Multi-target Tracking Based on the Collaboration of Mother-daughter Multi-UAV

In multi-target tracking by multiple UAVs (unmanned aerial vehicles), the obstacle positions are random, and multiple targets scatter and maybe move out of sensor perception range. To solve the problems, a multi-target joint coverage and tracking method based on mother-daughter multi-UAV system is proposed. Firstly, the minimum bounding circle generated by Welzl's algorithm is used to guide the mother UAV to plan a tracking trajectory for covering more targets. Then, the daughter UAVs are scheduled to approach the targets for one-on-one tracking, based on the positions of multiple targets predicted by the mother UAV. Finally, the mother and daughter UAVs continuously and stably observe multiple targets in the area through online collaborative tracking. Simulation results show that the proposed method can effectively enhance the observation capability of multiple UAVs for multiple targets, and the mother-daughter UAV multi-target tracking experiments in outdoor scene verify the effectiveness of this method deployed on actual multi-UAV platforms.