Multi-target GIMMJPDA tracking algorithm based on prior knowledge

摘要

In complex clutter environment, the traditional ground tracking model is mismatched and the association probability is inaccurate. Based on the prior knowledge of radar, target and environment, a multi-target ground interacting multiple models and joint probabilistic data association (GIMMJPDA) tracking algorithm is proposed. By dynamically adding and deleting the prior target motion trajectory in the traditional target model set, the algorithm is to solve the ground target tracking model mismatch problem and reduce the influence of clutter on the target by the clutter correlation probability estimation. The simulation results show that the tracking algorithm based on the prior knowledge can improve the tracking accuracy of the ground targets.