考虑反射强度的全自动运行列车障碍物检测算法研究

Obstacle Detection Algorithm of Fully Automatic Train Considering Reflection Intensity

摘要:

针对轨道交通全自动运行列车轨行区障碍物检测问题，提出了一种基于激光雷达的且考虑激光反射强度的障碍物检测算法。该算法使用欧式聚类法对点云进行聚类，并结合了自适应的反射强度阈值处理以及体素滤波器、聚类半径差异化等方法，提升算法的障碍物检测速度与精度。实验表明该算法具有良好的检测性能。

Abstract:

Aiming at the problem of obstacle detection in front of rail transit train， this paper proposes an obstacle detection algorithm based on LiDAR and considering laser reflection intensity for fully automatic train. This algorithm uses the Euclidean clustering method to cluster the point clouds， and combines the adaptive threshold processing of reflection intensity， VGF， clustering radius differentiation and other methods to improve the speed and accuracy of obstacle detection. The experiments show that the algorithm has a good detection performance.