The Center for Space Exploration Research at the University of California, Davis, is working with Proteus Space to launch a small satellite for the U.S. government. This satellite, about the size of a half refrigerator, has a special artificial intelligence (AI) payload designed to monitor its own condition.
The project is unique because it will be designed and launched in just 13 months, much faster than the years it usually takes for similar satellites. The launch is planned for October 2025 from Vandenberg Space Force Base in California.
The AI payload, created by Professor Stephen Robinson’s laboratory at UC Davis, is called a digital twin. A digital twin is a computer model that mimics the real-time condition of the satellite’s power system, like its batteries. Unlike most satellites, where ground control centers monitor health, this digital twin runs on the satellite itself. It uses sensors to check battery voltage and charge levels. The AI analyzes this data to understand the satellite’s current state and predict how it will work in the future.
Advanced technology for smarter satellites
This self-monitoring system allows the satellite to report its own health, a task usually done by people on Earth. As the satellite operates, the AI learns and improves its predictions about the battery’s performance.
"It should get smarter as it goes,” notes Robinson in a press release, “and be able to predict how it’s going to perform in the near future. Current satellites do not have this capability.”
This is a new feature, as current satellites do not have this ability. The satellite will carry other commercial and research payloads, but the digital twin is the key innovation. It will orbit in low Earth orbit, a region close to Earth, for up to 12 months. After about three years, it will naturally fall back to Earth and burn up in the atmosphere.
This project focuses on making the satellite’s AI smarter and more efficient, paving the way for future satellites to monitor themselves independently.