World’s First Robot Training Property: 300,000 Chinese Homes for Robots
Original: 全球首个机器人训练楼盘开盘：30万套中国住宅，机器人拎包入住
Kairos-Homeworld creates interactive Chinese home simulations for embodied AI robot training.
Daxiao Robot and CUHK MMLab introduced Kairos-Homeworld, an open project with 300,000 Chinese residential floor plans and 5,000 interactive 3D home scenes. It can generate full household environments from prompts, including layouts, furniture, objects, and physical properties. The article frames it alongside Kairos 3.0-4B as part of a broader embodied AI stack: world model, data, and environment.
QbitAI reports on Kairos-Homeworld, launched by Daxiao Robotics and CUHK's MMLab, describing it as a "training real-estate complex" for embodied-intelligence robots. The focus of this project is not house-hunting in real estate, but rather moving large numbers of Chinese family residences into an interactive digital simulation environment, so robots can practice navigation, cross-room tidying, and household tasks in scenes that closely resemble real home structures. The article notes that the team open-sourced two categories of resources: data on 300,000 real Chinese residential floor plans, and 5,000 fully interactive 3D home scenes. Compared with many indoor datasets based on Western households, Kairos-Homeworld emphasizes the spatial layouts common in Chinese residences—such as entryways, balconies, enclosed kitchens, and wet-dry separated bathrooms—aiming to reduce the scene gap robots face when deployed in Chinese homes.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on 量子位 QbitAI →Related
Summaries are AI-generated; the original article is authoritative.