Resilient Consensus Through Dynamic Event-Triggered Mechanism

In this brief, the resilient consensus problem for multi-agent systems (MASs) is addressed based on the dynamic event-triggered (DE) mechanism, when the network is subject to malicious attacks. A dynamic variable is introduced in the DE mechanism to adjust the threshold dynamically. Via the proposed...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on circuits and systems. II, Express briefs Jg. 71; H. 7; S. 3463 - 3467
Hauptverfasser: Liao, Zirui, Shi, Jian, Wang, Shaoping, Zhang, Yuwei, Sun, Zhiyong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.07.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Algorithms
dynamic variable
Electrical engineering
event-triggered mechanism
Heuristic algorithms
malicious attack
Multi-agent system
Multi-agent systems
Multiagent systems
Protocols
Resilience
resilient consensus
Technological innovation
ISSN:1549-7747, 1558-3791
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this brief, the resilient consensus problem for multi-agent systems (MASs) is addressed based on the dynamic event-triggered (DE) mechanism, when the network is subject to malicious attacks. A dynamic variable is introduced in the DE mechanism to adjust the threshold dynamically. Via the proposed dynamic event-triggered mean-subsequence-reduced (DE-MSR) algorithm, all cooperative agents are guaranteed to reach an agreement on the same consensus value, so that the MAS achieves resilient consensus despite the influence of some noncooperative agents in the network. Compared to existing event-based resilient algorithms, the proposed DE-MSR algorithm is superior in reducing communication overheads while maintaining a resilient consensus. Finally, a comparative case study is conducted to validate the theoretical findings.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1549-7747
1558-3791
DOI:10.1109/TCSII.2024.3364524