Chinese Lab Uses Orbital Satellites and OpenClaw to Control Ground-Based Humanoid Robots
What Happened
A joint laboratory operated by GuoXing Aerospace Technology and Shanghai Jiao Tong University successfully demonstrated remote control of ground-based humanoid robots using OpenClaw routed through orbiting satellites. In the experiment, an operator issued voice commands that OpenClaw transmitted to Chinese computing satellites in orbit. The satellites' onboard large language models processed the commands through in-orbit inference, then transmitted execution instructions back to Earth to control the robot's movements.
According to Xinhua, this marks the first deployment of AI token-calling services in space — meaning LLM inference was actually performed on orbital hardware rather than being relayed to ground-based data centers.
Why It Matters
This experiment pushes OpenClaw into territory no one anticipated when the project launched four months ago. The implications extend well beyond a tech demo: if AI agent inference can run reliably on satellites, it opens the door to autonomous systems that operate in environments where terrestrial networks are unreliable or nonexistent — disaster zones, remote industrial sites, maritime operations, and military applications. GuoXing Aerospace's ambitious plan to deploy 2,800 computing satellites by 2035 (with 1,000 by 2030) suggests this isn't just a proof of concept but a long-term infrastructure play.
What's Next
The space-computing angle represents a potential new frontier for OpenClaw's architecture. If orbital inference proves cost-effective at scale, it could reshape how autonomous agents operate in connectivity-challenged environments. The 12-satellite initial constellation is already operational, and the planned expansion to 1,000 satellites by 2030 will likely attract attention from defense and logistics sectors looking for reliable AI agent infrastructure beyond terrestrial networks.