Accelerated Progressive Gaussian Filtering with Applications to Extended Object Tracking},
 year = {2015}
}

@inproceedings{Fusion15_Zea,
 abstract = {Since the last years, Graphics Processing Units (GPUs) have massive parallel execution capabilities even
for non-graphic related applications. The field of nonlinear state estimation is no exception here.
Particle Filters have already been successfully ported to GPUs. In this paper,
we propose a GPU-accelerated variant of the Progressive Gaussian Filter (PGF). This allows
us to combine the advantages of the particle flow with the ability to process thousands of
measurements at once in order to improve state estimation quality. To get a meaningful
comparison between its CPU and GPU variants, we additionally propose a likelihood for tracking
a sphere and its extent in 3D based on noisy point measurements. The likelihood considers the
physical relationship between sensor, measurement, and sphere to best exploit the information of the received measurements.
We evaluate the GPU implementation of the PGF using the proposed likelihood in combination with tens of thousands of measurements.
Although the CPU implementation fully exploits parallelization techniques such as SSE and OpenMP, the GPU-accelerated PGF
reaches speedups over 20 and real-time tracking can nearly be achieved.},
 address = {Washington D.C., USA},
 author = {Antonio Zea and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th International Conference on Information Fusion (Fusion 2015)},
 month = {July},
 pdf = {Fusion15_Zea.pdf},
 title = {Exploiting Clutter: Negative Information for Enhanced Extended Object Tracking},
 year = {2015}
}

@inproceedings{ACC15_Dolgov,
 abstract = {In this paper, we consider infinite-horizon networked LQG control over multipurpose networks that do not provide acknowledgments
(UDP-like networks). The information communicated over the networks experiences transmission delays and losses that are modeled as stochastic
processes. In oder to mitigate the delays and losses in the controller-actuator channel, the controller transmits sequences of predicted control inputs
in addition to the current control input. To be able to mitigate delays and losses in the feedback channel, the estimator computes the estimate using
M last measurements. In this scenario, the separation principle does not hold and the optimal control law is in general nonlinear. However, assuming
that the controller and the estimator are linear and possess constant gains, we are able to derive the control law. The presented control law is
evaluated by means of simulations.},
 address = {Chicago, Illinois, USA},
 author = {Maxim Dolgov and Jörg Fischer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 American Control Conference (ACC 2015)},
 month = {July},
 pdf = {ACC2015_DolgovFischer.pdf},
 title = {Infinite-Horizon Sequence-based Networked Control without Acknowledgments},
 year = {2015}
}

@inproceedings{ACC15_Kurz,
 abstract = {In this paper, we present an open-loop Stochastic
Model Predictive Control (SMPC) method for discrete-time
nonlinear systems whose state is defined on the unit circle.
This modeling approach allows considering systems that include
periodicity in a more natural way than standard approaches
based on linear spaces. The main idea of this work is twofold:
(i) we model the quantities of the system, i.e., the state, the
measurements, and the noises, directly as circular quantities
described by circular probability densities, and (ii) we apply
deterministic sampling given in closed form to represent the
occurring densities. The latter allows us to make the prediction
required for solution of the SMPC problem tractable. We
evaluate the proposed control scheme by means of simulations.},
 address = {Chicago, Illinois, USA},
 author = {Gerhard Kurz and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2015 American Control Conference (ACC 2015)},
 month = {July},
 pdf = {ACC2015_KurzDolgov.pdf},
 title = {Nonlinear Stochastic Model Predictive Control in the Circular Domain},
 year = {2015}
}

@article{arXiv15_Dolgov,
 abstract = {In this paper, we consider stochastic optimal control of Markov Jump Linear Systems with state feedback 
but without observation of the jumping parameter. The proposed control law is assumed to be linear with constant gains 
that can be obtained from the necessary optimality conditions using an iterative algorithm. The proposed approach is 
demonstrated in a numerical example.},
 author = {Maxim Dolgov and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {July},
 title = {Infinite-horizon Linear Optimal Control of Markov Jump Systems without Mode Observation via State Feedback},
 url = {https://arxiv.org/abs/1507.00304},
 year = {2015}
}

@article{JAIF15_Faion,
 abstract = {In this work, we exploit geometric symmetries in extended
objects in order to improve Bayesian tracking algorithms that use
Spatial Distribution Models, Greedy Association Models as used in
curve fitting, and Random Hypersurface Models. The key idea is to
describe symmetric objects by solely modeling the non-redundant
part of the shape, while the remainder of the shape follows from
symmetry. Following this idea, we develop simplified versions for
the three models that take advantage of the symmetry. Exploiting
symmetries yields two major benefits. First, complex symmetric
shapes can be equivalently represented by a fraction of the original
shape parameters. Second, when using sample-based filters, such as
the widely used Unscented Kalman Filter, symmetry yields a higher
effective sample resolution. It is worth mentioning that estimating
even simple objects such as a stick, which only have one reflectional
symmetry, can be significantly improved.},
 author = {Florian Faion and Antonio Zea and Marcus Baum and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {June},
 number = {1},
 pages = {13--30},
 title = {Symmetries in Bayesian Extended Object Tracking},
 url = {https://confcats_isif.s3.amazonaws.com/web-files/journals/entries/433_1_art_9_24888.pdf},
 volume = {10},
 year = {2015}
}

@article{arXiv15_Steinbring,
 abstract = {Nonlinear Kalman Filters are powerful and widely-used 
techniques when trying to estimate the hidden state of a stochastic 
nonlinear dynamic system. In this paper, we extend the Smart Sampling 
Kalman Filter (S2KF) with a new point symmetric Gaussian sampling 
scheme. This not only improves the S2KF's estimation quality, but also 
reduces the time needed to compute the required optimal Gaussian samples 
drastically. Moreover, we improve the numerical stability of the sample 
computation, which allows us to accurately approximate a 
thousand-dimensional Gaussian distribution using tens of thousands of 
optimally placed samples. We evaluate the new symmetric S2KF by 
computing higher-order moments of standard normal distributions and 
investigate the estimation quality of the S2KF when dealing with 
symmetric measurement equations. Finally, extended object tracking based 
on many measurements per time step is considered. This high-dimensional 
estimation problem shows the advantage of the S2KF being able to use an 
arbitrary number of samples independent of the state dimension, in 
contrast to other state-of-the-art sample-based Kalman Filters.},
 author = {Jannik Steinbring and Martin Pander and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {June},
 title = {The Smart Sampling Kalman Filter with Symmetric Samples},
 url = {https://arxiv.org/abs/1506.03254},
 year = {2015}
}

@article{TSP15_Baum,
 abstract = {In multi-object estimation, the traditional minimum mean squared error (MMSE) objective
is unsuitable: a simple permutation of object identities can turn a very good estimate into
what is apparently a very bad one. Fortunately, a criterion tailored to sets-minimization of
the mean optimal sub-pattern assignment (MMOSPA)-has recently evolved. Aside from special cases,
exact MMOSPA estimates have seemed difficult to compute. But in this work we present the first
exact polynomial-time algorithms for calculating the MMOSPA estimate for probability densities
that are represented by particles. The key insight is that the MMOSPA estimate can be found by means
of enumerating the cells of a hyperplane arrangement, which is a traditional problem from computational
geometry. Although the runtime complexity is still high for the general case, efficient algorithms are
obtained for two special cases, i.e., (i) two targets with arbitrary state dimensions and (ii) an
arbitrary number of one-dimensional targets.},
 author = {Marcus Baum and Peter Willett and Uwe D. Hanebeck},
 doi = {10.1109/TSP.2015.2403292},
 journal = {IEEE Transactions on Signal Processing},
 month = {May},
 number = {10},
 pages = {2476 -- 2484},
 title = {Polynomial-Time Algorithms for the Exact MMOSPA Estimate of a Multi-Object Probability Density Represented by Particles},
 url = {https://ieeexplore.ieee.org/document/7041175/},
 year = {2015}
}

@article{AT15_Hanebeck,
 abstract = {This paper is concerned with the optimal approximation of a given multivariate Dirac mixture,
i.e., a density comprising weighted Dirac distributions on a continuous domain, by a Dirac mixture
with a reduced number of components. The parameters of the approximating density are calculated
by numerically minimizing a smooth distance measure, a generalization of the well-known
Cramer–von Mises-Distance to the multivariate case. This generalization is achieved by
defining an alternative to the classical cumulative distribution, the Localized Cumulative
Distribution (LCD), as a smooth characterization of discrete random quantities (on continuous domains).
The resulting approximation method provides the basis for various efficient nonlinear
estimation and control methods.},
 author = {Uwe D. Hanebeck},
 doi = {10.1515/auto-2015-0005},
 journal = {at -- Automatisierungstechnik, Oldenbourg Verlag},
 month = {April},
 number = {4},
 pages = {265--278},
 title = {Optimal Reduction of Multivariate Dirac Mixture Densities},
 url = {https://dx.doi.org/10.1515/auto-2015-0005},
 volume = {63},
 year = {2015}
}

@article{arXiv15_Kurz,
 abstract = {For recursive circular filtering based on circular statistics, we introduce 
a general framework for   estimation of a circular state based on different circular 
distributions, specifically the wrapped normal distribution and the von Mises distribution. 
We propose an estimation method for circular systems with nonlinear system and measurement 
functions. This is achieved by relying on efficient deterministic sampling techniques. 
Furthermore, we show how the calculations can be simplified in a variety of important special 
cases, such as systems with additive noise as well as identity system or measurement functions. 
We introduce several novel key components, particularly a distribution-free prediction algorithm, 
a new and superior formula for the multiplication of wrapped normal densities, and the ability 
to deal with non-additive system noise. All proposed methods are thoroughly evaluated and 
compared to several state-of-the-art solutions.},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {January},
 title = {Recursive Bayesian Filtering in Circular State Spaces},
 url = {https://arxiv.org/abs/1501.05151},
 year = {2015}
}

@article{JAIF14_Steinbring-S2KF,
 abstract = {An accurate Linear Regression Kalman Filter
(LRKF) for nonlinear systems called Smart Sampling Kalman
Filter (S²KF) is introduced. In order to get a better understanding
of this new filter, a general introduction to Nonlinear Kalman
Filters based on statistical linearization and LRKFs is given.
The S²KF is based on a new low-discrepancy Dirac mixture
approximation of Gaussian densities. This approximation com-
prises an arbitrary number of optimally and deterministically
placed samples in the relevant regions of the state space, so
that the filter resolution can be adapted to either achieve high-
quality results or to meet computational constraints. The S²KF
contains the UKF with equally weighted samples as a special case
when using the same amount of samples. With an increasing
number of samples, the new filter converges to the (typically
unfeasible) exact analytic statistical linearization. Hence, the
S²KF can be seen as the ultimate generalization of all LRKFs
such as the UKF, sigma-point filters, higher-order variants etc.,
as it homogeneously covers the state space with a freely chosen
number of samples. It is evaluated against state-of-the-art LRKFs
by performing nonlinear prediction and extended target tracking.},
 author = {Jannik Steinbring and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {December},
 number = {2},
 pages = {106--123},
 title = {LRKF Revisited: The Smart Sampling Kalman Filter (S2KF)},
 url = {https://confcats_isif.s3.amazonaws.com/web-files/journals/entries/441_1_art_11_17020.pdf},
 volume = {9},
 year = {2014}
}

@article{JAIF14_Kurz-Bingham,
 abstract = {This work considers filtering of uncertain data
defined on periodic domains, particularly the circle and the
manifold of orientations in 3D space. Filters based on the Kalman
filter perform poorly in this directional setting as they fail to take
the structure of the underlying manifold into account. We present
a recursive filter based on the Bingham distribution, which is
defined on the considered domains. The proposed filter can be
applied to circular filtering problems with 180 degree symmetry
and to estimation of orientations in three dimensional space.
It is easily implemented using standard numerical techniques
and suitable for real-time applications. We evaluate our filter
in a challenging scenario and compare it to a Kalman filtering
approach adapted to the particular setting.},
 author = {Gerhard Kurz and Igor Gilitschenski and Simon Julier and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {December},
 number = {2},
 pages = {90--105},
 pdf = {JAIF13_Kurz-Bingham.pdf},
 title = {Recursive Bingham Filter for Directional Estimation Involving 180 Degree Symmetry},
 volume = {9},
 year = {2014}
}

@inproceedings{CDC14_Baum,
 abstract = {Displaying tracks is an essential part of a multi-
target tracking system. Recently, it was proposed to extract
tracks with respect to the Optimal Sub-Pattern Assignment
(OSPA) metric, i.e., the traditionally used squared error loss
is replaced with an OSPA loss, which leads to the so-called
Minimum Mean OSPA (MMOSPA) estimate. So far, work
concentrated on traditional trackers that maintain probability
densities for the targets. In this paper, we aim at extracting
the MMOSPA estimate from a Probability Hypothesis Density
(PHD) as used within the PHD filter. We elaborate that the PHD
in general does not contain enough information to determine
the exact MMOSPA estimate. However, we then show that if
the loss function has a specific form, it is indeed possible to
extract point estimates from a PHD that are optimal w.r.t. the
underlying unknown random finite set. We discuss two specific
loss functions that fulfill this condition and are potentially close
to the OSPA loss, a nearest neighbor loss and a kernel distance
loss. It turns out that track extraction based on the nearest
neighbor loss can be performed with the well-known k-means
algorithm. Simulations show when the estimates based on the
nearest neighbor and the kernel loss are close to the MMOSPA
estimate.},
 address = {Los Angeles, California, USA},
 author = {Marcus Baum and Peter Willet and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 53rd IEEE Conference on Decision and Control (CDC 2014)},
 month = {December},
 pdf = {CDC14_Baum.pdf},
 title = {MMOSPA-based Track Extraction in the PHD Filter -- A Justification for k-Means Clustering},
 year = {2014}
}

@inproceedings{CDC14_Dolgov,
 abstract = {In this paper, we consider sequence-based LQG control
of stochastic linear systems with linear integral state and
input constraints over networks subject to stochastic packet
delays and losses. For this scenario, we derive a
novel closed-loop optimal control law that consists of a
feedback and a feedforward term. The feedback term depends
linearly on the state estimate, while the feedforward term
depends on the initial system state and the constraint functions.
The control law can be partially given in closed-form that
can be precomputed offline, and a numerical part which
demands a solution of a quadratic optimization procedure.
The number of the decision variables corresponds to the
number of constraints. The presented control law is
evaluated by means of a Monte-Carlo simulation.},
 address = {Los Angeles, California, USA},
 author = {Maxim Dolgov and Jörg Fischer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 53rd IEEE Conference on Decision and Control (CDC 2014)},
 month = {December},
 pdf = {CDC14_Dolgov.pdf},
 title = {Sequence-based LQG Control with Linear Integral Constraints over Stochastic Networks},
 year = {2014}
}

@inproceedings{CDC14_Kurz,
 abstract = {Estimation of angular quantities is a widespread
issue, but standard approaches neglect the true topology of the
problem and approximate directional with linear uncertainties.
In recent years, novel approaches based on directional statistics
have been proposed. However, these approaches have been
unable to consider arbitrary circular correlations between
multiple angles so far. For this reason, we propose a novel
recursive filtering scheme that is capable of estimating multiple
angles even if they are dependent, while correctly describing
their circular correlation. The proposed approach is based on
toroidal probability distributions and a circular correlation
coefficient. We demonstrate the superiority to a standard
approach based on the Kalman filter in simulations.},
 address = {Los Angeles, California, USA},
 author = {Gerhard Kurz and Igor Gilitschenski and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 53rd IEEE Conference on Decision and Control (CDC 2014)},
 month = {December},
 pdf = {CDC14_Kurz.pdf},
 title = {Bivariate Angular Estimation Under Consideration of Dependencies Using Directional Statistics},
 year = {2014}
}

@article{arXiv14_Hanebeck-Reduction,
 abstract = {This paper is concerned with the optimal approximation of a given 
multivariate Dirac mixture, i.e., a density comprising weighted Dirac 
distributions on a continuous domain, by an equally weighted Dirac mixture 
with a reduced number of components. The parameters of the approximating 
density are calculated by minimizing a smooth global distance measure, 
a generalization of the well-known Cramér-von Mises Distance to the 
multivariate case. This generalization is achieved by defining an alternative 
to the classical cumulative distribution, the Localized Cumulative Distribution (LCD), 
as a characterization of discrete random quantities (on continuous domains), 
which is unique and symmetric also in the multivariate case. The resulting 
approximation method provides the basis for various efficient nonlinear 
state and parameter estimation methods.},
 author = {Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {November},
 title = {Optimal Reduction of Multivariate Dirac Mixture Densities},
 url = {https://arxiv.org/abs/1411.4586},
 year = {2014}
}

@article{arXiv14_Baeuerle,
 abstract = {We consider a Hidden Markov Model (HMM) where the integrated 
continuous-time Markov chain can be observed at discrete time points 
perturbed by a Brownian motion. The aim is to derive a filter for the 
underlying continuous-time Markov chain. The recursion formula for the 
discrete-time filter is easy to derive, however involves densities which 
are very hard to obtain. In this paper we derive exact formulas for the 
necessary densities in the case the state space of the HMM consists of 
two elements only. This is done by relating the underlying integrated 
continuous-time Markov chain to the so-called asymmetric telegraph 
process and by using recent results on this process. In case the state 
space consists of more than two elements we present three different 
ways to approximate the densities for the filter. The first approach is 
based on the continuous filter problem. The second approach is to derive 
a PDE for the densities and solve it numerically and the third approach 
is a crude discrete time approximation of the Markov chain. All three 
approaches are compared in a numerical study.},
 author = {Nicole Bäuerle and Igor Gilitschenski and Uwe D. Hanebeck},
 journal = {arXiv preprint: Probability (math.PR)},
 month = {November},
 title = {Exact and Approximate Hidden Markov Chain Filters Based on Discrete Observations},
 url = {https://arxiv.org/abs/1411.0849},
 year = {2014}
}

@inproceedings{SDF14_Faion,
 abstract = {A popular approach when tracking extended objects
with elongated shapes, such as ships or airplanes, is to
approximate them as a line segment. Despite its simple shape,
the distribution of measurement sources on a line segment can
be characterized in many radically different ways. The spectrum
ranges from Spatial Distribution Models that assume a distinct
probability for each individual source, to Greedy Association
Models as used in curve fitting, which do not assume any
distribution at all. In between these border cases, Random
Hypersurface Models assume a distribution over subsets of all
sources. In this paper, we compare Bayesian estimators based
on these different models. We point out their advantages and
disadvantages and evaluate their performance by means of
illustrative examples with synthetic and real data using a Linear
Regression Kalman Filter.},
 address = {Bonn, Germany},
 author = {Florian Faion and Antonio Zea and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2014)},
 month = {October},
 pdf = {SDF14_Faion.pdf},
 title = {Bayesian Estimation of Line Segments},
 year = {2014}
}

@inproceedings{SDF14_Gilitschenski,
 abstract = {In this work, we propose a method for checking
whether two arbitrary-dimensional hyperellipsoids overlap without
making use of any optimization or root-finding methods.
This is achieved by formulating an overlap condition as a
polynomial root counting problem, which can be solved directly.
The addressed challenges involve the inversion of a polynomial
matrix using a direct method. The proposed approach extends one
of our earlier results, which was restricted to certain combinations
of ellipsoids and yields a fixed run-time for a fixed problem
dimensionality. Thus, for the first time, an algorithm for checking
overlap of arbitrary hyperellipsoids is proposed that can be
evaluated in closed form. That is, in the absence of cut-off errors,
the proposed method yields an exact result after a finite number
of steps.},
 address = {Bonn, Germany},
 author = {Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2014)},
 month = {October},
 pdf = {SDF14_Gilitschenski.pdf},
 title = {A Direct Method for Checking Overlap of Two Hyperellipsoids},
 year = {2014}
}

@inproceedings{SDF14_Kurz,
 abstract = {The wrapped normal distribution arises when the
density of a one-dimensional normal distribution is wrapped
around the circle infinitely many times. At first look, evaluation
of its probability density function appears tedious as an infinite
series is involved. In this paper, we investigate the evaluation
of two truncated series representations. As one representation
performs well for small uncertainties, whereas the other performs
well for large uncertainties, we show that in all cases a small
number of summands is sufficient to achieve high accuracy.},
 address = {Bonn, Germany},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2014)},
 month = {October},
 pdf = {SDF14_Kurz.pdf},
 title = {Efficient Evaluation of the Probability Density Function of a Wrapped Normal Distribution},
 year = {2014}
}

@inproceedings{SDF14_Zea,
 abstract = {As sensor resolution increases, the accuracy and robustness
of tracking algorithms can be improved by incorporating
more information about the shape of the target object. This raises
the need for simple and robust shape models capable of describing
detailed objects. In this paper we propose an approach based on
Random Hypersurface Models that interprets target shapes as
scaled extrusions. This is achieved by combining projection-based
models with probabilistic approaches, integrating the strengths
of both mechanisms. As extruded shapes such as bottles, boxes,
or containers can be extensively found in everyday situations,
this approach can be applied for tracking in a large variety of
environments.},
 address = {Bonn, Germany},
 annote = {Winner Best Paper Award Certificate (PDF)},
 author = {Antonio Zea and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2014)},
 month = {October},
 pdf = {SDF14_Zea.pdf},
 title = {Tracking Extended Objects using Extrusion Random Hypersurface Models},
 year = {2014}
}

@inproceedings{MFI14_Gilitschenski,
 abstract = {In this paper, we discuss computational efficiency
of a recursive estimator using the Bingham Distribution. The
Bingham distribution is defined directly on the unit hypersphere.
As such, it is able to describe both large and small uncertainties
in a unified framework. In order to tackle the challenging
computation of the normalization constant, we propose a method
using its Saddlepoint approximations and an approximate MLE
based on the Gauss-Newton method. In a set of simulation
experiments, we demonstrate that the Bingham filter not only
outperforms both Kalman and particle filters, but can also be
implemented efficiently.},
 address = {Beijing, China},
 author = {Igor Gilitschenski and Gerhard Kurz and Simon J. Julier and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2014 IEEE International Conference on Multisensor Fusion and Information Integration (MFI 2014)},
 month = {September},
 pdf = {MFI2014_Gilitschenski.pdf},
 title = {Efficient Bingham Filtering based on Saddlepoint Approximations},
 year = {2014}
}

@inproceedings{MFI14_Kurz,
 abstract = {We introduce a novel probability distribution on the
group of rigid motions SE(2) and we refer to this distribution
as the partially wrapped normal distribution. Describing probabilities
on the SE(2) is of interest in a wide range of areas, for
example robotics, autonomous vehicles, or information fusion. We
derive some important properties of this novel distribution and
derive an estimation scheme for its parameters based on moment
matching. Furthermore, we provide a comparison to a recently
published approach based on the Bingham distribution, and show
that there are complementary advantages and disadvantages of
the two approaches.},
 address = {Beijing, China},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2014 IEEE International Conference on Multisensor Fusion and Information Integration (MFI 2014)},
 month = {September},
 pdf = {MFI2014_Kurz-PWN.pdf},
 title = {The Partially Wrapped Normal Distribution for SE(2) Estimation},
 year = {2014}
}

@inproceedings{MFI14_Reinhardt,
 abstract = {In linear decentralized estimation, several nodes
concurrently aim to recursively estimate the state of a common
phenomenon by means of local measurements and data exchanges
subject to limited knowledge. In this contribution, an efficient
algorithm for consistent estimation in sensor networks under
Kalman filter assumptions is derived. The main theorems generalize
Covariance Intersection by means of an explicit consideration
of individual noise terms. We apply the results to linear
decentralized estimation and obtain covariance bounds with a
scalable precision between the exact covariances and the bounds
provided by Covariance Intersection subject to computation and
communication effort.},
 address = {Beijing, China},
 author = {Marc Reinhardt and Sanjeev Kulkarni and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2014 IEEE International Conference on Multisensor Fusion and Information Integration (MFI 2014)},
 month = {September},
 pdf = {MFI2014_Reinhardt.pdf},
 title = {Generalized Covariance Intersection based on Noise Decomposition},
 year = {2014}
}

@inproceedings{CURAC14_Kurz,
 abstract = {In cardiopulmonary bypass surgery, it is beneficial to avoid the use of a heart-lung machine and perform beating heart
surgery instead. This is a difficult task even for the most skilled surgeons. To eliminate the risks associated with cardio-
pulmonary bypass on a beating heart, a motion compensation system can be used. We place markers on the heart surface,
which we can use to track the complex heart motion and to produce still footage of the heart surface by applying one of
several stabilization algorithms to eliminate the motion. We compare six different stabilization algorithms, affine, B-
spline, piecewise linear and three types of radial basis functions. In this paper, we evaluate the results using three eval-
uation methods, pixel intensity average difference, optical flow, and stabilized marker tracking. All of these show a sig-
nificant reduction in motion after stabilization, especially for interpolation-based stabilization methods as opposed to the
affine approximation. We discuss advantages and disadvantages of the different evaluation methods.},
 address = {Munich, Germany},
 author = {Gerhard Kurz and Geneviève Foley and Péter Hegedüs and Gábor Szabó and Uwe D. Hanebeck},
 booktitle = {13. Jahrestagung der Deutschen Gesellschaft für Computer- und Roboterassistierte Chirurgie (CURAC14)},
 month = {September},
 pdf = {CURAC14_Kurz.pdf},
 title = {Evaluation of Image Stabilization Methods in Robotic Beating Heart Surgery},
 year = {2014}
}

