摘要
针对移动机器人在未知环境下自主导航时,仅利用单一传感器出现的不能及时定位、构建的地图不精确、鲁棒性不强等问题,提出一种双目视觉与激光雷达信息融合的组合导航方法,建立包含世界坐标系的路标数据库,并获得全局环境下的特征地图进而实现自主导航;通过在安装有开源机器人操作系统(ROS)的履带式移动机器人上进行实验。实验结果表明:通过上述方法进行多传感器融合,能够构建可靠性更高、更精确的2D栅格地图,提高了移动机器人同步定位与地图构建(SLAM)的鲁棒性。
Aiming at problem of inaccuracy and poor robustness of built maps and positioning is not timely when autonomous navigation of mobile robot is carried out by only using single sensor in unknow environment,an integrated navigation method is proposed based on information fusion combined binocular vision and laser radar,a landmark database containing the world coordinate system is built and the feature map in global environment is obtained so as to achieve autonomous navigation. Experiments are conducted by crawle-type mobile robot with open source robot operating system（ ROS）. Experiment results show that 2 D grid map with higher reliability and precision can be built by multi-sensor fusion and the robustness simultaneous localization and mapping（ SLAM） of mobile robot is improved.
作者
贺利乐
王消为
赵涛
HE Li-le;WANG Xiao-wei;ZHAO Tao(School of Mechanical and Electrical Engineering,Xi＇an University of Architecture ＆ Technology,Xi＇an 710055,China)
出处
《传感器与微系统》
CSCD
2018年第10期50-53,共4页
Transducer and Microsystem Technologies
基金
陕西省自然科学基础研究面上资助项目(2016JM5067)
西安市高校院所人才服务企业工程项目(CXYJZKD007)
关键词
移动机器人
自主导航
同步定位与地图构建
开源机器人操作系统
mobile robot
autonomous navigation
simultaneous localization and mapping （SLAM）
open source robot operating system（ROS）
