一种基于多个多线激光雷达分区障碍物检测方法

• Format: Patent
• Language: Chinese
• Published: 02.04.2024
• Subjects: ANALOGOUS ARRANGEMENTS USING OTHER WAVES; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION ORRERADIATION OF RADIO WAVES; MEASURING; PHYSICS; RADIO DIRECTION-FINDING; RADIO NAVIGATION; TESTING

本发明公开一种基于多个多线激光雷达分区障碍物检测方法，流程如下：接收三路激光雷达数据缓存和拼接，过滤无效点和畸变的点；分区体素下采样，根据不同的检测区域设置不同的像素；下采样完后进行数据集合并；基于渐进式形态学滤波算法进行地面拟合，检测出地面点；进行地面分割，分割出非地面点和地面点；基于欧几里得算法对非地面点云集合进行聚类，根据扫描距离的远近，设置不同的k-维树近邻搜索的搜索半径，进行障碍物检测；输出障碍物信息给融合层或者动态障碍物跟踪模块，包含障碍物的长度、宽度、高度和中心点信息。本发明实现目标物的高精度识别，保障每一帧障碍物检出时延小于40ms，检出率为98％。 The invention discloses a partition obstacle detection method based on multiple multi-line laser radars. The method comprises the following steps: receiving three paths of laser radar data for cachingand splicing, and filtering invalid points and distorted points; carrying out subregion voxel downsampling, and setting different pixels according to different detection areas; performing data set merging after down-sampling; performing ground fitting based on a progressive morphological filtering algorithm, and detecting ground points; carrying out ground segmentation to obtain non-ground pointsand ground points; clustering the non-ground point cloud set based on an Euclidean algorithm, setting different search radiuses of k-dimen