Session
Weekend Session V: Automation - Research and Academia
Location
Utah State University, Logan, UT
Abstract
Recent advances in space technology have prompted a surge in the deployment of small satellite constellations by companies such as Starlink and OneWeb. Now numbering in the thousands, these constellations have significantly increased the burden on satellite operators who must monitor and manage them to ensure their safe and reliable functioning. In response to this heightened operational demand, autonomous small satellite operations have become a focal point for investment, innovation and exploration. The latest hardware breakthroughs in Edge AI have paved the way for applying artificial intelligence (AI) techniques directly on board these small satellite systems, heralding a new era of AI-empowered autonomous satellites. Space agencies such as ESA and NASA have taken notice of these advancements and are actively pursuing the development of on-board AI systems. This paper investigates the application of AI to monitor satellite telemetry data on board small satellites to enable real-time detection and immediate response to anomalies. We have curated a unique dataset derived from EIRSAT-1, Ireland's first domestically produced satellite, as a testing and validation resource for these ML models and the future development of AI-enabled small satellites. This dataset consists of a training set developed during ground testing and containing artificial anomalies induced to train satellite operators, a validation dataset containing real anomalies encountered during the qualification campaign, and an early flight test dataset collected since the satellite was launched on December 1st, 2023. This paper presents an in-depth analysis of the efficacy of several ML techniques when applied to the EIRSAT-1 dataset using flight-ready hardware. This study not only showcases the capabilities of these ML techniques in an operational environment but also sets the stage for future research and development in autonomous satellite systems.
From Ground to Orbit: Enhancing Satellite Autonomy With AI-Powered Anomaly Detection
Utah State University, Logan, UT
Recent advances in space technology have prompted a surge in the deployment of small satellite constellations by companies such as Starlink and OneWeb. Now numbering in the thousands, these constellations have significantly increased the burden on satellite operators who must monitor and manage them to ensure their safe and reliable functioning. In response to this heightened operational demand, autonomous small satellite operations have become a focal point for investment, innovation and exploration. The latest hardware breakthroughs in Edge AI have paved the way for applying artificial intelligence (AI) techniques directly on board these small satellite systems, heralding a new era of AI-empowered autonomous satellites. Space agencies such as ESA and NASA have taken notice of these advancements and are actively pursuing the development of on-board AI systems. This paper investigates the application of AI to monitor satellite telemetry data on board small satellites to enable real-time detection and immediate response to anomalies. We have curated a unique dataset derived from EIRSAT-1, Ireland's first domestically produced satellite, as a testing and validation resource for these ML models and the future development of AI-enabled small satellites. This dataset consists of a training set developed during ground testing and containing artificial anomalies induced to train satellite operators, a validation dataset containing real anomalies encountered during the qualification campaign, and an early flight test dataset collected since the satellite was launched on December 1st, 2023. This paper presents an in-depth analysis of the efficacy of several ML techniques when applied to the EIRSAT-1 dataset using flight-ready hardware. This study not only showcases the capabilities of these ML techniques in an operational environment but also sets the stage for future research and development in autonomous satellite systems.
