Orbit Safe: An Integrated System for Space Object Collision Risk Assessment

This project focuses on developing a viable hybrid deep-learning system for NEO (Near Earth Object) monitoring and risk assessment, using TLE (Two-Line Element) data from open-source datasets such as NASA’s NEO Database and CelesTrak. The system’s performance will be assessed alongside other existing models using precision, accuracy, recall, validation loss, and F1 scores. Our proposed model aims to drastically reduce the danger posed by natural NEOs, such as asteroids orbiting the Earth, as well as artificial NEOs, such as space debris. These objects pose a significant risk given the exponential rise in humanity’s spacefaring activities, as well as the rapid rise in space debris caused as a consequence of the Kessler Syndrome effect. If this issue is left unaddressed, it will inevitably lead to catastrophic results. Our proposed model will not only enhance NEO monitoring and risk assessment, but it will also help ensure the essential goal of achieving sustainability in space, in alignment with the Space Debris Mitigation Guidelines set forth by UN COPUOS (Committee on the Peaceful Uses of Outer Space).