Coverage control of multi-agent systems for ergodic exploration

In this paper, according to the ergodic theory, the ergodic metric is defined as the distance of the time average statistics of the agent's trajectories from the spatial distribution probability of the terrain. The feedback control law of the agent's dynamic is designed using the ergodic metric to achieve uniform or non-uniform coverage of a given area. At the same time, in order to expand the detection range, and obtain more complete and accurate environmental information and performed different types of tasks, the multi-agent system covers in the form of formation, and can adapt to move along the boundary. In this paper, the combination of the leader-follower method and the artificial potential field method are used to control the formation of a multi-agent system.