Novel Multiple-Model Probability Hypothesis Density Filter for Multiple Maneuvering Targets Tracking

In this paper, we present a novel multiple-model probability hypothesis density (MMPHD) filter for multiple maneuvering targets tracking. In the proposed MMPHD filter, the multiple models are composed of two models, namely a constant velocity (CV) model and a "current" statistical (CS) model, and the PHD is approximated by a set of weighted random samples propagated over time using sequential Monte Carlo (SMC) methods. This resulting filter requires no knowledge of models and model transition probabilities for different maneuvering motions. Simulation results demonstrate that compared with the standard MMPHD filter, the proposed filter shows similar tracking performances but has faster processing rate.