Abstract:

As global investment in space continues to grow, the importance of space weather has gained significant commercial relevance, especially during the current solar maximum, which heightens space environment risks. By 2030, over 100,000 satellites are projected to orbit the Earth, making intelligent and resilient satellite design essential for space sustainability.

Space-based components of the F2T2EA (Find, Fix, Track, Target, Engage, and Assess) kill chain—such as sensors and satellites—must maintain performance under multiple, overlapping space weather impacts. This requires resilient designs incorporating health status monitors to ensure operational continuity.

JANUS business unit AER, in collaboration with AFRL, has a long-standing history in space weather research and development, recently integrating advanced artificial intelligence techniques. This collaboration has produced a comprehensive software and data framework to manage large radiation measurement datasets efficiently. Computationally intensive quantities—such as adiabatic invariant coordinates and pitch angle-dependent metrics—are precomputed and stored in a centralized dataset for streamlined access. Preprocessing and metadata tagging enriches the dataset with essential context, including noise levels, integration times, and quality indicators, optimizing AI/ML model training and enhancing data retrieval.

We present an advanced space anomaly simulation and prediction tool that leverages AI/ML to diagnose and assess the risks of satellite anomalies across specific constellations. By utilizing a common operational picture (COP) for a unified scene and employing AI fusion to integrate spatially offset yet temporally consistent tensors, our tool enables real-time inferences within a secure DevSecOps environment. The AI inference engine interfaces seamlessly with the anomaly tool, delivering actionable insights.

This innovative approach improves operational reliability in electronic warfare by enabling rapid anomaly detection and attribution. Our solution enhances the resilience of satellite constellations, ensuring they can withstand the increasing risks posed by the dynamic space environment.

Keywords: AI;ANALYTICS;BIG DATA

Menu

Ben Prince