@inproceedings{IFAC14_Fischer,
 abstract = {In networked control systems, sequence-based controllers are used to compensate
for transmission delays and losses in unreliable data networks. For this purpose, the controller
sends not only the current control input to the actuator but also a sequence of predicted control
inputs. The additional inputs can be used when subsequent transmissions get delayed or lost.
In this paper, the sequence-based method is applied to the problem of trajectory tracking over
an unreliable network and an optimal sequenced-based tracking controller is derived. The main
advantage of the presented approach is that future information on the reference trajectory can
optimally be embedded in the predicted control sequences. Furthermore, the controller can be
implemented offline. An interesting result is that the optimal controller can still be separated
into a feedback part and a feedforward part (as in standard optimal tracking control) despite of
both the unreliable network and the sequence-based method. The performance of the derived
tracking controller is demonstrated by Monte Carlo simulations with an inverted pendulum.},
 address = {Cape Town, South Africa},
 author = {Jörg Fischer and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th IFAC World Congress (IFAC 2014)},
 month = {August},
 pdf = {IFAC14_Fischer.pdf},
 title = {Optimal Sequence-Based Tracking Control over Unreliable Networks},
 year = {2014}
}

@inproceedings{IFAC14_Hanebeck,
 abstract = {Given a circular probability density function, called the true probability density
function, the goal is to find a Dirac mixture approximation based on some circular moments of
the true density. When keeping the locations of the Dirac points fixed, but almost arbitrarily
located, we are applying recent results on the circulant rational covariance extension problem
to the problem of calculating the weights. For the case of simultaneously calculating optimal
locations, additional constraints have to be deduced from the given density. For that purpose,
a distance measure for the deviation of the Dirac mixture approximation from the true density
is derived, which then is minimized while considering the moment conditions as constraints.
The method is based on progressive numerical minimization, converges quickly and gives well-
distributed Dirac mixtures that fulfill the constraints, i.e., have the desired circular moments.},
 address = {Cape Town, South Africa},
 author = {Uwe D. Hanebeck and Anders Lindquist},
 booktitle = {Proceedings of the 19th IFAC World Congress (IFAC 2014)},
 month = {August},
 pdf = {IFAC14_Hanebeck.pdf},
 title = {Moment-based Dirac Mixture Approximation of Circular Densities},
 year = {2014}
}

@inproceedings{IFAC14_Noack,
 abstract = {Distributed implementations of state estimation algorithms generally have in
common that each node in a networked system computes an estimate on the entire global
state. Accordingly, each node has to store and compute an estimate of the same state vector
irrespective of whether its sensors can only observe a small part of it. In particular, the task of
monitoring large-scale phenomena renders such distributed estimation approaches impractical
due to the sheer size of the corresponding state vector. In order to reduce the workload of the
nodes, the state vector to be estimated is subdivided into smaller, possibly overlapping parts.
In this situation, fusion does not only refer to the computation of an improved estimate but also
to the task of reassembling an estimate for the entire state from the locally computed estimates
of unequal state vectors. However, existing fusion methods require equal state representations
and, hence, cannot be employed. For that reason, a fusion strategy for estimates of unequal
and possibly overlapping state vectors is derived that minimizes the mean squared estimation
error. For the situation of unknown cross-correlations between local estimation errors, also a
conservative fusion strategy is proposed.},
 address = {Cape Town, South Africa},
 author = {Benjamin Noack and Joris Sijs and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 19th IFAC World Congress (IFAC 2014)},
 month = {August},
 pdf = {IFAC14_Noack.pdf},
 title = {Fusion Strategies for Unequal State Vectors in Distributed Kalman Filtering},
 year = {2014}
}

@article{arXiv14_Hanebeck,
 abstract = {We assume that a finite set of moments of a random vector is given. 
Its underlying density is unknown. An algorithm is proposed for efficiently 
calculating Dirac mixture densities maintaining these moments while providing 
a homogeneous coverage of the state space.},
 author = {Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {August},
 title = {Truncated Moment Problem for Dirac Mixture Densities with Entropy Regularization},
 url = {https://arxiv.org/abs/1408.7083},
 year = {2014}
}

@article{arXiv14_Munoz,
 abstract = {In this paper, a new synesthetic documentation method that contributes to 
media art conservation is presented. This new method is called e-Installation in analogy to the 
idea of the e-Book as the electronic version of a real book. An e-Installation is a virtualized 
media artwork that reproduces all synesthesia, interaction, and meaning levels of the artwork. 
Advanced 3D modeling and telepresence technologies with a very high level of immersion allow the virtual 
re-enactment of works of media art that are no longer performable or rarely exhibited. The virtual 
re-enactment of a media artwork can be designed with a scalable level of complexity depending on 
whether it addresses professionals such as curators, art restorers, and art theorists or the general public. 
An e-Installation is independent from the artwork's physical location and can be accessed via head-mounted 
display or similar data goggles, computer browser, or even mobile devices. In combination with informational and preventive 
conservation measures, the e-Installation offers an intermediate and long-term solution to archive, disseminate, 
and pass down the milestones of media art history as a synesthetic documentation when the original work can no 
longer be repaired or exhibited in its full function.},
 author = {Jesús Muñoz Morcillo and Florian Faion and Antonio Zea and Uwe D. Hanebeck and Caroline Y. Robertson-von Trotha},
 journal = {arXiv preprint: Other Computer Science (cs.OH)},
 month = {August},
 title = {e-Installation: Synesthetic Documentation of Media Art via Telepresence Technologies},
 url = {https://arxiv.org/abs/1408.1362v1},
 year = {2014}
}

@inproceedings{Fusion14_Ajgl,
 abstract = {The Covariance Intersection algorithm linearly
combines estimates when the cross-correlations between their
errors are unknown. It provides a fused estimate and an upper
bound of the corresponding mean square error matrix. The
weights of the linear combination are designed in order to
minimise the upper bound. This paper analyses the optimal
weights in relation to state estimation of dynamical systems.
It is shown that the use of the optimal upper bound in a
standard recursive filtering does not lead to optimal upper bounds
in subsequent processing steps. Unlike the fusion under full
knowledge, the fusion under unknown cross-correlations can fuse
the same information differently, depending on the independent
information that will be available in the future.},
 address = {Salamanca, Spain},
 author = {Jiří Ajgl and Miroslav Šimandl and Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Ajgl.pdf},
 title = {Covariance Intersection in State Estimation of Dynamical Systems},
 year = {2014}
}

@inproceedings{Fusion14_Duan,
 abstract = {In multi-sensor distributed estimation fusion, local
estimation errors are correlated in general. Two extreme ways to
handle this correlation is either to ignore them completely or to
have them fully considered. There is another case in the middle:
it admits the existence of the correlation, but does not know how
large it is. A sensible way is to set up an optimality criterion
and optimize it over all possible such correlations. This work
is a new development in the third class. First, a new general
objective function is introduced, which is the minimum sum
of statistical distances between the fused density and the local
posterior densities. Then it is shown that the new criterion leads
to a convex optimization problem if the Kullback-Leibler (KL)
divergence is used as the statistical distance between assumed
Gaussian densities. It is found that although the analytically
obtained fused estimate using the new criterion differs from
the simple convex combination rule only in mean squared error
(MSE) by a scaling factor N (the number of sensors used), it is
pessimistic semi-definite in MSE. Numerical examples illustrate
the effectiveness of the proposed distributed fuser by comparing
with several widely used distributed fusers.},
 address = {Salamanca, Spain},
 author = {Zhansheng Duan and Xiao-Rong Li and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Duan.pdf},
 title = {Multi-sensor Distributed Estimation Fusion Using Minimum Distance Sum},
 year = {2014}
}

@inproceedings{Fusion14_Chlebek,
 abstract = {In this work, we derive a distance measure for
the detection of changes in the behavior of linear dynamic
single-input-single-output (SISO) systems based on input-output
data. The distance is calculated as a function of the system
poles, which are directly estimated from the given data. Poles
represent a system as a set and have no identities, which is
analogous to the nature of association-free multi-target tracking.
This motivates the application of set distances known from
multi-target tracking, namely the optimal subpattern assign-
ment (OSPA) distance. Thus, the OSPA distance as well as a
modification, the MAX-OSPA distance, are formulated as pole-
distances between dynamic systems. In this formulation, the
OSPA distance finds the optimal assingment by minimizing over
the sum of distances between poles. The MAX-OSPA chooses
an optimal assignment by minimizing the maximum distance
between two poles. The proposed distances are evaluated in
several simulations comparing the deterministic OSPA and MAX-
OSPA to a state-of-the-art metric for autoregressive-moving-
average (ARMA) processes, as well as OSPA and MAX-OSPA
using the direct pole estimation and a two step-pole estimation
utilizing recursive ARX (AutoRegressive model with eXogenous
input) system identification.},
 address = {Salamanca, Spain},
 author = {Christof Chlebek and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Chlebek.pdf},
 title = {Pole-based Distance Measure for Change Detection in Linear Dynamic Systems},
 year = {2014}
}

@inproceedings{Fusion14_Faion,
 abstract = {This work considers the problem of estimating the
parameters of an extended object based on noisy point observations
from its boundary. The intention is to explore relationships
between common approaches by breaking them down into their
basic assumptions within the Bayesian framework. In doing so,
we find that distance-minimizing curve fitting algorithms can be
modeled by using a special Spatial Distribution Model, where
the source distribution is approximated by a greedy one-to-one
association of points to sources on the shape boundary. Based on
this insight, we explore the origin of the estimation bias, which is
a well-known issue of curve fitting algorithms. Furthermore, we
derive a general scheme to alleviate its effect for arbitrary shapes,
as well as for non-isotropic noise. This procedure is shown to be
a generalization of related special solutions.},
 address = {Salamanca, Spain},
 author = {Florian Faion and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Faion.pdf},
 title = {Reducing Bias in Bayesian Shape Estimation},
 year = {2014}
}

@inproceedings{Fusion14_GilitschenskiKurz,
 abstract = {This work proposes a novel way to represent uncertainty
on the Lie group of rigid-body motions in the plane. This is
achieved by using dual quaternions for representation of a planar
rigid-body motion and proposing a probability distribution from
the exponential family of distributions that inherently respects the
underlying structure of the representation. This is particularly
beneficial in scenarios involving strong measurement noise. A
relationship between the newly proposed distributional model
and the Bingham distribution is discussed. The presented results
involve formulas for computation of the normalization constant,
the mode, parameter estimation techniques, and a closed-form
Bayesian measurement fusion.},
 address = {Salamanca, Spain},
 annote = {Best Student Paper Award First Runner-Up Certificate (PDF)},
 author = {Igor Gilitschenski and Gerhard Kurz and Simon J. Julier and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_GilitschenskiKurz.pdf},
 title = {A New Probability Distribution for Simultaneous Representation of Uncertain Position and Orientation},
 year = {2014}
}

@inproceedings{Fusion14_GilitschenskiSteinbring,
 abstract = {In this work, we propose a novel way to approximating
mixtures of Gaussian distributions by a set of deterministically
chosen Dirac delta components. This approximation is performed
by adapting a method for approximating single Gaussian
distributions to the considered case. The proposed method turns
the approximation problem into an optimization problem by
minimizing a distance measure between the Gaussian mixture
and its Dirac mixture approximation. Compared to the simple
Gaussian case, the minimization criterion is much more complex
as multiple, non-standard Gaussian distributions have to be considered.},
 address = {Salamanca, Spain},
 author = {Igor Gilitschenski and Jannik Steinbring and Uwe D. Hanebeck and Miroslav Simandl},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_GilitschenskiSteinbring.pdf},
 title = {Deterministic Dirac Mixture Approximation of Gaussian Mixtures},
 year = {2014}
}

@inproceedings{Fusion14_Hanebeck,
 abstract = {For designing sample sets for nonlinear Kalman
filters, i.e., Linear Regression Kalman Filters (LRKFs), a new
method is introduced for approximating Gaussian densities by
discrete densities, so called Dirac Mixtures (DMs). This
approximating DM should maintain the mean and some higher-
order moments and should homogeneously cover the support
of the original density. Homogeneous approximations require
redundancy, which means there are more Dirac components than
necessary for fulfilling the moment constraints. Hence, some
sort of regularization is required as the solution is no longer
unique. Two types of regularizers are possible: The first type
ensures smooth approximations, e.g., in a maximum entropy
sense. The second type we pursue here ensures closeness of
the approximating density to the given Gaussian. As standard
distance measures are typically not well defined for discrete
densities on continuous domains, we focus on shifting the mass
distribution of the approximating density as close to the true
density as possible. Instead of globally comparing the masses as
in a previous paper, the key idea is to characterize individual
Dirac components by kernel functions representing the spread
of probability mass that is appropriate at a given location. A
distance measure is then obtained by comparing the deviation
between the true density and the induced kernel density. As a
result, the approximation problem is converted to an optimization
problem as we now minimize the distance under the desired
moment constraints.},
 address = {Salamanca, Spain},
 author = {Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Hanebeck.pdf},
 title = {Sample Set Design for Nonlinear Kalman Filters viewed as a Moment Problem},
 year = {2014}
}

@inproceedings{Fusion14_Sijs,
 abstract = {To reduce the amount of data transfer in networked
systems measurements are usually taken only when an event
occurs rather than periodically in time. However, a fundamental
assessment on the response of estimation algorithms receiving
event sampled measurements is not available. This research
presents such an analysis when new measurements are sampled
at well-designed events and sent to a Luenberger observer.
Conditions are then derived under which the estimation error
is bounded, followed by an assessment of two event sampling
strategies when the estimator encounters two different types of
disturbances: an impulse and a step disturbance. The sampling
strategies are compared via four performance measures, such as
estimation-error and communication resources. The result is a
clear insight of the estimation response in an event-based setup.},
 address = {Salamanca, Spain},
 author = {Joris Sijs and Leon Kester and Benjamin Noack},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Sijs.pdf},
 title = {A Study on Event Triggering Criteria for Estimation},
 year = {2014}
}

@inproceedings{Fusion14_KurzGilitschenski,
 abstract = {Circular estimation problems arise in many applications and can be
addressed with the help of circular distributions. In particular, the
wrapped normal and von Mises distributions are widely used in the
context of circular problems. To facilitate the development of nonlinear
filters, a deterministic sample-based approximation of these
distributions with a so-called wrapped Dirac mixture distribution is
beneficial. We propose a new closed-form solution to obtain a symmetric
wrapped Dirac mixture with five components based on matching the first
two circular moments. The proposed method is superior to
state-of-the-art methods, which only use the first circular moment to
obtain three Dirac components, because a larger number of Dirac
components results in a more accurate approximation.},
 address = {Salamanca, Spain},
 annote = {Winner Best Paper Award Certificate (PDF)},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_KurzGilitschenski.pdf},
 title = {Deterministic Approximation of Circular Densities with Symmetric Dirac Mixtures Based on Two Circular Moments},
 year = {2014}
}

@inproceedings{Fusion14_Kurz,
 abstract = {Image stabilization is relevant for various industrial and medical
applications. In particular, we consider the use of image stabilization
in robotic beating heart surgery. A robot, which is remotely controlled
by the surgeon, can automatically compensate for the motion of the
beating heart. To give the surgeon the illusion of operating on a
stationary heart, a stabilized image of the beating heart is shown to
the surgeon. Image stabilization cancels the unwanted motion of the
heart, but retains changes to color and texture, for example cuts on the
heart surface. In this paper, stabilization is first considered as a 2D
image transformation problem. Subsequently, it is extended to
stabilization of a 3D point cloud or surface. The proposed algorithms
are evaluated in both ex-vivo and in-vivo experiments. In the
evaluation, the stabilization quality achievable with several common
interpolation functions is compared.},
 address = {Salamanca, Spain},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Kurz.pdf},
 title = {2D and 3D Image Stabilization for Robotic Beating Heart Surgery},
 year = {2014}
}

@inproceedings{Fusion14_Noack,
 abstract = {The problem of fusing state estimates is encountered in many network-based multi-sensor applications.
The majority of distributed state estimation algorithms are designed to provide multiple estimates on the
same state, and track-to-track fusion then refers to the task of combining these estimates. While linear
fusion only requires the joint cross-covariance matrix to be known, dependencies between estimates in
nonlinear estimation problems have to be represented by high-dimensional probability density functions.
In general, storing and keeping track of nonlinear dependencies is too cumbersome. However, this paper
demonstrates that estimates represented by Gaussian mixtures prove to be an important exception to this rule.
The dependency structure can as well be characterized in terms of a higher-dimensional Gaussian mixture.
The different processing steps of distributed nonlinear state estimation, i.e., prediction, filtering,
and fusion, are studied in light of the joint density representation. The presented concept is complemented
with different simpler suboptimal representations of the dependency structure between Gaussian mixture densities.},
 address = {Salamanca, Spain},
 author = {Benjamin Noack and Marc Reinhardt and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Noack.pdf},
 title = {On Nonlinear Track-to-track Fusion with Gaussian Mixtures},
 year = {2014}
}

@inproceedings{Fusion14_Reinhardt,
 abstract = {Distributed Kalman filtering aims at optimizing an
estimate at a fusion center based on information that is gathered
in a sensor network. Recently, an exact solution based on local
estimation tracks has been proposed and an extension to cope
with packet losses has been derived. In this contribution, we
generalize both algorithms to packet delays. The key idea is
to introduce augmented measurement vectors in the sensors
that permit the optimization of local filter gains according to
time-dependent measurement capabilities at the fusion center.
In the most general form, the algorithm provides optimized
estimates in sensor networks with packets delays and losses. The
precision depends on the actual arrival patterns, and the results
correspond to those of the centralized Kalman filter when specific
assumptions about the measurement capability are satisfied.},
 address = {Salamanca, Spain},
 author = {Marc Reinhardt and Benjamin Noack and Sanjeev Kulkarni and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Reinhardt.pdf},
 title = {Distributed Kalman Filtering in the Presence of Packet Delays and Losses},
 year = {2014}
}

@inproceedings{Fusion14_Steinbring,
 abstract = {In this paper, we introduce a new sample-based
Gaussian filter. In contrast to the popular Nonlinear Kalman
Filters, e.g., the UKF, we do not rely on linearizing the
measurement model. Instead, we take up the Gaussian progressive
filtering approach introduced by the PGF 42 but explicitly rely
on likelihood functions. Progression means, we incorporate the
information of a new measurement gradually into the state
estimate. The advantages of this filtering method are on the one
hand the avoidance of sample degeneration and on the other
hand an adaptive determination of the number of likelihood
evaluations required for each measurement update. By this
means, less informative measurements can be processed quickly,
whereas measurements containing much information automatically
receive more emphasis by the filter. These properties allow
the new filter to cope with the demanding problem of very narrow
likelihood functions in an efficient way.},
 address = {Salamanca, Spain},
 author = {Jannik Steinbring and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Steinbring.pdf},
 title = {Progressive Gaussian Filtering Using Explicit Likelihoods},
 year = {2014}
}

@inproceedings{Fusion14_Zea,
 abstract = {As sensor resolution increases, estimators tracking
extended objects benefit from being able to closely model the
shape of the target. However, as more shape details are
incorporated, this usually leads to increasingly complex estimators.
A more useful approach is to describe these shapes as a combination
of simpler shapes connected to each other. In this paper, we
propose a modular approach to estimate these combined targets
in function of their simpler components. This allows the
characteristics of each component to be encapsulated, and permits
the combination of multiple filtering techniques as required by
each component shape. This approach can be applied to track
combined objects in a large variety of environments, such as
excavators, robotic arms, wagon trains, and many others.},
 address = {Salamanca, Spain},
 author = {Antonio Zea and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 17th International Conference on Information Fusion (Fusion 2014)},
 month = {July},
 pdf = {Fusion14_Zea.pdf},
 title = {Tracking Connected Objects Using Interacting Shape Models},
 year = {2014}
}

@inproceedings{ECC14_Chlebek,
 abstract = {In this work, a solution to the direct pole identification problem
of discrete-time autoregressive (AR) processes by general recursive
Bayesian estimation is presented. The poles of the transfer function
of an AR process are identified directly from the process output
data. Without intermediate estimation of the AR coefficient, the
AR process identification problem by means of its poles becomes nonlinear,
and thus cannot be solved exactly. A practical solution by application
of statistical linearization is given. The derived direct pole estimation
algorithm by statistical linearization is given in closed-form and
regression point based, by the so-called Linear Regression Kalman
Filter (LRKF). Two realizations of the LRKF algorithm are tested,
namely the Unscented Kalman Filter (UKF) for low computational complexity
and thus, for high update rates, and the Smart Sampling Kalman Filter
(S2KF) for high precision with faster convergence. Both, the UKF
and S2KF are compared to the Adaptive Pole Estimation (APE), a solution
by recursive nonlinear least squares minimizing the prediction error
gradient.},
 address = {Strasbourg, France},
 author = {Christof Chlebek and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2014 European Control Conference (ECC 2014)},
 month = {June},
 pdf = {ECC14_Chlebek.pdf},
 title = {Bayesian Approach to Direct Pole Estimation},
 year = {2014}
}

@inproceedings{SAM14_Baum,
 abstract = {This work is concerned with direction-of-arrival
(DOA) estimation of narrowband signals from multiple targets
using a planar antenna array. We illustrate the shortcomings of
Maximum Likelihood (ML), Maximum a Posteriori (MAP), and
Minimum Mean Squared Error (MMSE) estimation, issues that
can be attributed to the symmetry in the likelihood function that
must exist when there is no information about labeling of targets.
We proffer the recently introduced concept of Minimum Mean
OSPA (MMOSPA) estimation that is based on the optimal subpattern
assignment (OSPA) metric for sets and hence inherently
incorporates symmetric likelihood functions.},
 address = {A Coruña, Spain},
 author = {Marcus Baum and Peter Willett and Uwe D. Hanebeck},
 booktitle = {Proceedings of the Eighth IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM 2014)},
 month = {June},
 pdf = {SAM14_Baum.pdf},
 title = {MMOSPA-based Direction-of-Arrival Estimation for Planar Antenna Arrays},
 year = {2014}
}

@inproceedings{SAM14_Kurz,
 abstract = {Performing surgery on the beating heart has significant advantages for
the patient compared to traditional heart surgery on the stopped heart.
A remote-controlled robot can be used to automatically cancel out the
movement of the beating heart. This necessitates precise tracking of the
heart surface. For this purpose, we track 24 identical artificial
markers placed on the heart. This creates a data association problem,
because it is not known which measurement was obtained from which
marker. To solve this problem, we apply a multiple target tracking
method based on a symmetric kernel transformation. This method allows
efficient handling of the data association problem even for a reasonably
large number of targets. We demonstrate how to implement this method
efficiently. The proposed approach is evaluated on in-vivo data of a
real beating heart surgery performed on a porcine beating heart.},
 address = {A Coruña, Spain},
 author = {Gerhard Kurz and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the Eighth IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM 2014)},
 month = {June},
 pdf = {SAM14_Kurz.pdf},
 title = {Real-time Kernel-based Multiple Target Tracking for Robotic Beating Heart Surgery},
 year = {2014}
}

@inproceedings{SAM14_Zea,
 abstract = {When tracking extended objects, it is often the case that the shape of the
target cannot be fully observed due to issues of visibility, artifacts, or
high noise, which can change with time.
In these situations, it is a common approach to model targets as simpler
shapes instead, such as ellipsoids or cylinders.
However, these simplifications cause information loss from the original
shape, which could be used to improve the estimation results.
In this paper, we propose a way to recover information from these lost
details in the form of a stochastic boundary, whose parameters can be
dynamically estimated from received measurements.
The benefits of this approach are evaluated by tracking an object using
noisy, real-life RGBD data.},
 address = {A Coruña, Spain},
 author = {Antonio Zea and Florian Faion and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the Eighth IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM 2014)},
 month = {June},
 pdf = {SAM14_Zea.pdf},
 title = {Tracking Simplified Shapes Using a Stochastic Boundary},
 year = {2014}
}

@inproceedings{ACC14_Dolgov,
 abstract = {In this paper, we consider state estimation for Stochastic Hybrid
Systems (SHS). These are systems that possess both continuous-valued and
discrete-valued dynamics. For SHS with nonlinear hybrid dynamics and/or
non-Gaussian disturbances, state estimation can be implemented as an
Interacting Multiple Model (IMM) particle filter. However, a
disadvantage of particle filtering is the computational load caused by
the large number of particles required for a sufficiently good
estimation. We address this issue by first expressing the probability
density that describes the state of the SHS as a collection of densities
of the continuous-valued state only conditioned on the discrete-valued
state. Then, we deterministically approximate these individual densities
with Dirac mixtures. The employed approximation method places the
particles so that a so called modified Cramér-von Mises distance
between the true and the approximated density is minimized.
Deterministic approximation requires far less particles than the
stochastic sampling used by particle filters. To avoid particle
degeneration that can occur when a density is multiplied with the
likelihood, the filter uses progressive density correction. The
presented filter is demonstrated in a numerical maneuvering target
tracking example.},
 address = {Portland, Oregon, USA},
 author = {Maxim Dolgov and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2014 American Control Conference (ACC 2014)},
 month = {June},
 pdf = {ACC14_Dolgov.pdf},
 title = {State Estimation for Stochastic Hybrid Systems Based on Deterministic Dirac Mixture Approximation},
 year = {2014}
}

