Exploring the potential of neural networks in early detection of potentially hazardous near-earth objects
Open Access
- 24 April 2025
- journal article
- research article
- Published by Springer Nature in Celestial Mechanics and Dynamical Astronomy
- Vol. 137 (3) , 1-17
Abstract
We propose a novel method to estimate the minimum orbit intersection distance (MOID) of a near-earth object (NEO) based on artificial neural networks (NNs). The MOID is defined as the minimum distance between the two osculating Keplerian orbits of the Earth and the NEO as curves in the three-dimensional space; it is usually used as an indicator of the possibility of a collision between the asteroid and the Earth, at least for the period during which the Keplerian orbit of the asteroid provides a reliable approximation of the actual orbit. In the present work, the MOID is estimated with a multilayer feedforward NN that takes as input the coordinates of the asteroid at a specified epoch. The NN has been trained on an artificial dataset of about 800000 NEOs, generated with NEOPOP, and tested on the population of real near-earth asteroids (NEAs). The network exhibits near-instantaneous predictions of the MOID and achieves a mean absolute error of approximately $$ 10^{-3} $$ AU on the test set. This research represents a step forward in addressing the urgent need for effective impact monitoring techniques, partially answering the question of whether machine learning can serve as a preliminary filter for some problems of orbit determination.
Keywords
Funding Information
- Università degli Studi di Trento
This publication has 24 references indexed in Scilit:
- From perturbative methods to machine learning techniques in space scienceBollettino dell'Unione Matematica Italiana, 2024
- A Physic-Informed Neural Network Approach to Orbit DeterminationThe Journal of the Astronautical Sciences, 2023
- An investigation on space debris of unknown origin using proper elements and neural networksCelestial Mechanics and Dynamical Astronomy, 2023
- Machine learning applied to asteroid dynamicsCelestial Mechanics and Dynamical Astronomy, 2022
- Proper elements for space debrisCelestial Mechanics and Dynamical Astronomy, 2022
- An Algebraic Method to Compute the Critical Points of the Distance Function Between Two Keplerian OrbitsCelestial Mechanics and Dynamical Astronomy, 2005
- Distance Between Two Arbitrary Unperturbed OrbitsCelestial Mechanics and Dynamical Astronomy, 2005
- Orbit determination with very short arcs. I admissible regionsCelestial Mechanics and Dynamical Astronomy, 2004
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989
- The significance probability of the smirnov two-sample testArkiv för Matematik, 1958