Adaptive Control For Nonlinear Time-delay Systems And Analysis Of Fuzzy Membership Function Generation Based NNs

Adaptive control using on-line function approximators for feedback lineariz-able systems has proven to be a very e?ective way to design controllers based onapproximate knowledge of the system dynamics. Practical implementation of suchcontrollers on the aircraft have demonstrated their stability and performance charac-teristics, as well as superior fault tolerance when there is redundancy in the control.These real world experiments provide great momentum for theoretical research innonlinear adaptive control systems using neural networks to approximate unknownfunctions.In this thesis, adaptive neural control is proposed for a class of nonlinear un-known state time-varying delay systems in block-triangular control structure. Radialbasis function(RBF) neural networks (NNs) are utilized to estimate the unknowncontinuous functions.In Chapter 1 of this paper, we introduce the background of the adaptive controland neural networks.In Chapter 2 and Chapter 3, adaptive neural control is investigated for a classof nonlinear state time-varying delay systems with unknown virtual control coef-ficients, unknown nonlinear functions by using integral-type Lyapunov-Krasovskiifunctionals, neural networks. The main advantage of our result not only e?cientlyavoids the controller singularity, but also relaxes the restriction on unknown virtualcontrol coe?cients. Boundedness of all the signals in the closed-loop of nonlin-ear systems is achieved, while The outputs of the systems are proven to convergeto a small neighborhood of the desired trajectories. In Chapter 2,adaptive neuralcontrol is proposed for a class of single- input single-output (SISO) nonlinear Un-known Time Delays systems. Then in chapter 3, we expanded the second chapter'smethod, adaptive neural control is proposed for a class of multi-input multi-output(MIMO) nonlinear unknown state time-varying delay systems in block-triangularcontrol structure. In Chapter 4, a new scheme is proposed to generate fuzzy membership functionswith unsupervised learning using self-organizing feature map.The proposed schemeand experimental results o?er a neural network view (NN view) on the meaning offuzzy membership.