Machine-learning identification of asteroid groups

Abstract

Asteroid families are groups of asteroids that share a common origin. They can be the outcome of a collision or be the result of the rotational failure of a parent body or its satellites. Collisional asteroid families have been identified for several decades using hierarchical clustering methods (HCMs) in proper elements domains. In this method, the distance of an asteroid from a reference body is computed, and, if it is less than a critical value, the asteroid is added to the family list. The process is then repeated with the new object as a reference, until no new family members are found. Recently, new machine-learning clustering algorithms have been introduced for the purpose of cluster classification. Here, we apply supervised-learning hierarchical clustering algorithms for the purpose of asteroid families identification. The accuracy, precision, and recall values of results obtained with the new method, when compared with classical HCM, show that this approach is able to found family members with an accuracy above 89.5 per cent, and that all asteroid previously identified as family members by traditional methods are consistently retrieved. Values of the areas under the curve coefficients below Receiver Operating Characteristic curves are also optimal, with values consistently above 85 per cent. Overall, we identify 6 new families and 13 new clumps in regions where the method can be applied that appear to be consistent and homogeneous in terms of physical and taxonomic properties. Machine-learning clustering algorithms can, therefore, be very efficient and fast tools for the problem of asteroid family identification.

Using Mars co-orbitals to estimate the importance of rotation-induced YORP break-up events in Earth co-orbital space

“…Machine-learning techniques have sometimes been used to investigate the presence of coherent dynamical groups or genetic families within the populations of asteroids of the main belt (see e.g. Zappala et al 1990Zappala et al , 1994Masiero et al 2013;Smirnov & Markov 2017;Carruba, Aljbaae & Lucchini 2019;McIntyre 2019;Carruba et al 2020c). In this section, we apply unsupervised machine-learning in the form of clustering algorithms to the data discussed above.…”

Artificial Neural Network classification of asteroids in the M1:2 mean-motion resonance with Mars

“…Gallardo et al (2011) found that the three asteroids families most affected by the M1:2 mean-motion resonance were those of Nysa, Massalia, and Vesta. Here, following the approach of Carruba et al (2021), we use learning Hierarchical Clustering Method (HCM), as implemented in Carruba et al (2019), on a domain of proper elements for the group of M1:2 asteroids above described. The procedure used to implement this method was the same as that applied in Carruba et al (2021): a critical distance cutoff 1 2 d0 was obtained, and groups were identified for values of d0 ±5 m/s.…”

“…We selected groups that have at least 10 members at the critical distance cutoff value, 5 mem- bers at the lowest distance cutoff of d0 − 5 = 22.75 m/s, and were still identifiable at the highest distance cutoff of d0 − 5 = 32.75 m/s. Interested readers could find more details on the procedures used in Carruba et al (2019Carruba et al ( , 2021.…”

Machine learning classification of new asteroid families members

“…We can also define a Precision coefficient that yields the ability of the model to avoid predicting false data. This is given by (see equation 3 in Carruba et al 2019)…”