MULTI-AGENT UAV CONTROL SYSTEM WITH DISTRIBUTED VISUAL OBJECT RECOGNITION
DOI:
https://doi.org/10.34185/1991-7848.itmm.2026.01.099Keywords:
control systems, UAV, YOLOv8, signal, data processing, communication, information security, GPS spoofing, object recognition, deep learningAbstract
This paper presents a study of a multi-agent control system for unmanned aerial vehicles (UAVs) based on distributed visual object recognition using modern deep learning models. YOLOv8 is employed as the baseline model, adapted for deployment on embedded computing platforms of FPV drones. The primary focus is on ensuring system autonomy under conditions of signal loss, intentional jamming (electronic warfare), or spoofing of control and navigation signals. An approach based on the use of preloaded models, behavioral scenarios, and reference data is proposed, enabling a fully autonomous operational mode of UAVs. The obtained results confirm the effectiveness and promise of applying multi-agent approaches to enhance system efficiency and security.
References
WANG, Yongzhen; HE, Jiacong. A rapid concrete crack detection method based on improved YOLOv8. IEEE Access, 2025.
Lou H. et al. DC-YOLOv8: Small-size object detection algorithm based on camera sensor //Electronics. – 2023. – Т. 12. – №. 10. – С. 2323. URL: https://www.mdpi.com/2079-9292/12/10/2323
Syzonenko R., Klymenko S., Hnatushenko V. Utilization of cloud infrastructure for dataset markup. – 2025. URL: https://ceur-ws.org/Vol-4146/paper25.pdf
V. Hnatushenko, V. Kashtan, O. Kazymyrenko. Information technology for detecting cars on aerial imaging using a modified YOLO-OBB architecture. MoDaST 2025: Modern Data Science Technologies Doctoral Consortium, June, 15, 2025, Lviv, Ukraine, Pp. 293-304. URL: https://ceur-ws.org/Vol-4005/paper20.pdf.
GOODFELLOW, Ian, et al. Deep learning. Cambridge: MIT press, 2016. URL: https://synapse.koreamed.org/pdf/10.4258/hir.2016.22.4.351.
DAIMI, Kevin, et al. (ed.). Computer and network security essentials. Springer, 2018.
