Time-varying tube-based output feedback mpc for constrained linear systems with intermittently delayed data

Syed M. Hassaan, Tarun Pati, Qiang Shen, Sze Zheng Yong

Research output: Contribution to journalConference articlepeer-review

Abstract

This paper proposes a time-varying tube-based output feedback model predictive control (MPC) design for constrained linear systems in the presence of intermittently delayed observations, where the delayed/missing data patterns for each period satisfy a finite-length language. The design consists of a dynamic state estimator whose estimation errors satisfy equalized recovery (a weaker form of invariance with time-varying finite bounds), as well as an output feedback control law that extends existing tube-based output feedback MPC approaches to allow time-varying tubes for tightening the original state and input constraints. The resulting time-varying tube-based output feedback MPC design is robust to time-varying disturbances and errors, including when the observations are intermittently delayed. Further, we provide sufficient conditions for recursive feasibility and robust exponential stability of the proposed design. Simulation results demonstrate that the proposed approach is able to robustly stabilize and control a constrained linear system despite disturbances, noise and missing/delayed data.

Original languageEnglish (US)
Pages (from-to)103-108
Number of pages6
JournalIFAC-PapersOnLine
Volume54
Issue number5
DOIs
StatePublished - Jul 1 2021
Event7th IFAC Conference on Analysis and Design of Hybrid Systems, ADHS 2021 - Brussels, Belgium
Duration: Jul 7 2021Jul 9 2021

Keywords

  • Missing and delayed data
  • Optimal control and optimization
  • Output-based control
  • Robust model predictive control
  • Robust time-varying tubes

ASJC Scopus subject areas

  • Control and Systems Engineering

Fingerprint

Dive into the research topics of 'Time-varying tube-based output feedback mpc for constrained linear systems with intermittently delayed data'. Together they form a unique fingerprint.

Cite this