Intelligent path planning of robotic arm considering dynamics properties
L. Zhang, Q. Li, Shufang Wang · IET conference proceedings. · 2022
In order to improve the intelligence of robotic arm, this paper investigates an intelligent path planning method based on Deep Deterministic Policy Gradient. The dynamics model of a 2 degree-of-freedom robotic arm is established first. On this basis, the joint driven torque is chosen as the action space, and the set of joint angle, joint angular velocity and the vector from the end-effector position to the target position is chosen as the state space. Considering the continuous characteristics of the action space and the state space, Deep Deterministic Policy Gradient method is adopted to train the robotic arm agent. The training results prove that this method can improve the intelligence of the robotic arm.