Multi-sensor GIW-PHD filter for multiple extended target tracking

Abstract

Gaussian inverse Wishart probability hypothesis density (GIW-PHD) filter has proven to be a promising algorithm for multiple extended target tracking with shape estimation. However, as far as I know, this method only can be used in the single sensor tracking system, which cannot obtain the accurate state estimates for the complex tracking scenario. To solve this problem, we propose a multi-sensor GIW-PHD method by using the multiple sensor infusion technique, which is suitable to the multi-sensor tracking system for multiple extended target tracking. First, a novel measurement model of the extended target is constructed for multi-sensor in three-dimensional scenario, and then the fusion formulas of state update are derived. Simulation results show that the proposed algorithm has a better performance than that of the conventional GIW-PHD with a single sensor.