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 ›› 2022, Vol. 48 ›› Issue (6): 58-63.DOI: 10.16638/j.cnki.1671-7988.2023.06.012

• 智能网联汽车 • 上一篇    

车载激光雷达的应用算法

杜海林,张杭铖,王齐超   

  1. 长安大学 汽车学院
  • 出版日期:2023-03-30 发布日期:2023-03-30
  • 通讯作者: 杜海林
  • 作者简介:杜海林(1998—),男,硕士研究生,研究方向为自动驾驶、激光雷达、汽车电子等,E-mail: duhailin1998 @163.com。

Application Algorithm of Vehicle LIDAR

DU Hailin, ZHANG Hangcheng, WANG Qichao   

  1. School of Automobile, Chang'an University
  • Online:2023-03-30 Published:2023-03-30
  • Contact: DU Hailin

摘要: 伴随着无人驾驶领域的迅速发展,激光雷达(LIDRA)在这一行业上的应用越来越广 泛。激光雷达的快速发展,为智能网联汽车在目标跟踪与识别等方面提供了另一种方式。文 章以车载激光雷达在自动驾驶行业上的应用为切入点,介绍了车载激光雷达的三种应用算法。 目标跟踪与识别算法,通过目标跟踪算法对障碍物运动状态做出估计和预测,实时评估障碍 物和无人驾驶车辆的安全等级,作出相应的决策。即时定位与地图构建(SLAM)相关算法, 应用于解决机器人在未知环境中定位自身位置和姿态的一种高级算法。点云分割往往是物体 识别、地图构建的基础,通过对六种常用分割算法的描述,分析了算法各自的特点,为不同 应用场景算法的选择提供了一定参考。

关键词: 激光雷达;SLAM 算法;点云分割;地面分割

Abstract: With the rapid development of automatic driving, the application of light detection and ranging (LIDAR) in this field is more and more extensive. The fastest-growing of lidar provides another way for the intelligent connected vehicle in target tracking and recognition. Based on the application of vehicle lidar in the automatic driving industry, this paper introduces three application algorithms of vehicle lidar. Target tracking and recognition algorithm is used to estimate and predict the movement state of obstacles through target tracking algorithm, evaluate the safety level of obstacles and driverless vehicles in real time, and make corresponding decisions. Simultaneous localization and mapping (SLAM) algorithm is an advanced algorithm for robot to locate its position and pose in unknown environment. Point cloud segmentation is often the basis of object recognition and map construction. Through the description of six commonly used segmentation algorithms, the characteristics of the algorithms are analyzed, which provides a reference for the selection of algorithms in different application scenarios.

Key words: LIDAR; SLAM algorithm; Point cloud segmentation; Ground segmentation