Variational Bayesian approach for joint multitarget tracking of multiple detection systems

Different from the traditional single detection systems (SDS) assuming that one target generates at most one detection per scan, there exists a class of multiple detection systems (MDS) where each detection may originate from the interested target via one of multiple propagation modes or from the clutter, while the correspondence among targets, measurements, and propagation modes is unknown. The performance of MDS can be improved if multiple detections from the same target are effectively utilized, but suffers from two major challenges. The first is multimode detection that determines the optimal number of targets automatically. The second is multimode tracking that calculates the target-to-measurement-to-mode assignment matrices to estimate target states. This paper introduces a novel probabilistic joint detection and tracking algorithm (JDT-VB) that incorporates data association, mode association, state estimation and automatic track management based on the variational Bayesian framework. The relevant analytical solutions are derived in a closed-form iterative manner, which is effective for dealing with the coupling issue of multimode data association identification risk and state estimation error.