Highlights The use of ML-based approaches to estimate the ballistic coefficient is investigated. The sensitivity of the approach to relevant parameters is assessed. A simulation environment for the training, validation and testing phase is used. The approach application is shown considering synthetic and real-world test cases. Space weather effects are included in the training phase.

    Abstract The increasing number of Resident Space Objects poses a serious threat to the safe operation of satellites. Alongside with mitigation policies, it is fundamental to predict the trajectories of such objects which requires the accurate estimation of the physical characteristics that influence their orbits. An important role is played by the area-to-mass ratio, i.e., the ratio between the area exposed to the atmosphere/the Sun and the mass of the space object for the LEO/MEO and GEO region. Current literature proposes several approaches for the estimation of the area-to-mass ratio, ranging from semi-analytical to numerical methods. As regards the latter category, recent studies have focused on classification or regression algorithms for specific types of orbits (e.g., sun-synchronous, geostationary). In this context, this paper proposes a machine learning-based regression approach for the estimation of the ballistic coefficient (i.e., the product between the drag coefficient and the area-to-mass ratio) in Low Earth Orbit, covering a wide set of orbital parameters. Using a synthetic space catalogue, the performance of different types of machine learning techniques is evaluated and compared. A sensitivity analysis is carried out to analyse the effect on performance of the number of trainings, the propagation time, the number of objects of the training data set, and the frequency of the measurements. The applicability of the presented approach is tested and discussed using both real (i.e., based on publicly available catalogues of Two-Line Elements) and synthetic datasets.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Machine learning-based approach for ballistic coefficient estimation of resident space objects in LEO


    Contributors:
    Cimmino, N. (author) / Opromolla, R. (author) / Fasano, G. (author)

    Published in:

    Advances in Space Research ; 71 , 12 ; 5007-5025


    Publication date :

    2023-02-04


    Size :

    19 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English




    Ballistic Coefficient Prediction for Resident Space Objects

    Russell, R. | British Library Conference Proceedings | 2013


    Machine Learning Classification of Malicious Resident Space Objects

    Whited, Derick / Doyle, Daniel / Black, Jonathan | TIBKAT | 2022


    Machine Learning Classification of Malicious Resident Space Objects

    Whited, Derick / Doyle, Daniel / Black, Jonathan | AIAA | 2022


    Monocular Relative Pose Estimation Pipeline for Uncooperative Resident Space Objects

    Piazza, Massimo / Maestrini, Michele / Di Lizia, Pierluigi | AIAA | 2022


    Ballistic Coefficient Estimation for Low Altitude Debris Objects from Two-Line Element Data

    Afful, Michael / Gehly, Steven / Cai, Han et al. | AIAA | 2016