Neural network based manipulator control with time-optimal robot path tracking

A method aiming at implementing time-optimal central to enable robots under predictive neural network based control method to track paths to a prescribed tolerance is developed The paper considers the predictive control method for controlling robotic manipulators. A multi-layered neural network is provided as an identification model for the object to be controlled. The proposed method and a system for adaptive robot control provide connection to the tracking accuracy. Simulation results confirm the theories and suggest that the scheme can be applied in a robust fashion to achieve time-optimal path-tracking that is accurate to a prescribed tolerance and achieves good torque utilization most of the time.