An improvement to the linear jump Markov system Gaussian mixture probability hypothesis density filter for maneuvering target tracking

Abstract

An improvement approach to the linear Gaussian jump Markov system (LGJMS) Gaussian Mixture probability hypothesis density (GM-PHD) filter is designed for multiple maneuvering targets tracking. This method, which called mixture LGJMS GM-PHD (MLGJMS GM-PHD) filter, is based on the theory of generalized psuedo Bayes of the first order after the update step of LGJMS GM-PHD. Compared with the existing LGJMS GM-PHD filter, simulation results show that the designed filter weights over the original one.