研究者将 GPT-6 Astra 接入机器人清理陌生厨房
原标题：Researchers plug GPT-6 Astra directly into a robot and let it clean up an unfamiliar kitchen
AI 摘要
斯坦福和加州理工的研究人员构建了 HomeBody 系统，让 Unitree G1 机器人能够自主导航陌生厨房、整理物品并从抽屉取物。该系统去掉了语言模型与机器人之间通常的训练控制层，由可替换的视觉语言模型 GPT-6 Astra 直接调用可扩展技能库，机器人先探索房间、在 Nvidia Isaac Sim 中构建数字孪生并记录空间记忆。局限包括 Astra 延迟、手指伺服过热和高算力成本，代码已在 GitHub 发布。
正文节选
Researchers plug GPT-6 Astra directly into a robot and let it clean up an unfamiliar kitchen 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 direc