@inproceedings{ACC14_Kurz,
 abstract = {In this paper, we propose a novel progressive nonlinear measurement
update for circular states. This generalizes our previously published
circular filter that so far was limited to identity measurement
equations. The new update method is based on circular distributions in
order to capture the periodic properties of a circular system better
than conventional approaches that rely on standard Gaussian
distributions. Besides the progressive measurement update, we propose
two additional measurement updates that are obtained by adapting
traditional filters to the circular case. Simulations show the
superiority of the proposed progressive approach.},
 address = {Portland, Oregon, USA},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2014 American Control Conference (ACC 2014)},
 month = {June},
 pdf = {ACC14_Kurz.pdf},
 title = {Nonlinear Measurement Update for Estimation of Angular Systems Based on Circular Distributions},
 year = {2014}
}

@article{JAIF14_Kurz,
 abstract = {Surface estimation can be performed based on position or depth measurements.
We propose a method to fuse both types of measurements. Position measurements are obtained
from landmarks on the surface, i.e., they are fixed to a certain point on the surface. In
contrast, depth measurements reflect the depth measured along a line emanating from a depth
camera and are not fixed to a position on the surface. The proposed approach uses a mixture
of Cartesian and polar or spherical coordinate to treat both measurement types accordingly.
By doing so, the uncertainties associated with the different measurement types are explicitly
considered. The presented method represents the surface by a spline and is applicable to both
2D and 3D applications. Surface estimation is considered as a recursive filtering problem and
standard nonlinear filtering methods suchas the unscented Kalman filter can be used to obtain
surface estimates. We show a thorough evaluation of the proposed approach in simulations.},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 journal = {Journal of Advances in Information Fusion},
 month = {June},
 number = {1},
 pages = {13--26},
 pdf = {JAIF14_Kurz.pdf},
 title = {Dynamic Surface Reconstruction by Recursive Fusion of Depth and Position Measurements},
 url = {https://confcats_isif.s3.amazonaws.com/web-files/journals/entries/427_1_art_6_19271%5B1%5D.pdf},
 volume = {9},
 year = {2014}
}

@inproceedings{SPIE14_Baum,
 abstract = {The calculation of marginal association probabilities is
the major computational bottleneck in the Joint Probabilistic
Data Association Filter (JPDAF). In this paper, we investigate
approximations for the marginal associations that simplify the
(computational complex) original association model in order to
obtain efficient algorithms. In this context, we first discuss
the Bakhtiar-Alavi algorithm and the Linear Multitarget Integrated
Probabilistic Data Association (LMIPDA) algorithm. Second, we
propose a fast novel approximation that exploits systematic
combinations of the JPDAF measurement model with the
Probabilistic Multi-Hypothesis Tracker (PMHT) measurement model.
The discussed methods are evaluated by means of a tracking
scenario with a high number of closely-spaced targets.},
 address = {Baltimore, Maryland, USA},
 author = {Marcus Baum and Peter Willet and Yaakov Bar-Shalom and Uwe D. Hanebeck},
 booktitle = {SPIE -- Signal and Data Processing of Small Targets 2014},
 doi = {10.1117/12.2053431},
 month = {June},
 title = {Approximate Calculation of Marginal Association Probabilities using a Hybrid Data Association Model},
 url = {https://dx.doi.org/10.1117/12.2053431},
 year = {2014}
}

@article{arXiv14_Kurz,
 abstract = {The wrapped normal distribution arises when a the density of a 
one-dimensional normal distribution is wrapped around the circle infinitely 
many times. At first look, evaluation of its probability density function appears 
tedious as an infinite series is involved. In this paper, we investigate the 
evaluation of two truncated series representations. As one representation performs 
well for small uncertainties whereas the other performs well for large uncertainties, 
we show that in all cases a small number of summands is sufficient to achieve high accuracy.},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 journal = {arXiv preprint: Computation (stat.CO)},
 month = {May},
 title = {Efficient Evaluation of the Probability Density Function of a Wrapped Normal Distribution},
 url = {https://arxiv.org/abs/1405.6397},
 year = {2014}
}

@inproceedings{CISS14_Reinhardt,
 abstract = {In this paper, a sample representation of the estimation error is
utilized to reconstruct the joint covariance matrix in a distributed estimation
system. The key idea is to sample uncorrelated and fully correlated noise
according to different techniques at local estimators without knowledge about
the processing of other nodes in the network. This way, the correlation between
estimates is inherently linked to the representation of the corresponding
sample sets. We discuss the noise processing, derive key attributes, and
evaluate the precision of the covariance estimates.},
 address = {Princeton, New Jersey, USA},
 author = {Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 48th Annual Conference on Information Sciences and Systems (CISS 2014)},
 month = {March},
 pdf = {CISS14_Reinhardt.pdf},
 title = {Reconstruction of Joint Covariance Matrices in Networked Linear Systems},
 year = {2014}
}

@inproceedings{CISS14_Hanebeck,
 abstract = {This paper provides an efficient method for approximating a given
continuous probability density function (pdf) by a Dirac mixture density. Optimal
parameters are determined by systematically minimizing a distance measure. As
standard distance measures are typically not well defined for discrete densities
on continuous domains, we focus on shifting the mass distribution of the
approximating density as close to the true density as possible. Instead of globally
comparing the masses as in a previous paper, the key idea is to characterize
individual Dirac components by kernel functions representing the spread of
probability mass that is appropriate at a given location. A distance measure is
then obtained by comparing the deviation between the true density and the induced
kernel density. This new method for Dirac mixture approximation provides
high-quality approximation results, can handle arbitrary pdfs, allows considering
constraints for, e.g., maintaining certain moments, and is fast enough for online
processing.},
 address = {Princeton, New Jersey, USA},
 author = {Uwe D. Hanebeck},
 booktitle = {Proceedings of the 48th Annual Conference on Information Sciences and Systems (CISS 2014)},
 month = {March},
 pdf = {CISS14_Hanebeck.pdf},
 title = {Kernel-based Deterministic Blue-noise Sampling of Arbitrary Probability Density Functions},
 year = {2014}
}

@incollection{SPP14_Fischer,
 abstract = {In this chapter, we consider the problem of controlling networked
and distributed systems by means of model predictive control (MPC).
The basic idea behind MPC is to repeatedly solve an optimal control
problem based on a model of the system to be controlled. Every time
a new measurement is available, the optimization problem is solved
and the corresponding input sequence is applied until a new measurement
arrives. As explained in the sequel, the advantages of MPC over other
control strategies for networked systems are due to the fact that
a model of the system is available at the controller side, which
can be used to compensate for random bounded delays. At the same
time, for each iteration of the optimization problem an optimal input
sequence is calculated. In case of packet dropouts, one can reuse
this information to maintain closed-loop stability and performance.},
 author = {Jörg Fischer and Uwe D. Hanebeck},
 booktitle = {Control Theory of Digitally Networked Dynamic Systems},
 doi = {10.1007/978-3-319-01131-8_4},
 editor = {Jan Lunze},
 isbn = {978-3-319-01130-1},
 month = {January},
 pages = {111--167},
 pdf = {SPP14_Fischer.pdf},
 publisher = {Springer International Publishing},
 title = {Distributed and Networked Model Predictive Control},
 url = {https://dx.doi.org/10.1007/978-3-319-01131-8_4},
 year = {2014}
}

@article{TAES13_Baum_RHM,
 author = {Marcus Baum and Uwe D. Hanebeck},
 doi = {10.1109/TAES.2013.120107},
 journal = {IEEE Transactions on Aerospace and Electronic Systems},
 month = {January},
 pages = {149-159},
 pdf = {TAES13_Baum.pdf},
 title = {Extended Object Tracking with Random Hypersurface Models},
 volume = {50},
 year = {2014}
}

@inproceedings{CDC13_Fischer,
 abstract = {Sequence-based control is a well-established method applied in
Networked Control Systems (NCS) to mitigate the effect of time-varying
transmission delays and stochastic packet losses. The idea of this method
is that the controller sends sequences of predicted control inputs to the
actuator that can be applied in case a future transmission fails. In this paper,
the stability properties of sequence-based LQG controllers are analyzed
in terms of the boundedness of the long run average costs. On the one hand,
we derive sufficient conditions, each for the boundedness and unboundedness
of the costs. On the other hand, we give bounds on the minimal length of
the control input sequence needed to stabilize a system.},
 address = {Florence, Italy},
 author = {Jörg Fischer and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 52st IEEE Conference on Decision and Control (CDC 2013)},
 month = {December},
 pdf = {CDC13_Fischer.pdf},
 title = {On Stability of Sequence-Based LQG Control},
 year = {2013}
}

@inproceedings{CURAC13_Kurz,
 abstract = {This paper investigates the use of Kinect depth sensors as well as
inertial sensors in the context of beating heart surgery. In the past,
various sensors have been used in attempts to track the beating heart,
each with its own distinct set of advantages and disadvantages. With the
availability of affordable structured-light depth sensors such as the
Kinect and sufficiently small and low priced inertial sensors, the
question of their suitability for beating heart tracking arises. We
performed in-vivo experiments on a porcine heart in order to assess the
feasibility of beating heart tracking based on these sensors.},
 address = {Innsbruck, Austria},
 author = {Gerhard Kurz and Peter Hegedus and Gábor Szabó and Uwe D. Hanebeck},
 booktitle = {12. Jahrestagung der Deutschen Gesellschaft für Computer- und Roboterassistierte Chirurgie (CURAC13)},
 month = {November},
 pdf = {CURAC13_Kurz.pdf},
 title = {Experimental Evaluation of Kinect and Inertial Sensors for Beating Heart Tracking},
 year = {2013}
}

@article{arXiv13_Gilitschenski,
 abstract = {Orientation estimation for 3D objects is a common problem that is
usually tackled with traditional nonlinear filtering techniques such
as the extended Kalman filter (EKF) or the unscented Kalman filter
(UKF). Most of these techniques assume Gaussian distributions to
account for system noise and uncertain measurements. This distributional
assumption does not consider the periodic nature of pose and orientation
uncertainty. We propose a filter that considers the periodicity of
the orientation estimation problem in its distributional assumption.
This is achieved by making use of the Bingham distribution, which
is defined on the hypersphere and thus inherently more suitable to
periodic problems. Furthermore, handling of non-trivial system functions
is done using deterministic sampling in an efficient way. A deterministic
sampling scheme reminiscent of the UKF is proposed for the nonlinear
manifold of orientations. It is the first deterministic sampling
scheme that truly reflects the nonlinear manifold of the orientation.},
 author = {Igor Gilitschenski and Gerhard Kurz and Simon J. Julier and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {November},
 title = {Unscented Orientation Estimation Based on the Bingham Distribution},
 url = {https://arxiv.org/abs/1311.5796},
 year = {2013}
}

@inproceedings{IPIN13_Kurz,
 abstract = {In this paper, we present a novel approach for tracking objects whose
movement is constrained to a compact one-dimensional manifold, for
example a conveyer belt or a mobile robot whose movement is restricted
to tracks. Standard approaches either ignore the constraint at first and
retroactively move the estimate to lie on the manifold, or consider the
tracking problem on a manifold but falsely assume a Gaussian
distribution. Our method explicitly takes the actual topology into
account from the beginning and relies on special types of probability
distributions defined on the proper manifold. In particular, we consider
objects moving along a closed one-dimensional track, for example an
ellipse, a polygon, or similar closed shapes. This shape is transformed
to a circle with a homeomorphism. Thus, we can apply a recursive
circular filtering algorithm to the constrained tracking problem.
Finally, the estimate is transformed back to the original manifold. We
evaluate the proposed method in an experiment by tracking a toy train
moving along a track and comparing the results to those of traditional
approaches for this problem.},
 address = {Montbeliard, France},
 author = {Gerhard Kurz and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the Fourth IEEE GRSS International Conference on Indoor Positioning and Indoor Navigation (IPIN 2013)},
 month = {October},
 pdf = {IPIN13_Kurz.pdf},
 title = {Constrained Object Tracking on Compact One-dimensional Manifolds Based on Directional Statistics},
 year = {2013}
}

@inproceedings{NecSys13_Dolgov,
 abstract = {In Networked Control Systems (NCS), data networks not only
limit the amount of information exchanged by system components but
are also subject to stochastic packet delays and losses. In this
paper, we present a controller that simultaneously addresses these
problems by combining event-based and sequence-based control methods.
At every time step, the proposed controller calculates a sequence
of predicted control inputs and based on the expected future LQG
costs decides whether it transmits the control sequence to the actuator.
The proposed controller is evaluated with simulations.},
 address = {Koblenz, Germany},
 author = {Maxim Dolgov and Jörg Fischer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 4th IFAC Workshop on Distributed Estimation and Control in Networked Systems (NecSys 2013)},
 month = {September},
 pdf = {NecSys13_Dolgov.pdf},
 title = {Event-based LQG Control over Networks Subject to Random Transmission Delays and Packet Losses},
 year = {2013}
}

@inproceedings{ECC13_Sijs,
 abstract = {State fusion is a method for merging multiple estimates of the same
state into a single fused estimate. Dealing with multiple estimates
is one of the main concerns in distributed state estimation, where
an estimated value of the desired state vector is computed in each
node of a networked system. Most solutions for distributed state
estimation currently available assume that every node computes an
estimate of the (same) global state vector. This assumption is impractical
for systems observing large-area processes, due to the sheer size
of the process state. A more feasible solutions is one where each
node estimates a part of the global state vector, allowing different
nodes in the network to have overlapping state elements. Although
such an approach should be accompanied by a corresponding state fusion
method, existing solutions cannot be employed as they merely consider
fusion of two different estimates with equal state representations.
Therefore, an empirical solution is presented for fusing two state
estimates that have partially overlapping state elements. A justification
of the proposed fusion method is presented, along with an illustrative
case study for observing the temperature profile of a large rod,
though a formal derivation is future research.},
 address = {Zürich, Switzerland},
 author = {Joris Sijs and Uwe D. Hanebeck and Benjamin Noack},
 booktitle = {Proceedings of the 2013 European Control Conference (ECC 2013)},
 month = {July},
 pdf = {ECC13_Sijs.pdf},
 title = {An Empirical Method to Fuse Partially Overlapping State Vectors for Distributed State Estimation},
 year = {2013}
}

@inproceedings{Fusion13_Baum,
 abstract = {We present a novel method for tracking multiple targets, called Kernel-SME
filter, that does not require an enumeration of measurement-to-target
associations. This method is a further development of the symmetric
measurement equation (SME) filter that removes the data association
uncertainty of the original measurement equation with the help of
a symmetric transformation. The key idea of the Kernel-SME filter
is to define a symmetric transformation that maps the measurements
to a Gaussian mixture function. This transformation is scalable to
a large number of targets and allows for deriving a Gaussian state
estimator that only has a cubic runtime complexity in the number
of targets.},
 address = {Istanbul, Turkey},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Baum.pdf},
 title = {The Kernel-SME Filter for Multiple Target Tracking},
 year = {2013}
}

@inproceedings{Fusion13_Faion,
 abstract = {In this paper, we consider the problem of jointly tracking the pose
and shape of objects based on noisy data from cameras and depth sensors.
Our proposed approach formalizes object silhouettes from image data
as measurements within a Bayesian estimation framework. Projecting
object silhouettes from images back into space yields a visual hull
that constrains the object. In this work, we focus on the 2D case.
We derive a general equation for the silhouette measurement update
that explicitly considers segmentation uncertainty of each pixel.
By assuming a bounded error for the silhouettes, we can reduce the
complexity of the general solution to only consider uncertain edges
and derive an approximate measurement update. In simulations, we
show that the proposed approach dramatically improves point-cloud-based
estimators, especially in the presence of high noise.},
 address = {Istanbul, Turkey},
 author = {Florian Faion and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Faion.pdf},
 title = {Silhouette Measurements for Bayesian Object Tracking in Noisy Point Clouds},
 year = {2013}
}

@inproceedings{Fusion13_Gilitschenski,
 abstract = {In this paper, we introduce a novel approach for scheduled tracking
of a moving target based on bearings-only sensors. Unlike classical
approaches that are typically based on the extended or unscented
Kalman filter, we rely on circular statistics to describe probability
distributions for angular measurements more accurately. As the energy
available to sensors is limited in many scenarios, we introduce a
scheduling algorithm that selects a subset of two sensors to be active
at any given time step while minimizing the uncertainty of the state
estimate. This is done by anticipating possible future measurements.
We evaluate the proposed method in simulations and compare it to
an UKF-based solution. Our evaluation demonstrates the superiority
of the presented approach, particularly when high measurement uncertainty
makes consideration of the circular geometry necessary.},
 address = {Istanbul, Turkey},
 author = {Igor Gilitschenski and Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Gilitschenski.pdf},
 title = {Bearings-Only Sensor Scheduling Using Circular Statistics},
 year = {2013}
}

@inproceedings{Fusion13_Hanebeck,
 abstract = {A new Gaussian filter for estimating the state of nonlinear systems
is derived that relies on two main ingredients: i) the progressive
inclusion of the measurement information and ii) a tight coupling
between a Gaussian density and its deterministic Dirac mixture approximation.
No second Gaussian assumption for the joint density of state and
measurement is required, so that the performance is much better than
that of Linear Regression Kalman Filters (LRKFs), which heavily
rely on this assumption. In addition, the new filter directly works
with the generative system description. No Likelihood function is
required. It can be used as a plug -in replacement for standard Gaussian
filters such as the UKF.},
 address = {Istanbul, Turkey},
 author = {Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Hanebeck.pdf},
 title = {PGF 42: Progressive Gaussian Filtering with a Twist},
 year = {2013}
}

@inproceedings{Fusion13_Huber,
 abstract = {This paper proposes Gaussian filters for polynomial systems with efficient
solutions for both the prediction and the filter step. For the prediction
step, computationally efficient closed-form solutions are derived
for calculating the exact moments. In order to achieve a higher estimation
quality, the filter step is solved without the usual additional assumption
that state and measurement are jointly Gaussian distributed. As this
significantly complicates the required moment calculation, a homotopy
continuation method is employed that yields almost optimal results.},
 address = {Istanbul, Turkey},
 author = {Marco F. Huber and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Huber.pdf},
 title = {Gaussian Filtering for Polynomial Systems Based on Moment Homotopy},
 year = {2013}
}

@inproceedings{Fusion13_Kurz-Bingham,
 abstract = {Directional estimation is a common problem in many tracking applications.
Traditional filters like the Kalman filter perform poorly because
they fail to take the periodic nature of the problem into account.
We present a recursive filter for directional data based on the Bingham
distribution in two dimensions. The proposed filter can be applied
to circular filtering problems with 180 degree symmetry, i.e., rotations
by 180 degrees cannot be distinguished. It is easily implemented
using standard numerical techniques and suitable for real-time applications.
The presented approach is extensible to quaternions, which allow
tracking arbitrary three-dimensional orientations. We evaluate our
filter in a challenging scenario and compare it to a traditional
Kalman filtering approach.},
 address = {Istanbul, Turkey},
 annote = {Best Student Paper Award First Runner-Up Certificate (PDF)},
 author = {Gerhard Kurz and Igor Gilitschenski and Simon J. Julier and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Kurz-Bingham.pdf},
 title = {Recursive Estimation of Orientation Based on the Bingham Distribution},
 year = {2013}
}

@inproceedings{Fusion13_Kurz,
 abstract = {We propose an algorithm to combine both depth and position measurements
when estimating a continuous surface. Position measurements originate
from a fixed point on the surface, whereas depth measurements are
determined by the intersection of the surface with a line originating
from the depth sensor. Through fusion of both types of measurements,
it is possible to benefit from the advantages of different sensors.
The surface is obtained through interpolation of control points with
splines, which allows a compact representation of the surface. In
order to simplify the problem of intersecting the surface with lines
originating from the depth sensor, we propose the use of polar or
spherical coordinates in surface parameterization. The presented
algorithm can be applied in both 2D and 3D settings and is independent
of the particular choice of sensors. Our method can recursively include
new information as it is obtained by using nonlinear filtering and
it considers uncertainties associated with the measurements.},
 address = {Istanbul, Turkey},
 annote = {Winner Best Paper Award Certificate (PDF)},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Kurz.pdf},
 title = {Recursive Fusion of Noisy Depth and Position Measurements for Surface Reconstruction},
 year = {2013}
}

@inproceedings{Fusion13_Noack,
 abstract = {The federated Kalman filter embodies an efficient and easy-to-implement
solution for linear distributed estimation problems. Data from independent
sensors can be processed locally and in parallel on different nodes
without running the risk of erroneously ignoring possible dependencies.
The underlying idea is to counteract the common process noise issue
by inflating the joint process noise matrix. In this paper, the same
trick is generalized to nonlinear models and non-Gaussian process
noise. The probability density of the joint process noise is split
into an exponential mixture of transition densities. By this means,
the process noise is modeled to independently affect the local system
models. The estimation results provided by the sensor devices can
then be fused, just as if they were indeed independent.},
 address = {Istanbul, Turkey},
 author = {Benjamin Noack and Simon J. Julier and Marc Reinhardt and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Noack.pdf},
 title = {Nonlinear Federated Filtering},
 year = {2013}
}

@inproceedings{Fusion13_Pfaff,
 abstract = {For systems suffering from different types of uncertainties, finding
criteria for validating measurements can be challenging. In this
paper, we regard both stochastic Gaussian noise with full or imprecise
knowledge about correlations and unknown but bounded errors. The
validation problems arising in the individual and combined cases
are illustrated to convey different perspectives on the proposed
conditions. Furthermore, hints are provided for the algorithmic implementation
of the validation tests. Particular focus is put on ensuring a predefined
lower bound for the probability of correctly classifying valid data.},
 address = {Istanbul, Turkey},
 author = {Florian Pfaff and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Pfaff.pdf},
 title = {Data Validation in the Presence of Stochastic and Set-Membership Uncertainties},
 year = {2013}
}

@inproceedings{Fusion13_Reinhardt,
 abstract = {In this paper, linear distributed estimation is revisited on the basis
of the hypothesizing distributed Kalman filter and equations for
a flexible application of the algorithm are derived. We propose a
new approximation for the mean-squared-error matrix and present techniques
for automatically improving the hypothesis about the global measurement
model. Utilizing these extensions, the precision of the filter is
improved so that it asymptotically yields optimal results for time-invariant
models. Pseudo-code for the implementation of the algorithm is provided
and the lossless inclusion of out-of-sequence measurements is discussed.
An evaluation demonstrates the effect of the new extensions and compares
the results to state-of-the-art methods.},
 address = {Istanbul, Turkey},
 author = {Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Reinhardt.pdf},
 title = {Advances in Hypothesizing Distributed Kalman Filtering},
 year = {2013}
}

@inproceedings{Fusion13_Sijs,
 abstract = {To reduce the amount of data transfer in networked systems, measurements
are usually taken only when an event occurs rather than periodically
in time. However, this complicates estimation problems considerably
as it is not guaranteed that new sensor data will be sampled. Therefore,
an existing state estimator is extended so to cope with event-based
measurements successfully, i.e., curtail any diverging behavior in
the estimation results. To that extent, a general formulation of
event sampling is proposed. This formulation is used to set up a
state estimator combining stochastic as well as set-membership measurement
information according to a hybrid update: when an event occurs the
estimated state is updated using the stochastic measurement received
(positive information), while at periodic time instants no measurement
is received (negative information) and the update is based on knowledge
that the sensor value lies within a bounded subset of the measurement
space. An illustrative example further shows that the developed estimator
has an improved representation of estimation errors compared to a
purely stochastic estimator for various event sampling strategies.},
 address = {Istanbul, Turkey},
 author = {Joris Sijs and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Sijs.pdf},
 title = {Event-based State Estimation with Negative Information},
 year = {2013}
}

@inproceedings{Fusion13_Steinbring,
 abstract = {An accurate Linear Regression Kalman Filter (LRKF) for nonlinear systems
called Smart Sampling Kalman Filter (S²KF) is introduced. It is based
on a new low-discrepancy Dirac Mixture approximation of Gaussian
densities. The approximation comprises an arbitrary number of optimally
and deterministically placed samples in the entire state space, so
that the filter resolution can be adapted to either achieve high-quality
results or meet computational constraints. For two samples per dimension,
the S²KF comprises the UKF as a special case. With an increasing
number of samples, the new filter quickly converges to the (typically
infeasible) exact analytic LRKF. The S²KF can be seen as the ultimate
generalization of all sample-based LRKFs such as the UKF, sigma-point
filters, higher-order variants etc., as it homogeneously covers the
state space with an arbitrary number of samples. It is evaluated
by performing extended target tracking.},
 address = {Istanbul, Turkey},
 author = {Jannik Steinbring and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Steinbring.pdf},
 title = {S2KF: The Smart Sampling Kalman Filter},
 year = {2013}
}

