Conflict-free motion of multiple mobile robots based on decentralized motion planning and negotiation

Abstract

This article presents a novel approach to conflict-resolution for multiple mobile robots. By use of online motion planning, a framework for negotiation is developed, which permits quick decentralized and parallel decision-making. The key objective of the negotiation procedure is dynamic assignment of robot motion priorities. Thus, robots involved in a local conflict situation cooperate in planning and execution of the lowest cost motion paths without application of any centralized components. The features required for individual and cooperative motion are embedded in a hybrid control architecture. Results obtained from realistic simulation of a multirobot environment and also from experiments performed with two mobile robots demonstrate the flexibility and the efficiency of the proposed method.