New Concepts Of Automated Anomaly Detection In Space Operations Through ML-Based Techniques

Spacecraft health management is a key component to ensure the safety and mission operation life of a satellite complex system. The health monitoring task is pursued exploiting telemetry data, collected using various sensor reading fromonboard devices, that can be analyzed to retrieve and early detect anomalies which can lead to critical failures. The traditional monitoring methods, based on simple threshold checks, are now facing with lots of difficulties the increased complexity of the spacecraft, requiring updated and intelligent systems based on data-driven approaches. In this paper we propose different ML-based methods that contribute to the generation of an intelligent anomaly detector, that can face up the numerous telemetry data. Finally we focus on how to optimize and implement t he developed models on constrained hardware, representative of spacecraft processors.