Robust LQR via Bounded Data Uncertainties

Author

Ramos, C. ; Martinez, M. ; Sanchis, J. ; Salcedo, J.V.

Author_Institution

Dept. of Syst. Eng. & Control, Polytech. Univ. of Valencia, Valencia, Spain

fYear

2007

fDate

2-5 July 2007

Firstpage

2078

Lastpage

2083

Abstract

This work presents the tuning of a Linear Quadratic Regulator (LQR) via the Bounded Data Uncertainties (BDU) method in order to improve the system robustness. The BDU method considers models with bounded uncertainties and it is stated as a Min-Max problem where a solution which performs `best´ in the worst-possible scenario is sought. So a new guided way of tuning the LQR is offered, which takes into account the uncertainties bounds, and it results in the modification of the recursive Riccati equation. The application to multidimensional systems is not trivial due to the fact that the problem presents the form of a Two-Point Boundary Value Problem (TPBVP) and it is solved iteratively.

Keywords

Riccati equations; boundary-value problems; control system synthesis; linear quadratic control; minimax techniques; robust control; uncertainty handling; BDU method; TPBVP; bounded data uncertainties; linear quadratic regulator; min-max problem; recursive Riccati equation; robust LQR tuning; system robustness; two-point boundary value problem; Approximation methods; Equations; Mathematical model; Regulators; State-space methods; Uncertainty; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (ECC), 2007 European

Conference_Location

Kos

Print_ISBN

978-3-9524173-8-6

Type

conf

Filename

7068390