Motion planning by genetic algorithm for a redundant manipulator using an evaluation function based on criteria of skilled operators

This paper proposes a motion planning method to cut a three dimensional workpiece by a redundant manipulator with six degrees of freedom. The method applies a genetic algorithm to optimize the rotational angles of the end-effector on a path. For a fitness function, an evaluation function is defined based on references from skilled operators. The proposed method reduces the operator's labor, so that he only has to determine a path without considering redundant parameters. Both simulations and experiments show the effectiveness of the proposed method.