This work develops a Phase A operational concept for the IceBrain-1 mission payload, designed to demonstrate AI-based in-orbit image recognition on a GOMSpace CubeSat platform. The payload, centered around an Instrument Control Unit (ICU), integrates a high-performance imaging system that includes a camera with long focal length optics, an AI image processor, and a reconfigurable high performance number-crunching proces- sor. The ICU facilitates communication among payload’s components and with ground control via USB and Ethernet interfaces. The methodology follows a top-down approach, beginning with system-level operations and delving into component-level functionalities. Results include a comprehensive functional tree, an operational state machine, an opera- tional database detailing telemetry and telecommand packets, and a Python demonstrator that emulates the ICU’s core tasks. This research establishes a foundational framework for integrating AI into satellite payloads, providing insights into AI processing, COTS inte- gration, and best practices that support more autonomous, responsive, and cost-effective space missions aligned with New Space standards.
Questo lavoro sviluppa un concetto operativo di Fase A per il payload della missione IceBrain-1, progettato per dimostrare il riconoscimento di immagini in orbita basato su IA su una piattaforma CubeSat di GOMSpace. Il payload, incentrato su un’Unità di Con- trollo dello Strumento (ICU), integra un sistema di imaging ad alte prestazioni che include una fotocamera con elevata lunghezza focale, un processore di immagini IA e un processore riconfigurabile ad alte prestazioni per calcoli intensivi. L’ICU facilita la comunicazione tra i componenti del payload e con il controllo a terra tramite interfacce USB ed Ethernet. La metodologia segue un approccio dall’alto verso il basso, iniziando dalle operazioni a livello di sistema e approfondendo le funzionalità a livello di componente. I risultati includono un albero funzionale completo, una macchina a stati operativi, un database operativo che dettaglia i pacchetti di telemetria e telecomando, e un dimostratore in Python che emula i compiti principali dell’ICU. Questa ricerca stabilisce una struttura fondamentale per l’integrazione dell’IA nei payload satellitari, fornendo approfondimenti sull’elaborazione dell’IA, l’integrazione di componenti COTS e le migliori pratiche che supportano missioni spaziali più autonome, reattive ed economiche in linea con gli standard della New Space economy.
Phase A operational concept for the IceBrain-1 mission payload
Testa, Mariangela
2023/2024
Abstract
This work develops a Phase A operational concept for the IceBrain-1 mission payload, designed to demonstrate AI-based in-orbit image recognition on a GOMSpace CubeSat platform. The payload, centered around an Instrument Control Unit (ICU), integrates a high-performance imaging system that includes a camera with long focal length optics, an AI image processor, and a reconfigurable high performance number-crunching proces- sor. The ICU facilitates communication among payload’s components and with ground control via USB and Ethernet interfaces. The methodology follows a top-down approach, beginning with system-level operations and delving into component-level functionalities. Results include a comprehensive functional tree, an operational state machine, an opera- tional database detailing telemetry and telecommand packets, and a Python demonstrator that emulates the ICU’s core tasks. This research establishes a foundational framework for integrating AI into satellite payloads, providing insights into AI processing, COTS inte- gration, and best practices that support more autonomous, responsive, and cost-effective space missions aligned with New Space standards.File | Dimensione | Formato | |
---|---|---|---|
Phase A Operational Concept for the IceBrain-1 Mission Payload.pdf
accessibile in internet per tutti
Dimensione
7.06 MB
Formato
Adobe PDF
|
7.06 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/230077