The Design Of Model Reference Adaptive Switched Controller

Adaptive control has been a mature control method in the automation field for linear system especially. So if we still study adaptive control for linear systems, it may be not keep up with the rapid development of science and technology. In this paper, we choose a class of nonlinear systems to be controlled objects and design the adaptive controller then use matlab to simulate. And model reference adaptive control is the main control method in this paper because there are lots of ideal models which we want to achieve no matter in industrial design or other fields in reality. It is a method that the real system asymptotical tracks the reference model. So during the study of theory, we found that if the system join some new ideas to adjust the control strategy, the results will be better than former. At the same time, through reading a lot of relative papers and books, and doing many theoretical derivations and simulations, we decide to combine model reference adaptive control with switched systems.Switched systems are special complicated systems, according to those years’ study about switched systems, we found that there are few control strategies as its simple structures and they have already been studied deeply. But as the complicated of nonlinear systems, switched systems for nonlinear system is still a hot topic in control field. So if the system combines switched control strategy and adaptive control strategy, the performance of this system will improved increasingly as well as the system stability,it also can achieve the aim of control, which is the creation of this paper, we discuss details of the design of model reference adaptive controller for nonlinear switched system in the following chapters. While we received the control error asymptotically convergence, but parameter estimation did not get any results. When the parameter estimates converge to real value, the parameter estimation error is zero, the adaptive controller is equivalent to a complete model of the feedback linearization controller. Finally, through Matlab simulation experiments verify the error asymptotically convergence to estimate and the stability and the parameter estimation error convergence of model reference adaptive switching controller.