Distributed sweep coverage algorithm of multi-agent systems using workload memory

Abstract

This paper addresses the sweep coverage problem of multi-agent systems in the uncertain environment. A novel formulation of distributed sweep coverage is proposed for multiple agents to cooperatively complete the workload in the coverage region. To save the sweep time, each agent takes part in partitioning the whole region using its partition bar while sweeping its own subregion at a constant rate. The trajectories of partition bars form the boundaries between adjacent subregions. Essentially, the partition operation is carried out by means of workload memory in order to balance the workload in each subregion. In particular, it is proved that the dynamics of multi-agent system is input-to-state stable. Theoretical analysis is conducted to obtain the upper bound of the error between the actual sweep time and the optimal sweep time. Moreover, a sufficient condition is provided to avoid the collision of partition bars during the partition. Finally, numerical simulations demonstrate the effectiveness of the proposed approach.