Research On Methods Of Maneuvering Target Tracking Based On Particle Filter

Maneuvering Target Tracking is a study of the mobile target estimation problem which can not describe the target maneuverability accurately. Because its wide and effective applications and theories, it has been received more applications in both military and civil areas. However, due to more and more complex environment and the great change of maneuvering characteristics, more sophisticated tracking systems are desired for various applied systems. Therefore, looking for more effective real-time maneuvering target tracking method is of great significance.This dissertation research maneuvering target tracking based on particle filter, some real-time and effective algorithms have been proposed, to resolve the problems of complexity and low precision. Firstly, for real-time maneuvering target tracking, Multiple Mode-Quasi Monte Carlo-Gaussian Particle Filter (MM-QMC-GPF) and Iterative Multiple Model Particle Filter are proposed. The former algorithm have modified the state sampling method, QMC sampling insteads of the traditional MC sampling, the latter algorithm produces the proposal distribution according to iterative process. Simulation results show that the two new methods reduce the complexity of tracking maneuvering target, prove the real-time property effectively. Secondly, based on the GPF algorithm, a new algorithm which adaptively adjust the sampling interval is proposed, adjust the sampling interval adaptively according to the 2-norm of predicted states covariance matrix. Simulation results prove that the time used up is reduced distinctly without affect the precision. Finally, research an effective real-time variable rate particle filter (VRPF). This algorithm is of the precision and real-time property, have well application value for project practice.