Sakhanda Wire
NVDA $152.58 -2.28% MSFT $506.11 -2.29% GOOGL $279.81 -1.38% META $619.07 -1.17% AMZN $256.55 -0.47%
← Back to the news

GPT-6 Astra controls a robot cleaning an unfamiliar kitchen

Researchers from Stanford and Caltech developed HomeBody, a system that enabled a Unitree G1 humanoid robot to work autonomously in a kitchen it had not previously encountered. The robot tidied the room and retrieved objects from drawers, using GPT-6 Astra as its vision-language model.

HomeBody removes the conventional separately trained control layer between the language model and the robot. Instead, the model can call an expandable library of skills, including grasping, navigation and drawer opening. Before carrying out tasks, the robot surveys the environment, creates a digital twin in Nvidia's Isaac Sim and records objects and their positions in spatial memory. This allows it to locate items after they move out of view.

For a broad instruction such as cleaning the kitchen, the model determines the sequence of actions and corrects its plans when problems occur. The researchers identified several constraints: Astra responds slowly, the robot's finger servos can overheat, and the system requires substantial computing resources. The HomeBody code has been published on GitHub. Previous evaluations reported stronger spatial reasoning for Astra, while another assessment identified safety concerns when it controls a robot.

Written by Sakhanda Wire AI from reporting by The Decoder. Facts, quotes and full context are in the original article below.

Originally published by The Decoder on

Read the original on The Decoder ↗

The reporting and the image belong to The Decoder and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news