Loading...

Visual Simultaneous Localization and Mapping using an RGB-D Camera

Rashidi, Hossein | 2015

491 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47649 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Kasaei, Shohreh
  7. Abstract:
  8. Simultaneous localization and mapping (SLAM) is the action of detecting robot pose in an unknown environment and building the environment map by use of input data that captured from robot sensors. In visual SLAM, the input data for the robot, is limited to camera sensors. Nowadays, SLAM is one of the main challenges in robotic research.For autonomous action, we need robot pose in the map of the environment. The map production in the indoor environment, there is no GPS data, is one of the research issue in robotic community, in last decade. In this thesis, a new and efficient method is proposed for SLAM at the level of objects. The maps produced by state of the art methods don’t have a semantic concept and just build of low-level primitives. In contract of those methods, in the proposed approach is tried to build a map via localization of objects in the scene. At the first, objects in the captured frames are detected by the segmentation process and after achieving labeled segments, the pose of camera and the appropriate map updated by finding correspondences and alignment of them with previous detected objects in the database. The built map is consist of objects’ point cloud presented in the environment which is built gradually from information captured through the time. The produced map with proposed method is semantic map with respect to existing methods and it can be used in decision and scene understanding problems with robots. Another advantage of proposed method is handling the moving objects in the scene by classification of object to moving and static objects. The achieved results in this thesis indicates the efficacy of the proposed method in objective sense
  9. Keywords:
  10. Location ; Map Generation ; RGB-D Camera ; Depth Images ; Three Dimensional Reconstruction ; Uncertainty ; Visual Mapping

 Digital Object List

 Bookmark

No TOC