让机器人走路时稳稳端杯，不洒一滴
Hold My Beer: Learning Gentle Humanoid Locomotion and End-Effector Stabilization Control
- 分上下身双控制代理，分别以100Hz和50Hz运行
- 末端执行器加速度降低2-5倍，接近人类稳定水平
- 适合需要精细操作的机器人场景，如送杯、拍摄
能否让机器人在行走时稳稳递上一杯满杯啤酒而不洒？尽管人形机器人已能在舞蹈、搬运、崎岖地形中展示炫技，但行进中对末端执行器（EE）的精细稳定仍面临巨大挑战。根本原因在于任务动态不匹配：步行需要慢速、鲁棒的控制，而末端稳定则依赖快速高精度调整。为此，我们提出SoFTA——一种慢-快双代理框架，将上半身与下半身控制解耦为不同频率、不同奖励机制的独立智能体。上半身以100Hz运行，实现高精度末端控制；下半身以50Hz运行，保障步态鲁棒性。该设计有效缓解策略干扰，实现全身协调。实验表明，相比基线方法，末端加速度降低2-5倍，运动稳定性接近人类水平，可完成端近满杯、行走中持续拍摄视频、抗扰动保持末端稳定等精细任务。
原文摘要 · Abstract (English)
Can your humanoid walk up and hand you a full cup of beer, without spilling a drop? While humanoids are increasingly featured in flashy demos like dancing, delivering packages, traversing rough terrain, fine-grained control during locomotion remains a significant challenge. In particular, stabilizing a filled end-effector (EE) while walking is far from solved, due to a fundamental mismatch in task dynamics: locomotion demands slow-timescale, robust control, whereas EE stabilization requires rapid, high-precision corrections. To address this, we propose SoFTA, a Slow-Fast Two-Agent framework that decouples upper-body and lower-body control into separate agents operating at different frequencies and with distinct rewards. This temporal and objective separation mitigates policy interference and enables coordinated whole-body behavior. SoFTA executes upper-body actions at 100 Hz for precise EE control and lower-body actions at 50 Hz for robust gait. It reduces EE acceleration by 2-5x relative to baselines and performs much closer to human-level stability, enabling delicate tasks such as carrying nearly full cups, capturing steady video during locomotion, and disturbance rejection with EE stability.
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。