Session: CIE-07-01: CAPPD General

Paper Number: 114868

114868 - Selection of Inverse Kinematics Solution Type for Cooperative Robots and Singularity Avoidance Based on Reinforcement Learning 

In this paper, we propose a method to avoid singularity by selecting solution types of inverse kinematics for a six-degree-of-freedom (6-DOF) manipulator based on reinforcement learning. A general 6-DOF manipulator has eight solution types of inverse kinematics for any position and posture of the end-effector. Because of the complex structure of cooperative robots to prevent pinching during cooperative operations, inverse kinematics is often solved by numerical methods. Because numerical solution depends on the initial values, it is difficult to select the suitable solution types. According to the selection of solution types, the robot may pass through a singularity, causing some joints to rotate rapidly. To avoid this, the solution types must be selected considering the entire motion path of the robot. The proposed method uses Deep Q-Learning (DQN), a type of reinforcement learning, to select the solution types that minimize the angular velocity of each joint during the motion path. This is verified by a 6-DOF cooperative robot, where the robot is commanded to take a path through singularity, and the solution types of inverse kinematics is selected by DQN. As a result, the singularity is avoided by selecting suitable solution types, and the rotational speed of the joints is minimized.

Presenting Author: Daiki Kato Doshisha University

Presenting Author Biography: Daiki Kato is a doctoral course student in Japan. He has B.S. and M.F.A. in engineering from Doshisha University. His work focuses specifically on motion accuracy and control algorithms for industrial robots. Recently, he started working on cooperative robots.

Authors:

Daiki Kato Doshisha University
Naoki Maeda Doshisha University
Ayumu Takeuchi Doshisha University
Masataka Sekioka Doshisha University
Toshiki Hirogaki Doshisha University
Eiichi Aoyama Doshisha University