Distributed event-triggered algorithm for unconstrained convex optimisation over weight-balanced directed networks

In this study, the authors propose a distributed discrete-time algorithm for unconstrained optimisation with event-triggered communication over weight-balanced directed networks. They consider a multi-agent system where each agent has a state and an auxiliary variable for the estimates of the optima...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IET control theory & applications Jg. 14; H. 2; S. 253 - 261
Hauptverfasser: Hayashi, Naoki, Sugiura, Tomohiro, Kajiyama, Yuichi, Takai, Shigemasa
Format: Journal Article
Sprache:Englisch
Veröffentlicht: The Institution of Engineering and Technology 29.01.2020
Schlagworte:
auxiliary variables
convergence
convex programming
discrete time systems
discrete‐time algorithm
distributed control
distributed event‐triggered algorithm
event‐triggered communication
gradient methods
multiagent system
multi‐agent systems
optimal solution
optimisation
Research Article
trigger errors
unconstrained convex optimisation
unconstrained optimisation
weight‐balanced directed networks
multi-agent systems
discrete time systems
distributed control
convergence
auxiliary variables
convex programming
unconstrained convex optimisation
optimal solution
weight-balanced directed networks
multiagent system
distributed event-triggered algorithm
event-triggered communication
optimisation
discrete-time algorithm
unconstrained optimisation
trigger errors
gradient methods
ISSN:1751-8644, 1751-8652
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this study, the authors propose a distributed discrete-time algorithm for unconstrained optimisation with event-triggered communication over weight-balanced directed networks. They consider a multi-agent system where each agent has a state and an auxiliary variable for the estimates of the optimal solution and the average gradient of the entire cost function. Agents send the states and auxiliary variables to their neighbours when their trigger errors exceed thresholds. They derive a convergence rate of the proposed algorithm which shows faster convergence to the optimal solution compared to the subgradient-based method.
ISSN:1751-8644
1751-8652
DOI:10.1049/iet-cta.2019.0377