Adaptive quasi-sliding mode control for uncertain discrete nonlinear system with time delay

The problem of prediction quasi-sliding mode control for a class of uncertain discrete nonlinear system with time delay is studied in the paper. Based on the approximation capability of neural networks, an NN-based model is built and an NN-based d-step-ahead predictor is composed. A synthetical scheme of prediction quasi-sliding mode control is proposed. In order to achieve robust performance against system´s parameter variations and external disturbances, or reduce tracking errors, the NN rights both of model and predictor are regulated adaptively on line. By theoretical analysis and simulation results, the adaptive algorithm is verified to be convergent, the closed-loop prediction quasi-sliding mode control system is proven to be stable, with good convergence and robustness.