The Research On The Radar Multi-Target Tracking Methods Based On Particle Filter

With the rapid development of shipping in the 21st century, there is a growing emphasis on the radar security effectiveness in shipping. In the process of actual vessel sailing, there are usually multiple vessels rather a single ship, this proposed higher requirements to target tracking. In multiple target tracking, the joint probability data association is a better way to solve the data correlation. The particle filter is a better filtering method to solve nonlinear conditions. In this paper, the particle filter is combined with the joint probability data association to solve the problem of multi-target tracking. Therefore, the research of radar multi-target tracking method based on particle filter has theoretical significance and engineering application value.At first, this paper analyzes the multi-target tracking, including point trace admission, association processing, tracking filter and track management. The paper emphatically discusses the theory, the characteristics and classification of data correlation, analyzes the characteristics of the data association algorithms, such as: nearest neighbor data association (NNDA), probabilistic data association (PDA), multiple hypothesis tracking algorithm and joint probability data association (JPDA) etc. The effectiveness of each algorithm was verified after analyzing the theory and the experiment.Then this paper analyzes the basic theory and method of particle filter, including SIS, analyzes the problems it has and the solution strategy, and the simulation analysis to the influences of the particle number and noise were done on the particle filter, and gets the corresponding conclusions. The superiority of the particle filter algorithm to the EKF and UKF was verified by simulation experiments. The paper makes a detailed introduction to the joint probability data association. Finally, the paper combines the joint probability data association with particle filter to achieve reliable target tracking under the multi-target environment, and verifies its effectiveness after the simulation experiments.At last, this paper complete joint probability data association and particle filter algorithm under the Visual Studio 2010 environment, using real radar data to test the tracking result under the multi-target environment and display multi-target tracking result.