EUVIMEDEUVIMED entdecken
Crossref

Spatially Grouped Curriculum Learning for Multi-Agent Path Finding

Thomy Phan, Sven Koenig

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (35) · 29642-29650

DOI: 10.1609/aaai.v40i35.40208 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Thomy Phan und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multi-agent path finding (MAPF) is the challenging problem of finding conflict-free paths with minimal costs for multiple agents. While traditional MAPF solvers are centralized using heuristic search, reinforcement learning (RL) is becoming increasingly popular due to its potential to learn decentralized and generalizing policies. RL-based MAPF must cope with spatial coordination, which is often addressed by combining independent training with ad hoc measures like replanning and communication. Such ad hoc measures often complicate the approach and require knowledge beyond the actual accessible information in RL, such as the full map occupation or broadcast communication channels, which limits generalizability, effectiveness, and sample efficiency. In this paper, we propose Partitioned Attention-based Reverse Curricula for Enhanced Learning (PARCEL), considering a bounding region for each agent. PARCEL trains all agents with overlapping regions jointly via self-attention to avoid potential conflicts. By employing a reverse curriculum, where the bounding regions grow as the policies improve, all agents will eventually merge into a single coordinated group. We evaluate PARCEL in two simple coordination tasks and four MAPF benchmark maps. Compared with state-of-the-art RL-based MAPF methods, PARCEL demonstrates better effectiveness and sample efficiency without ad hoc measures.

Zitieren

APA

Phan, T., & Koenig, S. (2026). Spatially Grouped Curriculum Learning for Multi-Agent Path Finding. Proceedings of the AAAI Conference on Artificial Intelligence, 40(35), 29642-29650. https://doi.org/10.1609/aaai.v40i35.40208

Vancouver

Phan T, Koenig S. Spatially Grouped Curriculum Learning for Multi-Agent Path Finding. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(35):29642-29650. doi:10.1609/aaai.v40i35.40208

BibTeX
@article{Phan2026,
  author = {Phan, Thomy and Koenig, Sven},
  title = {Spatially Grouped Curriculum Learning for Multi-Agent Path Finding},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {35},
  pages = {29642--29650},
  doi = {10.1609/aaai.v40i35.40208}
}
Kostenlos starten →