Optimal path generation of a redundant manipulator with evolutionary programming

A kinematically redundant robot manipulator has a lot of useful properties, such as avoiding singularities, obstacles and optimized motions. The redundant manipulator can have those useful properties because it has an infinite number of solutions of inverse kinematics. Unfortunately, many solutions also make it hard to get the solution of inverse kinematics. In this paper we propose two schemes which can find the globally near optimal solution of redundant manipulator's inverse kinematics using evolutionary programming (EP). Evolutionary programming is a stochastic search method based on principles of evolution and hereditary. The two schemes are mainly different on chromosome representation method. The chromosome of the first scheme is composed of all joint angles while the chromosome of the second scheme is composed of only redundant joint angles. To make a complete trajectory, interpolation is needed between two knots and the bound of interpolation error is calculated. We make the globally optimized trajectory of a 3-DOF RRR-type planer manipulator.