Optimized task allocation in cooperative robot teams

Abstract

This paper presents a novel task allocation methodology based on supervisory control theory applicable to cooperative robot teams. A team of heterogeneous robots with different sensory capabilities is considered. The overall team's mission is decomposed into multiple tasks which are assigned to one or more robots. An evaluation function is utilized to provide optimal task allocation. The proposed controller design methodology demonstrates flexibility in task assignments and robot coordination, and it is tolerant to robot failures and repairs. A warehouse patrolling application is used to demonstrate the implementation aspect of the proposed methodology.