A matlab package for multiobjective control synthesis

X. Qi, Mustafa H. Khammash, Murti V. Salapaka

Research output: Contribution to journalArticlepeer-review

17 Scopus citations

Abstract

In this paper, a Matlab package, GMO 1.0, is introduced for synthesizing (sub-)optimal controllers for the general multiobjective (GMO) control problem involving ℓ1 norm, Η2 norm, Η norm, time-domain constraint (TDC), and controller structure constraints. Formulated in [1], the GMO framework encompasses a large class of multiobjective problems which can be effectively solved by using GMO 1.0 package (e.g. ℓ1, ℓ1/TDC, etc). While global convergence is guaranteed for the convex finite dimensional suboptimal problems, this set of routines yields a rational linear time-invariant controller delivering performance within any prescribed tolerance of the optimal cost. In this paper, several multiobjective design problems are solved to demonstrate the effectiveness of the framework and the software.

Original languageEnglish (US)
Pages (from-to)3991-3996
Number of pages6
JournalProceedings of the IEEE Conference on Decision and Control
Volume4
DOIs
StatePublished - 2001

Keywords

  • Linear matrix inequality (LMI)
  • Multiobjective control
  • Robust optimal control
  • Semidefinite programming (SDP)
  • Time-domain constraint
  • ℓ/Η/Η

Fingerprint

Dive into the research topics of 'A matlab package for multiobjective control synthesis'. Together they form a unique fingerprint.

Cite this