Skip to main navigation Skip to search Skip to main content

Performance Degradation of an Optimal LQ Controller based on a Reduced-Order Observer Designed via the Sylvester Algebraic Equation

  • Ajman University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper considers the problem of an optimal linear-quadratic (LQ) control based on a reduced-order observer (ROO) designed via the Sylvester equation. Since observers estimate state variables with certain errors, the optimal performance gets degraded in the case when observer estimates are used for feedback instead of the system states. In this work, the analytical expression is derived for the optimal LQ criterion loss for steady state optimization. The paper derives a formula based on the least square method for the selection of initial conditions of ROO that can drastically reduce the optimal performance loss. Since in some applications of ROO with LQ optimal controllers the performance loss can be very high it is mandatory that the corresponding applications be reexamined in the spirit of the results reported in this paper.

Original languageEnglish
Title of host publicationInternational Conference on Electrical and Computer Engineering Researches, ICECER 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665457569
DOIs
StatePublished - 2025
Event2025 International Conference on Electrical and Computer Engineering Researches, ICECER 2025 - Antananarivo, Madagascar
Duration: 6 Dec 20258 Dec 2025

Publication series

NameInternational Conference on Electrical and Computer Engineering Researches, ICECER 2025

Conference

Conference2025 International Conference on Electrical and Computer Engineering Researches, ICECER 2025
Country/TerritoryMadagascar
CityAntananarivo
Period6/12/258/12/25

Keywords

  • Reduced-order observer (ROO)
  • Sylvester equation design
  • linear-quadratic optimal performance loss

Fingerprint

Dive into the research topics of 'Performance Degradation of an Optimal LQ Controller based on a Reduced-Order Observer Designed via the Sylvester Algebraic Equation'. Together they form a unique fingerprint.

Cite this