Dottorato in INGEGNERIA AERONAUTICA E SPAZIALE

Titolo della tesi: Low-level human interaction pipeline for identification, monitoring and cataloguing orbital objects from optical observations

The exponential growth in satellite deployments and orbital debris necessitates advanced and automated methodologies for Space Surveillance and Tracking (SST). Current software solutions for extracting orbital parameters from SST images are heavily reliant on manual intervention, particularly in the detection and distinction of satellite tracks. This dependency introduces risks of misidentification and limits the integration of such tools into autonomous frameworks. This thesis addresses these limitations by developing a comprehensive system to extract right ascension (RA), declination (Dec) angles, and light-curve data from optical observations, which are crucial for enhancing orbital and attitude determination processes. The research focuses on improving orbital element accuracy for catalogued objects and enabling initial orbit determination for uncatalogued ones using optical SST data. Techniques for deriving light curves from optical observations are established to support attitude determination and behavioral analysis of Resident Space Objects (RSOs). A major contribution of this work is the development of software capable of extracting RA/Dec measurements and light-curve data from sequences of frames with minimal human intervention. Machine learning algorithms are integrated to enhance the accuracy and efficiency of detection processes. The research also investigates light curve inversion methods to optimize computational efficiency and precision in attitude determination using light curves. To support current analysis and future research, a structured SST database is designed to store observational frames, analysis outputs, and results. The methodologies proposed are grounded in extensive observational campaigns, incorporating optical systems such as AWARE, SCUDO, and Mini-SURGE. The thesis describes the installation, testing, and operation of these systems, as well as the design and deployment of a processing pipeline for frame analysis. The pipeline includes functionalities such as image calibration, tracking algorithms, light-curve extraction, and orbit determination techniques. Validation through observational campaigns demonstrates the effectiveness of the proposed methods and tools, highlighting their potential to advance automation and precision in SST operations. By addressing critical challenges in the field, this research significantly contributes to the development of autonomous systems for orbital and attitude determination, paving the way for future innovations in space situational awareness.