ESA Anomaly Detection Benchmark

• The Φsat-2 mission demonstrates advanced AI capabilities in Earth observation through a 6U satellite equipped with a multispectral instrument, launched in August 2024.

• KP Labs developed a CNN-based algorithm for automatic cloud detection, optimizing data transmission by filtering out cloud-covered images before downlink.

• The project implemented 8-bit quantization to ensure algorithm efficiency within CubeSat hardware limitations while maintaining effective real-time image processing capabilities.

‍

• The ORCHIDE project addresses satellite communication limitations by developing an edge computing ecosystem for autonomous data processing and AI integration in space.

• The system implements a software framework for in-orbit reprogramming and deploys machine learning models directly on satellites, enabling autonomous operations without constant ground contact.

• The solution focuses on real-time decision-making and decentralized computing architecture, supporting extended missions through enhanced system autonomy and resource optimization.

• The project addresses the challenge of efficiently handling growing satellite data volumes, as transferring all data to Earth for processing leads to delays and high bandwidth costs.

• KP Labs developed autonomous, in-orbit data processing systems for the Space-Based Data Centers project, designing distributed architectures for large-scale data storage and processing directly in space.

• The project validated machine learning models for space-based systems, explored innovative uses in space exploration, and enabled real-time processing of Earth observation data, reducing latency and bandwidth requirements.

• The PIGEON project develops advanced super-resolution algorithms to enhance hyperspectral satellite imagery while preserving spectral information.

• Using deep learning techniques, the project improves image clarity and detail interpretation for environmental and agricultural monitoring applications.

• The solution aims to enhance monitoring capabilities for applications like crop health assessment and environmental analysis through improved spatial resolution.

• The project addresses traditional soil monitoring challenges by developing satellite-based solutions using hyperspectral imaging and machine learning algorithms for soil composition analysis.

• GENESIS project aims to enable continuous monitoring of soil parameters through the Intuition-1 satellite platform, offering potential for large-scale data collection without ground sampling requirements.

• The proposed system is designed to provide insights for agricultural management through space-based soil analysis, supporting more precise and sustainable farming practices.

‍

‍

• The ESA Anomaly Detection Benchmark addresses inconsistent methods in satellite telemetry anomaly detection across the space industry by creating a standardized framework and comprehensive dataset.

• The project delivered a large-scale satellite telemetry anomaly dataset containing 31 GB of curated data from 3 missions, along with a reproducible evaluation pipeline tailored for operational needs.

• The initiative fostered industry collaboration through public dataset accessibility and introduced specialized algorithms for satellite telemetry anomaly detection, enabling effective comparison and assessment of detection methods.

• The project addresses the need for real-time anomaly detection in satellite telemetry data to enhance performance and reduce reliance on ground interventions.

• Utilizing ESA's OPS-SAT, the project implemented an onboard anomaly detection system with machine learning capabilities, featuring a RandomForest classifier with 95.7% accuracy.

• The system enables autonomous satellite monitoring and fast inference for immediate response, demonstrating the feasibility of onboard anomaly detection for mission autonomy.