• DocumentCode
    175532
  • Title

    An UAV navigation aided with computer vision

  • Author

    Zhang Chenghao ; Chen Jiabin ; Song Chunlei ; Xu Jianhua

  • Author_Institution
    Sch. of Autom., Beijing Inst. of Technol., Beijing, China
  • fYear
    2014
  • fDate
    May 31 2014-June 2 2014
  • Firstpage
    5297
  • Lastpage
    5301
  • Abstract
    Currently most of UAV navigation systems are composed of GPS/INS integrated navigation. To give a alternative navigation method when GPS doesn´t work on UAV, a machine vision-aided navigation method is proposed to navigate the UAV with INS. Image Stitching as the core technology is to build real-time ground panoramic image. Combined with GPS latitude and longitude information, certain data fusion method is used to build dynamic ground map. In case of failure of GPS positioning, scene matching techniques is used on the dynamic map to get real-time latitude and longitude information of UAV, to form a integrated navigation with pure inertial system. Also, UAVs can perform a variety of intelligent missions based on the method. In this paper, Visual C++ is applied to form complete visual/GPS/INS integrated navigation simulation system, the simulation experiments of proposed navigation method verify the feasibility of the program and high prospects.
  • Keywords
    C++ language; Global Positioning System; autonomous aerial vehicles; image matching; inertial navigation; robot vision; GPS latitude information; GPS longitude information; GPS-INS integrated navigation; UAV navigation systems; Visual C++; computer vision; dynamic ground map; image stitching; machine vision-aided navigation method; pure inertial system; real-time ground panoramic image; scene matching techniques; Aircraft navigation; Cameras; Equations; Global Positioning System; Mathematical model; Real-time systems; Image Stitching; Scene Matching; UAV; Vision Navigation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control and Decision Conference (2014 CCDC), The 26th Chinese
  • Conference_Location
    Changsha
  • Print_ISBN
    978-1-4799-3707-3
  • Type

    conf

  • DOI
    10.1109/CCDC.2014.6852209
  • Filename
    6852209