Visibility-Enhanced Elastic Tracker Algorithm for Stable and Flexible Aerial Tracking

This paper proposes the visibility-enhanced Elastic Tracker algorithm to address the line-of-sight occlusion problem presented in previous algorithms for target tracking with unmanned aerial vehicles (UAV). The path generated by the traditional A* algorithm often provides poor visibility for the UAV to track its target. To relieve this issue, we introduce a novel visibility heuristic function that can improve the visibility of the target along the tracking path. In addition, to further improve the visibility of the target, we first incorporate the occlusion effect (OE) metric against obstacles into the trajectory optimization process. Then, we devise two new visibility constraint terms, multi-target observation angle (MAO) and multi-target movement speed angle (MTSA), to enhance the visibility of the target and reduce the risk of tracking failure. Finally, the superiority of our method is verified through simulations by comparing it with existing methods. The comparative results show that our method can effectively relieve the issue of targets being occluded by obstacles and improve the tracking success rate.