Input-to-State Stability for Networked Predictive Control With Random Delays in Both Feedback and Forward Channels

Xi-Ming Sun*, Di Wu, Wei Wang, G-P Liu

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

The input-to-state stability (ISS) for a class of networked control systems with random delays and packet dropouts appearing simultaneously in both feedback and forward channels is thoroughly investigated in this paper. A new network predictive controller scheme is introduced in order to compensate the effect of transmission delays and packet dropouts. By making use of the small gain theorem, the stability criteria of the considered new system are derived. The proposed stability conditions are fairly easy to check and considerably less conservative than the existing ones. These criteria reveal that, if the original linear systems are controllable and observable, then, by adopting the proposed networked-predictive-control scheme, the ISS properties can be guaranteed for the overall system despite the effects of networking such as transmission bounded delays, packet dropouts, and possible disturbance inputs. When no disturbance inputs occur, the system stability can be guaranteed for random delays with a certain bound or else for any large constant delays. Results for two illustrative examples are given to validate the proposed control scheme, the second one being a laboratory-scale dc-motor rig.

Original languageEnglish
Pages (from-to)3519-3526
Number of pages8
JournalIEEE Transactions on Industrial Electronics
Volume61
Issue number7
DOIs
Publication statusPublished - Jul 2014

Keywords

  • Control synthesis
  • input-to-state stability (ISS)
  • networked control systems (NCSs)
  • networked predictive controller
  • random delays
  • CONTROL-SYSTEMS
  • LINEAR-SYSTEMS
  • NONLINEAR-SYSTEMS
  • STABILIZATION
  • DESIGN
  • MODEL

Fingerprint

Dive into the research topics of 'Input-to-State Stability for Networked Predictive Control With Random Delays in Both Feedback and Forward Channels'. Together they form a unique fingerprint.

Cite this