Search papers, labs, and topics across Lattice.
To eliminate the prohibitive warm-up times and label requirements of spacecraft monitoring, this work establishes an unsupervised, deployment-ready anomaly detection framework capable of operating from the mission's second month. The system integrates monthly incremental retraining and automated statistical model selection with adaptive Extreme Value Theory (EVT) thresholding to enforce rigorous false alarm control. Evaluated chronologically on the real-world ESA Anomalies Dataset, it achieves robust precision-weighted performance with an $F_{0.5}$ of 0.700 on Mission 1 and 0.698 on Mission 2 without any mission-specific tuning.
Spacecraft can now flag critical telemetry faults just 30 days post-launch without a single historical anomaly label or manual threshold adjustment.
Operational anomaly detection in spacecraft telemetry typically requires labeled historical anomalies or extended warm-up periods. These requirements are rarely met in practice. We propose an unsupervised, deployment-ready framework that produces predictions from the second month of operation without any labels, prior fault knowledge, or mission-specific tuning. The approach combines incremental monthly retraining, statistical model selection, and adaptive Extreme Value Theory (EVT) thresholding for false alarm control. On the ESA Anomalies Dataset (ESA-AD), it achieves $F_{0.5}=0.700$ on Mission~1 and $F_{0.5}=0.698$ on Mission~2 under strict chronological evaluation.
