In EDS ansehen

Decentralized Battery-Aware Connectivity Maintenance for Multi-UAV Missions

Gespeichert in:
Bibliographische Detailangaben
Titel: Decentralized Battery-Aware Connectivity Maintenance for Multi-UAV Missions
Autoren: Khawaja G. Alamdar, Tamara Petrović
Quelle: IEEE Access, Vol 13, Pp 83738-83751 (2025)
Verlagsinformationen: Institute of Electrical and Electronics Engineers (IEEE), 2025.
Publikationsjahr: 2025
Schlagwörter: multi-robot systems, Connectivity control, Electrical engineering. Electronics. Nuclear engineering, connectivity maintenance, fleet management, TK1-9971
Beschreibung: This study presents a novel battery-aware approach for connectivity maintenance for multi-agent systems. It extends a classical connectivity controller by incorporating battery states into the computation of control inputs, enhancing the network’s transition response in the event of agent removals by anticipating battery depletion. Additionally, strategies for agent addition are introduced to maintain the desired level of connectivity within the multi-agent network. The proposed approach is thoroughly tested in the Gymnasium (previously OpenAI Gym) environment using a specific mission scenario. The results demonstrate the controller’s ability to anticipate agent removals and maintain robust performance during transitional stages. The approach is further validated with Crazyflie UAVs in both simulation and real-world environments, with a fleet management system. The system is designed to be modular and easily adaptable to different flight controllers, demonstrating the applicability of the proposed algorithms to real-world multi-robot missions.
Publikationsart: Article
ISSN: 2169-3536
DOI: 10.1109/access.2025.3569206
Zugangs-URL: https://doaj.org/article/ed1718814fe64c76af3c301604135cc8
Rights: CC BY
Dokumentencode: edsair.doi.dedup.....59bbd1b57fb0cc566a127bf12c99b3fa
Datenbank: OpenAIRE
Beschreibung
Abstract:This study presents a novel battery-aware approach for connectivity maintenance for multi-agent systems. It extends a classical connectivity controller by incorporating battery states into the computation of control inputs, enhancing the network’s transition response in the event of agent removals by anticipating battery depletion. Additionally, strategies for agent addition are introduced to maintain the desired level of connectivity within the multi-agent network. The proposed approach is thoroughly tested in the Gymnasium (previously OpenAI Gym) environment using a specific mission scenario. The results demonstrate the controller’s ability to anticipate agent removals and maintain robust performance during transitional stages. The approach is further validated with Crazyflie UAVs in both simulation and real-world environments, with a fleet management system. The system is designed to be modular and easily adaptable to different flight controllers, demonstrating the applicability of the proposed algorithms to real-world multi-robot missions.
ISSN:21693536
DOI:10.1109/access.2025.3569206