LINEAR CONTROL OF THE ELECTRIC DRIVES OF BRUSHLESS DC MOTOR OF THE ROBOT MANIPULATOR USING A NEURAL TORQUE COMPENSATOR BASED ON ROBUST CONTROL
Presents a model of robot manipulator control based on an electrical drives of Brushless DC motor using an artificial neuron network (ANN) based on robust control as a torque compensator in the presence of uncertain dynamics due to unknown load and changes in system parameters. The structures of the neural torque compensator (NTC) are analyzed. The quality criterion of the control system was evaluated in combination with a nonlinear optimal torque compensator based on LQG control and a NTC based on Theta-D robust control. The learning process of a NTC is based on reference data generated by a robust controller operating on the basis of a dynamic system quality criterion. Optimal proportional-integral regulators of electric drives, including a torque loop with a feedback signal, are provided by a NTC of the manipulator torque, speed loop and position loop, provide the control system with the required dynamic characteristics.
Authors: T. D. Khoa, M. P. Belov
Direction: Electrical Engineering
Keywords: Robust control θ-D, neural compensator, artificial neural network
View full article