SCRAM (Satellite Collision Reduction / Avoidance Model )

Monte Del Sol Charter School 

Team # 

STEM Dragons

Gabriella Armijo, Adelina Baca, Emlee Taylor 

SCRAM

Satellite Collision Reduction / Avoidance Model 

Anything in orbit is at risk of collision. This is dangerous because it could potentially send thousands of shards of debris in multiple directions. Making matters worse, it brings into effect what is known as Kessler Syndrome, which is the theoretical scenario that says if a collision were to occur it would start a chain reaction of collisions that would lead to the downfall of all things in orbit. Even if the satellite does not produce large amounts of debris, damage could affect the instruments in the satellite rendering the satellite useless.

 This is a hard truth that all space researchers grapple with. Physics-based models exist to predict these satellite orbits; however, these commonly accepted models are problematic since they are based on mathematical equations and don’t account for past data. Known as simplified perturbation models (or cumulatively also called SGP4) have issues with precision as they only include one past data point. To improve prediction techniques, we will create a machine-learning model to include multiple past data points and provide a more accurately predicted orbit path. Both SGP4 and our machine learning model will use Two-Line Element data provided by spacetrack.org to make predictions. These predictions will then be compared to accurately predict future orbit paths.