Robust Adaptive Control Based On Neural Network For Nonlinear Uncertain Systems

Nonlinear system control is an important topic in control theory. Owing to the complexity of system description and some inevitable uncertainties including modeling error, parameter time delay and disturbance, we fail to acquire the satisfied solution and many conclusions of modern control theory can't play an important role in practical engineering application. Uncertain nonlinear system control research has been retaining certain difficulty up to now.Because neural network is of good capabilities in function approximation, it is common at present that most uncertain nonlinear system is set up based on neural network. Consequently, the corresponding adaptive control problem has received a great deal of attention in recent years. Some correlative issues in this area are studied in this paper, such as the controller design problem of uncertain nonlinear systems, which have BTT missile model and unknown control direction. An adaptive NN controller scheme which employs backstepping methodology based on Lyapunov stability theory is designed in this paper. A shifting mechanism based on variable structure control is put forward for unknown control direction.The main work in this paper is summarized as follows: First, a direct adaptive neural scheme is used in the nonlinear BTT missile controller. Second, an adaptive NN controller scheme combined with shifting mechanism is designed for a class of uncertain nonlinear systems with unknown control direction. Third, the simulation of some examples shows the validity of control algorithm in this paper.