• DocumentCode
    2137166
  • Title

    Redundant tracking of a robotic arm

  • Author

    Kravitz, Brian H. ; Madden, Brian C. ; Sayers, Craig P. ; Paul, Richard P.

  • Author_Institution
    Dept. of Comput. & Inf. Sci., Pennsylvania Univ., Philadelphia, PA, USA
  • fYear
    1997
  • fDate
    7-9 Jul 1997
  • Firstpage
    629
  • Lastpage
    633
  • Abstract
    This paper describes the development of a practical stereo vision system which is intended for use in visual tracking of man-made targets from a hovering subsea vehicle. Blob tracking, centroid computation, camera calibration, and stereo vision algorithms are combined to form a robust real-time implementation requiring minimal human intervention. The cameras are calibrated using images of a manipulator-mounted target. Using a less-than-state-of-the-art workstation (Sun IPX) without any special-purpose hardware, targets within a 400×150×400 millimeter viewing volume are tracked at rates exceeding 1 Hz (including the acquisition of 3 512×512 images) with a worst-case accuracy of better than 3 mm for both two- and three-camera configurations
  • Keywords
    calibration; cameras; computer vision; manipulators; marine systems; real-time systems; redundancy; stereo image processing; target tracking; tracking; 1 Hz; 150 mm; 262144 pixel; 400 mm; 512 pixel; Sun IPX; blob tracking; camera calibration; centroid computation; hovering subsea vehicle; man-made targets; manipulator-mounted target; redundant tracking; robotic arm; robust real-time implementation; stereo vision algorithms; stereo vision system; visual tracking; Calibration; Cameras; Computer vision; Humans; Robot vision systems; Robustness; Stereo vision; Target tracking; Vehicles; Workstations;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Advanced Robotics, 1997. ICAR '97. Proceedings., 8th International Conference on
  • Conference_Location
    Monterey, CA
  • Print_ISBN
    0-7803-4160-0
  • Type

    conf

  • DOI
    10.1109/ICAR.1997.620248
  • Filename
    620248