ROBUST NEURAL NETWORK CONTROL OF SPACE ROBOT SYSTEMWITH FLEXIBLE JOINTS

摘要

A robust neural network control scheme for the flexible-joint space robot system with uncertain parameters isdiscussed in this paper. With linear and angular momentumconservation of the system, the dynamic equations of therobot system are established by the Lagrange method. Then, a kind of joint flexibility compensator is introduced andthe singular perturbation approach is utilized to decompose the flexible-joint space robot system into two subsystems.After that, a torque differential feedback controller is given to stabilize the fast subsystem and a robust neuralnetwork control scheme is proposed for the slow subsystem to track the expected trajectory in joint space under theeffects of uncertain parameters. The convergence of the whole system is analyzed by Lyapunov theorem and verifiedby the numerical simulations. The theory analysis and simulation results show the feasibility and effectiveness of theproposed control scheme.