@inproceedings{Fusion13_Zea,
 abstract = {This paper presents a novel approach to track a non-convex shape approximation
of an extended target based on noisy point measurements. For this
purpose, a novel type of Random Hypersurface Model (RHM), called
Level-Set RHM is introduced that models the interior of a shape with
level-sets of an implicit function. Based on the Level-Set RHM, a
nonlinear measurement equation can be derived that allows to employ
a standard Gaussian state estimator for tracking an extended object
even in scenarios with high measurement noise. In this paper, shapes
are described using polygons and shape regularization is applied
using ideas from active contour models.},
 address = {Istanbul, Turkey},
 author = {Antonio Zea and Florian Faion and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 16th International Conference on Information Fusion (Fusion 2013)},
 month = {July},
 pdf = {Fusion13_Zea.pdf},
 title = {Level-Set Random Hypersurface Models for Tracking Non-Convex Extended Objects},
 year = {2013}
}

@article{AT13_Fischer,
 abstract = {In this paper, a unified approach to sequence-based control and estimation
of linear networked systems with multiple sensors is proposed. Time
delays and data losses in the controller-actuator link are compensated
by sending sequences of control inputs. The sequence-based design
paradigm is further extended to the sensor-controller connections
without increasing the load of the network. In this context, we present
a recursive solution based on the Hypothesizing Distributed Kalman
Filter (HKF) that is included in the overall sequence-based controller
design.},
 author = {Jörg Fischer and Marc Reinhardt and Uwe D. Hanebeck},
 doi = {10.1524/auto.2013.0039},
 journal = {at -- Automatisierungstechnik, Oldenbourg Verlag},
 month = {July},
 number = {7},
 pages = {467--476},
 pdf = {AT13_Fischer.pdf},
 title = {Optimal Sequence-Based Control and Estimation of Networked Linear Systems},
 url = {https://dx.doi.org/10.1524/auto.2013.0039},
 volume = {61},
 year = {2013}
}

@inproceedings{ACC13_Fischer,
 abstract = {This paper addresses the problem of sequence-based controller design
for Networked Control Systems (NCS), where control inputs and measurements
are transmitted over TCP-like network connections that are subject
to random transmission delays and packet losses. To cope with the
network effects, the controller not only sends the current control
input to the actuator, but also a sequence of predicted control inputs
at every time step. In this setup, we derive an optimal solution
to the Linear Quadratic Gaussian (LQG) control problem and prove
that the separation principle holds. Simulations demonstrate the
improved performance of this optimal controller compared to other
sequence-based approaches.},
 address = {Washington D.C., USA},
 author = {Jörg Fischer and Achim Hekler and Maxim Dolgov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2013 American Control Conference (ACC 2013)},
 month = {June},
 pdf = {ACC13_Fischer.pdf},
 title = {Optimal Sequence-Based LQG Control over TCP-like Networks Subject to Random Transmission Delays and Packet Losses},
 year = {2013}
}

@inproceedings{ACC13_Gilitschenski,
 abstract = {We propose an efficient method for approximating arbitrary Gaussian
densities by a mixture of Dirac components. This approach is based
on the modification of the classical Cramér-von Mises distance, which
is adapted to the multivariate scenario by using Localized Cumulative
Distributions (LCDs) as a replacement for the cumulative distribution
function. LCDs consider the local probabilistic influence of aprobability
density around a given point. Our modification of the Cramér-von
Mises distance can be approximated for certain special cases in closed-form.
The created measure is minimized in order to compute the positions
of the Dirac components for a standard normal distribution.},
 address = {Washington D.C., USA},
 author = {Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2013 American Control Conference (ACC 2013)},
 month = {June},
 pdf = {ACC13_Gilitschenski.pdf},
 title = {Efficient Deterministic Dirac Mixture Approximation},
 year = {2013}
}

@inproceedings{ACC13_Kurz,
 abstract = {Estimation of circular quantities is a widespread problem that occurs
in many tracking and control applications. Commonly used approaches
such as the Kalman filter, the extended Kalman filter (EKF), and
the unscented Kalman filter (UKF) do not take periodicity explicitly
into account, which can result in low estimation accuracy. We present
a filtering algorithm for angular quantities in nonlinear systems
that is based on circular statistics. The new filter switches between
three different representations of probability distributions on the
circle, the wrapped normal, the von Mises, and a Dirac mixture density.
It can be seen as a systematic generalization of the UKF to circular
statistics. We evaluate the proposed filter in simulations and show
its superiority to conventional approaches.},
 address = {Washington D.C., USA},
 author = {Gerhard Kurz and Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2013 American Control Conference (ACC 2013)},
 month = {June},
 pdf = {ACC13_Kurz.pdf},
 title = {Recursive Nonlinear Filtering for Angular Data Based on Circular Distributions},
 year = {2013}
}

@article{arXiv13_Kurz,
 abstract = {Directional estimation is a common problem in many tracking applications. 
Traditional filters such as the Kalman filter perform poorly because they fail to take 
the periodic nature of the problem into account. We present a recursive filter for directional 
data based on the Bingham distribution in two dimensions. The proposed filter can be applied 
to circular filtering problems with 180 degree symmetry, i.e., rotations by 180 degrees cannot 
be distinguished. It is easily implemented using standard numerical techniques and suitable for 
real-time applications. The presented approach is extensible to quaternions, which allow tracking 
arbitrary three-dimensional orientations. We evaluate our filter in a challenging scenario and 
compare it to a traditional Kalman filtering approach.},
 author = {Gerhard Kurz and Igor Gilitschenski and Simon J. Julier and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {April},
 title = {Recursive Estimation of Orientation Based on the Bingham Distribution},
 url = {https://arxiv.org/abs/1304.8019},
 year = {2013}
}

@article{arXiv13_Baum,
 abstract = {The Random Hypersurface Model (RHM) is introduced that allows for estimating 
a shape approximation of an extended object in addition to its kinematic state. 
An RHM represents the spatial extent by means of randomly scaled versions of the shape 
boundary. In doing so, the shape parameters and the measurements are related via a 
measurement equation that serves as the basis for a Gaussian state estimator. 
Specific estimators are derived for elliptic and star-convex shapes.},
 author = {Marcus Baum and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY), Draft accepted for publication in IEEE Transactions on Aerospace and Electronic Systems},
 month = {April},
 title = {Extended Object Tracking with Random Hypersurface Models},
 url = {https://arxiv.org/abs/1304.5084},
 year = {2013}
}

@inproceedings{CDC12_Baum,
 abstract = {Almost all multi-target tracking systems have to generate point estimates
for the targets, e.g., for displaying the tracks. The novel idea
in this paper is to  consider point estimates for multi-target states
that are optimal according to a  kernel distance measure. Because
the kernel distance is a metric on point sets and ignores the target
labels, shortcomings of  Minimum Mean Squared Error (MMSE) estimates
for multi-target states can be avoided. We show how the calculation
of  these point estimates can be casted as an optimization problem
and it turns out that it corresponds to the problem of reducing the
Probability Hypothesis Density (PHD) function   to a Dirac mixture
density. Finally, we  discuss a generalization of the kernel distance
called LCD distance, which does not require to choose a specific
kernel width. The presented methods are evaluated in a  Multiple-Hypotheses
Tracker (MHT) setting with up to ten targets.},
 address = {Maui, Hawaii, USA},
 author = {Marcus Baum and Patrick Ruoff and Dominik Itte and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 51st IEEE Conference on Decision and Control (CDC 2012)},
 month = {December},
 pdf = {CDC12_Baum.pdf},
 title = {Optimal Point Estimates for Multi-target States based on Kernel Distances},
 year = {2012}
}

@inproceedings{CDC12_Chlebek,
 abstract = {Increasing demand for Nonlinear Model Predictive Control with the
ability to handle highly noise-corrupted systems has recently given
rise to stochastic control approaches. Besides providing high-quality
results within a noisy environment, these approaches have one problem
in common, namely a high computational demand and, as a consequence,
generally a short prediction horizon. In this paper, we propose to
reduce the computational complexity of prediction and value function
evaluation within the control horizon by simplifying the system progressively
down to the deterministic case. Approximation of occurring probability
densities by a specific representation, the deterministic Dirac mixture
density, with a decreasing resolution (i.e., approximation quality)
leads via natural decomposition to a point estimate and thus, can
be treated in a deterministic manner. Hence, calculation of the remaining
time steps requires considerably less computation time.},
 address = {Maui, Hawaii, USA},
 author = {Christof Chlebek and Achim Hekler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 51st IEEE Conference on Decision and Control (CDC 2012)},
 month = {December},
 pdf = {CDC12_Chlebek.pdf},
 title = {Stochastic Nonlinear Model Predictive Control Based on Progressive Density Simplification},
 year = {2012}
}

@inproceedings{CDC12_Hekler,
 abstract = {In this paper, we address the problem of controlling a system over
an unreliable UDP-like network that is affected by time-varying delays
and randomly occurring packet losses. A major challenge of this setup
is that the controller just has uncertain information about the control
inputs actually applied by the actuator. The key idea of this work
is to model the uncertain control inputs by random variables, the
so-called virtual control inputs, which are characterized by discrete
probability density functions. Subject to this probabilistic description,
a novel, easy to implement sequencebased control approach is proposed
that extends any given state feedback controller designed without
consideration of the network-induced disturbances. The high performance
of the proposed controller is demonstrated by means of Monte Carlo
simulation runs with an inverted pendulum on a cart.},
 address = {Maui, Hawaii, USA},
 author = {Achim Hekler and Jörg Fischer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 51st IEEE Conference on Decision and Control (CDC 2012)},
 month = {December},
 pdf = {CDC12_Hekler.pdf},
 title = {Sequence-Based Control for Networked Control Systems Based on Virtual Control Inputs},
 year = {2012}
}

@inproceedings{CDC12_Noack,
 abstract = {In state estimation theory, two directions are mainly followed in
order to model disturbances and errors. Either uncertainties are
modeled as stochastic quantities or they are characterized by their
membership to a set. Both approaches have distinct advantages and
disadvantages making each one inherently better suited to model different
sources of estimation uncertainty. This paper is dedicated to the
task of combining stochastic and set-membership estimation methods.
A Kalman gain is derived that minimizes the mean squared error in
the presence of both stochastic and additional unknown but bounded
uncertainties, which are represented by Gaussian random variables
and ellipsoidal sets, respectively. As a result, a generalization
of the well-known Kalman filtering scheme is attained that reduces
to the standard Kalman filter in the absence of set-membership uncertainty
and that otherwise becomes the intersection of sets in case of vanishing
stochastic uncertainty. The proposed concept also allows to prioritize
either the minimization of the stochastic uncertainty or the minimization
of the set-membership uncertainty.},
 address = {Maui, Hawaii, USA},
 author = {Benjamin Noack and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 51st IEEE Conference on Decision and Control (CDC 2012)},
 month = {December},
 pdf = {CDC12_Noack.pdf},
 title = {Optimal Kalman Gains for Combined Stochastic and Set-Membership State Estimation},
 year = {2012}
}

@inproceedings{CDC12_Reinhardt,
 abstract = {A new method for globally optimal estimation in decentralized sensor-networks
is applied to the decentralized control problem. The resulting approach
is proven to be optimal when the nodes have access to all information
in the network. More precisely, we utilize an algorithm for optimal
distributed estimation in order to obtain local estimates whose combination
yields the globally optimal estimate. When the interconnectivity
is high, the local estimates are almost optimal, which motivates
the application of the principle of separation. Thus, we optimize
the controller and finally obtain a flexible algorithm, whose quality
is evaluated in different scenarios. In applications where the strong
requirements on a perfect communication cannot be guaranteed, we
derive quality bounds by help of a detailed evaluation of the algorithm.
When information is regularly exchanged, it is demonstrated that
the algorithm performs almost optimally and therefore, offers system
designers a flexible and easy to implement approach. The field of
applications lies within the area of strongly networked systems,
in particular, when communication disturbances cannot be foreseen
or when the network structure is too complicated to apply optimized
regulators.},
 address = {Maui, Hawaii, USA},
 author = {Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 51st IEEE Conference on Decision and Control (CDC 2012)},
 month = {December},
 pdf = {CDC12_Reinhardt.pdf},
 title = {Decentralized Control Based on Globally Optimal Estimation},
 year = {2012}
}

@article{arXiv12_Baum,
 abstract = {We present a novel method called Kernel-SME filter for tracking multiple targets 
when the association of the measurements to the targets is unknown. The method is a 
further development of the Symmetric Measurement Equation (SME) filter, which removes 
the data association uncertainty of the original measurement equation with the help of 
a symmetric transformation. The underlying idea of the Kernel-SME filter is to construct 
a symmetric transformation by means of mapping the measurements to a Gaussian mixture. 
This transformation is scalable to a large number of targets and allows for deriving a 
Gaussian state estimator that has a cubic time complexity in the number of targets.},
 author = {Marcus Baum and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {December},
 title = {The Kernel-SME Filter for Multiple Target Tracking},
 url = {https://arxiv.org/abs/1212.5882},
 year = {2012}
}

@inproceedings{CURAC12_Kurz,
 abstract = {Performing surgery on the beating heart has significant advantages
compared to cardiopulmonary bypass. However, when performed directly,
it is very demanding for the surgeon. As an alternative, using a
teleoperated robot for compensating the heart motion has been proposed.
As an addition, this paper describes how stabilized images are obtained
to create the illusion of operating on a stationary heart. For that
purpose, the heart motion is tracked with a stochastic physical model.
Based on correspondences obtained by motion tracking, image stabilization
is considered as a scattered data interpolation problem. The proposed
algorithms are evaluated on a heart phantom and in in-vivo experiments
on a porcine heart, which show that there is very little residual
motion in the stabilized images.},
 address = {Düsseldorf, Germany},
 author = {Gerhard Kurz and Uwe D. Hanebeck},
 booktitle = {11. Jahrestagung der Deutschen Gesellschaft für Computer- und Roboterassistierte Chirurgie (CURAC12)},
 month = {November},
 pdf = {CURAC12_Kurz.pdf},
 title = {Image Stabilization with Model-Based Tracking for Beating Heart Surgery},
 year = {2012}
}

@article{arXiv12_FischerLQG,
 abstract = {This paper addresses the problem of sequence-based controller design for Networked Control Systems (NCS), 
where control inputs and measurements are transmitted over TCP-like network connections that are subject to stochastic 
packet losses and time-varying packet delays. At every time step, the controller sends a sequence of predicted control 
inputs to the actuator in addition to the current control input. In this sequence-based setup, we derive an optimal 
solution to the Linear Quadratic Gaussian (LQG) control problem and prove that the separation principle holds. 
Simulations demonstrate the improved performance of this optimal controller compared to other sequence-based approaches.},
 author = {Jörg Fischer and Achim Hekler and Maxim Dolgov and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {November},
 title = {Optimal Sequence-Based LQG Control over TCP-like Networks Subject to Random Transmission Delays and Packet Losses},
 url = {https://arxiv.org/abs/1211.3020},
 year = {2012}
}

@article{TAES12_Baum,
 abstract = {A novel approach for extended object tracking is presented. In contrast
to existing approaches, no statistical assumptions about the location
of the measurement sources on the extended target object are made.
As a consequence, a combined set-theoretic and stochastic estimator
is obtained that is robust to systematic errors in the target model.
The benefits of the new approach is demonstrated by means of simulations.},
 author = {Marcus Baum and Uwe D. Hanebeck},
 doi = {10.1109/TAES.2012.6324680},
 issn = {0018-9251},
 journal = {IEEE Transactions on Aerospace and Electronic Systems},
 month = {October},
 number = {4},
 pages = {3103--3115},
 pdf = {TAES12_Baum.pdf},
 title = {Extended Object Tracking Based on Set-Theoretic and Stochastic Fusion},
 url = {https://dx.doi.org/10.1109/TAES.2012.6324680},
 volume = {48},
 year = {2012}
}

@inproceedings{IROS12_Faion,
 abstract = {This paper describes a method to intelligently schedule a network
of multiple RGBD sensors in a Bayesian object tracking scenario,
with special focus on Microsoft Kinect devices. These setups have
issues such as the large amount of raw data generated by the sensors
and interference caused by overlapping fields of view. The proposed
algorithm addresses these issues by selecting and exclusively activating
the sensor that yields the best measurement, as defined by a novel
stochastic model that also considers hardware constraints and intrinsic
parameters. In addition, as existing solutions to toggle the sensors
were found to be insufficient, the development of a hardware module,
especially designed for quick toggling and synchronization with the
depth stream, is also discussed. The algorithm then is evaluated
within the scope of a multi-Kinect object tracking scenario and compared
to other scheduling strategies.},
 address = {Vilamoura, Algarve, Portugal},
 author = {Florian Faion and Simon Friedberger and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2012)},
 month = {October},
 pdf = {IROS12_Faion.pdf},
 title = {Intelligent Sensor-Scheduling for Multi-Kinect-Tracking},
 year = {2012}
}

@inproceedings{MFI12_Baum,
 abstract = {This paper is about an experimental set-up for tracking a ground moving
mobile object from a bird's eye view. In this experiment, an RGB
and depth camera is used for detecting moving points. The detected
points serve as input for a probabilistic extended object tracking
algorithm that simultaneously estimates the kinematic parameters
and the shape parameters of the object. By this means, it is easy
to discriminate moving objects from the background and the probabilistic
tracking algorithm ensures a robust and smooth shape estimate. We
provide an experimental evaluation of a recent Bayesian extended
object tracking algorithm based on a so-called Random Hypersurface
Model and give a comparison with active contour models.},
 address = {Hamburg, Germany},
 annote = {Nominee Best Student Paper Award},
 author = {Marcus Baum and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2012)},
 month = {September},
 pdf = {MFI12_Baum.pdf},
 title = {Tracking Ground Moving Extended Objects using RGBD Data},
 year = {2012}
}

@inproceedings{MFI12_Reinhardt,
 abstract = {This paper deals with distributed information processing in sensor
networks. We propose the Hypothesizing Distributed Kalman Filter
that incorporates an assumption of the global measurement model into
the distributed estimation process. The procedure is based on the
Distributed Kalman Filter and inherits its optimality when the assumption
about the global measurement uncertainty is met. Recursive formulas
for local processing as well as for fusion are derived. We show that
the proposed algorithm yields the same results, no matter whether
the measurements are processed locally or globally, even when the
process noise is not negligible. For further processing of the estimates,
a consistent bound for the error covariance matrix is derived. All
derivations and explanations are illustrated by means of a new classification
scheme for estimation processes.},
 address = {Hamburg, Germany},
 annote = {Nominee Best Student Paper Award},
 author = {Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2012)},
 month = {September},
 pdf = {MFI12_Reinhardt.pdf},
 title = {The Hypothesizing Distributed Kalman Filter},
 year = {2012}
}

@article{arXiv12_Fischer,
 abstract = {In this paper, a unified approach to sequence-based control and estimation
of linear networked systems with multiple sensors is proposed. Time
delays and data losses in the controller-actuator-channel are compensated
by sending sequences of  control inputs. The sequence-based design
paradigm is further extended to the sensor-controller-channels without
increasing the load of the network. In this context, we present a
recursive solution based on the Hypothesizing Distributed Kalman
Filter (HKF) that is included in the overall sequence-based controller
design.},
 author = {Jörg Fischer and Marc Reinhardt and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {August},
 title = {Optimal Sequence-Based Control and Estimation of Networked Linear Systems},
 url = {https://arxiv.org/abs/1211.5086v1},
 year = {2012}
}

@article{SYSAES_Baum,
 abstract = {Target tracking algorithms are usually based on the assumption that
the target extent is small compared to the measurement noise; hence,
the target is modeled as a mathematical point. However, if the target
extent is rather large, the target may cause multiple sensor measurements
from different spatially distributed reflection centers. In this
case, the modeling of the target extent is essential. In particular,
the author looks at the random hypersurface model and its role in
tracking; the tracking method is evaluated using a Microsoft(R) Kinect(TM)
sensor as an example.},
 author = {Marcus Baum},
 doi = {10.1109/MAES.2012.6328840},
 issn = {0885-8985},
 journal = {IEEE Aerospace and Electronic Systems Magazine},
 month = {July},
 number = {7},
 pages = {42--44},
 title = {Student Research Highlight: Simultaneous Tracking and Shape Estimation of Extended Targets},
 url = {https://dx.doi.org/10.1109/MAES.2012.6328840},
 volume = {27},
 year = {2012}
}

@article{TAC12_Deisenroth,
 abstract = {We propose a principled algorithm for robust Bayesian filtering and
smoothing in nonlinear stochastic dynamic systems when both the transition
function and the measurement function are described by non-parametric
Gaussian process (GP) models. GPs are gaining increasing importance
in signal processing, machine learning, robotics, and control for
representing unknown system functions by posterior probability distributions.
This modern way of system identification is more robust than finding
point estimates of a parametric function representation. Our principled
filtering/smoothing approach for GP dynamic systems is based on analytic
moment matching in the context of the forward-backward algorithm.
Our numerical evaluations demonstrate the robustness of the proposed
approach in situations where other state-of-the-art Gaussian filters
and smoothers can fail.},
 author = {Marc Peter Deisenroth and Ryan Darby Turner and Marco F. Huber and Uwe D. Hanebeck and Carl Edward Rasmussen},
 doi = {10.1109/TAC.2011.2179426},
 issn = {0018-9286},
 journal = {IEEE Transactions on Automatic Control},
 keywords = {GP dynamic systems;analytic moment matching;control systems;forward-backward
algorithm;machine learning;measurement function;nonlinear stochastic
dynamic systems;nonparametric Gaussian process;parametric function
representation;point estimation;posterior probability distributions;robotics;robust
Bayesian filtering;robust Bayesian smoothing;signal processing;system
identification;transition function;unknown system function representation;Bayes
methods;Gaussian processes;identification;nonlinear dynamical systems;nonparametric
statistics;smoothing methods;statistical distributions;},
 month = {July},
 number = {7},
 pages = {1865--1871},
 pdf = {TAC12_Deisenroth.pdf},
 title = {Robust Filtering and Smoothing with Gaussian Processes},
 url = {https://dx.doi.org/10.1109/TAC.2011.2179426},
 volume = {57},
 year = {2012}
}

@inproceedings{Fusion12_Baum,
 abstract = {Extended target tracking deals with simultaneously tracking the shape
and the kinematic parameters of a target. In this work, we formulate
the extended target tracking problem as a state estimation problem
with both multiplicative and additive measurement noise. In case
of extended targets with known orientation, we show that the best
linear estimator is not consistent and, hence, is unsuitable for
this problem. In order to overcome this issue, we propose a quadratic
estimator for a recursive closed-form measurement update. Simulations
demonstrate the performance of the estimator.},
 address = {Singapore},
 author = {Marcus Baum and Florian Faion and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Baum-MultNoise.pdf},
 title = {Modeling the Target Extent with Multiplicative Noise},
 year = {2012}
}

@inproceedings{Fusion12_BaumWillett,
 abstract = {In this work, we present some exact algorithms for calculating the
minimum mean optimal sub-pattern assignment (MMOSPA) estimate for
probability densities that are represented with particles. First,
a polynomial-time algorithm for two targets is derived by reducing
the problem to the enumeration of the cells of a hyperplane arrangement.
Second, we present a linear-time algorithm for an arbitrary number
of scalar target states, which is based on the insight that the MMOSPA
estimate coincides with the mean of the order statistics.},
 address = {Singapore},
 author = {Marcus Baum and Peter Willett and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Baum-MMOSPA.pdf},
 title = {Calculating Some Exact MMOSPA Estimates for Particle Distributions},
 year = {2012}
}

@inproceedings{Fusion12_BenavoliNoack,
 abstract = {The paper focuses on the fundamental idea of Kalman's seminal paper:
how to solve the filtering problem from the only knowledge of the
first two moments of the noise terms. In this paper, by exploiting
set of distributions based filtering, we solve this problem without
introducing additional assumptions on the distributions of the noise
terms (e.g., Gaussianity) or on the final form of the estimator (e.g.,
linear estimator). Given the moments (e.g., mean and variance) of
random variable X, it is possible to define the set of all distributions
that are compatible with the moments information. This set of distributions
can be equivalently characterized by its extreme distributions which
is a family of mixtures of Dirac's deltas. The lower and upper expectation
of any function g of X are obtained in correspondence of these extremes
and can be computed by solving a linear programming problem. The
filtering problem can then be solved by running iteratively this
linear programming problem.},
 address = {Singapore},
 author = {Alessio Benavoli and Benjamin Noack},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_BenavoliNoack.pdf},
 title = {Pushing Kalman's Idea to the Extremes},
 year = {2012}
}

@inproceedings{Fusion12_Faion-CylinderTracking,
 abstract = {Depth sensors such as the Microsoft Kinect™ depth sensor
provide three dimensional point clouds of an observed scene. In this
paper, we employ Random Hypersurface Models (RHMs), which is a modeling
technique for extended object tracking, to point cloud fusion in
order to track a shape approximation of an underlying object. We
present a novel variant of RHMs to model shapes in 3D space. Based
on this novel model, we develop a specialized algorithm to track
persons by approximating their shapes as cylinders. For evaluation,
we utilize a Kinect network and simulations based on a stochastic
sensor model.},
 address = {Singapore},
 author = {Florian Faion and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Faion-CylinderTracking.pdf},
 title = {Tracking 3D Shapes in Noisy Point Clouds with Random Hypersurface Models},
 year = {2012}
}

