基于3D激光雷达道路边缘实时检测算法的研究与实现  

RESEARCH AND IMPLEMENTATION OF REAL-TIME ROAD EDGE DETECTION ALGORITHM BASED ON 3D LIDAR

作　　者：李强[1] 杜煜[1] 

机构地区：[1]北京联合大学信息学院,北京100101

出　　处：《计算机应用与软件》2017年第10期219-222,共4页Computer Applications and Software

基　　金：国家自然科学基金重大研究计划项目(91420202)

摘　　要：针对自然环境下无人驾驶车辆的道路边缘检测问题,提出一种基于3D激光雷达的实时道路边缘检测算法。对激光雷达的点云数据进行网格化处理,求出每个网格中的高度差,并针对道路边缘的高度特征,对网格数据进行阈值处理;再由近及远逐个提取左右道路边缘,利用最小二乘法对左右道路边缘网格进行曲线拟合平滑处理,得到左右道路边缘。实验结果表明,该道路边缘检测算法可靠性高,稳定性强,能够准确完成道路边缘检测,满足实时系统的要求。Aiming at the road edge detection problem of unmanned vehicle in natural environment,this paper proposes a real-time road edge detection algorithm based on 3 D lidar. The point cloud data of lidar are meshed to get the height difference of each grid,and the threshold is processed according to the height characteristic of the road edge.Then one by one from the near to the far to extract the left and right road edge,and we use the least square method to curve fitting the left and right road edge mesh,and get the right and left edge of the road. The experimental results show that the algorithm has high reliability and stability,and can finish the road edge detection accurately. It can meet the requirements of real-time system.

关 键 词：无人驾驶 激光雷达 道路边缘 最小二乘法 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]