研究一类具有输入滞后的离散非线性系统的准滑模控制问题 .根据滑模控制原理和神经网络的逼近能力 ,提出了一种基于神经网络预测的准滑模控制器设计方法 ,给出了神经网络预测器的自适应算法 .通过理论分析和仿真结果 ,证明了神经网络预测器的自适应算法是收敛的 ,闭环准滑模控制系统是稳定的 ,跟踪误差可收敛到零的一个邻域内 A class of quasi-sliding mode control problems for a class of discrete nonlinear systems with input hysteresis is studied. According to the principle of sliding mode control and the approximation capability of neural networks, a quasi-sliding mode controller design method based on neural network prediction is proposed. The adaptive algorithm of neural network predictor is proved by theoretical analysis and simulation results. It is proved that the adaptive algorithm of neural network predictor is convergent and the closed-loop quasi-sliding mode control system is stable and the tracking error can converge to a neighborhood of zero