Cooperative patrol planning of multi-robot systems by a competitive auction system

Abstract

This research focuses on how a multi-robot system can work cooperatively to complete patrol missions. The planning and assignment of patrol points to the robots are the major issues to tackle with. A cooperative auction system (CAS) is proposed to solve the problem of patrol planning. Here, each mobile robot picks its own patrol points via the cooperative auction system and the system will continuously re-auction, based on the team work performance, to further improve the performance of cooperation for a large scale multi-robot system. From the experimental results, the proposed approach demonstrates several advantages, such as less time complexity, lower routing path cost, better workload balancing among robots.