基于双目视觉的移动机器人地图构建研究-机械制造及其自动化专业论文.docxVIP

III万方数据 III 万方数据 Abstract Title: THE RESEARCH ON BINOCULAR VISIONBASED MAP- BUILDING FOR MOBILE ROBOT? Major：Mechanical Manufacturing and Automation Name：Wanping LI Signature: Supervisor：Prof.Weiping FU Signature: Abstract With the rapid development of science and technology and increasing human tasks need to be solved, human beings alone has been difficult to meet such requirements. As one of the most human thinking mode , robot has been the highest ideal tool of mankinds pursuit , therefore, the research of intelligent robot is expanding and deepening. The research of intelligent mobile robot in unknown environment, first to solve is mobile robots perception of their surroundings, namely the robot map building problem. Relying on the robot sensor to carry and for the environment information, building up accurate map of environment, the robots location and subsequent path planning has important meaning. This article mainly aims at the indoor unknown environment, mobile robot with binocular vision sensor gets environment information, the SIFT natural landmarks and through the extended kalman filtering to finish the robot in unknown environment map building and the orientation of the incremental, contour extraction combined with obstacles to make a more accurate description of the obstacle .The main work is as follows: First of all, to build a mobile robot based on binocular vision map to the overall structure of the implementation framework, and from the research of binocular stereo vision system, With the SIFT feature point extracting and binocular images matching, then matching error rejecting, matching point in the world coordinates of 3D content such as information retrieval, obtaining the natural reliable landmarks in the external environment of the robot as the basic elements of building map. Second, the extended kalman filtering fusion mobile robot positioning with maps to create SLAM algorithm at the same time. The characteristics of external envir