Методи роєвого штучного інтелекту в завданнях автономної навігації БПЛА
This paper presents a comparative analysis of nine swarm intelligence (SI) methods in terms of their suitability for onboard AI platforms in autonomous unmanned aerial vehicle (UAV) swarms. A set of key criteria is defined, including computational complexity, scalability, latency, robustness to agen...
Gespeichert in:
| Datum: | 2025 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2025
|
| Schlagworte: | |
| Online Zugang: | http://journal.iasa.kpi.ua/article/view/343081 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | System research and information technologies |
Institution
System research and information technologies| Zusammenfassung: | This paper presents a comparative analysis of nine swarm intelligence (SI) methods in terms of their suitability for onboard AI platforms in autonomous unmanned aerial vehicle (UAV) swarms. A set of key criteria is defined, including computational complexity, scalability, latency, robustness to agent loss, and adaptability. Decentralized Behavior Trees (BTs) are identified as the most balanced approach for the reactive behavior layer, while the global swarm optimization method GBestPSO proves effective for high-level planning. A hybrid two-layer cognitive architecture is proposed that integrates BTs and GBestPSO, with functional separation between layers and communication based on DDS/RTPS protocols. The architecture exhibits high autonomy, fault tolerance, modularity, and suitability for real-time embedded systems operating in dynamic or adversarial environments. The results were partially supported by the National Research Foundation of Ukraine, grant No. 2025.06/0022 “AI platform with cognitive services for coordinated autonomous navigation of distributed systems consisting of a large number of objects”. |
|---|