Machine Learning applied to asteroid dynamics: an emerging research field

V. Carruba, S. Aljbaae, R. C. Domingos, M. Huaman, W. Barletta

Published 2021-10-13, updated 2022-01-07Version 2

Machine Learning (ML) is the study of computer algorithms that can learn from data or data exposure to better themselves automatically. It is mainly divided into supervised learning, where the computer is presented with examples of entries, and the goal is to learn a general rule that maps inputs to outputs, and unsupervised learning, where no label is provided to the learning algorithm, leaving it alone to find structures. Deep learning is a branch of machine learning based on artificial neural networks, which are computing systems inspired by the biological neural networks that constitute animal brains. In asteroid dynamics, machine learning methods have been recently used to identify members of asteroid families, small bodies images in astronomical fields, and to identify resonant arguments images of asteroids in three-body resonances, among other applications. Here we will conduct a full review of available literature in the field, and classify it in terms of metrics recently used by other authors to assess the state of the art of applications of machine learning in other astronomical sub-fields. While applications of machine learning to Solar System bodies have already reached a state classified as progressing, with more established research communities and methodologies, and articles where the use of ML lead to the discovery of new celestial objects or features, ML applied to asteroid dynamics is still in the emerging phase, with smaller groups, methodologies still not well-established, and fewer papers producing new discoveries or insights. The constant development of more advanced ML algorithms and the need to use tools able to handle the very large datasets that will be produced by large observational surveys, like those conducted at the Vera C. Rubin Observatory, suggests that the need to develop ML algorithms to analyze these vast amounts of data will only increase with time.