Research On Adaptive Neural Network Tracking Control Of Uncertain Nonlinear Systems

Output tracking problem is an important research topic of control theory synthesis. The nonlinear characteristics of the real-world system and the inevitable uncertainties have a tremendous impact on the output tracking. Researches on the output tracking control of uncertain nonlinear systems have both theoretical and practical interests. Neural network has been proven to a powerful and effective method for controlling uncertain nonlinear systems for their abilities in nonlinear approximation, adaptation, generalization and associative memory and is herein the key technology permeating throughout this thesis. Based on the current research findings on intelligent control of nonlinear systems, this thesis studies adaptive neural network tracking control for uncertain nonlinear systems. The main contributions of this thesis are summarized as follows:(1) An adaptive neural netwok tracking control scheme is presented for a class of affine uncertain nonlinear matched SISO systems with zero dynamics by applying the Lyapunov theorem and gradient descent method. No robustifying control term is used in controller. Parameters in neural networks are updated using a gradient descent method which designed in order to minimize a quadratic cost function of the error between the unknown ideal implicit controller and the used neural networks controller. The final updated law is a nonlinear function of output error.. The convergence of parameters and the uniformly ultimately bounded of tracking error and all states of the corresponding closed-loop system are demonstrated by Lyapunov stability theorem.(2) a robust tracking control approach based on neural network disturbance observer is presented for a class of uncertain nonlinear MIMO systems under the unknown external disturbance and internal uncertainty to reduce the restrain conditions on the disturbance restrained condition.The compound disturbance consists of external disturbance , internal uncertainty and the cross-coupling of subsystems by applying"dominant input"concept.The distance observer is designed with the neural networks for monitoring the compound disturbance. The scheme combine neural network control with control and adaptive control. It is shown that the proposed schemes guarantee the the closed-loop system is stable and all the signals in the system are uniformly ultimately bounded and the influence of compound disturbance on the tracking error is attenuated to a prescribed level.(3) An output feedback tracking control algorithm using neural network for a class of uncertain affine nonlinear MIMO systems with external disturbances is presented, under the constraints that only the system output variables can be measured. By applying the"dominant input"concept, a MIMO system is divided into multiple SISO subsystems. Only the output error is used in control laws and weights update laws. No state observer or additional low-pass filter to make the estimation error dynamics strictly positive real(SPR) is employed in the algorithm. The design iof systems is simplified. The method has the significant value of theory and applications. The stability of closed-loop system and signals boundness are demonstrated by Lyapunov stability theorem.(4) An observer-based adaptive neural-network H∞tracking control scheme is presented for a class of nonaffine nonlinear systems with external disturbance and unavailable states. The controller consists of an equivalent controller and H∞controller. H∞controller is designed to attenuate the effect of external disturbance and approximation errors of the neural networks. The overall control scheme and the weights update laws based on Lyapunov theory can guarantee asymptotic convergence of the tracking error to zero and attenuate the effect of the disturbance to a prescribed level.(5)The theoretical results in the thesis are successfully applied to a two link rigid robot system.In the end, the main results of the thesis are summarized, and the fields for further investigation are expected.