EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GR-MAPPO Algorithm for Perimeter Defense Problem in Multi-Agent Systems

Huihui Tan, Shuang Zhang, Shiwei Lin iD, Bomin Huang iD

Entropy · 09.06.2026 · 28 (6) · 659

DOI: 10.3390/e28060659 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 09.06.2026. Verfasst von Huihui Tan und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer Engineering, Jimei University, Xiamen 361021, China. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Xiamen, China, Natural Science Foundation of Fujian Province, China, Startup Fund of Jimei University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multi-agent perimeter defense plays a critical role in cooperative defense scenarios in unmanned swarms. However, existing deep reinforcement learning approaches struggle to effectively exploit both coordination and temporal information under constrained local communication, and they lack generalization capability under dynamic variations in swarm size. To address these challenges, this paper proposes a multi-agent reinforcement learning strategy that integrates coordination under local communication constraints with spatiotemporal feature modeling. Specifically, a GraphSAGE-based spatial aggregation module is employed to enhance information exchange among defenders, while a GRU-based temporal encoding module processes historical observation sequences to improve coordination and anticipatory capability. Furthermore, to overcome scalability limitations, the inductive node-level aggregation mechanism enables agents to adapt to varying numbers of local neighbors, eliminating dependence on a fixed swarm size. Experimental results demonstrate that the proposed GR-MAPPO consistently improves capture performance under limited communication and exhibits better performance retention under cross-scale transfer across different swarm scales.

Schlagwörter

Stichwörter der Autor:innen

deep learningmulti-agent systemperimeter defensereinforcement learning

Autor:innen und Institutionen

  • College of Computer Engineering, Jimei University, Xiamen 361021, China.

ORCID-Profile: Shiwei Lin, Bomin Huang

Förderung

  • Natural Science Foundation of Xiamen, China
  • Natural Science Foundation of Fujian Province, China
  • Startup Fund of Jimei University

Zitieren

APA

Tan, H., Zhang, S., Lin, S., & Huang, B. (2026). GR-MAPPO Algorithm for Perimeter Defense Problem in Multi-Agent Systems. Entropy, 28(6), 659. https://doi.org/10.3390/e28060659

Vancouver

Tan H, Zhang S, Lin S, Huang B. GR-MAPPO Algorithm for Perimeter Defense Problem in Multi-Agent Systems. Entropy. 2026;28(6):659. doi:10.3390/e28060659

BibTeX
@article{Tan2026,
  author = {Tan, Huihui and Zhang, Shuang and Lin, Shiwei and Huang, Bomin},
  title = {GR-MAPPO Algorithm for Perimeter Defense Problem in Multi-Agent Systems},
  journal = {Entropy},
  year = {2026},
  volume = {28},
  number = {6},
  pages = {659},
  doi = {10.3390/e28060659}
}
Kostenlos starten →