A Dual-System Vision-Language-Action Architecture for Whole-Upper-Body Humanoid Control
暂无中文版，显示英文原文。
30 秒速读
A slow VLM planner (7–9 Hz) and a fast visuomotor policy (200 Hz) control a full humanoid upper body from language.
研究问题
How to get language understanding and fast reactive control in one deployable stack?
问题
Large VLMs are too slow for 200 Hz torque control.
既有方法
Scripted skills or single-rate policies.
新方法
Two systems communicating via a latent vector; trained on ~500 h of teleop.
结果
Multi-robot collaboration and picking of thousands of novel household items.
局限
No public benchmark; single-embodiment; teleop data cost.
产业影响
The architecture pattern is now widely copied; watch actuator bandwidth as the next bottleneck.