European Space Agency Benchmark for Anomaly Detection in Satellite Telemetry

2024/06/25 by Krzysztof Kotowski, Christoph Haskamp, Kotowski, Krzysztof +20 · 1 voice · 2 citations
Computer Science · Engineering · Physics and Astronomy · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #GNSS positioning and interference #Machine Learning (cs.LG) #Space exploration and regulation #Spacecraft Design and Technology #cs.AI #cs.LG

paper · pdf · doi:10.48550/arxiv.2406.17826

openalex publication_date 2024/06/25 · arxiv published 2024/06/25 · openalex created_date 2024/06/28 · arxiv updated 2025/08/17 · openalex updated_date 2026/07/29

Abstract

A Python library that evaluates time-series anomaly detection the way the recent literature recommends: corrected event-wise precision/recall/F-beta and affiliation-based precision/recall (canonical KDD-2022 reference implementation, vendored and maintained), behind a validated ingestion contract that raises typed errors instead of producing numbers from leaky or malformed evaluations — including a train/test-window leakage guard that is on by default. Includes ESA-ADB and TimeEval format readers, scikit-learn-convention metric wrappers, and deterministic JSON/Markdown reports. Core dependencies: numpy and pandas only.

Cited by

• An Improved Time Series Anomaly Detection by Applying Structural Similarity
• CAPMix: Robust Time Series Anomaly Detection Based on Abnormal Assumptions with Dual-Space Mixup

Discussions

• European Space Agency Benchmark for Anomaly Detection in Satellite Telemetry [hn, 3 points, 0 comments]

Related

• The OPS-SAT benchmark for detecting anomalies in satellite telemetry
• Machine learning-driven Anomaly Detection and Forecasting for Euclid Space Telescope Operations
• Convolutional Autoencoders for Data Compression and Anomaly Detection in Small Satellite Technologies
• Automating the Deep Space Network Data Systems; A Case Study in Adaptive Anomaly Detection through Agentic AI
• A Data-Driven Health Monitoring Method for Satellite Housekeeping Data Based on Probabilistic Clustering and Dimensionality Reduction