This paper introduces a novel method using chaser spacecraft image processing and Convolutional Neural Networks (CNNs) to detect structural markers on the European Space Agency's (ESA) Environmental Satellite (ENVISAT) for safe de-orbiting. Advanced image pre-processing techniques, including noise addition and blurring, are employed to improve marker detection accuracy and robustness. Initial results show promising potential for autonomous space debris removal, supporting proactive strategies for space sustainability. The effectiveness of our approach suggests that our estimation method could significantly enhance the safety and efficiency of debris removal operations by implementing more robust and autonomous systems in actual space missions.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Markers Identification for Relative Pose Estimation of an Uncooperative Target


    Beteiligte:
    Candan, Batu (Autor:in) / Servadio, Simone (Autor:in)

    Erscheinungsdatum :

    2024


    Anmerkungen:

    2024 AAS/AIAA Astrodynamics Specialist Conference


    Medientyp :

    Preprint


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Monocular Relative Pose Estimation Pipeline for Uncooperative Resident Space Objects

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



    Mapping and Pose Estimation of an Uncooperative Rotating Target for Spacecraft Proximity Operations

    Jinzhen, Mu / Shuqing, Cao / Yu, Zhang et al. | Springer Verlag | 2021


    Uncooperative pose estimation with a LIDAR-based system

    Opromolla, Roberto | Online Contents | 2014