测绘通报 ›› 2017, Vol. 0 ›› Issue (3): 52-57,90.doi: 10.13474/j.cnki.11-2246.2017.0083

• 学术研究 • 上一篇    下一篇

一种基于里程参数的道路平面几何解析模型

陶璐1, 朱敦尧1,2,3, 王军德2,3, 张攀3   

  1. 1. 武汉大学卫星导航定位技术研究中心, 湖北 武汉 430079;
    2. 武汉理工大学智能交通系统研究中心, 湖北 武汉 430063;
    3. 武汉光庭信息技术股份有限公司, 湖北 武汉 430073
  • 收稿日期:2016-06-23 修回日期:2017-01-15 出版日期:2017-03-25 发布日期:2017-03-31
  • 作者简介:陶璐(1991-),男,硕士生,主要从事高精度导航地图研究。E-mail:ltao@whu.edu.cn
  • 基金资助:
    测绘地理信息公益性行业科研专项(201512007)

A Road Plane Geometry Analytical Model Based on Mileage Parameter

TAO Lu1, ZHU Dunyao1,2,3, WANG Junde2,3, ZHANG Pan3   

  1. 1. GNSS Research Center, Wuhan University, Wuhan 430079, China;
    2. Intelligent Transportation Systems Center, Wuhan University of Technology, Wuhan 430063, China;
    3. Wuhan KOTEI Informatics Co. Ltd., Wuhan 430073, China
  • Received:2016-06-23 Revised:2017-01-15 Online:2017-03-25 Published:2017-03-31

摘要: 随着主动安全技术、自动驾驶技术的发展,高精度导航地图得到越来越多的运用。本文提出了一种面向高精度导航地图的、基于里程参数的道路平面几何解析模型。通过动态地构造点集上的正交多项式系和约束最小二乘法,根据拟合最大误差动态地计算道路的几何描述参数。使用该模型进行了120 km高速道路高精度地图数据生产,以三段典型道路为实例进行试验,表明该方法能对道路几何进行准确的描述,保证道路几何的连续与光滑,同时最大限度地减少数据冗余。该模型对高精度地图道路几何描述具有一定优势。

关键词: 道路几何, 解析模型, 高精度地图, 正交多项式, 约束最小二乘

Abstract: With the development of active safety technology and autonomous driving technology, high precision map has been used in many applications. A road geometry analytical model based on mileage parameter was proposed for high precision map. By constructing the orthogonal polynomials based on the data set and using restricted least-square method, the geometric description parameters were dynamically calculated based on maximum fitting error. The model can precisely describe the road geometry, and minimize the data redundancy. It can also ensure the continuity and smoothness of the road geometry. The model was used for high precision map production of 120 km highway. Taking three typical roads as examples, the advantages of proposed model were validated.

Key words: road geometry, analytical model, high precision map, orthogonal polynomials, restricted least-square method

中图分类号: