Robust Finite-Time Parameter Estimation Using a Hybrid Systems Framework

Abstract

This technical note investigates the problem of parameter estimation for a class of nonlinear systems. We show that by placing a member of this class of systems into a hybrid dynamical system, we can identify its parameter exactly and in finite time. Furthermore, by using tools developed for hybrid systems, we show that the estimation has robust properties. In particular, the hybrid system contains a compact globally asymptotically stable set where the stability is robust to small perturbations. We also present conditions that ensure convergence of the parameter estimate.