@inproceedings{Fusion12_Faion-RecursiveCalibration,
 abstract = {In this paper we present a recursive Bayesian method to calibrate
rigidly linked depth sensors with non-overlapping fields of view.
The extrinsic parameters of this setup are obtained by rotating and
translating both cameras, estimating the local transformations using
point feature correspondences, and finally using these values to
recursively find a solution to the matrix equation $\matA_k\matX=\matX\matB_k$.
The algorithm is based on a Bayesian estimator, which allows the
consideration of camera-specific measurement noise and permits the
system to adapt naturally to changes in the extrinsic parameters.
The derived equations were carefully chosen to be free from singularities.
This paper also includes a thorough evaluation based on synthetic
and real data to show the effectiveness of the algorithm.},
 address = {Singapore},
 author = {Florian Faion and Patrick Ruoff and Antonio Zea and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Faion.pdf},
 title = {Recursive Bayesian Calibration of Depth Sensors with Non-Overlapping Views},
 year = {2012}
}

@inproceedings{Fusion12_Fischer,
 abstract = {We consider the problem of state estimation in a Networked Control
System, where measurements and control inputs are transmitted via
a communication network. The network is subject to time-varying delays
and stochastic data losses and does not provide acknowledgments of
successfully transmitted data packets. A challenge that arises in
this configuration is that the estimator has only uncertain information
about the actually applied control inputs. In this paper, we derive
a multiple-model based estimator that uses the state measurements
to estimate the applied control inputs so that the overall state
estimation is improved. The efficiency of the proposed approach is
demonstrated by means of Monte-Carlo-Simulation runs with an inverted
pendulum on a cart.},
 address = {Singapore},
 author = {Jörg Fischer and Achim Hekler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Fischer.pdf},
 title = {State Estimation in Networked Control Systems},
 year = {2012}
}

@inproceedings{Fusion12_Gilitschenski,
 abstract = {On-line fault-detection in uncertain measurement and estimation systems
is of particular interest in many applications. In certain systems
based on the Kalman filter, this test can be performed by checking
whether hyperellipsoids overlap. This test can be applied to detecting
failure in the system itself or in the sensors used to determine
the system state. To facilitate the practical application of such
tests, we describe a simple condition for overlap of two ellipsoids
and propose an efficient algorithmic implementation for testing this
condition. There are applications in many other areas, such as collision
avoidance or computer graphics. Our proposal makes use of Leverriere's
algorithm and Sturm's theorem, a result of algebraic geometry. Thus,
no approximative methods, such as root finding or minimization are
needed. Furthermore, the complexity of the algorithm is fixed for
a fixed problem dimension.},
 address = {Singapore},
 author = {Igor Gilitschenski and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Gilitschenski.pdf},
 title = {A Robust Computational Test for Overlap of Two Arbitrary-dimensional Ellipsoids in Fault-Detection of Kalman Filters},
 year = {2012}
}

@inproceedings{Fusion12_Hanebeck,
 abstract = {In this paper, we propose a progressive Gaussian filter, where the
measurement information is continuously included into the given prior
estimate (although we perform observations at discrete time steps).
The key idea is to derive a system of ordinary first-order differential
equations (ODE) that is used for continuously tracking the true non-Gaussian
posterior by its best matching Gaussian approximation. Calculating
the required moments of the true posterior is performed based on
corresponding Dirac Mixture approximations. The performance of the
new filter is evaluated in comparison with state-of-the-art filters
by means of a canonical benchmark example, the discrete-time cubic
sensor problem.},
 address = {Singapore},
 author = {Uwe D. Hanebeck and Jannik Steinbring},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Hanebeck.pdf},
 title = {Progressive Gaussian Filtering Based on Dirac Mixture Approximations},
 year = {2012}
}

@inproceedings{Fusion12_Hekler,
 abstract = {Time delays and data losses arising from an unreliable communication
between the components of a control loop decrease the quality of
control and thus, have to be incorporated explicitly in the control
decision. In this paper, a novel concept, the so-called virtual control
inputs, is presented, which extends the well-established control
technique of sending sequences of future inputs by incorporating
also the potential effects of previously transmitted sequences on
the future system behavior. The key idea of this paper is to model
the unknown future inputs as random variables characterized by probability
density functions over the finite set of potential future inputs.
Subject to this probabilistic description of the future inputs, the
controller determines the optimal open-loop sequence over a finite
horizon. The high capacity of the proposed approach is demonstrated
by simulations, in which a sensor manager schedules sensors for tracking
a mobile object.},
 address = {Singapore},
 author = {Achim Hekler and Jörg Fischer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Hekler.pdf},
 title = {Control over Unreliable Networks Based on Control Input Densities},
 year = {2012}
}

@inproceedings{Fusion12_Noack,
 abstract = {In state estimation theory, stochastic and set-membership approaches
are generally considered separately from each other. Both concepts
have distinct advantages and disadvantages making each one inherently
better suited to model different sources of estimation uncertainty.
In order to better utilize the potentials of both concepts, the core
element of this paper is a Kalman filtering scheme that allows for
a simultaneous treatment of stochastic and set-membership uncertainties.
An uncertain quantity is herein modeled by a set of Gaussian densities.
Since many modern applications operate in networked systems that
may consist of a multitude of local processing units and sensor nodes,
estimates have to be computed in a distributed manner and measurements
may arrive at high frequency. An algebraic reformulation of the Kalman
filter, the information filter, significantly eases the implementation
of such distributed fusion architectures. This paper explicates how
stochastic and set-membership uncertainties can simultaneously be
treated within this information form and compared to the Kalman filter,
it becomes apparent that the quality of some required approximations
is enhanced.},
 address = {Singapore},
 author = {Benjamin Noack and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Noack.pdf},
 title = {Combined Stochastic and Set-Membership Information Filtering in Multisensor Systems},
 year = {2012}
}

@inproceedings{Fusion12_Packi,
 abstract = {Indoor localization is a field in research with many competing technologies
using different kinds of media. A common challenge faced by most
systems is dealing with Non-Line-of-Sight (NLOS) conditions. We are
addressing this issue with focus on sound in the frequency range
above 20 kHz, as we encountered severe occurrence of outliers due
to multipath propagation, by reflections, and from occlusion. The
proper discrimination of erroneous signals is of special concern
during initialization time of the tracking system. During run time,
the computationally demanding process can be spared, if motion is
modelled and stochastic filtering techniques are applied. This paper
depicts solutions for both cases, and demonstrates that a combined
use of static and dynamic localization methods delivers increased
robustness at an affordable computational cost.},
 address = {Singapore},
 author = {Ferdinand Packi and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Packi.pdf},
 title = {Robust NLOS Discrimination for Range-Based Acoustic Pose Tracking},
 year = {2012}
}

@inproceedings{Fusion12_Reinhardt,
 abstract = {The distributed processing of measurements and the subsequent data
fusion is called Track-to-Track fusion. Although a solution for the
Track-to-Track fusion that is equivalent to a central processing
scheme has been proposed, this algorithm suffers from strict requirements
regarding the local availability of knowledge about utilized models
of the remote nodes. By means of simple examples, we investigate
the effects of incorrectly assumed models and trace the errors back
to a bias, which we derive in closed form. We propose an extension
to the exact Track-to-Track fusion algorithm that corrects the bias
after arbitrarily many time steps. This new approach yields optimal
results when the assumptions about the measurement models are correct
and otherwise still provides the exact value for the mean-squared-error
matrix. The performance of this algorithm is demonstrated and applications
are presented that, e.g.,~allow the employment of nonlinear filter
methods.},
 address = {Singapore},
 author = {Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Reinhardt.pdf},
 title = {On Optimal Distributed Kalman Filtering in Non-ideal Situations},
 year = {2012}
}

@inproceedings{Fusion12_Reinhardt-FastCI,
 abstract = {The fusion under unknown correlations is an important technique in
sensor-network information processing as the cross-correlations between
different estimates remain often unknown to the nodes. Covariance
intersection is a wide-spread and efficient algorithm to fuse estimates
under such uncertain conditions. Although different optimization
criteria have been developed, the trace or determinant minimization
of the fused covariance matrix seems to be most meaningful. However,
this minimization requires numeric solutions of a convex optimization
problem. We derive an algorithm to reduce this nonlinear optimization
to the well-known polynomial root-finding problem. This allows us
to present closed-form solutions for the determinant criterion when
the dimension of the occurring covariance matrices is at most~4 and
for the trace criterion when the dimension of the covariance matrices
is at most~3. We demonstrate the effectiveness of the approach by
means of a speed evaluation.},
 address = {Singapore},
 author = {Marc Reinhardt and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 15th International Conference on Information Fusion (Fusion 2012)},
 month = {July},
 pdf = {Fusion12_Reinhardt-FastCI.pdf},
 title = {Closed-form Optimization of Covariance Intersection for Low-dimensional Matrices},
 year = {2012}
}

@inproceedings{ACC12_Hekler,
 abstract = {The main problem of stochastic nonlinear model predictive control
(SNMPC) is that the equations for state prediction and calculation
of the expected reward are in general not solvable in closed form.
A popular approach is to approximate the occurring continuous probability
density functions by a discrete density representation, which allows
an analytical solution of the SNMPC equations. In this paper, we
propose to draw the samples not randomly as in Monte Carlo based
methods, but systematically by minimizing a distance measure. In
doing so, fewer components are generally required to represent the
underlying probability density while achieving the same approximation
quality. Especially if the evaluation of the expected reward is computationally
expensive, this property affects the complexity of computation significantly.
By means of a path planning problem, we have substantiated this statement
with several simulation runs.},
 address = {Montréal, Canada},
 author = {Achim Hekler and Christof Chlebek and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2012 American Control Conference (ACC 2012)},
 month = {June},
 pdf = {ACC12_Hekler.pdf},
 title = {Open-Loop Feedback Control of Nonlinear Stochastic Systems Based on Deterministic Dirac Mixture Densities},
 year = {2012}
}

@inproceedings{ACC12_Lyons,
 abstract = {We consider stochastic model predictive control of a multi-agent systems
with constraints on the probabilities of inter-agent collisions.
First, we propose a method based on sample average approximation
of the collision probabilities to make the stochastic control problem
computationally tractable. Its approximation error vanishes for fixed
control inputs as the number of samples goes to infinity. However,
empirical results indicate that the complexity of the resulting optimization
problem can be too high to be solved under under real-time requirements.
To alleviate the computational burden we propose a second approach
that uses probabilistic bounds to determine regions of increased
probability of presence for each agent and introduce constraints
for the control problem prohibiting overlap of these regions. We
prove that the resulting problem is conservative for the original
problem, i.e., every control strategy that is feasible under our
new constraints will automatically be feasible for the true original
problem. Furthermore, we present simulations demonstrating better
run-time performance of our second approach compared to the sample
average approximation. Finally, we empirically show that our approach
is better suited than robust control approaches for situations in
which the systems under control are affected by stochastic disturbances.},
 address = {Montréal, Canada},
 author = {Daniel Lyons and Jan-Peter Calliess and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2012 American Control Conference (ACC 2012)},
 month = {June},
 pdf = {ACC12_Lyons.pdf},
 title = {Chance Constrained Model Predictive Control for Multi-Agent Systems with Coupling Constraints},
 year = {2012}
}

@article{arXiv12_Hekler,
 abstract = {In this paper, we address the problem of controlling a system over
an unreliable connection that is affected by time-varying delays
and randomly occurring packet losses. A novel sequence-based approach
is proposed that extends a given controller designed without consideration
of the network-induced disturbances. Its key idea is to model the
unknown future control inputs by random variables, the so-called
virtual control inputs, which are characterized by discrete probability
density functions. Subject to this probabilistic description, the
actual sequence of future control inputs is determined and transmitted
to the actuator. The high performance of the proposed approach is
demonstrated by means of Monte Carlo simulation runs with an inverted
pendulum on a cart and by a detailed comparison to standard NCS approaches.},
 author = {Achim Hekler and Jörg Fischer and Uwe D. Hanebeck},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {June},
 title = {Sequence-Based Control for Networked Control Systems Based on Virtual
Control Inputs},
 url = {https://arxiv.org/abs/1206.0549},
 year = {2012}
}

@inproceedings{SPIE12_FischerBaum,
 abstract = {In this article we present an evaluation of different target tracking
methods based on various simulated scenarios in the maritime domain.
We implemented well known algorithms (JIPDA, Linear Multi Target
PDA, Linear Joint PDA, Monte Carlo Markov Chain Data Association)
and integrated them into a data fusion architecture. The algorithms
have been compared based on extensions of the Optimal Subpattern
Assignment metric. Also further performance measures are used to
get a single score for each algorithm. As no single algorithm is
equally well fitted to all tested scenarios, our results show which
algorithms fits best for specific scenarios.},
 address = {Baltimore, Maryland, USA},
 author = {Yvonne Fischer and Marcus Baum and Fabian Flohr and Uwe D. Hanebeck and Jürgen Beyerer},
 booktitle = {Signal Processing, Sensor Fusion, and Target Recognition XXI (Proceedings of SPIE)},
 doi = {10.1117/12.919234},
 month = {April},
 title = {Evaluation of Tracking Methods for Maritime Surveillance },
 url = {https://dx.doi.org/10.1117/12.919234},
 year = {2012}
}

@article{arXiv12_Deisenroth,
 abstract = {We propose a principled algorithm for robust Bayesian filtering and
smoothing in nonlinear stochastic dynamic systems when both the transition
function and the measurement function are described by non-parametric
Gaussian process (GP) models. GPs are gaining increasing importance
in signal processing, machine learning, robotics, and control for
representing unknown system functions by posterior probability distributions.
This modern way of "system identification" is more robust than finding
point estimates of a parametric function representation. In this
article, we present a principled algorithm for robust analytic smoothing
in GP dynamic systems, which are increasingly used in robotics and
control. Our numerical evaluations demonstrate the robustness of
the proposed approach in situations where other state-of-the-art
Gaussian filters and smoothers can fail.},
 author = {Marc Peter Deisenroth and Ryan Turner and Marco F. Huber and Uwe D. Hanebeck and Carl Edward Rasmussen},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {March},
 title = {Robust Filtering and Smoothing with Gaussian Processes},
 url = {https://arxiv.org/abs/1203.4345},
 year = {2012}
}

@article{arXiv12_Hanebeck,
 abstract = {In this paper, we propose a progressive Bayesian procedure, where
the measurement information is continuously included into the given
prior estimate (although we perform observations at discrete time
steps). The key idea is to derive a system of ordinary first-order
differential equations (ODE) by employing a new coupled density representation
comprising a Gaussian density and its Dirac Mixture approximation.
The ODE is used for continuously tracking the true non-Gaussian posterior
by its best matching Gaussian approximation. The performance of the
new filter is evaluated in comparison with state-of-the-art filters
by means of a canonical benchmark example, the discrete-time cubic
sensor problem.},
 author = {Uwe D. Hanebeck and Jannik Steinbring},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {March},
 title = {Progressive Gaussian Filtering},
 url = {https://arxiv.org/abs/1204.0133},
 year = {2012}
}

@inproceedings{TRB12_Kretz,
 abstract = {In this article a new way to collect data of pedestrian dynamics is
introduced. A virtual reality system consisting of an extended range
telepresence system and a microscopic pedestrian simulation is used
to simplify data collection. The extended range telepresence system
allows a user to move through a virtual environment by natural walking
instead of by using conventional input devices, like a joystick.
The telepresence system is connected to a pedestrian simulation which
produces real time 3D animated output which is presented to the user
with a head-mounted display (HMD) capable of showing 3D imagery.
The simulated pedestrians react to the user of the telepresence as
if it were another simulated pedestrian. With this system data about
pedestrian dynamics can be collected in experiments in which not
all participants need to be real people, but some - ideally all except
for one - can be simulated. This allows the general collection of
data about pedestrian dynamics but also to calibrate model specific
parameters. In this paper three experiments are introduced. However,
the focus of the contribution is to give an idea and an overview
of the combined telepresence-simulation system as data collection
tool.},
 address = {Washington D.C., USA},
 author = {Tobias Kretz and Antonia Pérez Arias and Simon Friedberger and Uwe D. Hanebeck},
 booktitle = {Proceedings of the Transportation Research Board 91st Annual Meeting (TRB 2012)},
 month = {January},
 pdf = {TRB12_Kretz.pdf},
 title = {Using Extended Range Telepresence to Collect Data of Pedestrian Dynamics},
 year = {2012}
}

@inproceedings{CDC11_Baum,
 abstract = {This paper presents a novel method for tracking multiple extended
objects. The shape of a single extended object is modeled with a
recently developed approach called Random Hypersurface Model (RHM)
that assumes a varying number of measurement sources to lie on scaled
versions of the shape boundaries. This approach is extended by introducing
a so-called Mixture Random Hypersurface Model (Mixture RHM), which
allows for modeling multiple extended targets. Based on this model,
a Gaussian-assumed Bayesian tracking method that provides the means
to track and estimate shapes of multiple extended targets is derived.
Simulations demonstrate the performance of the new approach.},
 address = {Orlando, Florida, USA},
 author = {Marcus Baum and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 50th IEEE Conference on Decision and Control (CDC 2011)},
 month = {December},
 pdf = {CDC11_Baum.pdf},
 title = {Random Hypersurface Mixture Models for Tracking Multiple Extended Objects},
 year = {2011}
}

@inproceedings{CDC11_Hekler,
 abstract = {In model predictive control, a high quality of control can only be
achieved, if the model of the system reflects the real-world process
as precisely as possible. Therefore, the controller should be capable
of both handling a nonlinear system description and systematically
incorporating uncertainties affecting the system. Since stochastic
nonlinear model predictive control (SNMPC) problems in general cannot
be solved in closed form, either the system model or the occurring
densities have to be approximated. In this paper, we present an SNMPC
framework, which approximates the densities and the reward function
by their wavelet expansions. Due to the few requirements on the shape
and family of the densities or reward function, the presented technique
can be applied to a large class of SNMPC problems. For accelerating
the optimization, we additionally present a novel thresholding technique,
the so-called dynamic thresholding, which neglects coefficients that
are insignificant, while at the same time guaranteeing that the optimal
control input is still chosen. The capabilities of the proposed approach
are demonstrated by simulations with a path planning scenario.},
 address = {Orlando, Florida, USA},
 author = {Achim Hekler and Martin Kiefel and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 50th IEEE Conference on Decision and Control (CDC 2011)},
 month = {December},
 pdf = {CDC11_Hekler.pdf},
 title = {Stochastic Nonlinear Model Predictive Control with Guaranteed Error Bounds Using Compactly Supported Wavelets},
 year = {2011}
}

@inproceedings{ICCV11_Kretz,
 abstract = {In this contribution we present the results of a pilot study in which
an Extended Range Telepresence System is used to calibrate parameters
of a pedestrian model for simulation. The parameters control a model
element that is intended to make simulated agents walk in the direction
of the esti- mated smallest remaining travel time. We use this to,
first, show that that an Extended Range Telepresence System can serve
for such a task in general and second to actually find simulation
parameters that yield realistic results.},
 address = {Barcelona, Spain},
 author = {Tobias Kretz and Stefan Hengst and Vidal Roca and Antonia Pérez Arias and Simon Friedberger and Uwe D. Hanebeck},
 booktitle = {Proceedings of the first IEEE Workshop on Modeling, Simulation and Visual Analysis of Large Crowds in conjunction with the 13th International Conference on Computer Vision (ICCV 2011)},
 month = {November},
 pdf = {ICCV11_Kretz.pdf},
 title = {Calibrating Dynamic Pedestrian Route Choice with an Extended Range Telepresence System},
 year = {2011}
}

@article{arXiv11_Kretz,
 abstract = {A combination of a telepresence system and a microscopic traffic simulator
is introduced. It is evaluated using a hotel evacuation scenario.
Four different kinds of supporting information are compared, standard
exit signs, floor plans with indicated exit routes, guiding lines
on the floor and simulated agents leading the way. The results indicate
that guiding lines are the most efficient way to support an evacuation
but the natural behavior of following others comes very close. On
another level the results are consistent with previously performed
real and virtual experiments and validate the use of a telepresence
system in evacuation studies. It is shown that using a microscopic
traffic simulator extends the possibilities for evaluation, e.g.
by adding simulated humans to the environment.},
 author = {Tobias Kretz and Stefan Hengst and Antonia Pérez Arias and
Simon Friedberger and Uwe D. Hanebeck},
 journal = {arXiv preprint: Human-Computer Interaction (cs.HC)},
 month = {November},
 title = {Using a Telepresence System to Investigate Route Choice Behavior},
 url = {https://arxiv.org/abs/1111.1103},
 year = {2011}
}

@inproceedings{Humanoids11_Rybok,
 abstract = {Human action and activity recognition from videos has attracted an
increasing number of researchers in recent years. However, most of
the works aim at multimedia retrieval and surveillance applications,
but rarely at humanoid household robots, even though the robotic
perception of human activities would allow a more natural human-robot
interaction (HRI). To encourage future studies in this domain, we
present in this work a novel data set specifically designed for the
application in HRI scenarios. This Robo-kitchen data set consists
of 14 typical kitchen activities recorded in two different stereo-camera
setups, and each performed by 17 subjects. To establish a baseline
for future work, we extend a state-of-the-art action recognition
method to be applicable on the activity classification problem and
evaluate it on the Robo-kitchen data set showing promising results.},
 address = {Bled, Slovenia},
 author = {Lukas Rybok and Simon Friedberger and Uwe D. Hanebeck and Rainer Stiefelhagen},
 booktitle = {Proceedings of the 2011 11th IEEE-RAS International Conference on Humanoid Robots (Humanoids 2011)},
 month = {October},
 pdf = {Humanoids11_Rybok.pdf},
 title = {The KIT Robo-Kitchen Data set for the Evaluation of View-based Activity Recognition Systems},
 year = {2011}
}

@inproceedings{SDF11_Huber,
 abstract = {Many information fusion tasks involve the processing of Gaussian mixtures
with simple underlying shape, but many components. This paper addresses
the problem of reducing the number of components, allowing for faster
density processing. The proposed approach is based on identifying
components irrelevant for the overall density's shape by means of
the curvature of the density's surface. The key idea is to minimize
an upper bound of the curvature while maintaining a low global reduction
error by optimizing the weights of the original Gaussian mixture
only. The mixture is reduced by assigning zero weights to reducible
components. The main advantages are an alleviation of the model selection
problem, as the number of components is chosen by the algorithm automatically,
the derivation of simple curvature-based penalty terms, and an easy,
efficient implementation. A series of experiments shows the approach
to provide a good trade-off between quality and sparsity.},
 address = {Berlin, Germany},
 author = {Marco Huber and Peter Krauthausen and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2011)},
 month = {October},
 pdf = {SDF11_Huber.pdf},
 title = {Superficial Gaussian Mixture Reduction},
 year = {2011}
}

@inproceedings{IROS11_Baum,
 abstract = {Fitting conic sections, e.g., ellipses or circles, to noisy data points
is a fundamental sensor data processing problem, which frequently
arises in robotics. In this paper, we introduce a new procedure for
deriving a recursive Gaussian state estimator for fitting conics
to data corrupted by additive Gaussian noise. For this purpose, the
original exact implicit measurement equation is reformulated with
the help of suitable approximations as an explicit measurement equation
corrupted by multiplicative noise. Based on stochastic linearization,
an efficient Gaussian state estimator is derived for the explicit
measurement equation. The performance of the new approach is evaluated
by means of a typical ellipse fitting scenario.},
 address = {San Francisco, California, USA},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011)},
 month = {September},
 pdf = {IROS11_Baum.pdf},
 title = {Fitting Conics to Noisy Data Using Stochastic Linearization},
 year = {2011}
}

@inproceedings{IROS11_Gehrig,
 abstract = {In this paper, a multi-level approach to intention, activity, and
motion recognition for a humanoid robot is proposed. Our system processes
images from a monocular camera and combines this information with
domain knowledge. The recognition works on-line and in real-time,
it is independent of the test person, but limited to predefined view-points.
Main contributions of this paper are the extensible, multi-level
modeling of the robot's vision system, the efficient activity and
motion recognition, and the asynchronous information fusion based
on generic processing of mid-level recognition results. The complementarity
of the activity and motion recognition renders the approach robust
against misclassifications. Experimental results on a real-world
data set of complex kitchen tasks, e.g., Prepare Cereals or Lay Table,
prove the performance and robustness of the multi-level recognition
approach.},
 address = {San Francisco, California, USA},
 author = {Dirk Gehrig and Peter Krauthausen and Lukas Rybok and Hildegard Kühne and Tanja Schultz and Uwe D. Hanebeck and Rainer Stiefelhagen},
 booktitle = {Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011)},
 month = {September},
 pdf = {IROS11_Gehrig.pdf},
 title = {Combined Intention, Activity, and Motion Recognition for a Humanoid Household Robot},
 year = {2011}
}

