Integral variable structure control of nonlinear system using CMAC-based learning approach

A CMAC-based controller with a compensating neural network and an update rule is proposed to design the integral variable structure control (IVSC) of nonlinear system. The control scheme comprises a stabilizer controller and a CMAC neural network. Based on the Lyapunov theorem, the stabilizer controller guarantees the global stability of the system. The CMAC neural network performs the equivalent control by a real-time learning algorithm. The proposed control scheme is globally stable in the sense that all signals involved are bounded. The new IVSC control scheme reduced the dependency to system parameters. Simulation results of numerical example demonstrate the effectiveness and robustness of the proposed controller.