Multi-agent snake-like motion with reactive obstacle avoidance

Abstract

This chapter presents a geometric constraint satisfaction approach to serial linkage (chain) motion and reactive obstacle avoidance, based on a multi-agent architecture. Application of this architecture to achieve real-time motion planning for serial manipulators and mobile robot snake-line swarm formations is then suggested. A hierarchy of “master–slave relationships is used, with the event of an autonomous motion of a controller agent propagating to its two neighboring in the formation chain and progressively further on toward the two endpoints of the chain. At each propagation step, a constraint preservation mechanism enforces the respecting of minimum and maximum distance constraints between pairs of consecutive chain agents. The emergent behavior of the multi-agent system equals to having the moving part of the chain push or pull the two subparts of the chain it connects. To cater for fixed base manipulators, and to allow replaning in case some slave part of the chain formation cannot adapt to its master's motion, being trapped in some obstacles or malfunctioning, the notion of a “master-vetoable slave” relationship is used, where a slave part can object (veto) to the motion of its master part.