@inproceedings{IROS11_Perez,
 abstract = {This paper presents the control concept of a semimobile haptic interface
for extended range telepresence that enables the user to explore
spatially unrestricted target environments even from a small user
environment. The semi-mobile haptic interface consists of a haptic
manipulator mounted on a large grounded Cartesian robot, the prepositioning
unit. The prepositioning unit is controlled in such a way that the
haptic manipulator is kept off its workspace limits. At the same
time, the control algorithm allows the optimal utilization of the
available space in the user environment and guarantees the safety
of the user. The proposed control method is based on the position
and velocity of the end-effector and also takes the position of the
user into account. Moreover, it is robust against noisy measurements
of the user position or outliers due, for example, to occlusions
in the tracking system. Experimental results show the suitability
of the proposed control to provide haptic interaction in extended
range telepresence.},
 address = {San Francisco, California, USA},
 author = {Antonia Pérez Arias and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011)},
 month = {September},
 pdf = {IROS11_Perez.pdf},
 title = {Motion Control of a Semi-mobile Haptic Interface for Extended Range Telepresence},
 year = {2011}
}

@inproceedings{TGF11_Kretz,
 abstract = {A combination of a telepresence system and a microscopic traffic simulator
is introduced. It is evaluated using a hotel evacuation scenario.
Four different kinds of supporting information are compared, standard
exit signs, floor plans with indicated exit routes, guiding lines
on the floor and simulated agents leading the way. The results indicate
that guiding lines are the most efficient way to support an evacuation
but the natural behavior of following others comes very close. On
another level the results are consistent with previously performed
real and virtual experiments and validate the use of a telepresence
system in evacuation studies. It is shown that using a microscopic
traffic simulator extends the possibilities for evaluation, e.g.
by adding simulated humans to the environment.},
 address = {Moscow, Russia},
 author = {Tobias Kretz and Stefan Hengst and Antonia Pérez Arias and Simon Friedberger and Uwe D. Hanebeck},
 booktitle = {Proceedings of the Traffic and Granular Flow Conference 2011 (TGF 2011)},
 month = {September},
 pdf = {TGF11_Kretz.pdf},
 title = {Using Extended Range Telepresence to Investigate Route Choice Behavior},
 year = {2011}
}

@inproceedings{IFAC11_Huber,
 abstract = {In nonlinear filtering, special types of Gaussian mixture filters
are a straightforward extension of Gaussian filters, where linearizing
the system model is performed individually for each Gaussian component.
In this paper, two novel types of linearization are combined with
Gaussian mixture filters. The first linearization is called analytic
stochastic linearization, where the linearization is performed analytically
and exactly, i.e., without Taylor-series expansion or approximate
sample-based density representation. In cases where a full analytical
linearization is not possible, the second approach decomposes the
nonlinear system into a set of nonlinear subsystems that are conditionally
integrable in closed form. These approaches are more accurate than
fully applying classical linearization.},
 address = {Milan, Italy},
 author = {Marco F. Huber and Frederik Beutler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th IFAC World Congress (IFAC 2011)},
 month = {August},
 pdf = {IFAC11_Huber.pdf},
 title = {(Semi-)Analytic Gaussian Mixture Filter},
 year = {2011}
}

@inproceedings{IFAC11_Noack,
 abstract = {Especially in the field of sensor networks and multi-robot systems,
fully decentralized estimation techniques are of particular interest.
As the required elimination of the complex dependencies between estimates
generally yields inconsistent results, several approaches, e.g.,
covariance intersection, maintain consistency by providing conservative
estimates. Unfortunately, these estimates are often too conservative
and therefore, much less informative than a corresponding centralized
approach. In this paper, we provide a concept that conservatively
decorrelates the estimates while bounding the unknown correlations
as closely as possible. For this purpose, known independent quantities,
such as measurement noise, are explicitly identified and exploited.
Based on tight covariance bounds, the new approach allows for an
intuitive and systematic derivation of appropriate tailor-made filter
equations and does not require heuristics. Its performance is demonstrated
in a comparative study within a typical SLAM scenario.},
 address = {Milan, Italy},
 author = {Benjamin Noack and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 18th IFAC World Congress (IFAC 2011)},
 month = {August},
 pdf = {IFAC11_Noack.pdf},
 title = {Automatic Exploitation of Independencies for Covariance Bounding in Fully Decentralized Estimation},
 year = {2011}
}

@inproceedings{Fusion11_Baum-RHM,
 abstract = {This paper is about tracking an extended object or a group target,
which gives rise to a varying number of measurements from different
measurement sources. For this purpose, the shape of the target is
tracked in addition to its kinematics. The target extent is modeled
with a new approach called Random Hypersurface Model (RHM) that assumes
varying measurement sources to lie on scaled versions of the shape
boundaries. In this paper, a star-convex RHM is introduced for tracking
star-convex shape approximations of targets. Bayesian inference for
star-convex RHM is performed by means of a Gaussian-assumed state
estimator allowing for an efficient recursive closed-form measurement
update. Simulations demonstrate the performance of this approach
for typical extended object and group tracking scenarios.},
 address = {Chicago, Illinois, USA},
 annote = {Winner Best Student Paper Award Certificate (PDF)},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Baum-RHM.pdf},
 title = {Shape Tracking of Extended Objects and Group Targets with Star-Convex RHMs},
 year = {2011}
}

@inproceedings{Fusion11_Baum-USF,
 abstract = {This paper is about the use of symmetric state transformations for
multi-target tracking. First, a novel method for obtaining point
estimates for multi-target states is proposed. The basic idea is
to apply a symmetric state transformation to the original state in
order to compute a minimum mean-square-error (MMSE) estimate in a
transformed state. By this means, the known shortcomings of MMSE
estimates for multi-target states can be avoided. Second, a new multi-target
tracking method based on state transformations is suggested, which
entirely performs the time and measurement update in a transformed
space and thus, avoids the explicit calculation of data association
hypotheses and removes the target identity from the estimation problem.
The performance of the new approach is evaluated by means of tracking
two crossing targets.},
 address = {Chicago, Illinois, USA},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Baum-USF.pdf},
 title = {Using Symmetric State Transformations for Multi-Target Tracking},
 year = {2011}
}

@inproceedings{Fusion11_Baum,
 abstract = {This paper is about tracking multiple targets with the so-called Symmetric
Measurement Equation (SME) filter. The SME filter uses symmetric
functions, e.g., symmetric polynomials, in order to remove the data
association uncertainty from the measurement equation. By this means,
the data association problem is converted to a nonlinear state estimation
problem. In this work, an efficient optimal Gaussian filter based
on analytic moment calculation for discrete-time multi-dimensional
polynomial systems corrupted with Gaussian noise is derived, and
then applied to the polynomial system resulting from the SME filter.
The performance of the new method is compared to an UKF implementation
by means of typical multiple target tracking scenarios.},
 address = {Chicago, Illinois, USA},
 author = {Marcus Baum and Benjamin Noack and Frederik Beutler and Dominik Itte and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Baum.pdf},
 title = {Optimal Gaussian Filtering for Polynomial Systems Applied to Association-free Multi-Target Tracking},
 year = {2011}
}

@inproceedings{Fusion11_Bogatyrenko,
 abstract = {Visual stabilization proposed in this paper compensates changes of
the scene caused by motion and deformation of an observed object.
This is of high importance in computer-assisted beating heart surgery,
where the views of the beating heart should be stabilized. The proposed
model-based method defines visual stabilization as a transformation
of the current image sequence to a stabilized image sequence. This
transformation incorporates physical model of the observed object
and model of the measurement process. In contrast to standard approaches,
the quality of the visual stabilization is continuously evaluated
and improved in two aspects. On the one hand, discretization errors
are reduced. On the other hand, the parameters of the underlying
models are adjusted. The performance of the proposed method is evaluated
in an experiment with a pressure-regulated artificial heart. Compared
with standard methods, the model-based method provides higher accuracy,
which is additionally improved by a feedback mechanism.},
 address = {Chicago, Illinois, USA},
 author = {Evgeniya Bogatyrenko and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Bogatyrenko.pdf},
 title = {Adaptive Model-Based Visual Stabilization of Image Sequences Using Feedback},
 year = {2011}
}

@inproceedings{Fusion11_Krauthausen,
 abstract = {In this paper, the estimation of conditional densities between continuous
random variables from noisy samples is considered. The conditional
densities are modeled as heteroscedastic Gaussian mixture densities
allowing for closed-form solution of Bayesian inference with full-densities.
The main contributions of this paper are an improved generalization
quality of the estimates by the introduction of a superficial regularizer,
the consideration of model uncertainty relative to local data densities
by means of adaptive covariances, and the proposition of an efficient
distance-based estimation algorithm. This algorithm corresponds to
an iterative nested optimization scheme, optimizing hyper-parameters,
component placement, and mixture weights. The obtained solutions
are sparse, smooth, and generalize well as benchmark experiments,
e.g., in nonlinear filtering show.},
 address = {Chicago, Illinois, USA},
 author = {Peter Krauthausen and Patrick Ruoff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Krauthausen.pdf},
 title = {Sparse Mixture Conditional Density Estimation by Superficial Regularization},
 year = {2011}
}

@inproceedings{Fusion11_Noack,
 abstract = {Many modern fusion architectures are designed to process and fuse
data in networked systems. Alongside the advantages, such as scalability and
robustness, distributed fusion techniques particularly have to tackle the problem of dependencies
between locally processed data. In linear estimation problems, uncertain
quantities with unknown cross-correlations can be fused by means
of the covariance intersection algorithm, which avoids overconfident
fusion results. However, for nonlinear system dynamics and sensor
models perturbed by arbitrary noise, it is not only a problem to
characterize and parameterize dependencies between estimates, but
also to find a proper notion of consistency. This paper addresses
these issues by transforming the state estimates to a different state
space, where the corresponding densities are Gaussian and only linear
dependencies between estimates, i.e., correlations, can arise. These
pseudo Gaussian densities then allow the notion of covariance consistency
to be used in distributed nonlinear state estimation.},
 address = {Chicago, Illinois, USA},
 author = {Benjamin Noack and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Noack.pdf},
 title = {Covariance Intersection in Nonlinear Estimation Based on Pseudo Gaussian Densities},
 year = {2011}
}

@inproceedings{Fusion11_Reinhardt,
 abstract = {In data fusion theory, multiple estimates are combined to yield an
optimal result. In this paper, the set of all possible results is
investigated, when two random variables with unknown correlations
are fused. As a first step, recursive processing of the set of estimates
is examined. Besides set-theoretic considerations, the lack of knowledge
about the unknown correlation coefficient is modeled as a stochastic
quantity. Especially, a uniform model is analyzed, which provides
a new optimization criterion for the covariance intersection algorithm
in scalar state spaces. This approach is also generalized to multi-dimensional
state spaces in an approximative, but fast and scalable way, so that
consistent estimates are obtained.},
 address = {Chicago, Illinois, USA},
 author = {Marc Reinhardt and Benjamin Noack and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Reinhardt.pdf},
 title = {Analysis of Set-theoretic and Stochastic Models for Fusion under Unknown Correlations},
 year = {2011}
}

@inproceedings{Fusion11_Ruoff,
 abstract = {Since the advent of Monte-Carlo particle filtering, particle representations
of densities have become increasingly popular due to their flexibility
and implicit adaptive resolution. In this paper, an algorithm for
the multiplication of a systematic Dirac mixture (DM) approximation
with a continuous likelihood function is presented, which applies
a progressive correction scheme, in order to avoid the particle degeneration
problem. The preservation of sample regularity and therefore, representation
quality of the underlying smooth density, is ensured by including
a new measure of smoothness for Dirac mixtures, the DM energy, into
the distance measure. A comparison to common correction schemes in
Monte-Carlo methods reveals large improvements especially in cases
of small overlap between the likelihood and prior density, as well
as for multi-modal likelihoods.},
 address = {Chicago, Illinois, USA},
 author = {Patrick Ruoff and Peter Krauthausen and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 14th International Conference on Information Fusion (Fusion 2011)},
 month = {July},
 pdf = {Fusion11_Ruoff.pdf},
 title = {Progressive Correction for Deterministic Dirac Mixture Approximations},
 year = {2011}
}

@inproceedings{ACC11_Bogatyrenko,
 abstract = {In order to assist a surgeon by operating on a beating heart, visual
stabilization makes the beating heart appear still to a surgeon by
providing the current heart view as stationary and non-moving. In
this way, the surgeon is not disturbed during an operation by a motion
of the heart and can get an impression of performing conventional
surgery. In contrast to existing methods for visual stabilization,
the proposed approach involves a model-based transformation of image
sequences provided by a camera system. This transformation incorporates
the knowledge of physical characteristics of the heart in form of
a mathematical model of the heart surface. Its main advantage is
that the uncertainties of the model and measurements are considered.
This occurs by estimating the parameters of the transformation. Furthermore,
the quality of the visual stabilization is additionally improved
by adapting the parameters of the underlying physical model. A performance
of the proposed approach is evaluated in an experiment with a pressure-regulated
artificial heart. In comparison to standard approaches, it provides
superior results illustrating the high quality of the visual stabilization.},
 address = {San Francisco, California, USA},
 author = {Evgeniya Bogatyrenko and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2011 American Control Conference (ACC 2011)},
 month = {June},
 pdf = {ACC11_Bogatyrenko.pdf},
 title = {Visual Stabilization of Beating Heart Motion by Model-based Transformation of Image Sequences},
 year = {2011}
}

@inproceedings{ACC11_Huber,
 abstract = {For Gaussian Assumed Density Filtering based on moment matching, a
framework for the efficient calculation of posterior moments is proposed
that exploits the structure of the given nonlinear system. The key
idea is a careful discretization of some dimensions of the state
space only in order to decompose the system into a set of nonlinear
subsystems that are conditionally integrable in closed form. This
approach is more efficient than full discretization approaches. In
addition, the new decomposition is far more general than known Rao-Blackwellization
approaches relying on conditionally linear subsystems. As a result,
the new framework is applicable to a much larger class of nonlinear
systems.},
 address = {San Francisco, California, USA},
 author = {Marco F. Huber and Frederik Beutler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2011 American Control Conference (ACC 2011)},
 month = {June},
 pdf = {ACC11_Huber.pdf},
 title = {Semi-Analytic Gaussian Assumed Density Filter},
 year = {2011}
}

@inproceedings{ACC11_Krauthausen,
 abstract = {In this paper, the problem of sparse nonparametric conditional density
estimation based on samples and prior knowledge is addressed. The
prior knowledge may be restricted to parts of the state space and
given as generative models in form of mean-function constraints or
as probabilistic models in the form of Gaussian mixtures. The key
idea is the introduction of additional constraints and a modified
kernel function into the conditional density estimation problem.
This approach to using prior knowledge is a generic solution applicable
to all nonparametric conditional density estimation approaches phrased
as constrained optimization problems. The quality of the estimates,
their sparseness, and the achievable improvements by using prior
knowledge are shown in experiments for both Support-Vector Machine-based
and integral distance-based conditional density estimation.},
 address = {San Francisco, California, USA},
 author = {Peter Krauthausen and Masoud Roschani and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2011 American Control Conference (ACC 2011)},
 month = {June},
 pdf = {ACC11_Krauthausen.pdf},
 title = {Incorporating Prior Knowledge into Nonparametric Conditional Density Estimation},
 year = {2011}
}

@inproceedings{ACC11_Noack,
 abstract = {The information filter has evolved into a key tool for distributed
and decentralized multisensor estimation and control. Essentially,
it is an algebraical reformulation of the Kalman filter and provides
estimates on the information about an uncertain state rather than
on a state itself. Whereas many practicable Kalman filtering techniques
for nonlinear system and sensor models have been developed, approaches
towards nonlinear information filtering are still scarce and limited.
In order to deal with nonlinear systems and sensors, this paper derives
an approximation technique for arbitrary probability densities that
provides the same distributable fusion structure as the linear information
filter. The presented approach not only constitutes a nonlinear version
of the information filter, but it also points the direction to a
Hilbert space structure on probability densities, whose vector space
operations correspond to the fusion and weighting of information.},
 address = {San Francisco, California, USA},
 author = {Benjamin Noack and Daniel Lyons and Matthias Nagel and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2011 American Control Conference (ACC 2011)},
 month = {June},
 pdf = {ACC11_Noack.pdf},
 title = {Nonlinear Information Filtering for Distributed Multisensor Data Fusion},
 year = {2011}
}

@inproceedings{WHC11_Perez,
 abstract = {In this work, we propose a plenhaptic guidance function that systematically
describes the haptic information for guiding the user in the target
environment. The plenhaptic guidance function defines the strength
of the guidance at any position in space, at any direction, and at
any time, and takes the geometry of the target environment as well
as all possible goals into account. The plenhaptic guidance function,
which can be rendered as active and passive guidance, is sampled
and displayed to the user through a haptic interface in the user
environment. The benefits of the plenhaptic guidance function for
guiding the user to several simultaneous goals while avoiding the
obstacles in a large target environment are demonstrated in real
experiments.},
 address = {Istanbul, Turkey},
 author = {Antonia Pérez Arias and Henning P. Eberhardt and Florian Pfaff and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE World Haptics Conference (WHC 2011)},
 month = {June},
 pdf = {WHC11_Perez.pdf},
 title = {The Plenhaptic Guidance Function for Intuitive Navigation in Extended Range Telepresence Scenarios},
 year = {2011}
}

@inproceedings{ARMS_AAMAS11_Lyons,
 abstract = {We present an auction-flavored multi-robot planning mechanism where
coordination is to be achieved on the occupation of atomic resources
modeled as binary inter-robot constraints. Introducing virtual obstacles,
we show how this approach can be combined with particlebased obstacle
avoidance methods, offering a decentralized, auction-based alternative
to previously established centralized approaches for multirobot open-loop
control. We illustrate the effectiveness of our new approach by presenting
simulations of typical spatially-continuous multirobot path-planning
problems and derive bounds on the collision probability in the presence
of uncertainty.},
 address = {Taipei, Taiwan},
 author = {Jan-P. Calliess and Daniel Lyons and Uwe D. Hanebeck},
 booktitle = {Autonomous Robots and Multirobot Systems (ARMS) Workshop at AAMAS 2011},
 month = {May},
 pdf = {ARMS11_CalliessLyons.pdf},
 title = {Lazy auctions for multi-robot collision avoidance and motion control under uncertainty},
 year = {2011}
}

@article{arXiv11_Lyons,
 abstract = {We consider stochastic model predictive control of a multi-agent systems
with constraints on the probabilities of inter-agent collisions.
We first study a sample-based approximation of the collision probabilities
and use this approximation to formulate constraints for the stochastic
control problem. This approximation will converge as the number of
samples goes to infinity, however, the complexity of the resulting
control problem is so high that this approach proves unsuitable for
control under real-time requirements. To alleviate the computational
burden we propose a second approach that uses probabilistic bounds
to determine regions with increased probability of presence for each
agent and formulate constraints for the control problem that guarantee
that these regions will not overlap. We prove that the resulting
problem is conservative for the original problem with probabilistic
constraints, ie. every control strategy that is feasible under our
new constraints will automatically be feasible for the original problem.
Furthermore we show in simulations in a UAV path planning scenario
that our proposed approach grants significantly better run-time performance
compared to a controller with the sample-based approximation with
only a small degree of sub-optimality resulting from the conservativeness
of our new approach},
 author = {Daniel Lyons and Jan-P. Calliess and Uwe D. Hanebeck},
 ee = {https://arxiv.org/abs/1104.5384},
 journal = {arXiv preprint: Systems and Control (cs.SY)},
 month = {April},
 title = {Chance-constrained Model Predictive Control for Multi-Agent Systems},
 url = {https://arxiv.org/abs/1104.5384},
 year = {2011}
}

@inproceedings{EWSN11_Schmid,
 abstract = {In this paper, the localization of persons by means of a Wireless
Sensor Network (WSN) is considered. Persons carry on-body sensor
nodes and move within a WSN. The location of each person is calculated
on this node and communicated through the network to a central data
sink for visualization. Applications of such a system could be found in
mass casualty events, firefighter scenarios, hospitals or retirement homes for example.
For the location estimation on the sensor node, three derivatives of the
Kalman Filter and a closed-form solution (CFS) are applied, compared,
and evaluated in a real-world scenario. A prototype 65-node ZigBee WSN
is implemented and data are collected in in- and outdoor environments
with differently positioned on-body nodes. The described estimators are
then evaluated off-line on the experimentally collected data.
The goal of this paper is to present a comprehensive real-world evaluation of methods for
person localization in a WSN based on received signal strength (RSS) range measurements.
It is concluded that person localization in in- and outdoor environments is possible
under the considered conditions with the considered filters. The compared methods
allow for suffciently accurate localization results and are robust against
inaccurate range measurements.},
 address = {Bonn, Germany},
 author = {Johannes Schmid and Frederik Beutler and Benjamin Noack and Uwe D. Hanebeck and Klaus D. Müller-Glaser},
 booktitle = {Proceedings of the 8th European Conference on Wireless Sensor Networks (EWSN 2011)},
 doi = {10.1007/978-3-642-19186-2_10},
 editor = {Pedro José Marrón and Kamin Whitehouse},
 month = {February},
 pages = {147--162},
 pdf = {EWSN11_Schmid.pdf},
 publisher = {Springer},
 title = {An Experimental Evaluation of Position Estimation Methods for Person Localization in Wireless Sensor Networks},
 url = {https://dx.doi.org/10.1007/978-3-642-19186-2_10},
 volume = {6567},
 year = {2011}
}

@inproceedings{CDC10_Hekler,
 abstract = {Bayesian estimation for nonlinear systems is still a challenging problem, as
in general the type of the true probability density changes and the
complexity increases over time. Hence, approximations of the occurring
equations and/or of the underlying probability density functions are
inevitable. In this paper, we propose an approximation of the conditional
densities by wavelet expansions. This kind of representation allows a sparse
set of characterizing coefficients, especially for smooth or piecewise
smooth density functions. Besides its good approximation properties, fast
algorithms operating on sparse vectors are applicable and thus, a good
trade-off between approximation quality and run-time can be achieved.
Moreover, due to its highly generic nature, it can be applied to a large
class of nonlinear systems with a high modeling accuracy. In particular, the
noise acting upon the system can be modeled by an arbitrary probability
distribution and can influence the system in any way.},
 address = {Atlanta, Georgia, USA},
 author = {Achim Hekler and Martin Kiefel and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE Conference on Decision and Control (CDC 2010)},
 month = {December},
 pdf = {CDC10_HeklerKiefel.pdf},
 title = {Nonlinear Bayesian Estimation with Compactly Supported Wavelets},
 year = {2010}
}

@article{TM10_Noack,
 abstract = {Die systematische Behandlung von Unsicherheiten stellt eine wesentliche
Herausforderung in der Informationsfusion dar. Einerseits müssen
geeignete Darstellungsformen für die Unsicherheiten bestimmt
werden und andererseits darauf aufbauend effiziente Schätzverfahren
hergeleitet werden. Im Allgemeinen wird zwischen stochastischen und
mengenbasierten Unsicherheitsbeschreibungen unterschieden. Dieser
Beitrag stellt ein Verfahren zur Zustandsschätzung vor, welches
simultan stochastische und mengenbasierte Fehlergrößen berücksichtigen
kann, indem unsichere Größen nicht mehr durch eine einzelne
Wahrscheinlichkeitsdichte, sondern durch eine Menge von Dichten repräsentiert
werden. Besonderes Augenmerk liegt hier auf den Vorteilen und Anwendungsmöglichkeiten
dieser Unsicherheitsbeschreibung.},
 author = {Benjamin Noack and Vesa Klumpp and Daniel Lyons and Uwe D. Hanebeck},
 doi = {10.1524/teme.2010.0087},
 editor = {Klaus-Dieter Sommer and Fernando Puente León and Michael Heizmann},
 journal = {tm -- Technisches Messen, Oldenbourg Verlag},
 month = {October},
 number = {10},
 pages = {544--550},
 pdf = {TM10_Noack.pdf},
 title = {Modellierung von Unsicherheiten und Zustandsschätzung mit Mengen von Wahrscheinlichkeitsdichten},
 url = {https://dx.doi.org/10.1524/teme.2010.0087},
 volume = {77},
 year = {2010}
}

@article{TM10_Kuwertz,
 abstract = {Bewegte Quellen können durch Emission räumlich
ausgedehnte Phänomene wie beispielsweise Schadstoff- oder
Temperaturverteilungen erzeugen. Zur Lokalisierung von Quellen
mit unbekannter Position stehen in vielen Aufgabenstellungen
Informationen nur indirekt durch die verteilte Vermessung des
induzierten Phänomens zur Verfügung - etwa unter Verwendung
stationärer oder mobiler Sensoren. Dieser Beitrag stellt
modellbasierte Verfahren für eine echtzeitfähige Lokalisierung
und Verfolgung von bewegten Quellen vor. Zur gezielten Maximierung
des Informationsgehalts der Messungen wird dabei eine vorausschauende
Sensoreinsatzplanung genutzt, welche eine hohe Lokalisierungsgüte bei
geringem Aufwand ermöglicht.},
 author = {Achim Kuwertz and Marco F. Huber and Felix Sawo and Uwe D. Hanebeck},
 doi = {10.1524/teme.2010.0088},
 editor = {Klaus-Dieter Sommer and Fernando Puente León and Michael Heizmann},
 journal = {tm -- Technisches Messen, Oldenbourg Verlag},
 month = {October},
 number = {10},
 pages = {551--557},
 pdf = {TM10_Kuwertz.pdf},
 title = {Modellbasierte Quellenverfolgung in räumlich ausgedehnten Phänomenen mittels Sensoreinsatzplanung},
 url = {https://dx.doi.org/10.1524/teme.2010.0088},
 volume = {77},
 year = {2010}
}

