A Branching Particle-based Nonlinear Filter for Multi-target Tracking
D. Ballantyne, Hubert Y. Chan, Michael A. Kouritzin · 2001
Abstract – A branching particle-based filter is used to detect and track multiple simulated maneuvering ships in a region of water. The ship trajectories exhibit nonlinear dynamics and interact in a nonlinear manner so that the ships do not collide. There is no a priori knowledge of the number of ships in the region. Observations model a sensor tracking the ships from above the region as in a low observable problem. The branching filter simulates particles, each of which is a sample from the domain of possible combinations of ship number and the state of those ships, and each of which is evolved independently using the stochastic law of the signal between observations. The branching filter employs these particles to provide the approximated con-ditional distribution of the signal in the combined domain, given all observations. Quantitative results recording the capacity of the branching filter to determine the number of ships in the region and the location of each ship are presented.