TY - JOUR
T1 - 3D Point Clouds Data Super Resolution Aided LiDAR Odometry for Vehicular Positioning in Urban Canyons
AU - Yue, Jiang
AU - Wen, Weisong
AU - Han, Jin
AU - Hsu, Li Ta
N1 - Funding Information:
Manuscript received May 9, 2020; revised August 27, 2020 and February 1, 2021; accepted March 14, 2021. Date of publication March 29, 2021; date of current version June 9, 2021. This work was supported in part by the Hong Kong PolyU internal Grant on the project ZVKZ, “Navigation for Autonomous Driving Vehicle using Sensor Integration” and in part by the National Natural Science Foundation of China under Grant 61601225. The review of this article was coordinated by Prof. Z. Ma. (Corresponding author: Weisong Wen.) Jiang Yue is with the Nanjing University of Science and Technology, Nanjing, Jiangsu 210014, China, and also with the Intelligent Positioning and Navigation Lab, The Hong Kong Polytechnic University 999077, Hong Kong (e-mail: [email protected]).
Publisher Copyright:
© 1967-2012 IEEE.
PY - 2021/5
Y1 - 2021/5
N2 - LiDAR odometry algorithms are extensively studied for vehicular positioning. However, achieving high-precision positioning using low-cost 16-channel LiDAR in urban canyons remains a challenging problem due to the limited point cloud density from low-cost LiDAR and excessive dynamic surrounding objects. To fill this gap, this paper proposes enriching sparse 3D point clouds to denser clouds via a novel deep learning-based superresolution (SR) algorithm before its utilization in 3D LiDAR odometry. We validate the effectiveness of the proposed method using the KITTI dataset and a challenging dataset collected in an urban canyon (with complex environmental structures and dynamic objects) of Hong Kong. We conclude that significantly denser point clouds are achieved with considerable accuracy. In addition, significantly improved performance of 3D LiDAR odometry is obtained in the evaluated dataset collected in an urban canyon of Hong Kong.
AB - LiDAR odometry algorithms are extensively studied for vehicular positioning. However, achieving high-precision positioning using low-cost 16-channel LiDAR in urban canyons remains a challenging problem due to the limited point cloud density from low-cost LiDAR and excessive dynamic surrounding objects. To fill this gap, this paper proposes enriching sparse 3D point clouds to denser clouds via a novel deep learning-based superresolution (SR) algorithm before its utilization in 3D LiDAR odometry. We validate the effectiveness of the proposed method using the KITTI dataset and a challenging dataset collected in an urban canyon (with complex environmental structures and dynamic objects) of Hong Kong. We conclude that significantly denser point clouds are achieved with considerable accuracy. In addition, significantly improved performance of 3D LiDAR odometry is obtained in the evaluated dataset collected in an urban canyon of Hong Kong.
KW - LiDAR
KW - LiDAR odometry
KW - NDT
KW - convolutional neural network
KW - sparse point clouds
KW - superresolution
KW - vehicular positioning
UR - http://www.scopus.com/inward/record.url?scp=85103797780&partnerID=8YFLogxK
U2 - 10.1109/TVT.2021.3069212
DO - 10.1109/TVT.2021.3069212
M3 - Journal article
AN - SCOPUS:85103797780
SN - 0018-9545
VL - 70
SP - 4098
EP - 4112
JO - IEEE Transactions on Vehicular Technology
JF - IEEE Transactions on Vehicular Technology
IS - 5
M1 - 9388922
ER -