斯坦福与 Caltech 用 GPT-6 Astra 直驱 Unitree G1 机器人整理陌生厨房

The Decoder:AI News(RSS)·2026-09-27 18:59·2小时前·Matthias Bastian
AI 导读

斯坦福和 Caltech 构建了 HomeBody 系统,让 Unitree G1 机器人无需传统训练控制层,由可替换的 VLM(此处为 GPT Astra)直接调用抓取、导航、开抽屉等技能库,自主探索陌生厨房并完成整理和取物。

The Decoder:AI News(RSS)
47AI 编辑部评分,满分 100

斯坦福与 Caltech 用 GPT-6 Astra 直驱 Unitree G1 机器人整理陌生厨房

2026-09-27 18:59· 2小时前· Matthias Bastian
AI 导读

斯坦福和 Caltech 构建了 HomeBody 系统,让 Unitree G1 机器人无需传统训练控制层,由可替换的 VLM(此处为 GPT Astra)直接调用抓取、导航、开抽屉等技能库,自主探索陌生厨房并完成整理和取物。

Researchers from Stanford and Caltech have built HomeBody, a system that lets a Unitree G1 robot autonomously navigate an unfamiliar kitchen, tidy up, and fetch items from drawers. It's yet another test showing GPT-6 Astra works well with robots.

HomeBody drops the typical trained control layer between language model and robot. Instead, a swappable vision-language model (VLM), here GPT Astra, calls directly into an extensible skill library for grasping, navigating, or opening drawers. The robot explores the room first, builds a digital twin in Nvidia's Isaac Sim, and logs objects and locations in spatial memory. That lets it find items even after they leave its field of view. For tasks like "clean up the kitchen," the language model plans each step and self-corrects on errors.

视频 · 前往原文观看

Limitations include Astra's latency, overheating finger servos, and high compute costs. The code is on GitHub. Earlier benchmarks showed Astra's greatly improved spatial reasoning, while another flagged safety issues when Astra controls a robot. OpenAI has already announced plans to get back into robotics, including for personal use.

HomeBody

来源:The Decoder:AI News(RSS)· the-decoder.com