摘要
车道线检测是无人驾驶环境感知模块的重要组成部分。纯追踪算法是一种基于几何追踪的方法,广泛应用于无人驾驶车辆的研究。为实现对无人小车循迹过程中的精准控制,本文提出了一种基于纯追踪算法和车道线检测的预瞄点检测方法。算法对车道线原始图像进行二值化、逆透视变换以及ROI区域选取等获得车道线逆透视特征图,通过滑动窗口方法提取车道线特征点并用二次函数拟合车道线,最后利用纯追踪算法得到预瞄点。实验结果表明,该算法能够在准确的识别出车道线的基础之上,实时地得到预瞄点的位置,可以达到无人小车实时循迹的要求。
Lane line detection is an important part of the driverless environment perception module.Pure tracking algorithm is a method based on geometric tracking,which is widely used in the research of unmanned vehicles.In order to achieve precise control of the unmanned car’s tracking process,a look-ahead point detection method based on pure pursuit algorithm and lane line recognition is proposed.The algorithm performs binarization,inverse perspective transformation,and ROI region selection on the original image of the lane line to obtain the lane line inverse perspective feature map.The lane line feature points are extracted through the sliding window method and the lane line is fitted with a quadratic function.Finally,a pure pursuit algorithm is used to get look-ahead points.The experimental results show that the algorithm can obtain the position of the look-ahead point in real time on the basis of accurately identifying the lane line,which can satisfy the requirements of real-time tracking of unmanned vehicles.
作者
苏诗
陈壮豪
尚小庆
郭风朋
Su Shi;Chen Zhuanghao;Shang Xiaoqing;Guo Fengpeng(School of Mechanical and Power Engineering,Shanghai Jiaotong University,Shanghai 200240)
出处
《传动技术》
2020年第3期3-6,共4页
Drive System Technique
关键词
纯追踪算法
车道线检测
逆透视变换
pure-pursuit algorithm
lane line recognition
inverse perspective transformation
