针对一类非线性系统 ,将其变换为易于设计变结构控制规律的规范型。利用小脑关节神经网络( CMAC)估计系统中的不确定函数 ,利用 CMAC神经网络和多面滑模技术估计出变换后系统的状态 ,最后利用变结构控制技术设计出控制器 ,特点是无需已知不确定性函数及其各阶导数的上界 ,与经典设计方法相比 ,所提出的方案允许非参数化不确定性。最后将此方法应用于空空弹控制系统的设计中 ,仿真结果表明了该方法的有效性 For a class of nonlinear systems, it is transformed into a canonical form that is easy to design variable structure control law. The CMAC neural network (CMAC) is used to estimate the uncertain function in the system. The CMAC neural network and the multi-surface sliding mode are used to estimate the state of the transformed system. Finally, the controller is designed by using the variable structure control technology. The upper bound of the deterministic function and its derivatives is compared with that of the classical design method. The proposed scheme allows non-parametric uncertainties. Finally, this method is applied to the design of air-to-air bomb control system. The simulation results show the effectiveness of the method