A Graph-Based Nonparametric Drivable Road Region Segmentation Approach for Driverless Car Based on LIDAR Data
Road surface detection is one of the essential issues for autonomous driving.In this paper,we present a drivable road region detection approach using LIDAR data,where a nonparametric connected region representation is used to describe such a road region.First,an imaging of LIDAR data is conducted.Second,the gradient feature of road surface is extracted.Third,the flood fill algorithm is adopted to segment connected regions that express road surfaces.Using graph-based algorithm to deal with LIDAR data is a tentative way to segment drivable regions for driverless car.The experimental results show that the proposed approach is able to reliably detect drivable road regions under normal road conditions.
Drivable road region segmentation Laser point clouds Nonparametric Driverless car
Xiaolong Liu Zhidong Deng
State Key Laboratory of Intelligent Technology and Systems,Tsinghua National Laboratory for Information Science and Technology,Department of Computer Science,Tsinghua University,Beijing 100084,China
国际会议
The 2015 Chinese Intelligent Automation Conference(2015中国智能自动化会议)
福州
英文
431-439
2015-05-08（万方平台首次上网日期，不代表论文的发表时间）