Leader–Follower Interactive Potential for Target Enclosing of Perception-Limited UAV Groups

摘要

Target enclosing is representative of the multirobot intelligence and coordination as it is pervasively involved in many practical scenarios. However, the limited perceptions of an individual robot bring in many challenges for the enclosing control such as the requirements to connectivity preservation and improvement of robustness. Motivated by this topic, a general control law based on an enclosing potential combined with a consensus-based target observer is presented for the target enclosing of unmanned aerial vehicles with limited perceptual abilities. Asymptotic stability can be guaranteed under some assumptions of target's property and network connectivity, whereas the enclosing control law may lose efficacy when capturing an agile target. Then, a novel group of leader–follower interactive potentials combined with a modified observer are derived to promote the robustness by strengthening the interactions between informed and uninformed nodes in a bicyclic spatial layout. Comparative results show the advantages of the proposed algorithms over state-of-the-art approaches on the strong stabilities of enclosing layouts with limited sensing.