基于LabVIEW和单线激光雷达的障碍物识别算法研究

Research on Barrier Recognition Algorithms Based on LabVIEW and Single Line Lidar

为实现对动态障碍物的检测和跟踪,提出一种基于LabVIEW和单线激光雷达自主动态障碍物的检测方法。通过改进的DBSCAN自适应动态阈值算法对雷达点云进行聚类,使用最近邻算法的思想进行连续两帧雷达点云的数据关联,采用卡尔曼滤波对障碍物轨迹进行跟踪。实验结果表明:在低速驾驶的条件下,该方法能够有效识别车辆前方30 m的移动障碍物并检测其位置和速度。

In order to detect and track dynamic obstacles,a method based on LabVIEW and single line lidar is proposed.The radar point cloud is clustered through the improved DBSCAN adaptive dynamic threshold algorithm,the idea of nearest neighbor algorithm is used to correlate the data of two consecutive frames of radar point cloud,and Kalman filter is used to track the obstacle trajectory.The experimental results show that the method can effectively identify the moving obstacles thirty meters in front of the vehicle,and detect their position and speed under the condition of low-speed driving.