主办：陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊：1976年
汽车实用技术

汽车实用技术 ›› 2022, Vol. 47 ›› Issue (20): 29-33.DOI: 10.16638/j.cnki.1671-7988.2022.020.007

• 智能网联汽车 • 上一篇    

基于激光雷达的轻量化定位与建图方法

龚柯阳，侯宝龙，史志飞，王孝宇   

  1. 长安大学 汽车学院
  • 出版日期:2022-10-30 发布日期:2022-10-30
  • 通讯作者: 龚柯阳
  • 作者简介:龚柯阳（1997—），男，硕士研究生，研究方向为无人驾驶智能车辆系统技术，E-mail:352191543@qq.com。

LiDAR-based Lightweight Localization and Mapping Method

GONG Keyang, HOU Baolong, SHI Zhifei, WANG Xiaoyu   

  1. School of Automobile, Chang'an University
  • Online:2022-10-30 Published:2022-10-30
  • Contact: GONG Keyang

PDF

摘要： 随着科技的进步，自动驾驶的发展如火如荼，作为其关键技术之一的定位与建图方法 目前主要依赖于全球定位系统（Global Positioning System, GPS），这类方法易受天气以及高层 建筑物的影响。考虑到现有方法的局限性，文章提出了基于激光雷达的轻量化定位与建图方 法，该方法主要由前端配准、回环检测、后端优化、建图等四个部分组成，通过以上几个部 分对采集到的周围环境的数据进行提取、匹配、识别、优化，得到兼具精度和鲁棒性的定位 与建图效果，为自动驾驶的感知、规划、决策、控制等建立基础。

关键词: 激光雷达；自动驾驶；轻量化定位；建图方法；场景识别

Abstract: With the advancement of science and technology, the development of autonomous driving is in full swing. As one of its key technologies, localization and mapping methods mainly rely on Global Positioning Ststem (GPS), which are easily affected by weather and high-rise buildings. Considering the limitations of existing methods, this paper proposes a lightweight localization and mapping method based on Lidar. The method is mainly composed of four parts: front-end registration, loop closure detection, back- end optimization and mapping. Each part extracts, matches, identifies, and optimizes the collected data of the surrounding environment to obtain both accurate and robust localization and mapping effects, and establishes the basis for perception, planning, decision-making, and control of autonomous driving.

Key words: Lidar; Autonomous driving; Lightweight localization; Mapping method; Scene recognition