@inproceedings{IROS10_Beutler,
 abstract = {In range-based localization, the trajectory of a
mobile object is estimated based on noisy range measurements
between the object and known landmarks. In order to deal
with this uncertain information, a Bayesian state estimator
is presented, which exploits optimal stochastic linearization.
Compared to standard state estimators like the Extended
or Unscented Kalman Filter, where a point-based Gaussian
approximation is used, the proposed approach considers the
entire Gaussian density for linearization. By employing the common
assumption that the state and measurements are jointly
Gaussian, the linearization can be calculated in closed form
and thus analytic expressions for the range-based localization
problem can be derived.},
 address = {Taipei, Taiwan},
 author = {Frederik Beutler and Marco F. Huber and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
 month = {October},
 pdf = {IROS10_Beutler.pdf},
 title = {Optimal Stochastic Linearization for Range-based Localization},
 year = {2010}
}

@inproceedings{IROS10_Bogatyrenko,
 abstract = {A reliable estimation of heart surface motion
is an important prerequisite for the synchronization of surgical
instruments in robotic beating heart surgery. In general, only
an imprecise description of the heart dynamics and measurement
systems is available. This means that the estimation of heart
motion is corrupted by stochastic and systematic uncertainties.
Without consideration of these uncertainties, the obtained results
will be inaccurate and a safe robotic operation cannot be guaranteed.
Until now, existing approaches for estimating the motion of the
heart surface are either deterministic or treat only stochastic
uncertainties. The proposed method extends the heart motion
estimation to the simultaneous consideration of stochastic and
unknown but bounded systematic uncertainties. It computes dynamic
bounds in order to provide the surgeon with a guidance by
constraining the motion of the surgical instruments and thereby
protecting sensitive tissue.},
 address = {Taipei, Taiwan},
 author = {Evgeniya Bogatyrenko and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
 month = {October},
 pdf = {IROS10_BogatyrenkoNoack.pdf},
 title = {Reliable Estimation of Heart Surface Motion under Stochastic and Unknown but Bounded Systematic Uncertainties},
 year = {2010}
}

@inproceedings{IROS10_Packi,
 abstract = {Extended range telepresence aims at enabling a
user to experience virtual or remote environments, taking his
own body movements as an input to define walking speed and
viewing direction. Therefore, localization and tracking of the
users pose (position and orientation) is necessary to perform
a body-centered scene rendering. Visual and acoustic feedback
is provided to the user by a head mounted display (HMD).
To allow for free movement within the user environment, the
tracking system is supposed to be user-wearable and entirely
wireless. Consequently, a lightweight design is presented fea-
turing small dimensions to fit into a conventional 13"laptop
backpack, which satisfies the above stated demands for highly
immersive extended range telepresence scenarios. Dedicated
embedded hardware combined with off-the-shelf components
is employed to form a robust, low-cost telepresence system that
can be easily installed in any living room.},
 address = {Taipei, Taiwan},
 author = {Ferdinand Packi and Antonia Pérez Arias and Frederik Beutler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
 month = {October},
 pdf = {IROS10_PackiBeutler.pdf},
 title = {A Wearable System for the Wireless Experience of Extended Range Telepresence},
 year = {2010}
}

@inproceedings{IROS10_Perez,
 abstract = {In this paper, we present a novel use of haptic
information in extended range telepresence, the wide-area haptic
guidance. It consists of force and position signals applied to the
user's hand in order to improve safety, accuracy, and speed in
some telepresent tasks. Wide-area haptic guidance assists the
user in reaching a desired position in a remote environment of
arbitrary size without degrading the feeling of presence. Several
methods for haptic guidance are analyzed. With active haptic
guidance, the user is guided by superimposed forces that pull
him into the desired direction of motion, whereas under passive
haptic guidance, the movement of the user is lightened in the
preferred direction and constrained in the other directions. By
using closed-loop haptic guidance instead of open-loop haptic
guidance, not only is the user guided to his target but also
the deviation from the desired target path is reduced. The
proposed guidance methods were tested with a haptic interface
specifically designed for extended range telepresence.},
 address = {Taipei, Taiwan},
 author = {Antonia Pérez Arias and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
 month = {October},
 pdf = {IROS10_Perez.pdf},
 title = {Wide-Area Haptic Guidance: Taking the User by the Hand},
 year = {2010}
}

@inproceedings{IROS10_Krauthausen,
 abstract = {Estimating a user's intention is central to close
human-robot cooperation. In this paper, the problem of per-
forming intention recognition with tree-structured Dynamic
Bayesian Networks for large environments with many features
is addressed. The proposed approach reduces the computational
complexity of inference O(b^s) for tree-structured measurement
models with an average branching factor b and tree height s
to O((b)s), where b~ << b. The key idea is to switch between a
finite set of reduced system and measurement models in order
to restrict inference to the most important features. A model
predictive approach to online switching between the reduced
models is proposed that exploits an upper bound of the distances
of the reduced models to the full model. The effectiveness of
the proposed algorithm is validated in the intention recognition
for a humanoid robot using a telepresent household scenario.},
 address = {Taipei, Taiwan},
 author = {Peter Krauthausen and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
 month = {October},
 pdf = {IROS10_Krauthausen.pdf},
 title = {A Model-Predictive Switching Approach To Efficient Intention Recognition},
 year = {2010}
}

@inproceedings{SDF10_Baum,
 abstract = {Based on previous work of the authors, this paper provides a comparison
of two different tracking methodologies for extended objects and
group targets, where the true shape of the extent is approximated by
an ellipsoid. Although both methods exploit usual sensor data, i.e.,
position measurements of varying scattering centers, the distinctions
are a consequence of the different modeling of the extent as a symmetric
positive definite random matrix on the one hand and an elliptic random
hypersurface model on the other. Besides analyzing the fundamental
assumptions and a comparison of the properties of these tracking methods,
simulation results are presented based on a static tracking environment
to highlight especially the differences in the update step for the
extension estimate.},
 address = {Leipzig, Germany},
 author = {Marcus Baum and Michael Feldmann and Dietrich Fränken and Uwe D. Hanebeck and Wolfgang Koch},
 booktitle = {Proceedings of the IEEE ISIF Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2010)},
 month = {October},
 pdf = {SDF10_BaumFeldmann.pdf},
 title = {Extended Object and Group Tracking: A Comparison of Random Matrices and Random Hypersurface Models},
 year = {2010}
}

@book{KI10_Hanebeck,
 editor = {Rüdiger Dillmann and Jürgen Beyerer and Uwe D. Hanebeck and Tanja Schultz},
 month = {September},
 publisher = {Springer},
 series = {Lecture Notes in Computer Science / Lecture Notes in Artificial Intelligence},
 title = {KI 2010: Advances in Artificial Intelligence, Proceedings of the 33rd Annual German Conference on AI, Karlsruhe, Germany},
 year = {2010}
}

@inproceedings{KI10_Krauthausen,
 abstract = {Recognizing human intentions is part of the decision
process in many technical devices. In order to achieve
natural interaction, the required estimation quality and
the used computation time need to be balanced. This becomes
challenging, if the number of sensors is high and measurement
systems are complex. In this paper, a model predictive approach
to this problem based on online switching of small,
situation-specific Dynamic Bayesian Networks is proposed.
The contributions are an efficient modeling and inference
of situations and a greedy model predictive switching algorithm
maximizing the mutual information of predicted situations. The
achievable accuracy and computational savings are demonstrated
for a household scenario by using an extended range telepresence system.},
 address = {Karlsruhe, Germany},
 author = {Peter Krauthausen and Uwe D. Hanebeck},
 booktitle = {33rd Annual German Conference on Artificial Intelligence (KI 2010)},
 month = {September},
 pdf = {KI10_Krauthausen.pdf},
 publisher = {Gesellschaft für Informatik e.V.},
 title = {Situation-Specific Intention Recognition for Human-Robot-Cooperation},
 year = {2010}
}

@inproceedings{IPIN10_Packi,
 abstract = {Telepresence systems enable a user to experience
virtual or distant environments by providing sensory feedback.
Appropriate devices include head mounted displays (HMD) for
visual perception, headphones for auditory response, or even
haptic displays for tactile sensation and force feedback. While
most common designs use dedicated input devices like joysticks
or a space mouse, the approach followed in the present work
takes the user's position and viewing direction as an input, as he
walks freely in his local surroundings. This is achieved by using
acoustic tracking, where the user's pose (position and orientation)
is estimated on the basis of ranges measured between a set
of wall-fastened loudspeakers and a microphone array fixed on
the user's HMD. To allow for natural user motion, a wearable,
fully wireless telepresence system is introduced. The increase in
comfort compared to wired solutions is obvious, as the user's
awareness of distracting cables is taken away during walking.
Also the lightweight design and small dimensions contribute to
ergonomics, as the whole assembly fits well into a small backpack.},
 address = {Zürich, Switzerland},
 author = {Ferdinand Packi and Frederik Beutler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Indoor Positioning and Indoor Navigation (IPIN 2010)},
 month = {September},
 pdf = {IPIN10_Packi.pdf},
 title = {Wireless Acoustic Tracking for Extended Range Telepresence},
 year = {2010}
}

@inproceedings{VECIMS10_Baum,
 abstract = {This contribution is about an information management and storage system for modeling the
environment of autonomous systems. The three pillars of the system consist of prior
knowledge, environment model and sensory information. The main pillar is the environment model,
which supplies the autonomous system with relevant information about its current environment.
For this purpose, an abstract representation of the real world is created, where instances
with attributes and relations serve as virtual substitutes of entities (persons and objects)
of the real world. The environment model is created based on sensory information about
the real world. The gathered sensory information is typically uncertain in a stochastic
sense and is represented in the environment model by means of Degree-of-Belief (DoB) distributions.
The prior knowledge contains all relevant background knowledge (e.g., concepts organized in ontologies)
for creating and maintaining the environment model. The concept of the three pillar information system
has previously been published. Therefore this contribution focuses on further central properties
of the system. Furthermore, the development status and possible applications as well as evaluation
scenarios are discussed.},
 address = {Taranto, Italy},
 author = {Ioana Gheta and Marcus Baum and Andrey Belkin and Jürgen Beyerer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Virtual Environments, Human-Computer Interfaces and Measurement Systems (VECIMS 2010)},
 month = {September},
 pdf = {VECIMS10_GhetaBaum.pdf},
 title = {Three Pillar Information Management System for Modeling the Environment of Autonomous Systems},
 year = {2010}
}

@inproceedings{ASE10_Baum,
 address = {Antwerp, Belgium},
 author = {Reiner Hähnle and Marcus Baum and Richard Bubel and Marcel Rothe},
 booktitle = {Proceedings of the 25th IEEE/ACM International Conference on Automated Software Engineering (ASE 2010)},
 month = {September},
 title = {A Visual Interactive Debugger Based on Symbolic Execution},
 year = {2010}
}

@inproceedings{MFI10_Baum,
 abstract = {This paper introduces a new concept for tracking closely spaced targets in Cartesian
space based on position measurements corrupted with additive Gaussian noise.
The basic idea is to select a special state representation that eliminates the target identity
and avoids the explicit evaluation of association probabilities.
One major advantage of this approach is that the resulting likelihood function for this special problem is unimodal.
Hence, it is especially suitable for closely spaced targets.
The resulting estimation problem can be tackled with a standard nonlinear estimator.
In this work, we focus on two targets in two-dimensional Cartesian space.
The Cartesian coordinates of the targets are represented by means of  extreme values, i.e.,
minima and maxima. Simulation results demonstrate the feasibility of the new approach.},
 address = {Salt Lake City, Utah, USA},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_Baum.pdf},
 title = {Association-free Tracking of Two Closely Spaced Targets},
 year = {2010}
}

@inproceedings{MFI10_BaumRect,
 abstract = {In this paper, a novel Bayesian estimator for the minimum bounding axis-aligned rectangle of a point set based on noisy measurements is
derived. Each given measurement stems from an unknown point  and is corrupted with additive Gaussian noise.
Extreme value theory is applied in order to derive a linear measurement equation for the problem.
The new estimator is applied to the problem of group target and extended object tracking.
Instead of estimating each single group member or point feature explicitly, the basic idea is to track a summarizing shape, namely the minimum bounding rectangle, of the
group. Simulation results demonstrate the feasibility of the estimator.},
 address = {Salt Lake City, Utah, USA},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_BaumRect.pdf},
 title = {Tracking a Minimum Bounding Rectangle based on Extreme Value Theory},
 year = {2010}
}

@inproceedings{MFI10_BaumGheta,
 abstract = {This contribution introduces a three pillar information storage and management
system for modeling the environment of autonomous systems. The main
characteristics is the separation of prior knowledge, environment model and
sensor information. In the center of the system is the environment model, which
provides the autonomous system with information about the current state of the
environment. It consists of instances with attributes and relations as virtual
substitutes of entities (persons and objects) of the real world.
Important features are the representation of uncertain information by means
of Degree-of-Belief (DoB) distributions,
the information exchange between the three pillars as well as creation,
deletion and update of instances, attributes and relations in the environment
model. In this work, a Bayesian method for fusing new observations to the
environment model is introduced. For this purpose, a Bayesian data association
method is derived. The main question answered here is the
observation-to-instance mapping
and the decision mechanisms for creating a new instance or
updating already existing instances in the environment model.},
 address = {Salt Lake City, Utah, USA},
 author = {Marcus Baum and Ioana Gheta and Andrey Belkin and Jürgen Beyerer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_BaumGheta.pdf},
 title = {Data Association in a World Model for Autonomous Systems},
 year = {2010}
}

@inproceedings{MFI10_Beutler,
 abstract = {In range-based pose tracking, the translation and
rotation of an object with respect to a global coordinate system
has to be estimated. The ranges are measured between the
target and the global frame. In this paper, an intelligent decomposition
is introduced in order to reduce the computational
effort for pose tracking. Usually, decomposition procedures only
exploit conditionally linear models. In this paper, this principle
is generalized to conditionally integrable substructures and
applied to pose tracking. Due to a modified measurement
equation, parts of the problem can even be solved analytically.},
 address = {Salt Lake City, Utah, USA},
 author = {Frederik Beutler and Marco F. Huber and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_Beutler.pdf},
 title = {Semi-Analytic Stochastic Linearization for Range-Based Pose Tracking},
 year = {2010}
}

@inproceedings{MFI10_Bogatyrenko,
 abstract = {Most existing approaches for tracking of the
beating heart motion assume known cardiac kinematics and
material parameters. However, these assumptions are not realistic
for application in beating heart surgery. In this paper,
a novel probabilistic tracking approach based on a physical
model of the heart surface is presented. In contrast to existing
approaches, the physical information about heart kinematics
and material properties is incorporated and considered in
an estimation of the heart behavior. An additional advantage
is that the time-dependencies and uncertainties of the heart
parameters are efficiently handled by exploiting simultaneous
state and parameter estimation. Furthermore, by decomposing
the state into linear and nonlinear substructures, the computational
complexity of the estimation problem is reduced. The
experimental results demonstrate the high performance of the
method proposed in this paper. The solution of the parameter
identification problem allows a personalized physical model and
opens up possibilities to apply the physics-based tracking of the
heart surface motion in a clinical environment.},
 address = {Salt Lake City, Utah, USA},
 author = {Evgeniya Bogatyrenko and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_Bogatyrenko.pdf},
 title = {Simultaneous State and Parameter Estimation for Physics-Based Tracking of Heart Surface Motion},
 year = {2010}
}

@inproceedings{MFI10_KrauthausenEberhardt,
 abstract = {In this paper, a novel distance-based density
estimation method is proposed, which considers the overall
density function in the goodness-of-fit. In detail, the parameters
of Gaussian mixture densities are estimated from samples,
based on the distance of the cumulative distributions over
the entire state space. Due to the ambiguous definition of the
standard multivariate cumulative distribution, the Localized
Cumulative Distribution and a modified Cramér-von Mises
distance measure are employed. A further contribution is the
derivation of a simple-to-implement optimization procedure
for the optimization problem. The proposed approach's good
performance in estimating arbitrary Gaussian mixture densities
is shown in an experimental comparison to the Expectation
Maximization algorithm for Gaussian mixture densities.},
 address = {Salt Lake City, Utah, USA},
 author = {Peter Krauthausen and Henning Eberhardt and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_KrauthausenEberhardt.pdf},
 title = {Multivariate Parametric Density Estimation Based On The Modified Cramér-von Mises Distance},
 year = {2010}
}

@inproceedings{MFI10_Krauthausen,
 abstract = {In this paper, a distance-based method for both
multivariate non-parametric density and conditional density
estimation is proposed. The contributions are the formulation
of both density estimation problems as weight optimization
problems for Gaussian mixtures centered about samples with
identical parameters. Furthermore, the minimization is based
on the modified Cramér-von Mises distance of the Localized
Cumulative Distributions, removing the ambiguity of the defi-
nition of the multivariate cumulative distribution function. The
minimization problem is amended with a regularization term
penalizing the densities' roughness to avoid overfitting. The
resulting estimation problems for both densities and conditional
densities are shown to be phrasable in the form of readily
implementable quadratic programs. Experimental comparison
against EM, SVR, and GPR based on the log-likelihood and
performance in benchmark recursive filtering applications show
high quality of the densities and good performance at less
computational cost, i.e., the density representations are sparser.},
 address = {Salt Lake City, Utah, USA},
 author = {Peter Krauthausen and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_Krauthausen.pdf},
 title = {Regularized Non-Parametric Multivariate Density and Conditional Density Estimation},
 year = {2010}
}

@inproceedings{MFI10_LyonsHekler,
 abstract = {Closed-loop model predictive control of nonlinear systems,
whose internal states are not completely accessible, incorporates
the impact of possible future measurements into the planning
process. When planning ahead in time, those measurements
are not known, so the closed-loop controller accounts for
the expected impact of all potential measurements. We propose a novel
conservative closed-loop control approach that does not calculate the
expected impact of all measurements, but solely considers the single
future measurement that has the worst impact on the control objective.
In doing so, the model predictive controller guarantees robustness
even in the face of high disturbances acting upon the system. Moreover,
by considering only a single dedicated measurement, the complexity of
closed-loop control is reduced significantly. The capabilities of our
approach are evaluated by means of a path planning problem for a mobile robot.},
 address = {Salt Lake City, Utah, USA},
 annote = {Nominee Best Paper Award},
 author = {Daniel Lyons and Achim Hekler and Markus Kuderer and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)},
 month = {September},
 pdf = {MFI10_LyonsHekler.pdf},
 title = {Robust Model Predictive Control with Least Favorable Measurements},
 year = {2010}
}

@inproceedings{MSC10_HeklerLyonsNoack,
 abstract = {In Model Predictive Control, the quality of control
is highly dependent upon the model of the system under control.
Therefore, a precise deterministic model is desirable. However,
in real-world applications, modeling accuracy is typically limited
and systems are generally affected by disturbances. Hence,
it is important to systematically consider these uncertainties
and to model them correctly. In this paper, we present a
novel Nonlinear Model Predictive Control method for systems
affected by two different types of perturbations that are
modeled as being either stochastic or unknown but bounded
quantities. We derive a formal generalization of the Nonlinear
Model Predictive Control principle for considering both types
of uncertainties simultaneously, which is achieved by using
sets of probability densities. In doing so, a more robust and
reliable control is obtained. The capabilities and benefits of
our approach are demonstrated in real-world experiments with
miniature walking robots.},
 address = {Yokohama, Japan},
 author = {Achim Hekler and Daniel Lyons and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE Multi-Conference on Systems and Control (MSC 2010)},
 month = {September},
 pdf = {MSC10_HeklerLyonsNoack.pdf},
 title = {Nonlinear Model Predictive Control Considering Stochastic and Systematic Uncertainties with Sets of Densities},
 year = {2010}
}

@article{IJCARS10_Bogatyrenko,
 abstract = {Purpose: Tracking of beating heart motion in a robotic
surgery system is required for complex cardiovascular interventions.
Methods: A heart surface motion tracking method is developed,
including a stochastic physics-based heart surface
model and an efficient reconstruction algorithm. The algorithm
uses the constraints provided by the model that exploits
the physical characteristics of the heart. The main advantage
of the model is that it is more realistic than most standard
heartmodels. Additionally, no explicit matching between the
measurements and the model is required. The application of
meshless methods significantly reduces the complexity of
physics-based tracking.
Results: Based on the stochastic physical model of the heart
surface, this approach considers the motion of the intervention
area and is robust to occlusions and reflections. The
tracking algorithm is evaluated in simulations and experiments
on an artificial heart. Providing higher accuracy than
the standardmodel-based methods, it successfully copes with
occlusions and provides high performance even when all
measurements are not available.
Conclusions: Combining the physical and stochastic description
of the heart surface motion ensures physically correct
and accurate prediction. Automatic initialization of the physics-based
cardiac motion tracking enables system evaluation
in a clinical environment.},
 author = {Evgeniya Bogatyrenko and Pascal Pompey and Uwe D. Hanebeck},
 doi = {doi:10.1007/s11548-010-0517-5},
 issn = {1861-6410},
 journal = {International Journal of Computer Assisted Radiology and Surgery (IJCARS 2010)},
 month = {August},
 note = {doi:10.1007/s11548-010-0517-5},
 number = {3},
 pages = {387--399},
 pdf = {IJCARS10_Bogatyrenko.pdf},
 publisher = {Springer Berlin / Heidelberg},
 title = {Efficient Physics-Based Tracking of Heart Surface Motion for Beating Heart Surgery Robotic Systems},
 url = {https://dx.doi.org/10.1007/s11548-010-0517-5},
 volume = {6},
 year = {2010}
}

@inproceedings{Fusion10_BaumKlumpp,
 abstract = {This paper introduces a novel recursive Bayesian
estimator for the center and radius of a circle based on
noisy points. Each given point is assumed to be a noisy measurement
of an unknown true point on the circle that is corrupted with known
isotropic Gaussian noise. In contrast to existing approaches, the
novel method does not make  assumptions about the  true points on
the circle, where the measurements stem from. Closed-form expressions
for the measurement update step are derived. Simulations show that
the novel method outperforms standard Bayesian approaches for
circle fitting.},
 address = {Edinburgh, United Kingdom},
 author = {Marcus Baum and Vesa Klumpp and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_BaumKlumpp.pdf},
 title = {A Novel Bayesian Method for Fitting a Circle to Noisy Points},
 year = {2010}
}

@inproceedings{Fusion10_BaumNoack,
 abstract = {This paper provides new results and
insights for tracking an  extended target object
modeled with an  Elliptic Random Hypersurface Model (RHM).
An Elliptic RHM specifies the relative squared Mahalanobis
distance of a measurement source to the center of the
target object by means of a one-dimensional random scaling
factor. It is shown that uniformly distributed measurement
sources on an ellipse lead to a uniformly distributed
squared scaling factor. Furthermore, a Bayesian inference
mechanisms tailored to elliptic shapes is introduced, which
is also suitable for scenarios with high measurement noise.
Closed-form expressions for the measurement update in case
of Gaussian and uniformly distributed squared scaling factors are derived.},
 address = {Edinburgh, United Kingdom},
 author = {Marcus Baum and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_BaumNoack.pdf},
 title = {Extended Object and Group Tracking with Elliptic Random Hypersurface Models},
 year = {2010}
}

@inproceedings{Fusion10_Beutler,
 abstract = {In multilateration tracking, an object, e.g., an airplane, emits a known
reference signal, which is received by several base stations (sensors) located at
known positions. The receiving times of the signal at the sensors correspond to the times of
arrival (TOA) plus an unknown offset, because the emission time is unknown.
Usually, for estimating the position of the object, the receiving times are
converted to a larger number of time differences of arrival (TDOA) in order
to eliminate the unknown offset. To avoid this conversion, the proposed
approach directly uses the receiving times. This is achieved by 1. determining the optimal offset from the redundant measurements in closed
form and 2. by considering a modified measurement equation. As a result,
position estimation can be performed by optimal stochastic linearization.},
 address = {Edinburgh, United Kingdom},
 author = {Frederik Beutler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_Beutler.pdf},
 title = {A Two-Step Approach for Offset and Position Estimation from Pseudo-Ranges Applied to Multilateration Tracking},
 year = {2010}
}

