A Particle PHD Filter with Improved Resampling Design for Multiple Target Tracking

Abstract

Multi-target tracking is a complex problem including time-varying number of targets and their states in the presence of data association uncertainty and clutter. In this article, we develop a novel implementation of Sequential Monte Carlo filter with a new improved partial resampling strategy in random finite sets framework. This algorithm provides an approach to increase diversity of particles and keep accuracy of filtering performance. Simulation results verify that for the MTT problems, the proposed algorithm could achieve better performance than the standard particle PHD filter.