For most tracking applications the Kalman filter is reliable and efficient, but it is limited to a relatively restricted class of linear Gaussian problems. To solve problems beyond this restricted class, particle filters are proving to be dependable methods for stochastic dynamic estimation. Packed with 867 equations, this cutting-edge book introduces the latest advances in particle filter theory, discusses their relevance to defense surveillance systems, and examines defense-related applications of particle filters to nonlinear and non-Gaussian problems. With this hands-on guide, you can develop more accurate and reliable nonlinear filter designs and more precisely predict the performance of these designs. You can also apply particle filters to tracking a ballistic object, detection and tracking of stealthy targets, tracking through the blind Doppler zone, bi-static radar tracking, passive ranging (bearings-only tracking) of maneuvering targets, range-only tracking, terrain-aided tracking of ground vehicles, and group and extended object tracking.
Part I - Theoretical Concepts.; IntroductionNonlinear Filtering. The Problem and Its Conceptual Solutions. Optimal Algorithms. Multiple Switching Dynamic Models. Basics of Target Tracking. Summary.; Suboptimal Nonlinear FiltersAnalytic Approximations. Numerical Methods. Gaussian Sum Filters. Unscented Kalman Filter. Summary.; A Tutorial on Particle FiltersMonte Carlo Integration. Sequential Importance Sampling. Resampling. Selection of Importance Density. Versions of Particle Filters. Summary.; Cramer-Rao Bounds for Nonlinear FilteringBackground. General Recursive Calculations. Special Cases. Multiple Switching Dynamic Models. Probability of Detection Less than 1. Summary.; Part II - Tracking Applications.; Tracking a Ballistic Object -Introduction. Target Dynamics and Measurements. Cramer-Rao Bound. Tracking Filters Numerical Results. Concluding Remarks.; Bearings-Only Tracking -Introduction. Problem Formulation. Cramer-Rao Lower Bounds. Tracking Algorithms. Simulation Results. Summary. Appendix: Linearized Transition Matrix for MP-EKF.; Range-Only Tracking -Introduction. Problem Description. Cramer-Rao Bounds. Tracking Algorithms. Algorithm Performance and Comparison. Application to Ingara ISAR Data. Summary.; Bistatic Radar Tracking -Introduction. Problem Formulation. Cramer-Rao Bounds. Tracking Algorithms. Algorithm Performance. Summary.; Tracking Targets Through Blind Doppler ?Introduction. Problem Formulation. EKF-Based Track Maintenance. Particle Filter-Based Solution. Simulation Results. Summary.; Terrain Aided Tracking -Introduction. Problem Description and Formulation. Variable Structure IMM. Variable Structure Multiple-Model Particle Filter. Simulation Results. Conclusions.; Detection and Tracking of Stealthy Targets -Introduction. Target and Sensor Models. Conceptual Solution in the Bayesian Framework. A Particle Filter for Track-Before-Detect. A Numerical Example. Performance Analysis. Summary and Extension.; Group and Extended Object Tracking -Introduction. Tracking Model. Formal Bayesian Solution. Affine Model. Particle Filters. Simulation Example. Concluding Remarks.; Epilogue. Appendix. List of Acronyms. About the Authors. Index.;
-
Sanjeev Arulampalam
Sanjeev Arulampalam is a senior research scientist in the Submarine Combat Systems Group, Maritime Operations Division of DSTO, Edinburgh, Australia. In 2000 he was awarded the Anglo-Australian postdoctoral fellowship by the Royal Academy of Engineering, London. He earned his Ph.D. in electrical and electronics engineering at the University of Melbourne, Australia.
-
Neil Gordon
Neil Gordon is a senior research scientist in the Tracking and Sensor Fusion Group at the ISR Division of DSTO, Edinburgh, Australia. Dr Gordon earned his Ph.D. in statistics at the Imperial College, University of London.
-
Branko Ristic
Branko Ristic is a senior research scientist in the Tracking and Sensor Fusion Group at the ISR Division of DSTO, Edinburgh, Australia. In 2002 he was awarded the Defence Science Fellowship by the Information Sciences Laboratory of DSTO. He earned his Ph.D. at the Signal Processing Research Centre of Queensland University of Technology, Australia.