H 2/H ∞ formulation of LQR controls based on LMI for continuous-time uncertain systems
| Autor(a) principal: | |
|---|---|
| Data de Publicação: | 2021 |
| Outros Autores: | , |
| Tipo de documento: | Artigo |
| Idioma: | eng |
| Título da fonte: | Repositório Institucional da UNESP |
| Texto Completo: | http://dx.doi.org/10.1080/00207721.2020.1836283 http://hdl.handle.net/11449/208078 |
Resumo: | The classical theory of linear matrix inequality (LMI)-based linear quadratic regulator (LQR) control that is well established in literature does not provide, in the controller project phase, the occurrence of exogenous inputs acting on the plants. Therefore, this study proposes sufficient conditions for the robust synthesis of a mixed (Formula presented.) / (Formula presented.) control of the LQR problem. A new mixed polytopic and norm-bounded uncertainties representation for linear time-invariant uncertain systems allows to express different uncertainty models in one single LMI by using the S-procedure. As an additional objective, we propose the study of the robustness of controllers concerned with the sensibility of their coefficients, in such a way that the variations in controller gain matrix are described by norm-bounded uncertainties. A practical implementation in Quanser (Formula presented.) active suspension evaluates the control projects through the rejection level of vibrations caused by track irregularities. |