• DocumentCode
    681594
  • Title

    Detecting, locating and crossing a door for a wide indoor surveillance robot

  • Author

    Dawei Dai ; Guolai Jiang ; Junbo Xin ; Xiang Gao ; Liangliang Cui ; Yongsheng Ou ; Guoqiang Fu

  • Author_Institution
    Guangdong Provincial Key Lab. of Robot. & Intell. Syst., Shenzhen Inst. of Adv. Technol., Shenzhen, China
  • fYear
    2013
  • fDate
    12-14 Dec. 2013
  • Firstpage
    1740
  • Lastpage
    1746
  • Abstract
    Crossing a door is usually a necessary action for autonomous patrol of an indoor surveillance robot. However, it is difficult for a large robot to cross narrow doors. In this paper, a method for detecting, locating and crossing a door for an indoor surveillance robot with Kinect is presented. Firstly, doors are detected and located based on RGB-D data collected form Kinect. Then, the conversion between robot coordinate system and world coordinate system is discussed. Thirdly, a nonlinear adaptive controller is designed for the robot to move perpendicularly through the door. Experiments were taken on a surveillance robot and the results illustrated the effectiveness of this method.
  • Keywords
    adaptive control; control system synthesis; image sensors; nonlinear control systems; object detection; path planning; robot vision; Kinect; RGB-D data; autonomous patrol; door crossing; door detection; door location; narrow doors; nonlinear adaptive controller design; red-green-blue-depth data; robot coordinate system; wide indoor surveillance robot; world coordinate system; Cameras; Image edge detection; Mobile robots; Robot kinematics; Robot sensing systems; Surveillance; indoor navigation; nonlinear control; robot;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Robotics and Biomimetics (ROBIO), 2013 IEEE International Conference on
  • Conference_Location
    Shenzhen
  • Type

    conf

  • DOI
    10.1109/ROBIO.2013.6739719
  • Filename
    6739719