Experiments on a time-optimal trajectory planning method based on neural networks

Operating robots along time-optimal trajectories can significantly increase the productivity of robot systems. To plan realistic optimal trajectories, the robot dynamics have to be described precisely. In this paper, a neural network (NN) based algorithm for time-optimal trajectory planning is introduced. This method utilises neural networks for representing the inverse dynamics of the robot. As the proposed neural networks can be trained using data obtained from exciting the robot with given torque inputs, they will capture the complete dynamics of the robot system. Therefore, the optimal trajectories generated by using the neural network model will be more realistic than those obtained using robot dynamic equations with nominal parameters. Time-optimal trajectories are generated for a PUMA robot to demonstrate the proposed method.