中国机械工程

• 机械基础工程 • 上一篇    下一篇

基于斜率差的激光雷达环境特征提取方法

刘朋;任工昌;杨力鹏;胡小龙   

  1. 陕西科技大学机电工程学院,西安,710021
  • 出版日期:2019-07-25 发布日期:2019-07-30
  • 基金资助:

    国家自然科学基金资助项目(51175314);
    陕西省科学技术研究发展计划资助项目(2014GY2-04);
    陕西省教育厅专项科研计划资助项目(15JK1079)

A Method for Feature Extraction of Environments of Laser Radars Based on Slope Difference

LIU Peng;REN Gongchang;YANG Lipeng;HU Xiaolong   

  1. College of Mechanical and Electrical Engineering, Shaanxi University of Science and Technology, Xi'an, 710021
  • Online:2019-07-25 Published:2019-07-30

摘要: 在利用激光雷达获取数据进行特征提取时,大多采用迭代计算的方法,且对阈值敏感,计算复杂度高,计算量较大。针对该问题,提出了一种通过计算相邻扫描点的斜率差进行直线段分割与特征点提取的算法。通过实验验证,该算法在直线段分割和特征点提取方面可以获得与手工分割相同的结果,而且对阈值不敏感。与实际测量结果的对比分析结果表明,该算法无论是提取特征线段的长度还是斜率均可获得较高的正确率。

关键词: 自主机器人, 激光雷达, 特征提取, 斜率差

Abstract: The most existing methods for feature extraction of the data obtained from laser radars were based on iterative calculation, which were sensitive to the thresholds and had high computational complexity.To overcome these drawbacks, a new algorithm was proposed to split the lines and extract the feature points by calculating the slope difference of adjacent scanning points.Experimental results show that the developed algorithm may obtain the same results as manual segmentation and is insensitive to the thresholds.Comparing with the results of measurement, the method may obtain higher accuracy in both of length and slope of the feature line segments.

Key words: autonomous robot, laser radar, feature extraction, slope difference

中图分类号: