• DocumentCode
    3586975
  • Title

    Indoor visual navigation system based on paired-landmark for small UAVs

  • Author

    Tianmiao Wang ; Yicheng Zhang ; Chaolei Wang ; Jianhong Liang ; Han Gao ; Miao Liu ; Qinpu Guan ; Anqi Sun

  • Author_Institution
    BeiHang Univ., Beijing, China
  • fYear
    2014
  • Firstpage
    1703
  • Lastpage
    1708
  • Abstract
    This paper has presented an implementation of visual based non-GNSS navigation method for small VTOL (Vertically Taking-off and Landing) UAV (Unmanned Aerial Vehicle). A novel method of constructing landmarks is described in detail. It uses a pair of landmarks instead of the general landmark which usually is a single geometry or coding image. The paired-landmark method inherits general landmark methods´ advantages of low-cost, high-efficiency and the ability of absolute positioning. And it also solves the problem of requiring too much different characteristic landmarks when handling large area navigation. The paired-landmark method has been verified by a real parameter simulation environment. Real flight experiment is also carried out to test the effectiveness and performance of the paired-landmark navigation method. It can be concluded that, besides achieved the expected positioning precision, the proposed system also reduced the consumption of calculation and memory space.
  • Keywords
    autonomous aerial vehicles; geometry; image coding; indoor navigation; position control; robot vision; satellite navigation; VTOL; coding image; geometry; indoor visual navigation system; non-GNSS navigation method; paired-landmark method; small UAV; unmanned aerial vehicle; vertically taking-off and landing; Aircraft; Aircraft navigation; Atmospheric modeling; Cameras; Estimation; Visualization;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Robotics and Biomimetics (ROBIO), 2014 IEEE International Conference on
  • Type

    conf

  • DOI
    10.1109/ROBIO.2014.7090580
  • Filename
    7090580