Title: Anomaly detection for Spacecraft TelemetRy dAta using Artificial Intelligence (ASTRA-AI)

Duration: 2023-2025

Funding scheme: MCST Space Upstream Programme

Budget: ca EUR 125,000

Principal investigator: Prof. Ing. Gianluca Valentino

Co-investigator: Dr Robert Camilleri

Detecting anomalies in telemetry data captured on-board a spacecraft is critical to ensure its safe operation, and enables operators to respond to various failures and hazards more quickly. The ASTRA-AI project proposes to use advanced deep learning neural network architectures to detect such anomalies in a precise manner. In addition, a surrogate model will be developed to build a digital twin, which can be used for diagnostic and prognostic purposes. The anomaly detection models will be deployed on a single-board computer to evaluate whether they can effectively be run onboard a spacecraft. A tool will also be developed to visualize the anomalies.