State-dependent Riccati equation control with predicted trajectory

A. Dutka, M.J. Grimble

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

5 Citations (Scopus)
5 Downloads (Pure)


A modified State-Dependent Riccati Equation method is used which takes into account future variations in the system model dynamics. The system in the state dependent coefficient form, together with the prediction of the future trajectory, may be considered to be approximated by known time-varying system. For such a system the optimal control solution may be obtained for a discrete time system by solving the Riccati Difference Equation. The minimisation of the cost function for a predicted time-varying system is achieved by considering the prediction horizon as a combination of infinite and finite horizon parts. The infinite part is minimised by solving the Algebraic Riccati Equation and the finite part by the Riccati Difference Equation. The number of future prediction steps depends upon the problem and is a fixed variable chosen during the controller design. A comparison of results is provided with other design methods, which indicates that there is considerable potential for the technique.
Original languageEnglish
Title of host publicationProceedings of the 2004 American Control Conference
Place of PublicationNew York
Number of pages6
ISBN (Print)0780383354
Publication statusPublished - Jun 2004
EventAmerican Control Conference 2004 - Boston, United States
Duration: 30 Jun 20042 Jul 2004

Publication series

NameProceedings of the American Control Conference
ISSN (Print)0743-1619


ConferenceAmerican Control Conference 2004
Country/TerritoryUnited States


  • state-dependent
  • riccati equation control
  • predicted trajectory
  • riccati equations
  • time-varying systems
  • optimal control
  • minimisation
  • infinite horizon
  • discrete time systems
  • difference equations
  • control system synthesis


Dive into the research topics of 'State-dependent Riccati equation control with predicted trajectory'. Together they form a unique fingerprint.

Cite this