针对一类带有死区输入的未建模动态非线性系统,借助多层神经网络的逼近能力,提出了一种自适应控制方案.该方案放宽了未建模动态的假设限制,去掉了传统多层神经网络中的一个调节参数,消除了未建模动态对Lyapunov函数影响.理论证明了闭环系统所有信号半全局一致终结有界.计算机仿真验证了该方案的有效性。 For a class of unmodeled dynamic nonlinear systems with dead-zone inputs, an adaptive control scheme is proposed based on the approximation ability of multi-layer neural networks, which relaxes the assumption of unmodeled dynamics and removes the traditional An adjustment parameter in multi-layer neural network eliminates the influence of unmodeled dynamics on Lyapunov function. It is proved theoretically that all signals in a closed-loop system are semi-globally uniformly terminated and bounded. Computer simulations verify the effectiveness of the proposed scheme.