Variable structure control of unknown parameters DC servo systems using CMAC-based learning approach

A CMAC-based controller with a compensating neural network and an update rule is proposed to design the variable structure control (VSC) of unknown parameters DC servo systems. By introducing a stabilizer controller and a CMAC neural network to construct the VSC control law, the new control scheme performs the equivalent control by a real-time learning algorithm. The stabilizer controller is designed by using the Lyapunov stability theory and the updating rule of the CMAC weights is obtained by using the gradient descent method. Simulation results of a simplified robot link model demonstrate the effectiveness and robustness of the proposed controller.