Motion Planning for Dual-Arm Robot Based on Soft Actor-Critic

In this paper, a motion planning method based on the Soft Actor-Critic (SAC) is designed for dual-arm robot with two 7-Degree-of-Freedom (7-DOF) arms so that can effectively avoid self-collision and at same time joint limits singularities of arm. The left-arm right-arm each have neural network to control its position orientation. Dual-agent training, distributed training structure, progressive environment are used train networks. During process, one arm regarded as other arm, agents trained time. input part proposed method, all parameters come from angle axis kinematic calculation, no additional sensors needed, easier transplant different robots. With some appropriate inputs reward functions design, perform expected avoidance Finally, experiments simulation tests in Gazebo simulator actual laboratory-made presented illustrate SAC-based feasible practicable self-collision, limits, singularities.