We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Optimization and Control

Title: Distributed and Localized Model Predictive Control via System Level Synthesis

Abstract: We present the Distributed and Localized Model Predictive Control (DLMPC) algorithm for large-scale structured linear systems, wherein only local state and model information needs to be exchanged between subsystems for the computation and implementation of control actions. We use the System Level Synthesis (SLS) framework to reformulate the MPC problem as an optimization problem over closed loop system responses, and show that this allows us to naturally impose localized communication constraints between sub-controllers, such that only local state and system model information needs to be exchanged for both computation and implementation of closed loop MPC control policies. In particular, we show that the structure of the resulting optimization problem can be exploited to develop an Alternating Direction Method of Multipliers (ADMM) based algorithm that allows for distributed and localized computation of control decisions. Moreover, our approach can accommodate constraints and objective functions that couple the behavior of different subsystems, so long as the coupled systems are able to communicate directly with each other, allowing for a broader class of MPC problems to be solved via distributed optimization. We conclude with numerical simulations to demonstrate the usefulness of our method, and in particular, we demonstrate that the computational complexity of the subproblems solved by each subsystem in DLMPC is independent of the size of the global system.
Comments: Extended version of CDC 2020 submission
Subjects: Optimization and Control (math.OC); Systems and Control (eess.SY)
Cite as: arXiv:1909.10074 [math.OC]
  (or arXiv:1909.10074v4 [math.OC] for this version)

Submission history

From: Carmen Amo Alonso [view email]
[v1] Sun, 22 Sep 2019 19:18:27 GMT (2134kb,D)
[v2] Tue, 1 Oct 2019 15:21:36 GMT (1056kb,D)
[v3] Mon, 18 May 2020 18:32:37 GMT (1830kb,D)
[v4] Fri, 11 Sep 2020 00:38:26 GMT (2111kb,D)

Link back to: arXiv, form interface, contact.