Particle Filters for Random Set Models

Branko Ristic is at the Defence Science and Technology Organisation, AustraliaDefence Science and Technology Organisation, Australia

From the book reviews:“The book realizes a happy union between theory and practice. Of high interest are the Algorithms for which their pseudo-codes are presented. We think we are faced with an excellent book that will have a great success and audience between those interested for new approaches in filtering theory.” (Dumitru Stanomir, zbMATH 1306.93002, 2015)

• Introduction.- References.- Background.- A brief review of particle filters.- Online sensor control.- Non-standard measurements.- Imprecise measurements.- Imprecise measurement function.- Uncertain implication rules.- Particle filter implementation.- Applications.- Multiple objects and imperfect detection.- Random finite sets.- Multi-object stochastic filtering.- OSPA metric.- Specialized multi-object filters.- Bernoulli filter.- PHD and CPHD filter.- References.- Applications involving non-standard measurements.- Estimation using imprecise measurement models.- Localization using the received signal strength.- Prediction of an epidemic using syndromic data.- Summary.- Fusion of spatially referring natural language statements.- Language, space and modelling.- An illustrative example.- Classification using imprecise likelihoods.- Modelling.- Classification results.- References.- object particle filters.- Bernoulli particle filters.- Standard Bernoulli particle filters.- Bernoulli box-particle filter.- PHD/CPDH particle filters with adaptive birth intensity.- Extension of the PHD filter.- Extension of the CPHD filter.- Implementation.- A numerical study.- State estimation from PHD/CPHD particle filters.- Particle filter approximation of the exact multi-object filter.- References.- Sensor control for random set based particle filters.- Bernoulli particle filter with sensor control.- The reward function.- Bearings only tracking in clutter with observer control.- Target Tracking via Multi-Static Doppler Shifts.- Sensor control for PHD/CPHD particle filters.- The reward function.- A numerical study.- Sensor control for the multi-target state particle filter.- Particle approximation of the reward function.- A numerical study.- References.- Multi-target tracking.- OSPA-T: A performance metric for multi-target tracking.- The problem and its conceptual solution.- The base distance and labeling of estimated tracks.- Numerical examples.- Trackers based on random set filters.- Multi-target trackers based on the Bernoulli PF.- Multi-target trackers based on the PHD particle filter.- Error performance comparison using the OSPA-T error.- Application: Pedestrian tracking.- Video dataset and detections.- Description of Algorithms.- Numerical results.- References.- Advanced topics.- Filter for extended target tracking.- Mathematical models.- Equations of the Bernoulli filter for an extended target.- Numerical Implementation.- Simulation results.- Application to a surveillance video.- Calibration of tracking systems.- Background and problem formulation.- The proposed calibration algorithm.- Importance sampling with progressive correction.- Application to sensor bias estimation.- References.- Index.