上海交通大学学报

上海交通大学学报（自然版）

• 自动化技术、计算机技术 • 上一篇    下一篇

基于自然柱状特征地图的智能车定位

伍舜喜，杨明   

  1. (上海交通大学 电子信息与电气工程学院，上海 200240)
  • 收稿日期:2007-10-25 修回日期:1900-01-01 出版日期:2008-10-28 发布日期:2008-10-28
  • 通讯作者: 杨明

Natural Polemap Dased Localization for Intelligent Vehicles

WU Shun-xi, YANG Ming   

  1. (School of Electronic Information and Electrical Engineering,
    Shanghai Jiaotong University, Shanghai 200240, China)
  • Received:2007-10-25 Revised:1900-01-01 Online:2008-10-28 Published:2008-10-28
  • Contact: YANG Ming

摘要： 针对室外环境中手工测量柱状自然特征建立环境的特征地图(简称柱图)工作量大、精度低的问题，提出了一种基于激光雷达的地图自动生成方法.在此基础上，采用基于柱图的定位方法，使用迭代最近点算法进行地图匹配，提高了定位精度.仿真数据和实际数据的实验结果表明，该方法具有精度高、速度快等特点.

关键词: 车辆定位, 激光雷达, 迭代最近点

Abstract: In order to overcome the big workload and low accuracy of measuring the polemap manually in outdoor environment, a laser radar based mapping method was presented. On the basis of this, a polemap based localization was introduced which does mapmatching using iterative closest points and can enhance the accuracy of localization. This method was tested on both synthetic and real range data collected in natural environment. The experimental results demonstrate its high accuracy, low computation and wide applicability.

中图分类号: