AI System To Monitor Satellite Health In New ESA-Backed Research Program
Spire Global and SATE Will Spend 18 Months Testing Whether Satellites Can Diagnose Their Own Problems
A new research program will test whether artificial intelligence can let satellites detect and diagnose their own technical problems in real time, under an 18-month effort backed by the European Space Agency.
“STRAIDE brings artificial intelligence on-board for satellite autonomy.”
John Ward, Spire Global
Spire Global, based in Vienna, Virginia, is partnering with Systems and Advanced Technologies Engineering (SATE), an engineering firm based in Venice, Italy, on the program, called STRAID. SATE describes itself as a leading provider of AI-based software for advanced health monitoring and predictive diagnostics. The work is funded through subscriptions to ESA’s General Support Technology Program held by the Italian Space Agency and the UK Space Agency, with programmatic support from ESA itself, according to SATE’s own announcement of the partnership.
According to a news release, the program aims to build a system for real-time satellite health monitoring, anomaly diagnosis and predictive failure analysis, according to the release. Neither company disclosed a dollar figure for the program’s funding.
Onboard artificial intelligence is not new to European satellite programs. ESA and the German Research Center for Artificial Intelligence set up a lab in 2021 to develop AI systems for satellite autonomy and collision avoidance, and the agency’s FSSCat mission, launched in 2020, was Europe’s first Earth observation mission to carry AI onboard. NASA has pursued similar ground, using its ASPEN system to help plan mission operations while other AI algorithms monitor spacecraft health and predict failures, according to Capitol Technology University. STRAIDE would add Spire and SATE to that list of companies and agencies working on autonomous fleet health monitoring.
Spire said it will contribute its satellite platform, onboard processing capabilities and operational flight data to the effort, along with a ground-based hardware replica of its satellite systems. SATE will develop CLUE, an AI-based diagnostic system designed to monitor satellite “vital signs,” detect early warning signals, diagnose anomalies and estimate time-to-failure for critical components, the release said. CLUE is designed to run both onboard satellites and on the ground.
“As Spire’s satellite constellation grows in scale, the operational complexity of managing each asset demands a fundamentally different approach,” John E Ward, Spire’s senior director of research and development, said in the release. “STRAIDE brings artificial intelligence on-board for satellite autonomy, detecting anomalies earlier, anticipating failures before they occur, and freeing operators to oversee larger fleets with greater confidence.”
Chiara Brighenti, SATE’s sole administrator, said in the company’s own announcement that the program “represents a significant step forward in enabling highly robust and autonomous satellite constellations, enhancing operational efficiency and service availability.” She said CLUE is designed to deliver early warnings of emerging anomalies and support autonomous mitigation and recovery.
According to the release, Spire has deployed satellites with onboard processing capabilities since 2022.
The partners aim to reach a validated end-to-end demonstration of the system by the end of the 18-month program, according to Spire. The company said the work is intended to inform how AI-driven diagnostics could eventually be built into its satellite architecture. No timeline for operational deployment was given.