Global adaptive control of a class of uncertain nonlinear systems using neural networks

The paper considers the problem of global adaptive tracking for a class of uncertain nonlinear systems in which the uncertainty is impossible to be parameterized. With the help of the technique of unit partition in differential topology, a result on global approximation of function using neural networks is proved. Based the result, a method of global adaptive neural network control for the uncertain nonlinear system is presented. The method ensures that the tracking error converges to an arbitrarily given small neighborhood of zero. When the tracked signal is constant, the tracking error converges to zero.