东南大学学报（自然科学版）2013，Vol.43Issue(z1)：183-187,5.DOI:10.3969/j.issn.1001-0505.2013.S1.038
基于Kinect传感器的移动机器人室内环境三维地图创建
Kinect-based 3D indoor environment map building for mobile robot
摘要
Abstract
For the 3D perception problem of mobile service robots in unknown indoor environments,a practical approach to building 3D maps using a low-cost Kinect sensor is proposed.Successive frames of RGB-D (red-green-blue-depth) information are captured during the robot's movements.First,SURF(speeded up robust features) detector is applied to color images for extracting and matching stable feature points.Combined with depth images,initial registration procedure is then performed by using the RANSAC (random sample consensus) algorithm for eliminating possible mistake matching points and thus the relative transformation between frames can be estimated.Finally,a generalized ICP(iterative closest point) algorithm is employed to perform fine registration on captured depth images,which finally produces 3D point cloud mosaic.In addition,an application system of mobile robot 3D mapping is developed.Experimental results validate the practicability and effectiveness of the approach.关键词
Kinect传感器/SURF/地图创建/广义ICP/移动机器人Key words
Kinect sensor/ SURF(speeded up robust features) / map building/ generalized ICP(iterative closest point) / mobile robot分类
信息技术与安全科学引用本文复制引用
杨鸿,钱堃,戴先中,马旭东,房芳..基于Kinect传感器的移动机器人室内环境三维地图创建[J].东南大学学报（自然科学版）,2013,43(z1):183-187,5.基金项目
国家自然科学基金资助项目(61075090,61005092,61105094,61175091)、东南大学复杂工程系统测量与控制教育部重点实验室开放课题基金资助项目(MCCSE2012B02). （61075090,61005092,61105094,61175091）
