UAVs operating in GNSS-denied environments face significant challenges, including maintaining navigation accuracy, avoiding collisions, and ensuring swarm cohesion. These limitations are particularly pronounced in scenarios where traditional centralized control systems or single-leader strategies are employed, as they are prone to single points of failure and lack adaptability to dynamic conditions. To overcome these challenges, we propose a collaborative framework for UAV swarms, lever-aging distributed leader-follower dynamics and reinforcement learning. By enabling UAVs to dynamically share leadership roles and collaborate through cooperative reward structures, the framework enhances swarm cohesion and navigation reliability. Collaboration allows UAVs to effectively compensate for the absence of GNSS signals, improving their ability to navigate, avoid collisions, and maintain formation in complex environments. Simulation results demonstrate the efficacy of the proposed approach, showing up to 56% improvement in navigation performance, ${74\%}$ reduction in collision incidents, and 33% increase in mission success rates compared to existing leader-follower systems. These findings highlight the transformative potential of collaboration in enabling UAV swarms to operate effectively in GNSS-denied scenarios.
Distributed Cooperative UAV Leader-Follower Strategies in GNSS-Denied Environments
09.06.2025
551570 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Cooperative Navigation for an UAV Tandem in GNSS Denied Environments
British Library Conference Proceedings | 2018
|DataCite | 2023
|