arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.04463cs.AIcs.RO
收藏

多智能体强化学习在仓库机器人中的应用

MARL Warehouse Robots

Department of Computer Science, Oral Roberts University(计算机科学系,奥尔巴尼罗尔斯大学)

Price Allman, Lian Thang, Dre Simmons, Salmon Riaz

AI总结:

本文比较了MARL算法在仓库机器人中的性能,发现QMIX在奖励回报上显著优于IPPO,但需大量超参数调优,适用于小规模部署但存在扩展挑战。

摘要:

我们对多智能体强化学习(MARL)算法在协作仓库机器人中的应用进行了比较研究。我们在Robotic Warehouse(RWARE)环境和一个定制的Unity 3D模拟中评估了QMIX和IPPO。我们的实验表明,QMIX的价值分解显著优于独立学习方法(平均回报达到3.25,而先进的IPPO为0.38),但需要大量的超参数调优——特别是扩展的epsilon退火(500万步以上)用于稀疏奖励发现。我们在Unity ML-Agents中展示了成功的部署,经过100万次训练步骤后实现了稳定的包裹交付。虽然MARL在小规模部署(2-4台机器人)中显示出前景,但显著的扩展挑战仍然存在。代码和分析:https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/

英文摘要:

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/

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