Space Debris Characterization Using Machine Learning Methods

The project focuses on characterizing space debris using advanced machine learning methods and 3D scanning techniques. Currently, NASA tracks over 23,000 debris objects, and the increasing number of satellites in Low Earth Orbit raises collision risks, potentially leading to a dangerous cascade of debris. The proposal aims to enhance debris classification by employing high-resolution 3D scans to gather extensive shape characteristics, which are then analyzed using machine learning algorithms like Principal Component Analysis and K-Means Clustering. This approach seeks to address the limitations of existing models that rely on basic shape factors and decision trees. As the materials and designs of satellites evolve, the need for rapid and accurate debris characterization becomes critical. The project will utilize a dataset of debris from hypervelocity impacts to create detailed 3D models, facilitating better data processing and classification. Safety measures will be in place to handle debris carefully, minimizing risks associated with sharp edges. Outreach activities at the Prescott Regional SciTech Festival will help disseminate the research findings. Ultimately, the project aims to improve understanding of orbital debris and contribute to developing better collision avoidance strategies and manufacturing techniques, benefiting both the research community and future space missions.

This content was originally posted on my projects website here. The above summary was made by the Kagi Summarizer