For the first time, a satellite has used a vision-language model to find what it was looking for on its own. The onboard software package, built by NASA's Jet Propulsion Laboratory, identified areas of interest in response to natural language queries. This milestone marks the first reported use of a vision-language model in orbit and could fundamentally change what space-based sensors are capable of.
The demonstration is significant as it could make space sensors more useful by doing initial data triage on orbit, reducing the flood of raw data that analysts currently have to wade through. Longer term, it's a proof point toward running larger-scale AI infrastructure in space.
The goal is to build out the constellation to ensure real-time coverage of anywhere on Earth, which would take somewhere between 50 and 100 satellites like Yam-9.
