Event-triggered consensus adaptive filters for target localization

Distributed filters show strong robustness by using certain resources of communications and calculations to collaboratively estimate or track an unknown dynamic process of interest over a sensor network. In this paper, an event-triggered mechanism (ETM) is introduced for least mean square (LMS)-base...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of the Franklin Institute Jg. 362; H. 1; S. 107413
Hauptverfasser: Peng, Chen, Deng, Bo, Xie, Siyu
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Inc 01.01.2025
Schlagworte:
Consensus least mean square algorithms
Distributed filters
Event triggers
Sensor networks
Target localization
Unmanned aerial vehicles
Sensor networks
Event triggers
Consensus least mean square algorithms
Unmanned aerial vehicles
Target localization
Distributed filters
ISSN:0016-0032
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Distributed filters show strong robustness by using certain resources of communications and calculations to collaboratively estimate or track an unknown dynamic process of interest over a sensor network. In this paper, an event-triggered mechanism (ETM) is introduced for least mean square (LMS)-based consensus adaptive filters to deal with applications with communication resource constraints. An upper bound of the estimation errors for the proposed event-triggered consensus adaptive filters is established under a cooperative information condition without independent or stationary signal assumptions. To verify the effectiveness and resource saving properties of the proposed event-triggered consensus LMS-based filters, numerical simulations for target localization using bearing-only measurements of multiple unmanned aerial vehicles are provided. It is proved that the ETM provides settable thresholds to artificially adjust the proportions between the estimation accuracy and the resource consumption. Finally, experimental results are given to further show the performance and applicability of the proposed algorithm in practical engineering. •An event-triggered mechanism is introduced into LMS-based filters to solve the constraint of limited resources, which is inevitable in a distributed problem of sensor networks. Theoretical boundedness is analyzed without independent or stationary signal assumptions since the practical signals are often correlated caused by multi-path effect or feedback.•The proposed algorithm is applied in multi-UAVs cooperative localization to estimate the locations of stationary and slow-moving targets.•The ability of saving resources is verified by both numerical simulations and practical experiments.
ISSN:0016-0032
DOI:10.1016/j.jfranklin.2024.107413