Adaptive Sliding-Mode Control of Nonlinear Systems Using Neural Network Approach

This paper is concerned with the adaptive sliding-mode control of nonlinear dynamic systems with model uncertainties. The proposed control method combines the advantages of sliding-mode control and backstepping methodology, such that the requirement of the restrictive matching condition is removed, which seriously clams the application of sliding-mode control. In the control scheme, networks of Gaussian radial basis functions with variable weights are used to compensate the model uncertainties. The adaptive law developed using the Lyapunov synthesis approach guarantees the stability of the control scheme. The performance is illustrated by experimental studies with a flexible-joint manipulator.