Research on Tracking and Recovery of Mobile Robot Target Following System
Kernel correlation filtering(KCF)algorithm is widely used in target following tasks of mobile robots be-cause of its high efficiency,but traditional KCF often fails due to target occlusion and other reasons.Therefore,a tracking quality assessment method is proposed in this paper,and combined with the YOLOv4-tiny algorithm,the chassis motion control strategy is designed according to the target position information,and the experiment is carried out on the robot platform.The results show that the improved algorithm can give the robot the ability of sensing and recovery in the case of tracking failure.
mobile RobotsKCFtracking quality assessmenttracking recoveryYOLOv4-tiny
万方数据
维普
