A Real-Time Curb Detection Method for Vehicle by Using a 3D-LiDAR Sensor
- Delivery
- Provide download link
- Format
- Price
- Non-members (tax incl.)：￥6,600 Members (tax incl.)：￥5,280
- Paper/Info type
- SAE Paper
No.2021-01-0076
- Pages
- 1-10(Total 10 p)
- Date of publication
- Apr 2021
- Publisher
- SAE International
- Language
- English
- Event
- SAE WCX Digital Summit 2021
Detailed Information
| Author(E) | 1) Zhiqiang Du, 2) Jian Wu, 3) Rui He, 4) Guojun Wang, 5) Shuai Li, 6) Jinghua Zhang, 7) Guosheng Chen |
|---|---|
| Affiliation(E) | 1) Jilin University, 2) Jilin University, 3) Jilin University, 4) Jilin University, 5) Jilin University, 6) Jilin University, 7) Jilin University |
| Abstract(E) | Effectively detecting road boundaries in real time is critical to the applications of autonomous vehicles, such as vehicle localization, path planning and environmental understanding. To precisely extract the road boundaries from the 3D-LiDAR data, a dedicated algorithm consisting of four steps is proposed in this paper. The steps are as follows: Firstly, the 3D-LiDAR data is pre-processed by employing the RANSAC method, the ground points are quickly separated from the original 3D-LiDAR point cloud to reduce the disturbance from the obstacles on the road, this greatly decreases the size of the point cloud to be processed. Secondly, based on the principle of 3D-LiDAR scanning, the ground points are divided into scan layers. And the road boundary points of each scan layer are detected by using three spatial features based on sliding window. Thirdly, based on the improved beam model, the road type (straight road or curved road) where the vehicle is located is predicted, and then the edge points are subdivided into different areas by the road type. Finally, we use the distance filtering and RANSAC filtering to filter out false road boundary points caused by obstacles and obtain accurate road boundaries. Compared to other methods, we can effectively reduce the wrong classification of road edge points caused by obstacles by using the improved beam model. And the false points caused by obstacles are effectively reduced by using distance filtering and RANSAC filtering. The performance of the proposed method is verified through experiments with a vehicle driving on campus roads and extensive tests with the KITTI data set the experimental results demonstrate the accuracy and robustness of the proposed method. |
