| 초록 |
LiDAR sensors attached to the vehicle are vulnerable to contamination due to external factors. If a blind spot occurs due to contamination of the LiDAR surface, the surrounding environment and objects are not detected, causing problems in the safety of passengers, pedestrians, and surrounding vehicles. Therefore, this paper explains how to determine LiDAR surface contamination in real time. The LiDAR data type was used in the form of a spherical coordinate system to efficiently detect missing data parts. In addition, data omission factors were classified into pollution factors, environmental factors, and structural factors. And it was removed using an algorithm suitable for the characteristics of each data omission. In this paper, a bitmap containing data omission information was used to guarantee real-time determination of contamination in the range of 1 to 10. To verify the performance of the algorithm, experimental results are analyzed obtained from the LiDAR sensor artificially contaminatied on the surface while driving on the road. |
