The deployment of drone swarms is expected to increase significantly in the coming years, especially for monitoring and inspection scenarios, as well as hazardous situations, such as rescue missions in hostile or disaster-stricken areas. Smart Unmanned Aerial Vehicles (UAVs) are well-suited for these tasks due to their agility and maneuverability. However, their limited battery capacity presents substantial challenges, particularly during missions that demand complete coverage of extensive areas within a short time frame. This paper presents a Multi-Agent Reinforcement Learning-based approach for computing the coverage path of large regions by swarms of UAVs supported by mobile battery swapping stations. By implementing advanced path optimization algorithms, it is possible to enhance the efficiency and effectiveness of UAV operations. These algorithms can reduce travel distance, optimize battery swapping during the mission, minimize energy consumption, and ensure comprehensive area coverage. Incorporating real-time data and adaptive strategies can further refine path planning, enabling UAV swarms to adapt to environmental changes and mission requirements dynamically. This approach not only maximizes the operational lifespan of the UAVs, but also ensures the timely and accurate completion of critical tasks in challenging environments.
Supporting UAVs Swarm Missions by Multi-Agent Reinforcement Learning
2025-02-17
1028778 byte
Conference paper
Electronic Resource
English
Wiley | 2012
|