@inproceedings{Fusion10_DunauPacki,
 abstract = {Multilateration systems operate by deter-
mining distances between a signal transmitter and a
number of receivers. In aerial surveillance, radio sig-
nals are emitted as Secondary Surveillance Radar (SSR)
by the aircraft, representing the signal transmitter. A
number of base stations (sensors) receive the signals at
different times. Most common approaches use time dif-
ference of arrival (TDOA) measurements, calculated by
subtracting receiving times of one receiver from another.
As TDOAs require intersecting hyperboloids, which is
considered a hard task, this paper follows a different ap-
proach, using raw receiving times. Thus, estimating the
signal's emission time is required, captured as a com-
mon offset within an augmented version of the system
state. This way, the multilateration problem is reduced
to intersecting cones. Estimation of the aircraft's posi-
tion based on a nonlinear measurement model and an
underlying linear system model is achieved using a lin-
ear regression Kalman filter [1, 2]. A decomposed com-
putation of the filter step is introduced, allowing a more
efficient calculation.},
 address = {Edinburgh, United Kingdom},
 author = {Patrick Dunau and Ferdinand Packi and Frederik Beutler and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_DunauPacki.pdf},
 title = {Efficient Multilateration Tracking with Concurrent Offset Estimation using Stochastic Filtering Techniques},
 year = {2010}
}

@inproceedings{Fusion10_EberhardtKlumpp,
 abstract = {In this paper, a new class of nonlinear Bayesian
estimators based on a special space partitioning structure, generalized
Octrees, is presented. This structure minimizes memory and calculation
overhead. It is used as a container framework for a set of node functions
that approximate a density piecewise. All necessary operations are derived
in a very general way in order to allow for a great variety of Bayesian
estimators. The presented estimators are especially well suited for
multi-modal nonlinear estimation problems. The running time performance
of the resulting estimators is first analyzed theoretically and then backed
by means of simulations. All operations have a linear running time in
the number of tree nodes.},
 address = {Edinburgh, United Kingdom},
 author = {Henning Eberhardt and Vesa Klumpp and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_EberhardtKlumpp.pdf},
 title = {Density Trees for Efficient Nonlinear State Estimation},
 year = {2010}
}

@inproceedings{Fusion10_Klumpp-SGMF,
 abstract = {In this paper, a novel nonlinear/non-linear model
decomposition for the Sliced Gaussian Mixture Filter is presented.
Based on the level of nonlinearity of the model, the overall estimation
problem is decomposed into a severely nonlinear and a slightly
nonlinear part, which are processed by different estimation techniques.
To further improve the efficiency of the estimator, an adaptive state
decomposition algorithm is introduced that allows decomposition
according to the linearization error for nonlinear system and
measurement models. Simulations show that this approach has orders of
magnitude less complexity compared to other state of the art
estimators, while maintaining comparable estimation errors.},
 address = {Edinburgh, United Kingdom},
 author = {Vesa Klumpp and Frederik Beutler and Uwe D. Hanebeck and Dietrich Fränken},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_Klumpp-SGMF.pdf},
 title = {The Sliced Gaussian Mixture Filter with Adaptive State Decomposition Depending on Linearization Error},
 year = {2010}
}

@inproceedings{Fusion10_Klumpp,
 abstract = {In estimation theory, mainly set-theoretic or
stochastic uncertainty is considered. In some cases, especially when
some statistics of a distribution are not known or additional
stochastic information is used in a set-theoretic estimator, both
types of uncertainty have to be considered. In this paper, two
estimators that cope with combined stoachastic and set-theoretic
uncertainty are compared, namely the Set-theoretic and Statistical
Information filter, which represents the uncertainty by means of
random sets, and the Credal State filter, in which the state
information is given by sets of probability density functions.
The different uncertainty assessment in both estimators leads to
different estimation results, even when the prior information and
the measurement and system models are equal. This paper explains
these differences and states directions, when which estimator
should be applied to a given estimation problem.},
 address = {Edinburgh, United Kingdom},
 author = {Vesa Klumpp and Benjamin Noack and Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_Klumpp.pdf},
 title = {Combined Set-Theoretic and Stochastic Estimation: A Comparison of the SSI and the CS Filter},
 year = {2010}
}

@inproceedings{Fusion10_KrauthausenHuber,
 abstract = {A non-parametric conditional density
estimation algorithm for nonlinear stochastic dynamic
systems is proposed. The contributions are a novel sup-
port vector regression for estimating conditional den-
sities, modeled by Gaussian mixture densities, and an
algorithm based on cross-validation for automatically
determining hyper-parameters for the regression. The
conditional densities are employed with a modi?ed axis-
aligned Gaussian mixture filter. The experimental va-
lidation shows the high quality of the conditional densi-
ties and good accuracy of the proposed filter.},
 address = {Edinburgh, United Kingdom},
 author = {Peter Krauthausen and Marco F. Huber and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_KrauthausenHuber.pdf},
 title = {Support-Vector Conditional Density Estimation for Nonlinear Filtering},
 year = {2010}
}

@inproceedings{Fusion10_Noack,
 abstract = {Applying the Kalman filtering scheme to linearized system dynamics and observation models does in general not yield optimal state estimates.
More precisely, inconsistent state estimates and covariance matrices are caused by neglected linearization errors.
This paper introduces a concept for systematically predicting and updating bounds for the linearization errors within the Kalman filtering framework.
To achieve this, an uncertain quantity is not characterized by a single probability density anymore, but rather by a set of densities and accordingly,
the linear estimation framework is generalized in order to process sets of probability densities. By means of this generalization,
the Kalman filter may then not only be applied to stochastic quantities, but also to unknown but bounded quantities.
In order to improve the reliability of Kalman filtering results, the last-mentioned quantities are utilized to bound the typically neglected nonlinear parts of a linearized mapping.},
 address = {Edinburgh, United Kingdom},
 author = {Benjamin Noack and Vesa Klumpp and Nikolay Petkov and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)},
 month = {July},
 pdf = {Fusion10_Noack.pdf},
 title = {Bounding Linearization Errors with Sets of Densities in Approximate Kalman Filtering},
 year = {2010}
}

@inproceedings{SUTC10_Lyons,
 abstract = {In distributed sensor networks, computational and energy resources are
in general limited.  Therefore, an intelligent selection of sensors for
measurements is of great importance to ensure both high estimation
quality and an extended lifetime of the network. Methods from the theory
of model predictive control together with information theoretic measures
have been employed to pick sensors yielding measurements with high
information value. We present a novel information measure that originates from a
scalar product on a class of continuous probability densities and apply it
to the field of sensor management. Aside from its mathematical justifications
for quantifying the information content of probability densities, the most
remarkable property of the measure, an analogon of the triangle inequality
under Bayesian information fusion, is deduced. This allows for deriving
computationally cheap upper bounds for the model predictive sensor selection
algorithm and for comparing the performance of planning over different lengths of time horizons.},
 address = {Newport Beach, California, USA},
 author = {Daniel Lyons and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Proceedings of the IEEE International Conference on Sensor Networks, Ubiquitous, and Trustworthy Computing (SUTC 2010)},
 month = {June},
 pdf = {SUTC10_Lyons.pdf},
 title = {A Log-Ratio Information Measure for Stochastic Sensor Management},
 year = {2010}
}

@inproceedings{ACC10_Eberhardt,
 abstract = {The sample-based recursive prediction of discrete-time nonlinear
stochastic dynamic systems requires a regular reapproximation of the Dirac mixture
densities characterizing the state estimate with an exponentially increasing number
of components. For that purpose, a systematic approximation method is proposed that
is deterministic and guaranteed to minimize a new type distance measure, the so
called modified Cramér-von Mises distance. A huge increase in approximation
performance is achieved by exploiting structural independencies usually occurring
between the random variables used as input to the system. The corresponding prediction
step achieves optimal performance when no further assumptions can be made about the
system function. In addition, the proposed approach shows a much better convergence
compared to the prediction step of the particle filter and by far fewer Dirac components
are required for achieving a given approximation quality. As a result, the new
approximation method opens the way for the development of new fully deterministic and
optimal stochastic state estimators for nonlinear dynamic systems.},
 address = {Baltimore, Maryland, USA},
 author = {Henning Eberhardt and Vesa Klumpp and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 American Control Conference (ACC 2010)},
 keywords = {Stochastic systems, Filtering},
 month = {June},
 pdf = {ACC10_EberhardtKlumpp.pdf},
 title = {Optimal Dirac Approximation by Exploiting Independencies},
 year = {2010}
}

@inproceedings{ACC10_Krauthausen,
 abstract = {In this paper, the first algorithm for learning hybrid Bayesian
Networks with Gaussian mixture and Dirac mixture conditional densities from data
given their structure is presented. The mixture densities to be learned allow for
nonlinear dependencies between the variables and exact closedform inference. For
learning the network's parameters, an incremental gradient ascent algorithm is derived.
Analytic expressions for the partial derivatives and their combination with messages are
presented. This hybrid approach subsumes the existing approach for purely discrete-valued
networks and is applicable to partially observable networks, too. Its practicability is
demonstrated by a reference example.},
 address = {Baltimore, Maryland, USA},
 author = {Peter Krauthausen and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2010 American Control Conference (ACC 2010)},
 month = {June},
 pdf = {ACC10_Krauthausen.pdf},
 title = {Parameter Learning for Hybrid Bayesian Networks With Gaussian Mixture and Dirac Mixture Conditional Densities},
 year = {2010}
}

@incollection{VMS10_Lyons,
 abstract = {Bei der Beobachtung eines räumlich verteilten Phänomens mit einer
Vielzahl von Sensoren ist die intelligente Auswahl von Messkonfigurationen aufgrund von
beschränkten Rechen- und Kommunikationskapazitäten entscheidend für die
Lebensdauer des Sensornetzes. Mit der Sensoreinsatzplanung kann die im nächsten
Zeitschritt anzusteuernde Messkonfiguration dynamisch mittels einer stochastischen
modell-prädiktiven Planung über einen endlichen Zeithorizont bestimmt werden.
Dabei wird als Gütekriterium die Maximierung des zu erwartenden Informationsgewinns
durch zukünftige Messungen unter sparsamer Verwendung der Energieressourcen gewählt.
In diesem Artikel wird ein neues Maß für kontinuierliche Wahrscheinlichkeitsdichten
vorgestellt, das sich kanonisch aus der Konstruktion eines Vektorraums für
Wahrscheinlichkeitsdichten ergibt. Dieses Maß wird als Gütefunktion in der
vorausschauenden Sensoreinsatzplanung zur Bewertung des informationstheoretischen Einfluß
von Messungen auf die aktuelle Zustandsschätzung verwendet.},
 author = {Daniel Lyons and Achim Hekler and Benjamin Noack and Uwe D. Hanebeck},
 booktitle = {Verteilte Messsysteme},
 editor = {Fernando Puente León and Klaus-Dieter Sommer and Michael Heizmann},
 month = {March},
 pages = {121--132},
 publisher = {KIT Scientific Publishing},
 title = {Maße für Wahrscheinlichkeitsdichten in der informationstheoretischen Sensoreinsatzplanung},
 url = {https://publikationen.bibliothek.kit.edu/1000015670},
 year = {2010}
}

@incollection{VMS10_Noack,
 abstract = {Die systematische Behandlung von Unsicherheiten stellt eine wesentliche
Herausforderung in der Informationsfusion dar. Einerseits müssen geeignete Darstellungsformen
für die Unsicherheiten bestimmt werden und andererseits darauf aufbauend effiziente
Schätzverfahren hergeleitet werden. Im Allgemeinen wird zwischen stochastischen und
mengenbasierten Unsicherheitsbeschreibungen unterschieden. Dieser Beitrag stellt ein Verfahren
zur Zustandsschätzung vor, welches simultan stochastische und mengenbasierte Fehlergrößen
berücksichtigen kann, indem unsichere Größen nicht mehr durch eine einzelne
Wahrscheinlichkeitsdichte, sondern durch eine Menge von Dichten repräsentiert werden.
Besonderes Augenmerk liegt hier auf den Vorteilen und Anwendungsmöglichkeiten dieser
Unsicherheitsbeschreibung.},
 author = {Benjamin Noack and Vesa Klumpp and Daniel Lyons and Uwe D. Hanebeck},
 booktitle = {Verteilte Messsysteme},
 editor = {Fernando Puente León and Klaus-Dieter Sommer and Michael Heizmann},
 month = {March},
 pages = {167--178},
 publisher = {KIT Scientific Publishing},
 title = {Systematische Beschreibung von Unsicherheiten in der Informationsfusion mit Mengen von Wahrscheinlichkeitsdichten},
 url = {https://publikationen.bibliothek.kit.edu/1000015670},
 year = {2010}
}

@incollection{VMS10_Kuwertz,
 abstract = {Räumlich ausgedehnte Phänomene wie Schadstoffverteilungen in Gewässern
oder Temperaturverteilungen in Räumen werden vielfach durch unbekannte, aber gegebenenfalls
sich bewegende Quellen erzeugt. Allerdings stehen in vielen praktisch relevanten Aufgabenstellungen
Informationen zur Lokalisierung einer derartigen Quelle nur indirekt durch eine Vermessung des
induzierten Phänomens zur Verfügung, welche den Einsatz eines verteilten Messsystems erfordert.
Die Messungen können dabei beispielsweise von einem stationären Sensornetz oder von mobilen
Sensoren stammen. In diesem Beitrag werden modellbasierte Verfahren zu echtzeitfähigen Lokalisierung
und schritthaltenden Verfolgung von Quellen vorgestellt, welche gezielt räumlich und zeitlich
verteilte Messungen einsetzen. Um den Informationsgewinn und somit den Nutzen verteilter Messungen zu
maximieren, spielt bei diesem Verfahren neben einer mathematischen Modellierung auch eine vorausschauende
Sensoreinsatzplanung eine zentrale Rolle. Das in diesem Beitrag vorgeschlagene Planungsverfahren
ermöglicht dabei die effiziente und ressourcenschonende Verfolgung beweglicher Quellen bei gleichzeitig
hoher Lokalisierungsgenauigkeit.},
 author = {Achim Kuwertz and Marco F. Huber and Felix Sawo and Uwe D. Hanebeck},
 booktitle = {Verteilte Messsysteme},
 editor = {Fernando Puente León and Klaus-Dieter Sommer and Michael Heizmann},
 month = {March},
 pages = {179--191},
 publisher = {KIT Scientific Publishing},
 title = {Sensoreinsatzplanung zur Verfolgung von Quellen räumlich ausgedehnter Phänomene},
 url = {https://publikationen.bibliothek.kit.edu/1000015670},
 year = {2010}
}

@inproceedings{PED10_Perez,
 abstract = {In this contribution, we propose a new framework to evaluate pedestrian
simulations by using Extended Range Telepresence. Telepresence is
used as a virtual reality walking simulator, which provides the user
with a realistic impression of being present and walking in a virtual
environment that is much larger than the real physical environment,
in which the user actually walks. The validation of the simulation
is performed by comparing motion data of the telepresent user with
simulated data at some points of the simulation. The use of haptic
feedback from the simulation makes the framework suitable for training
in emergency situations.},
 address = {Gaithersburg, Maryland, USA},
 author = {Antonia Pérez Arias and Tobias Kretz and Peter Ehrhardt and Stefan Hengst and Peter Vortisch and Uwe D. Hanebeck},
 booktitle = {Proccedings of the 5th International Conference on Pedestrian and Evacuation Dynamics (PED 2010), Springer-Verlag},
 month = {March},
 pdf = {PED10_Perez.pdf},
 title = {Extended Range Telepresence for Evacuation Training in Pedestrian Simulations},
 year = {2010}
}

@article{arXiv10_Perez,
 abstract = {In this contribution, we propose a new framework to evaluate pedestrian
simula-tions by using Extended Range Telepresence. Telepresence is
used as a virtual reality walking simulator, which provides the user
with a realistic impression of being present and walking in a virtual
environment that is much larger than the real physical environment,
in which the user actually walks. The validation of the simulation
is performed by comparing motion data of the telepresent user with
simulated data at some points of the simulation. The use of haptic
feedback from the simulation makes the framework suitable for training
in emergency situations.},
 author = {Antonia Pérez Arias and Uwe D. Hanebeck and Peter Ehrhardt
and Stefan Hengst and Tobias Kretz and Peter Vortisch},
 journal = {arXiv preprint: Human-Computer Interaction (cs.HC)},
 month = {February},
 title = {Extended Range Telepresence for Evacuation Training in Pedestrian
Simulations},
 url = {https://arxiv.org/abs/1002.3770},
 year = {2010}
}

@inproceedings{NIPS09_Deisenroth,
 address = {Whistler, Canada},
 author = {Ryan Turner and Marc Peter Deisenroth and Carl Edward Rasmussen},
 booktitle = {Nonparametric Bayes Workshop at NIPS 2009},
 month = {December},
 title = {System Identification in Gaussian Process Dynamical Systems},
 year = {2009}
}

@inproceedings{ISSPIT09_Baum,
 abstract = {Target tracking algorithms usually assume that the received measurements
stem from a point source. However, in many scenarios  this assumption
is not feasible so that measurements may stem from different locations,
named measurement sources, on the target surface. Then, it is necessary
to incorporate the target extent into the estimation procedure in
order to obtain robust and precise estimation results. This paper
introduces the novel concept of Random Hypersurface Models for extended
targets. A Random Hypersurface Model assumes that each measurement
source is an element of a randomly generated hypersurface. The applicability
of this approach is demonstrated by means of an elliptic target shape.
In this case, a Random Hypersurface Model specifies the random (relative)
Mahalanobis distance of a measurement source to the center of the
target object. As a consequence, good estimation results can be obtained
even if the true target shape significantly differs from the modeled
shape. Additionally, Random Hypersurface Models are computationally
tractable with standard nonlinear stochastic state estimators.},
 address = {Ajman, United Arab Emirates},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 9th IEEE International Symposium on Signal Processing and Information Technology (ISSPIT 2009)},
 month = {December},
 pdf = {ISSPIT09_Baum.pdf},
 timestamp = {2009.10.13},
 title = {Random Hypersurface Models for Extended Object Tracking},
 year = {2009}
}

@inproceedings{CDC09_HanebeckHuber,
 abstract = {For the optimal approximation of multivariate
Gaussian densities by means of Dirac mixtures, i.e., by means of
a sum of weighted Dirac distributions on a continuous domain,
a novel systematic method is introduced. The parameters of
this approximate density are calculated by minimizing a global
distance measure, a generalization of the well-known Cramér-
von Mises distance to the multivariate case. This generalization
is obtained by defining an alternative to the classical cumulative
distribution, the Localized Cumulative Distribution (LCD). In
contrast to the cumulative distribution, the LCD is unique
and symmetric even in the multivariate case. The resulting
deterministic approximation of Gaussian densities by means of
discrete samples provides the basis for new types of Gaussian
filters for estimating the state of nonlinear dynamic systems
from noisy measurements.},
 address = {Shanghai, China},
 author = {Uwe D. Hanebeck and Marco F. Huber and Vesa Klumpp},
 booktitle = {Proceedings of the 2009 IEEE Conference on Decision and Control (CDC 2009)},
 month = {December},
 pdf = {CDC09_Hanebeck.pdf},
 title = {Dirac Mixture Approximation of Multivariate Gaussian Densities},
 year = {2009}
}

@inproceedings{IROS09_Bogatyrenko,
 abstract = {A novel heart surface motion estimation frame-
work for a robotic surgery on a stabilized beating heart is
proposed. It includes an approach for the reconstruction and
prediction of heart surface motion based on a novel physical
model of the intervention area described by a distributed-
parameter system. Instead of conventional element methods, a
meshless method is used for a spatial and temporal decomposi-
tion of this system. This leads to a finite-dimensional state-space
form. Furthermore, the state of the resulting lumped-parameter
system, which provides an approximation of the deflection and
velocity of the heart surface, is dynamically estimated under
consideration of uncertainties both occurring in the system
and arising from noisy camera measurements. By using the
estimation results, an accurate reconstruction of heart surface
motion for the synchronisation of the surgical instruments is
also achieved at occluded or non-measurement points.},
 author = {Evgeniya Bogatyrenko and Uwe D. Hanebeck and Gábor Szabó},
 booktitle = {Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2009)},
 month = {October},
 pdf = {IROS09_Bogatyrenko.pdf},
 title = {Heart Surface Motion Estimation Framework for Robotic Surgery Employing Meshless Methods},
 year = {2009}
}

@inproceedings{SDF09_Baum,
 abstract = {In many tracking applications, the extent of the target object is neglected
and it is assumed that the received measurements stem from a point source. However,
modern sensors are able to supply several measurements from different scattering cen-
ters on the target object due to their high-resolution capability. As a consequence, it
becomes necessary to incorporate the target extent into the estimation procedure. This
paper introduces a new method for tracking the smallest enclosing rectangle of an ex-
tended object with an unknown shape. At each time step, a finite set of noisy position
measurements that stem from arbitrary, unknown measurement sources on the target
surface may be available. In contrast to common approaches, the presented approach
does not have to make any statistical assumptions on the measurement sources.},
 address = {Lübeck, Germany},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {4th German Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF 2009), 39th Annual Conference of the Gesellschaft für Informatik e.V. (GI)},
 month = {October},
 pdf = {SDF09_Baum.pdf},
 title = {Tracking an Extended Object Modeled as an Axis-Aligned Rectangle},
 year = {2009}
}

@inproceedings{Deisenroth2009,
 address = {Hluboka nad Vltavou, Czech Republic},
 author = {Marc P. Deisenroth and Carl E. Rasmussen},
 booktitle = {Proceedings of the 10th International PhD Workshop on Systems and Control},
 month = {September},
 title = {Efficient Reinforcement Learning for Motor Control},
 year = {2009}
}

@book{MFI09_Hanebeck,
 editor = {Uwe D. Hanebeck and Thomas C. Henderson},
 month = {September},
 number = {3},
 pages = {237--344},
 pdf = {MFI09_Hanebeck.pdf},
 publisher = {ScienceDirect},
 series = {Selected papers from 2006 IEEE International Conference on Multisensor Fusion and Integration (MFI 2006), 2006 IEEE International Conference on Multisensor Fusion and Integration},
 title = {Robotics and Autonomous Systems},
 volume = {57},
 year = {2009}
}

@inproceedings{CCA09_RauhHanebeck,
 abstract = {In this paper, the Prior Density Splitting Mixture
Estimator (PDSME), a new Gaussian mixture filtering
algorithm for nonlinear dynamical systems and nonlinear
measurement equations, is introduced. This filter reduces the
linearization error which typically arises if nonlinear state and
measurement equations are linearized to apply linear filtering
techniques. For that purpose, the PDSME splits the prior probability
density into several components of a Gaussian mixture
with smaller covariances. The PDSME is applicable to both
prediction and filter steps. A measure for the linearization error
similar to the Kullback-Leibler distance is introduced allowing
the user to specify the desired estimation quality. An upper
bound for the computational effort can be given by limiting
the maximum number of Gaussian mixture components.},
 author = {Andreas Rauh and Kai Briechle and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 2009 IEEE International Conference on Control Applications (CCA 2009)},
 month = {July},
 pdf = {CCA09_RauhHanebeck.pdf},
 title = {Nonlinear Measurement Update and Prediction: Prior Density Splitting Mixture Estimator},
 year = {2009}
}

@inproceedings{Fusion09_Baum,
 abstract = {In this paper, a novel approach for tracking
extended objects is presented. The target object is
modeled as a circular disc such that the center and
extent of the target object can be estimated. At each
time step, a finite set of position measurements that
are corrupted with stochastic noise may be available.
Each position measurement stems from an unknown measurement
source on the extended object. In contrast to existing
approaches, no statistical assumptions about the distribution
of the measurement sources on the extended object are made.
As a consequence, it is necessary to deal with stochastic
and set-valued uncertainties. For this purpose, a novel
combined stochastic and set-theoretic estimator that employs
random hyperboloids to express the uncertainties about the
true circular disc is derived.},
 address = {Seattle, Washington, USA},
 author = {Marcus Baum and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 12th International Conference on Information Fusion (Fusion 2009)},
 month = {July},
 pdf = {Fusion09_Baum.pdf},
 timestamp = {2009.04.21},
 title = {Extended Object Tracking based on Combined Set-Theoretic and Stochastic Fusion},
 year = {2009}
}

@inproceedings{Fusion09_Beutler,
 abstract = {State estimation for nonlinear systems generally
requires approximations of the system or the probability
densities, as the occurring prediction and filtering equations
cannot be solved in closed form. For instance, Linear Regression
Kalman Filters like the Unscented Kalman Filter
or the considered Gaussian Filter propagate a small set of
sample points through the system to approximate the posterior
mean and covariance matrix. To reduce the number of
sample points, special structures of the system and measurement
equation can be taken into account. In this paper, two
principles of system decomposition are considered and applied
to the Gaussian Filter. One principle exploits that only
a part of the state vector is directly observed by the measurement.
The second principle separates the system equations
into linear and nonlinear parts in order to merely approximate
the nonlinear part of the state. The benefits of both
decompositions are demonstrated on a real-world example.},
 address = {Seattle, Washington, USA},
 author = {Frederik Beutler and Marco F. Huber and Uwe D. Hanebeck},
 booktitle = {Proceedings of the 12th International Conference on Information Fusion (Fusion 2009)},
 month = {July},
 pdf = {Fusion09_Beutler.pdf},
 timestamp = {2009.04.20},
 title = {Gaussian Filtering using State Decomposition Methods},
 year = {2009}
}