A sampling-based path planner for dual-arm manipulation

Abstract

This paper presents a method to compute collision-free coordinated manipulation paths for multi-arm robot systems. The method extends previous work by incorporating an explicit treatment of singular configurations, which permit to find solution paths requiring the robot reconfiguration. Such reconfiguration motions are neglected in most of related work. The performance of the planner is analyzed through experiments with academic examples. The generality of the approach is demonstrated by its application to a model of the complex DLRpsilas Justin robot.