Brief Rbust control of linear systems with real parametric uncertainty

Authors:
I. E. Kose;F. Jabbari
Affiliations:
Department of Mechanical Engineering, Bogazici University, Bebek, Istanbul 80815, Turkey;Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92697, USA
Venue:
Automatica (Journal of IFAC)
Year:
1999

Citing 7
Cited 1

Optimal, constant I/O similarity scaling for full-information and state-feedback control problems

Systems & Control Letters
Gain scheduling via linear fractional transformations

Systems & Control Letters
All controllers for the general H∞ control problem: LMI existence conditions and state space formulas

Automatica (Journal of IFAC)
Robust performance of linear parametrically varying systems using parametrically-dependent linear feedback

Systems & Control Letters
Robust performance of systems with structured uncertainties in state space

Automatica (Journal of IFAC)
Self-scheduled H∞ control of linear parameter-varying systems: a design example

Automatica (Journal of IFAC)
Low-order control design for LMI problems using alternating projection methods

Automatica (Journal of IFAC)

Robust output-feedback controller design via local BMI optimization

Automatica (Journal of IFAC)

Quantified Score

Hi-index	22.15

Visualization

Abstract

We discuss the problem of robust controller synthesis to provide quadratic stability and a desirable disturbance attenuation level (through an appropriately small L"2 gain) for systems with time-varying, real parametric uncertainty. Through the use of skewsymmetric matrices, the conservatism of standard scaled H"~ approach is reduced. While the full state feedback problem is convex, the output feedback problem is not. A set of conditions under which the design of output feedback controllers can be broken into two sequential convex problems is presented. The results are compared with recent results concerning the design of parameter-varying controllers and a simple result regarding the mixed problem (where some of the unknown parameters can be measured on-line) is discussed.