This paper presents the use of stereoscopic vision as a means of sensing and detecting obstacles and other aircraft for collision avoidance system for a multicopter. A ZED stereo camera is mounted on a DJI S900 Hexacopter UAS to generate depth maps. A NVIDIA Jetson TX1 board is used for onboard processing of the depth maps and collision avoidance algorithm. The board communicates with the Pixhawk autopilot, which transmits data to the ground control station via XBee radios. By using the ZED software development kit (SDK), it is possible to obtain depth maps directly from the camera and use them in the implementation of obstacle avoidance. The algorithm that is used partitions the depth maps into multiple sections in order to find the section of the image that represents the objects furthest away. This selected section is the section that is free of obstacles. The hexacopter can maneuver itself to that selected section autonomously, thus avoiding obstacles in its path. The developed algorithm and flight test results are discussed.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Autonomous Collision Avoidance System for a Multicopter using Stereoscopic Vision


    Beteiligte:
    Perez, Erwin (Autor:in) / Winger, Alexander (Autor:in) / Tran, Alexander (Autor:in) / Garcia-Paredes, Carlos (Autor:in) / Run, Niran (Autor:in) / Keti, Nick (Autor:in) / Bhandari, Subodh (Autor:in) / Raheja, Amar (Autor:in)


    Erscheinungsdatum :

    2018-06-01


    Format / Umfang :

    1641883 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch






    Collision Avoidance System using Stereoscopic Vision for Unmanned Aerial Systems

    Bhandari, Subodh / Srinivasan, Thirupathi / Gray, Justin et al. | AIAA | 2016


    Collision Detection and Avoidance using Optical Flow for Multicopter UAVs

    Urieva, Natallia / McDonald, Jeffrey / Uryeva, Tatsiana et al. | IEEE | 2020