Multi-Target Tracking Method for Collaborative RSUs
Yi Zhang, Wenfeng Du, Youfang Hong, Song Xu, Ming Cen · 2022 34th Chinese Control and Decision Conference (CCDC) · 2022
Road Side Unit (RSU) based target tracking takes a key role in Cooperative Vehicle Infrastructure System. In current solution, each RSU implements target detection and tracking independently and lacks effective utilization of scenario information from other RSUs. To improve the tracking performance of road side sensors, a multi-target tracking method for collaborative RSUs is proposed in this paper. The RSU receives the target trajectory information broadcasted by neighboring RSUs as potential targets firstly. Then the measurements of targets detected by road side sensors are associated with existing target trajectories and potential target trajectories. The associated potential target trajectories are considered as existing target trajectories, and the unassociated measurements are utilized for tracking initiation. Finally, the associated existing target trajectories are updated by tracking filter and broadcasted out. The experimental results show that comparing with conventional solution, the proposed method can improve the tracking accuracy and convergence rate of the road side tracking filter.