斯坦福和加州理工將GPT-6 Astra直接植入機器人，用於廚房清理

researchers from stanford and caltech have developed homebody, a system that integrates gpt-6 astra with a unitree g1 robot to autonomously navigate and clean an unfamiliar kitchen. the system eliminates the typical trained control layer between the language model and robot, instead using gpt-6 astra to directly call into an extensible skill library for tasks like grasping and navigating. the robot first explores the room, builds a digital twin in nvidia's isaac sim, and logs objects and locations in spatial memory, allowing it to find items even after they leave its field of view. however, limitations include astra's latency, overheating finger servos, and high compute costs. the code is available on github.