Immersion And Invariance-Based Adaptive Control Design Of Switched Nonlinear Systems

The increasing complexity of modern industrial production has led to higher and higher requirements for the performance of control systems.The actual system is restricted by various conditions,it is inevitable that there is a certain nonlinearity.It is difficult for linear control methods to meet the requirements of control,so nonlinear control has become one of the most challenging subjects in the study of control theory.When the system modeling has unknown parameters,the research of adaptive control is particularly important.The traditional adaptive control has certain limitations based on the principle of deterministic equivalence,however,the thesis is based on immersion and invariant adaptive control method.The method allows a prespecified consistently stable dynamic to be assigned to the parameter estimation error,resulting in a modular design that is easier to adjust than traditional adaptive control.In addition,switched systems have received increasing attention in recent years because most actual systems can be modeled as switched systems,such as power systems,mechanical systems,transportation systems.It is an important way to study it by using the switched theory.In addition,sliding mode control is widely used because of its good robustness and high reliability.At present,the research on non-switched system immersion and invariant adaptive control has yielded a lot of results,but the research on switched system has little results.The thesis focuses on nonlinear switched systems with unknown parameters and disturbances.The immersion and invariant-based adaptive control problems with dwell time is studied.The detail is as follows:First,the problem of adaptive stabilization via immersion and invariance for a class of switched nonlinear systems in strict-feedback form with unknown constant parameters under dwell time is investigated.By adopting common Lyapunov function and the immersion and invariance(I&I)technique which does not rely on certainty equivalence principle,error estimator and continuous state feedback controllers for subsystems are designed.By the generalized Barbalat’s lemma,the switched system is proved to be globally asymptotically stable.By exploiting the backstepping method,a common virtual control function is designed to avoid different coordinate transformations for different subsystems.Then,two examples,which include a mass-spring-damping system,are provided to demonstrate the effectiveness of the proposed design method.Second,the problem of adaptive tracking control based on immersion and invariance for nonlinear switched systems is studied with uncertain external disturbances.A near space vehicle model with uncertain external disturbances is established.Then the immersion and invariant method is used to estimate the unknown disturbances.The manifold of the estimation error is established and the adaptive law is designed.It is guaranteed that the estimation errors for disturbance are bounded.Also,a sliding mode controller is designed based on the method of backstepping and sliding mode control.The common Lyapunov function method is used to prove that all signals of the system are bounded under arbitrary switching.The effectiveness of the design method is also proved by simulation.Finally,conclusions and perspectives are presented in the end of the thesis.