A Submodular Receding Horizon Control Strategy to Distributed Persistent Monitoring

This paper investigates the multi-agent persistent monitoring problem via a novel distributed submodular receding horizon control approach. In order to approximate global performance, with definition of sub-modularity, original objective is divided into several local objectives in framework, and optimal trajectories each agent are obtained by taking account neighborhood information. Specifically, optimization derived from target states information received their neighboring agents. Based on sub-modularity objective, greedy algorithm proposed. As result, coordinates agents asynchronously optimizes its trajectory independently, which reduces computational complexity while achieving performance as much possible. The conditions established ensure estimation error converges bounded performance. Finally, simulation results show effectiveness proposed method.