AI tool detects unusual satellite behavior to improve space safety

A new AI model has been developed to monitor satellites and detect anomalies in their behavior, aiming to improve space safety. The tool, created by researchers from the Alan Turing Institute's Defence AI Research Centre, is the first to predict satellite motion by analyzing light reflections. The model is trained on large quantities of satellite brightness readings, or "light curves," and can identify unusual or unexpected light curves 88% of the time. It can also distinguish between different satellite behaviors, such as spinning versus tumbling, which is essential for in-orbit servicing and satellite longevity. The project, led by Professor Massimiliano Vasile, is part of a UK Space Agency international bilateral fund consortium across several countries.

The AI model is designed to monitor satellites and detect anomalies in their behavior, reducing the risk of in-orbit collisions. The tool has been trained on real-time or recent light curves from ground-based observatories and is fed into the system to flag anomalies for human investigation. The model's ability to identify unusual satellite behavior and predict their motion is crucial for space traffic management and collision avoidance. The research highlights the growing challenge of safeguarding thousands of satellites, with over 4,000 new satellites launched in 2025, compared to 159 in 2000. The AI tool demonstrates the potential for real-time anomaly detection and allows human operators to quickly investigate and take action to avoid collisions.

The research is part of the AI4 Space Safety and Sustainability project, which is led by Professor Vasile and involves multiple institutions and industry partners. The tool's success is seen as a key step toward a complete and systematic analysis of the behavior of any resident space object. The project's next steps include researching the potential for multimodal systems to include radar data, hyperspectral data, and satellite orbit data, which could provide even greater insights for satellite monitoring and safeguarding. The tool's ability to detect anomalies and predict satellite motion is a significant advancement in space safety and could help address the increasing number of satellites in orbit.

AI-generated summary of the original article — see the source for the full story.