MARL Warehouse Robots

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Allman, Price, Thang, Lian, Simmons, Dre, Riaz, Salmon
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915661727399936
author Allman, Price
Thang, Lian
Simmons, Dre
Riaz, Salmon
author_facet Allman, Price
Thang, Lian
Simmons, Dre
Riaz, Salmon
contents We present a comparative study of multi-agent reinforcement learning (MARL) algorithms for cooperative warehouse robotics. We evaluate QMIX and IPPO on the Robotic Warehouse (RWARE) environment and a custom Unity 3D simulation. Our experiments reveal that QMIX's value decomposition significantly outperforms independent learning approaches (achieving 3.25 mean return vs. 0.38 for advanced IPPO), but requires extensive hyperparameter tuning -- particularly extended epsilon annealing (5M+ steps) for sparse reward discovery. We demonstrate successful deployment in Unity ML-Agents, achieving consistent package delivery after 1M training steps. While MARL shows promise for small-scale deployments (2-4 robots), significant scaling challenges remain. Code and analyses: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
format Preprint
id arxiv_https___arxiv_org_abs_2512_04463
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MARL Warehouse Robots
Allman, Price
Thang, Lian
Simmons, Dre
Riaz, Salmon
Artificial Intelligence
Robotics
We present a comparative study of multi-agent reinforcement learning (MARL) algorithms for cooperative warehouse robotics. We evaluate QMIX and IPPO on the Robotic Warehouse (RWARE) environment and a custom Unity 3D simulation. Our experiments reveal that QMIX's value decomposition significantly outperforms independent learning approaches (achieving 3.25 mean return vs. 0.38 for advanced IPPO), but requires extensive hyperparameter tuning -- particularly extended epsilon annealing (5M+ steps) for sparse reward discovery. We demonstrate successful deployment in Unity ML-Agents, achieving consistent package delivery after 1M training steps. While MARL shows promise for small-scale deployments (2-4 robots), significant scaling challenges remain. Code and analyses: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
title MARL Warehouse Robots
topic Artificial Intelligence
Robotics
url https://arxiv.org/abs/2512.04463