Target Tracking based on Improved Unscented Particle Filter with Markov Chain Monte Carlo
Ramazan Havangi · IETE Journal of Research · 2017
In this paper, a target-tracking algorithm based on improved unscented particle filter with the Markov chain Monte Carlo (MCMC) is proposed. In the proposed method, the improved unscented Kalman filter (UKF) is used to generate the proposal distribution, and particle swarm optimization (PSO) integrates into the UKF proposal. Moreover, the sample impoverishment created by resampling step is restrained with MCMC move step after the resampling. Experiments are presented to evaluate the performance of the proposed algorithm. The results show that the proposed algorithm has more significant advantages in tracking accuracy than other classical algorithms.