A novel non-collision path planning strategy for multi-manipulator cooperative manufacturing systems

Abstract

Analogous to the definition of human–robot interaction, the case of multiple manipulators with shared workspace, asynchronous manufacturing tasks, and independent objects is named as a multi-manipulator cooperative manufacturing system. Multi-manipulator cooperation is increasingly used in modern industrial manufacturing systems and requires collision-free path planning as a vital issue in terms of safety and efficiency. This study proposes a novel method called Sampling-based Operation Space Map Search method, which combines a map search method with a time-sampling-based method. Two candidate position determination methods are proposed to establish local planning maps for all manipulators individually during each sampling time interval. After a specific search map simplification process, the optimal local path fragments within each sampling period can be determined by the map search method. Then, all chronologically planned path segments can be glued together to generate collision-free paths for all manipulators in a multi-manipulator cooperative manufacturing system. The simulation results demonstrated that the proposed strategy could successfully achieve collision avoidance of dual manipulator system whilst meeting the real-time requirements for cooperative assembling scenarios. Compared with the conventional map search method, this proposal is highly effective at the cost of forgoing the global optimum. Further satisfactory simulation results for triple manipulators indicate that our algorithm can be extended to multi-manipulator cooperative manufacturing applications.