Document Type
Manuscript
Publication Date
Fall 2024
School or Division
Graduate School of Engineering and Management
Abstract
Approximately 12% of satellites and other objects launched into outer space have not been registered with the United Nations (UN) as required by international law. To predict whether States will register a launched space object and understand what factors influence a registration decision, data from a UN online index of space objects was used to train and select the best machine learning model. After preparation, the dataset had 1938 datapoints with 11 features, with categorical features simplified and converted to binary.
Multiple variations of classical logistic regression models were compared to multiple variations of dense neural network models. The best model was a logistic regression model using p-value feature selection for binary classification, which balanced performance and simplicity. It had a precision metric of 0.91 and a recall metric of 0.90 with only 6 input features, besting the trivial model’s precision of 0.73 and 0.85.
The best model predicted registration decisions with 90% accuracy. From its results, inferences can be drawn that States which generate a national designator for their space objects are more likely to register those objects, while space objects that decay or deorbit within five years after launch are less likely to be registered.
Recommended Citation
Sawmiller, J. K. (2024). Using machine learning to predict state compliance with international legal obligations for registration of space objects: Comparative performance of logistic regression and dense neural network models. Student Publications series. https://scholar.afit.edu/studentpub/52
Included in
Air and Space Law Commons, Artificial Intelligence and Robotics Commons, Data Science Commons
Comments
Author note: Jonathan Sawmiller was a student in AFIT's Data Analytics Certificate program at the time of this paper.
Approved for public release by USSPACECOM.