MW-MADDPG: a meta-learning based decision-making method for collaborative UAV swarm

Unmanned Aerial Vehicles (UAVs) have gained popularity due to their low lifecycle cost and minimal human risk, resulting in their widespread use in recent years. In the UAV swarm cooperative decision domain, multi-agent deep reinforcement learning has significant potential. However, current approach...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Frontiers in neurorobotics Ročník 17; s. 1243174
Hlavní autori: Zhao, Minrui, Wang, Gang, Fu, Qiang, Guo, Xiangke, Chen, Yu, Li, Tengda, Liu, XiangYu
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Lausanne Frontiers Research Foundation 21.09.2023
Frontiers Media S.A
Predmet:
Algorithms
Cluster analysis
Collaboration
Decision making
Deep learning
Efficiency
Evacuations & rescues
Learning
MADDPG
meta learning
Model Agnostic Meta Learning (MAML)
multi-agent reinforcement learning (MARL)
Neuroscience
Reinforcement
Scheduling
UAV
Unmanned aerial vehicles
ISSN:1662-5218, 1662-5218
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Unmanned Aerial Vehicles (UAVs) have gained popularity due to their low lifecycle cost and minimal human risk, resulting in their widespread use in recent years. In the UAV swarm cooperative decision domain, multi-agent deep reinforcement learning has significant potential. However, current approaches are challenged by the multivariate mission environment and mission time constraints. In light of this, the present study proposes a meta-learning based multi-agent deep reinforcement learning approach that provides a viable solution to this problem. This paper presents an improved MAML-based multi-agent deep deterministic policy gradient (MADDPG) algorithm that achieves an unbiased initialization network by automatically assigning weights to meta-learning trajectories. In addition, a Reward-TD prioritized experience replay technique is introduced, which takes into account immediate reward and TD-error to improve the resilience and sample utilization of the algorithm. Experiment results show that the proposed approach effectively accomplishes the task in the new scenario, with significantly improved task success rate, average reward, and robustness compared to existing methods.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Reviewed by: Mu Hua, University of Lincoln, United Kingdom; Yan Fang, Kennesaw State University, United States; Pengyu Yuan, Google, United States
Edited by: Ming-Feng Ge, China University of Geosciences Wuhan, China
ISSN:1662-5218
1662-5218
DOI:10.3389/fnbot.2023.1243174