A Multi-UAV Joint Search Method with Repeated Coverage Avoidance Based on Potential Game

In this paper, a multiple unmanned aerial vehicles (UAVs) joint search method based on potential game is proposed, which takes into account the repeated coverage problem in the search process. The UAV action strategy is determined by the coverage benefit of the mission area covered by the UAV and its neighbors to realize the cooperative movement of the swarm. Then the two-stage learning algorithm is proposed to solve the joint motion problem of the UAV swarm. In the first learning stage, the UAV swarm adopts the greedy strategy to make the most of the exploratory capability of the UAV swarm. In the second stage, the UAV swarm adopts the binary log-linear learning (BLL) algorithm, which gradually makes the deployment of the UAV swarm converge to a Nash equilibrium. Finally, the validity and the efficiency of the proposed method is verified by simulations.