First-Order Synergies for Motion Planning of Anthropomorphic Dual-Arm Robots

This paper addresses the problem of designing a planning algorithm for anthropomorphic dual-arm robotic systems to find paths that mimics the movements of real human beings by using first-order synergies (correlations between joint velocities). The key idea of the proposal is to convert captured human movements into a vector field of velocities, defined in the configuration space of the robot, and use it to guide the search of a solution path. The motion planning is solved using the proposed algorithm, called FOS-BKPIECE, that is a bidirectional version of the KPIECE planner working with an improved version of the extension procedure of the VF-RRT planner. The obtained robot movements follow the directions of the defined vector field and hence allow the robot to solve the task in a human-like fashion. The paper presents a description of the proposed approach as well as results from conceptual and application examples, the latter using a real anthropomorphic dual-arm robotic system. A thorough comparison with other previous planning algorithms shows that the proposed approach obtains better results.