Star Dust Intelligence Uses AOS to Leverage General AI Benefits Without Directly Competing with Large Models
不与大模型对抗，星尘智能想用AOS接住通用智能的红利
Star Dust Intelligence founder Lai Jie proposes a partnership between general-purpose and embodied AI models, facilitated by Astribot OS (AOS). The company's three-tier architecture integrates self-developed Lumo models, the AOS operating system, and hardware. Recent hackathons demonstrated that AOS lowers development barriers, enabling non-experts to create robot applications using natural language, with 16 of 17 finalist apps successfully tested on real machines.
Source: Chinese robotics — Robot learning research · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
robot_app_runs
16
Reported trials
development_speed_increase
3–5
View original evidence
工具链接通之后，星尘内部的开发速度提升了3到5倍，新来的应届生和实习生在开发速度上也能接近老手，因为他们不必先花几年熟悉机器人硬件。
Open source S23robot_teams
17
Reported trials
elapsed_hours
36 hours
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
工具链接通之后，星尘内部的开发速度提升了3到5倍，新来的应届生和实习生在开发速度上也能接近老手，因为他们不必先花几年熟悉机器人硬件。
Open source S23
决赛现场的17支队伍在36小时里将想法推进到真机测试，16个App跑通真机，上真机率达94%。
Open source S18
更值得注意的是，其中不少参赛者此前没有机器人开发经验。
Open source S19
Source：Chinese robotics — Robot learning research · news.qq.com