Multi-target tracking algorithm of vehicle-mounted millimeter-wave radar based on IMM-JPDA-STUKF

Abstract:
In order to enhance the tracking performance of vehicle-mounted millimeter-wave radar on multi-target vehicles on the road and improve the safety of vehicle driving, this paper proposes a multi-target tracking algorithm of vehicle-mounted millimeter-wave radar based on interactive multi-model joint probabilistic data association strong tracking unscented Kalman filter (IMM-JPDA-STUKF). The scene simulation is carried out according to the change of the common vehicle motion state on the road. The simulation experiment shows that the algorithm can effectively improve the distance and speed tracking accuracy of the vehicle in the multi-target tracking state compared with the IMM-JPDA-UKF algorithm.