Adaptive Neural Control for a Class of Pure-Feedback Nonlinear Systems Via Dynamic Surface Technique
Zongcheng Liu,Xinmin Dong,Jianping Xue,Hongbo Li,Yong Chen
DOI: https://doi.org/10.1109/tnnls.2015.2462127
IF: 14.255
2016-01-01
IEEE Transactions on Neural Networks and Learning Systems
Abstract:This brief addresses the adaptive control problem for a class of pure-feedback systems with nonaffine functions possibly being nondifferentiable. Without using the mean value theorem, the difficulty of the control design for pure-feedback systems is overcome by modeling the nonaffine functions appropriately. With the help of neural network approximators, an adaptive neural controller is developed by combining the dynamic surface control (DSC) and minimal learning parameter (MLP) techniques. The key features of our approach are that, first, the restrictive assumptions on the partial derivative of nonaffine functions are removed, second, the DSC technique is used to avoid "the explosion of complexity" in the backstepping design, and the number of adaptive parameters is reduced significantly using the MLP technique, third, smooth robust compensators are employed to circumvent the influences of approximation errors and disturbances. Furthermore, it is proved that all the signals in the closed-loop system are semiglobal uniformly ultimately bounded. Finally, the simulation results are provided to demonstrate the effectiveness of the designed method.