Estimação robusta de parâmetros variantes no tempo sujeita a estados estimados: teoria e aplicações em sistemas lpv

Abstract

This study addresses the problem of the parametric estimation of time-varying parameters with unavailable states and the effect of bounded disturbances in the context of Linear Parameter-Varying (LPV) systems. A robust filter with the generalized H2 norm is synthesized to estimate the system states, based on Linear Matrix Inequalities (LMIs), in conjunction with a direct robust adaptive law, to overcome the challenges posed by unavailable states, time-varying parameters, and bounded disturbances simultaneously. The joint convergence and stability of the filter and the adaptive law are analyzed and proven to be globally uniformly ultimately bounded (GUUB) under the well-known Persistence of Excitation (PE) condition of the system regressor matrix. Simulated experiments are conducted to demonstrate the advantages and disadvantages of the proposed procedure and to analyze its behavior.