E. Smirnov “The machine-learning methods in the asteroids dynamics”

Abstract. In asteroid dynamics, many problems require numerical integration of asteroids
orbits. This approach consumes enough computer resources, especially when we try to
analyse the dynamics of hundreds of thousands of asteroids. Any improvement in the
orbit of the asteroid requires additional computer resources to be applied. Therefore,
within the context of the increasing volume of new information, fast new methods should
be applied to work with big data.
Artificial intelligence and machine-learning (ML) methods have become popular in
recent years. In this study we apply the modern ML methods to the classical problems
of the dynamics of the asteroid: the identification of the resonances, families, non-regular
objects. It is shown, that such methods provides acceptable accuracy and requires much
less computational resources.

Leave a Comment

Your email address will not be published. Required fields are marked *