Obstacle avoidance of 7-DOF Redundant Manipulators

Abstract

This paper mainly focuses on the problem with obstacle avoidance of 7-DOF redundant robots. Firstly, the end-effector Jacobian is given to reflect the relationship between joint angle rates and end-effector rates. From geometric features of the manipulator, the Jacobian matrix of the arm angle is built to solve the kinematics of the redundant manipulator. Furthermore a minimum intersection distance is proposed through the intersection test between the oriented bounding box of links and the obstacles. The minimum distance is directly obtained by using the target point coordinate value. In the end, the minimum distance is calculated for modified RRT* algorithm to the redundant robot, null-space matrix is used for path planning. The experiment is demonstrated by YuMi left arm. The results of simulations strongly prove the validity and efficiency of this method.