Modular Reinforcement Learning For Cooperative Swarms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shtossel, Erel, Kaminka, Gal A.
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915984039739392
author Shtossel, Erel
Kaminka, Gal A.
author_facet Shtossel, Erel
Kaminka, Gal A.
contents A cooperative robot swarm is a collective of computationally-limited robots that share a common goal. Each robot can only interact with a small subset of its peers, without knowing how this affects the collective utility. Recent advances in distributed multi-agent reinforcement learning have demonstrated that it is possible for robots to learn how to interact effectively with others, in a manner that is aligned with the common goal, despite each robot learning independently of others. However, this requires each robot to represent a potentially combinatorial number of interaction states, challenging the memory capabilities of the robots. This paper proposes an alternative approach for representing spatial interaction states for multi-robot reinforcement learning in swarms. A modular (decomposed) representation is used, where each feature of the state is handled by a separate learning procedure, and the results aggregated. We demonstrate the efficacy of the approach in numerous experiments with simulated robot swarms carrying out foraging.
format Preprint
id arxiv_https___arxiv_org_abs_2605_04939
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Modular Reinforcement Learning For Cooperative Swarms
Shtossel, Erel
Kaminka, Gal A.
Robotics
Artificial Intelligence
A cooperative robot swarm is a collective of computationally-limited robots that share a common goal. Each robot can only interact with a small subset of its peers, without knowing how this affects the collective utility. Recent advances in distributed multi-agent reinforcement learning have demonstrated that it is possible for robots to learn how to interact effectively with others, in a manner that is aligned with the common goal, despite each robot learning independently of others. However, this requires each robot to represent a potentially combinatorial number of interaction states, challenging the memory capabilities of the robots. This paper proposes an alternative approach for representing spatial interaction states for multi-robot reinforcement learning in swarms. A modular (decomposed) representation is used, where each feature of the state is handled by a separate learning procedure, and the results aggregated. We demonstrate the efficacy of the approach in numerous experiments with simulated robot swarms carrying out foraging.
title Modular Reinforcement Learning For Cooperative Swarms
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2605.04939