Clustering-Based Lateral Longitudinal Target Recognition of In-Vehicle LIDAR Data
Chao Deng, Wu, Nengchao Lyu, Ze Li · 2016
In order to improve the accuracy of lateral and vertical target recognition in a car-following situation, the box plot is used to analyze and filter discrete points to overcome the LIDAR point data error; application of the modified adaptive K-means clustering algorithm which is based on the clustering evaluation index is applied to process the LIDAR point from LUX4; the candidate targets are output by clustering results. The test results show that the obstacle detection algorithm is more robust and reliable in the car-following situation.