A broken-track association method for robust multi-target tracking adopting multi-view Doppler measurement information
Jiaqi Zhang,Cao Zeng,Haihong Tao,Yuhong Zhang,Shihua Zhao,Qirui Wu
DOI: https://doi.org/10.1016/j.sigpro.2024.109815
IF: 4.729
2024-11-28
Signal Processing
Abstract:Due to reasons such as target maneuvering and track crossing, mistaken track association may be caused. When the target is in Doppler blind zone, it often leads to the loss of measurement information, which is reflected in the situation as track breakage. In response to the demand of track repair in the case of track breakage, we propose a robust multi-target broken-track association method that comprehensively utilizes multi-view Doppler measurement information. Firstly, based on the derivation of measurement coordinate transformation bias, the multi-sensor Doppler measurement after error correction is fused to obtain the heading velocity and heading acceleration measurement of the target. Secondly, the multi-scan association cost function based on the heading velocity and heading acceleration measurement is constructed, and the optimal correlation sequence is obtained by minimizing the complexity of the cost function. Then, for the case of missing measurements, the optimal association sequence is used to extrapolate the missing measurements, thereby accomplishing the correlation among the broken track segment, the extrapolated measurement and the optimal correlation sequence. Thirdly, we design a multi-scan GLMB smoother to perform forward prediction and backward smoothing on the above correlation results to improve the smoothness of the track. Simulation and actual experimental results suggest that our proposed method can effectively deal with the track breakage situation of dense targets, particularly in track integrity, track accuracy and robustness compared with the existing approaches.
engineering, electrical & electronic