【摘要】 针对激光雷达采集行驶车辆的三维点云数据中包含过多畸变数据，影响车辆定位效果的问题，本文研究一种基于激光雷达和特征地图的车辆智能定位方法。激光雷达利用基于飞行时间的激光测距法，采集车辆及其行驶环境的三维激光点云数据，去除激光点云数据中的畸变数据。利用正态分布变换方法，优化删除畸变数据的点云集的正态分布概率值，配准三维激光点云数据。从完成配准后的三维激光点云数据中，提取柱状物体的圆形特征，构建车辆行驶的自然柱状特征地图。利用卡尔曼滤波算法，结合自然柱状特征地图信息，实现高精度的车辆智能定位。实验结果证明：该方法可以精准定位车辆目标，车辆智能定位精度较高，最高可达到97%,定位效率较好，最短可在5 s时间内完成定位，具有一定应用价值。更多还原

【Abstract】 Aiming at the problem that the 3D point cloud data collected by laser radar contains too much distorted data, which affects the vehicle location effect, this paper studies an intelligent vehicle location method based on laser radar and feature map. The laser radar uses time-of-flight laser ranging method to collect the three-dimensional laser point cloud data of vehicles and their driving environment and remove the distortion data in the laser point cloud data. Using the normal distribution transformation method, the normal distribution probability value of the point cluster of distorted data is optimized to register the 3D laser point cloud data. From the 3D laser point cloud data after registration, the circular features of cylindrical objects are extracted, and the natural cylindrical feature map of vehicle driving is constructed. Using Kalman filter algorithm and natural columnar feature map information, high precision vehicle intelligent location is realized. The experimental results show that this method can accurately locate the vehicle target, and the vehicle intelligent positioning accuracy is high, up to 97%, the positioning efficiency is good, and the shortest time to complete the positioning is 5 s, which has certain application value.更多还原