Finite-Time Parameter Estimation of Switched Systems

An estimator is proposed for switched systems to estimate their linear unknown parameters within finite time. With the excitation condition satisfied, we can calculate the accurate value of the unknown parameters by constructing a switched estimator and some auxiliary variables. In the absence of excitation condition, a switched estimation law is given to ensure convergence of the state estimation error and boundedness of the parameter estimation error. Using a simulation, we show how the estimator is constructed for a model reference adaptive control system and how to enhance the adaptive control performance by resetting the control parameter when the accurate estimation is available.