Robust non-minimal order filter and smoother design for discrete-time uncertain systems

Carregando...
Imagem de Miniatura

Data

Título da Revista

ISSN da Revista

Título de Volume

Editor

Universidade Federal de Minas Gerais

Descrição

Tipo

Artigo de periódico

Título alternativo

Primeiro orientador

Membros da banca

Resumo

This paper is concerned with the design of robust non-minimal order H∞ filters for uncertain discrete-time linear systems. The uncertainty is assumed to be time-invariant and to belong to a polytope. The novelty is that a convex filtering design procedure with Linear Matrix Inequality constraints is proposed to synthesize guaranteed-cost filters with order greater than the order of the system. An H∞-norm bound for the transfer-function from the system input to the filtering error is adopted as performance criterion. The non-minimal order filters proposed generalize other existing filters with augmented structures from the literature and can provide better performance. An extension to the problem of robust smoothing is proposed as well. The procedure is illustrated by a numerical example.

Abstract

Assunto

Controle automático, Sistemas de tempo discreto

Palavras-chave

new approach for designing robust non-minimal order H∞ filters and fixed-lag smoothers for uncertain linear time-invariant discrete-time systems was proposed. The method does not impose any a priori structure on the matrices of the filter realization, which are parametrized as a full-order filter for an auxiliary system augmented with a non-observable subsystem., Parameter-dependent Lyapunov matrices and slack variables lead to improved upper bounds on the guaranteed H∞ performance of the filter. The method is extended to deal with the problem of robust non-minimal order fixed-lag smoothing. A numerical example illustrates and compares the improvements with other robust filters and smoothers in the literature.

Citação

Curso

Endereço externo

https://onlinelibrary.wiley.com/doi/10.1002/rnc.3592

Avaliação

Revisão

Suplementado Por

Referenciado Por