Instant distributed MPC with reference governor

Martin Figura, Lanlan Su, Masaki Inoue, Vijay Gupta

Research output: Contribution to journalArticlepeer-review


Model predictive control is a popular choice for systems that must satisfy prescribed constraints on states and control inputs. Although much progress has been made in distributed model predictive control, existing algorithms tend to be computationally expensive. This limits their use in systems with fast dynamics. In this paper, we propose a new distributed model predictive control algorithm that we term as instant distributed model predictive control (iDMPC). The proposed algorithm employs a realisation of the primal-dual algorithm in the controller dynamics for fast computation. We show that the closed-loop system trajectories with the proposed iDMPC algorithm converge asymptotically to a desired reference. To ensure the satisfaction of the state constraints during the transient, we also include an explicit reference governor in the feedback loop.

Original languageEnglish
JournalInternational Journal of Control
Publication statusAccepted/In press - 2023


  • dissipativity
  • Distributed model predictive control
  • reference governor
  • suboptimal control

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications


Dive into the research topics of 'Instant distributed MPC with reference governor'. Together they form a unique fingerprint.

Cite this