本程序为基于RBF神经网络上界自适应学习的滑模控制，适用于实际无法测量上界值的情况。-This procedure is based on the RBF neural network to learn the upper bound of adaptive sliding mode control, can not be applied to the actual measurement of the upper bound values.