This paper presents an autonomous decision-making framework for drone swarms using deep reinforcement learning, dividing UAVs into “reconnaissance” and “executive” roles. The framework includes a situational communication layer, task decision and planning layers, and a control layer. It employs a semi-centralized, semi-distributed mode in the communication layer and integrates Deep Deterministic Policy Gradient (DDPG) and Multi-Agent Deep Deterministic Policy Gradient (MADDPG) in the decision and planning layers. An enhanced MADDPG algorithm is developed, resulting in a drone swarm autonomous decision-making method based on deep reinforcement learning. Simulations show that this approach outperforms in UAV swarm tasks.
Autonomous Decision-Making of Drone Swarm Based on Deep Reinforcement Learning
Lect. Notes Electrical Eng.
International Conference on Guidance, Navigation and Control ; 2024 ; Changsha, China August 09, 2024 - August 11, 2024
04.03.2025
10 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch