Less conservative robustness analysis of linear parameter varying systems using integral quadratic constraints

Harald Pfifer, Peter J Seiler Jr

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

This paper considers the robustness of a feedback connection of a known linear parameter varying system and a perturbation. A sufficient condition is derived to bound the worst-case gain and ensure robust asymptotic stability. The input/output behavior of the perturbation is described by multiple integral quadratic constraints (IQCs). The analysis condition is formulated as a dissipation inequality. The standard approach requires a non-negative definite storage function and the use of ‘hard’ IQCs. The term ‘hard’ means that the IQCs can be specified as time-domain integral constraints that hold over all finite horizons. The main result demonstrates that the dissipation inequality condition can be formulated requiring neither a non-negative storage function nor hard IQCs. A key insight used to prove this result is that the multiple IQCs, when combined, contain hidden stored energy. This result can lead to less conservative robustness bounds. Two simple examples are presented to demonstrate this fact.

Original languageEnglish (US)
Pages (from-to)3580-3594
Number of pages15
JournalInternational Journal of Robust and Nonlinear Control
Volume26
Issue number16
DOIs
StatePublished - Nov 10 2016

Fingerprint

Asymptotic stability
Feedback

Keywords

  • linear parameter varying systems
  • robust control
  • uncertain systems

Cite this

@article{d73632944d964f22a146c078c7577852,
title = "Less conservative robustness analysis of linear parameter varying systems using integral quadratic constraints",
abstract = "This paper considers the robustness of a feedback connection of a known linear parameter varying system and a perturbation. A sufficient condition is derived to bound the worst-case gain and ensure robust asymptotic stability. The input/output behavior of the perturbation is described by multiple integral quadratic constraints (IQCs). The analysis condition is formulated as a dissipation inequality. The standard approach requires a non-negative definite storage function and the use of ‘hard’ IQCs. The term ‘hard’ means that the IQCs can be specified as time-domain integral constraints that hold over all finite horizons. The main result demonstrates that the dissipation inequality condition can be formulated requiring neither a non-negative storage function nor hard IQCs. A key insight used to prove this result is that the multiple IQCs, when combined, contain hidden stored energy. This result can lead to less conservative robustness bounds. Two simple examples are presented to demonstrate this fact.",
keywords = "linear parameter varying systems, robust control, uncertain systems",
author = "Harald Pfifer and {Seiler Jr}, {Peter J}",
year = "2016",
month = "11",
day = "10",
doi = "10.1002/rnc.3521",
language = "English (US)",
volume = "26",
pages = "3580--3594",
journal = "International Journal of Robust and Nonlinear Control",
issn = "1049-8923",
publisher = "John Wiley and Sons Ltd",
number = "16",

}

TY - JOUR

T1 - Less conservative robustness analysis of linear parameter varying systems using integral quadratic constraints

AU - Pfifer, Harald

AU - Seiler Jr, Peter J

PY - 2016/11/10

Y1 - 2016/11/10

N2 - This paper considers the robustness of a feedback connection of a known linear parameter varying system and a perturbation. A sufficient condition is derived to bound the worst-case gain and ensure robust asymptotic stability. The input/output behavior of the perturbation is described by multiple integral quadratic constraints (IQCs). The analysis condition is formulated as a dissipation inequality. The standard approach requires a non-negative definite storage function and the use of ‘hard’ IQCs. The term ‘hard’ means that the IQCs can be specified as time-domain integral constraints that hold over all finite horizons. The main result demonstrates that the dissipation inequality condition can be formulated requiring neither a non-negative storage function nor hard IQCs. A key insight used to prove this result is that the multiple IQCs, when combined, contain hidden stored energy. This result can lead to less conservative robustness bounds. Two simple examples are presented to demonstrate this fact.

AB - This paper considers the robustness of a feedback connection of a known linear parameter varying system and a perturbation. A sufficient condition is derived to bound the worst-case gain and ensure robust asymptotic stability. The input/output behavior of the perturbation is described by multiple integral quadratic constraints (IQCs). The analysis condition is formulated as a dissipation inequality. The standard approach requires a non-negative definite storage function and the use of ‘hard’ IQCs. The term ‘hard’ means that the IQCs can be specified as time-domain integral constraints that hold over all finite horizons. The main result demonstrates that the dissipation inequality condition can be formulated requiring neither a non-negative storage function nor hard IQCs. A key insight used to prove this result is that the multiple IQCs, when combined, contain hidden stored energy. This result can lead to less conservative robustness bounds. Two simple examples are presented to demonstrate this fact.

KW - linear parameter varying systems

KW - robust control

KW - uncertain systems

UR - http://www.scopus.com/inward/record.url?scp=84957692482&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957692482&partnerID=8YFLogxK

U2 - 10.1002/rnc.3521

DO - 10.1002/rnc.3521

M3 - Article

AN - SCOPUS:84957692482

VL - 26

SP - 3580

EP - 3594

JO - International Journal of Robust and Nonlinear Control

JF - International Journal of Robust and Nonlinear Control

SN - 1049-8923

IS - 16

ER -