考虑了一类具有零动态的非仿射非线性不确定系统的神经网络直接自适应跟踪控制问题.控制信号由神经网络系统直接产生,无需另外设计系统估计器及鲁棒控制项.采用梯度下降方法以最小化神经网络控制器与未知理想控制器的误差代价函数产生神经网络自适应参数更新律.应用Lyapunov方法证明了闭环系统的稳定性及跟踪误差和相应闭环系统的所有状态最终一致有界性.最后针对带有外部扰动的非仿射非线性系统的仿真结果验证了该文方法的有效性.

The paper deals with direct adaptive neural network tracking control for a class of uncertain nonaffine nonlinear systems with zero dynamics. Within this scheme, a neural network system is used to generate directly the control input signal without system estimation and robust control term. In the used neural network system the adjustable adaptive parameters are updated by using the gradient descent algorithm which is designed to minimize a quadratic cost function of the rror between the used neural network control law and the unknown ideal implicit controller. The stability analysis of the closed-loop system is performed by using a Lyapunov approach. It is shown that the tracking error and all states of the corresponding closed-loop system are uniformly ultimately bounded. The availability of the proposed method is demonstrated through the simulation of a nonaffine nonlinear system with disturbance.