对比MARL算法在仓库机器人协作中的表现，发现QMIX更优但需大量调参。
MARL Warehouse Robots
- 采用价值分解机制的QMIX优于独立学习方法。
- QMIX达3.25平均回报，远超高级IPPO的0.38。
- 适合小规模机器人协同，但扩展性仍有挑战。
我们对多智能体强化学习（MARL）算法在协作式仓库机器人中的应用进行了对比研究。在Robotic Warehouse（RWARE）环境及自定义Unity 3D仿真中评估了QMIX与IPPO算法。实验表明，QMIX的价值分解机制显著优于独立学习方法（平均回报达3.25，而先进IPPO仅为0.38），但需大量超参数调优，尤其需要超过500万步的ε退火以发现稀疏奖励。我们在Unity ML-Agents中成功部署，训练100万步后实现稳定包裹配送。尽管MARL在小规模部署（2-4台机器人）中展现潜力，但大规模扩展仍面临显著挑战。代码与分析详见：https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
原文摘要 · Abstract (English)
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/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。