Monday 23/March/2026 – 12:25 PM
A Chinese laboratory has succeeded in a pioneering experiment to control a human robot using space computing, paving the way for broad future applications of artificial intelligence and autonomous systems in various parts of the world.
The experiment was conducted in collaboration between GuoXing and Shanghai Jiao Tong University, where the team combined a ground robot, an open source artificial intelligence agent, and on-orbit processing capabilities into a closed-loop control system.
According to a report published by Interesting Engineering, the system begins by issuing voice commands from an operator on the ground, which are sent directly to orbiting satellites, and Alibaba’s Qwen3 artificial intelligence model analyzes the commands and makes decisions about moving the robot.
These instructions then return to Earth, where the OpenClaw AI agent translates them into actual movements that the robot accurately executes. This is the first successful case of hosting entire AI “reasoning” operations in orbit.
This achievement demonstrates the ability of space computing to reliably operate robots and autonomous systems, especially in areas where terrestrial networks such as 5G and fiber optics are weak, from humanoid robots and robot dogs to unmanned vehicles and drones.
The experiment also proves that chip-based systems are able to operate efficiently in harsh space environments, laying the foundation for using satellites as an “orbital brain” to operate artificial intelligence and manage autonomous systems on a global scale.
