Trajectory planning for saving energy of a flexible manipulator using soft computing methods

Abstract

This paper presents a trajectory planning method for saving the operating energy of a flexible manipulator in point-to-point (PTP) motion. An artificial neural network (ANN) is employed to generate the desired joint angle, and then, particle swarm optimization (PSO) is used as the learning algorithm. The sum of the motor torques is adopted as the objective function in the PSO algorithm. By operating the manipulator along the trajectory obtained using the proposed method, residual vibrations can also be suppressed. The applicability and effectiveness of the proposed trajectory planning method are confirmed by performing numerical simulation and verified by experimental results.