Controle adaptativo de sistemas discretos baseado em aproximadores fuzzy Takagi-Sugeno: Mínimos quadrados recursivo com regularização da covariância

Abstract

In order to deal with the parameter drift problem in discrete-time adaptive control, this work proposes a time-varying regularization of the covariance matrix in the recursive least squares method with a forgetting factor. A set of synthesis conditions, based on Linear Matrix Inequalities, is presented for the control law that guarantees that the closed loop system is uniformly ultimately bounded when employed together with the proposed adaptation law. Throughout all of the analysis, the approximation error is taken into consideration, since it always exists in the context of universal approximators. Computational simulations are performed to assess the efficacy of the proposed approach, by means of comparisons against traditional recursive least squares methods.