The Tall Lunar Tower project at the NASA Langley Research Center is focused on the design, modeling, fabrication, and testing of a supervised autonomously assembled engineering development unit for tall lunar towers. The lunar south pole environment poses many challenges for robotic assembly of the tall tower, particularly to computer vision camera systems due to a high-contrast lighting environment. This paper will present an Unreal Engine 5 video game engine Lunar South Pole Lighting Testbed to simulate realistic lunar lighting conditions for synthetic image generation. The fidelity of the simulation environment is investigated by comparing the accuracy of computer vision models trained using synthetic image data and trained from real image data collected in a lunar analog environment.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen


    Exportieren, teilen und zitieren



    Titel :

    Unreal Engine Testbed for Computer Vision of Tall Lunar Tower Assembly


    Beteiligte:
    B. Notosubagyo (Autor:in) / M. K. Mahlin (Autor:in) / J. T. Cassady (Autor:in)

    Erscheinungsdatum :

    2023


    Format / Umfang :

    19 pages


    Medientyp :

    Report


    Format :

    Keine Angabe


    Sprache :

    Englisch




    Unreal Engine Testbed for Computer Vision of Tall Lunar Tower Assembly

    Brian Notosubagyo / Matthew K. Mahlin / Jacob T. Cassady | NTRS


    Unreal Engine Testbed for Computer Vision of Tall Lunar Tower Assembly

    Notosubagyo, Brian / Mahlin, Matthew / Cassady, Jacob | AIAA | 2023



    Unreal Engine Testbed for Computer Vision of Tall Lunar Tower Assembly

    Brian Notosubagyo / Matthew K. Mahlin / Jacob T. Cassady | NTRS