METHOD FOR DETECTING DEFECTS IN THE 3D LIDAR SENSOR USING POINT CLOUD DATA

摘要

The present invention relates to a method for detecting defects in a 3D LIDAR sensor using point cloud data, and particularly, when a failure occurs in a 3D LiDAR sensor, relates to a technology for detecting a defect in a sensor through Point Cloud Data with another 3D LiDAR sensor. The present invention includes the steps of generating point cloud data corresponding to image coordinates of information on a subject acquired by a three-dimensional lidar sensor; Calculating a rotation (R) and translation (T) capable of minimizing an error in each direction by applying an Iterative Closest Point (ICP) algorithm to the point cloud data; And 3D LIDAR sensor using point cloud data comprising the step of calculating the 2-norm of the error from the rotation and translation for each direction to detect a LiDAR sensor in which an error has occurred among the three-dimensional LiDAR sensors. Provides a method for detecting defects. According to the present invention having the configuration as described above, by detecting an object based on the calibrated 3D LiDAR data, it is possible to quickly detect a defect in the 3D LiDAR sensor and prevent the occurrence of a dangerous situation due to abnormal object detection. There is an advantage of ensuring the safety of the vehicle even in a defective situation.