Application of machine analysis methods for classification of resonant motion of small bodies of the Solar System

T.Yu. Galushina, I. V. Tomilova, E. A. Pakhomov, O.N. Letner, O. M. Syusina
  1. Tomsk State University, Lenina, Tomsk, Russia
DOI10.17184/eac.9178 Pages267–276 Published21 December 2024 Cite Read the PDF · 10 p.

Abstract

This paper examines the possibility of using machine analysis methods to study the features of the resonance dynamics of asteroids and artificial Earth satellites. The paper considers two options for automatic classification of resonant arguments: using convolutional neural networks and by analyzing the argument values. In the course of experimental studies, it was shown that both programs are capable of identifying and classifying the resonant arguments of small bodies, which makes it possible to speed up the process of analyzing the dynamics of these objects. However, the second option is more visual, while the neural network is essentially a black box. The paper also presents the examples of application of the developed software system to the analysis of the dynamics of asteroids and artificial Earth satellites.

Keywords
Asteroidsartificial Earth satellitesorbital resonancesclassification of reso- nancesconvolutional neural networks

Read the article

First page of the article

PDF of 10 pages, as printed in the journal, pages 267 to 276. Open access.

Read the PDF