Tracking of Multiple Targets using GM-PHD Filter with Nonlinear Measurement Model

In this paper, a method of multi-target tracking based on Gaussian mixture probability hypothesis density(GM-PHD) filter is investigated against targets that have drastic velocity change during their maneuvering. GM-PHD filter, a recursive algorithm to track a number of targets, applies the probability hypothesis density recursion to propagate the posterior intensity. To consider practical environments, nonlinearity of an observation model is considered and techniques of extended Kalman filter(EKF) and unscented Kalman filter(UKF) are applied. In a multi-target environment when targets may appear and disappear, performance of multiple-target tracking problem is shown by the simulation.