Application of repulsive force and genetic algorithm to multi-manipulator collision avoidance

Abstract

Manipulator collision avoidance using genetic algorithms is presented. Control gains in the collision avoidance control model are selected based on genetic algorithms. A repulsive force is artificially created using the distances between the robot links and obstacles, which are generated by a distance computation algorithm. Real-time manipulator collision avoidance control has achieved. A repulsive force gain is introduced through the approaches for definition of link coordinate frames and kinematics computations. The safety distance between objects is affected by the repulsive force gain. This makes the safety zone adjustable and provides greater intelligence for robotic tasks under the ever-changing environment.