A Curb-Detection Network with a Tri-Plane BEV Encoder Module for Autonomous Delivery Vehicles

Author:

Zhang Lu1,Wang Jinzhu1,Zhu Xichan1,Ma Zhixiong1ORCID

Affiliation:

1. School of Automotive Studies, Tongji University, Shanghai 201804, China

Abstract

Curb detection tasks play a crucial role in the perception of the autonomous driving environment for logistics vehicles. With the popularity of multi-modal sensors under the BEV (Bird’s Eye View) paradigm, curb detection tasks are increasingly being integrated into multi-task perception networks, achieving robust detection results. This paper modifies and integrates the tri-plane spatial feature representation method of the EG3D network from the field of 3D reconstruction into a BEV-based multi-modal sensor detection network, including LiDAR, pinhole cameras, and fisheye cameras. The system collects a total of 24,350 frames of data under real road conditions for experimentation, proving the effectiveness of the proposed method.

Funder

National Key R and D Program of China

Publisher

MDPI AG

Reference28 articles.

1. Siegemund, J., Pfeiffer, D., Franke, U., and Farstner, W. (2010, January 21–24). Curb reconstruction using Conditional Random Fields. Proceedings of the IEEE Intelligent Vehicles Symposium (IV), La Jolla, CA, USA.

2. Road Curb Detection and Localization with Monocular Forward-View Vehicle Camera;Panev;IEEE Trans. Intell. Transp. Syst.,2018

3. Automated 3D Road Boundary Extraction and Vectorization Using MLS Point Clouds;Mi;IEEE Trans. Intell. Transp. Syst.,2021

4. LIDAR based detection of road boundaries using the density of accumulated point clouds and their gradients;Rato;Robot. Auton. Syst.,2021

5. Deac, S.E.C., Giosan, I., and Nedevschi, S. (2019, January 27–30). Curb detection in urban traffic scenarios using LiDARs point cloud and semantically segmented color images. Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC), Auckland, New Zealand.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3