TY - GEN
T1 - Semantic and Moving Object Segmentation-assisted LiDAR Odometry and Mapping
AU - Wang, Fei
AU - Sun, Chao
AU - Zhong, Guoqi
AU - Liang, Weiqiang
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - For autonomous vehicles, Simultaneous localization and mapping (SLAM) is one of the fundamental capabilities. Accurate and reliable SLAM are important for autonomous vehicles. In this work, we propose a novel LiDAR odometry and mapping method assisted by semantic segmentation and moving object segmentation. First, to acquire semantic information of point clouds and distinguish moving objects, a framework for segmenting LiDAR point clouds is proposed. Then an effective method for integrating semantic information and moving object information into feature-based LiDAR SLAM is proposed. With the assistance of semantic information and moving object information, moving points are filtered out, and semantic constrains are added in feature extraction and pose estimation to improve the localization accuracy. The experiment results on public datasets show that, compared to the baseline, the average relative pose estimation error of our proposed method is reduced by21.4% in rotation and 29.4% in translation.
AB - For autonomous vehicles, Simultaneous localization and mapping (SLAM) is one of the fundamental capabilities. Accurate and reliable SLAM are important for autonomous vehicles. In this work, we propose a novel LiDAR odometry and mapping method assisted by semantic segmentation and moving object segmentation. First, to acquire semantic information of point clouds and distinguish moving objects, a framework for segmenting LiDAR point clouds is proposed. Then an effective method for integrating semantic information and moving object information into feature-based LiDAR SLAM is proposed. With the assistance of semantic information and moving object information, moving points are filtered out, and semantic constrains are added in feature extraction and pose estimation to improve the localization accuracy. The experiment results on public datasets show that, compared to the baseline, the average relative pose estimation error of our proposed method is reduced by21.4% in rotation and 29.4% in translation.
KW - LiDAR SLAM
KW - autonomous vehicles
KW - convolutional neural networks
KW - moving object segmentation
KW - semantic segmentation
UR - http://www.scopus.com/inward/record.url?scp=85164261060&partnerID=8YFLogxK
U2 - 10.1109/EECR56827.2023.10150083
DO - 10.1109/EECR56827.2023.10150083
M3 - Conference contribution
AN - SCOPUS:85164261060
T3 - 2023 9th International Conference on Electrical Engineering, Control and Robotics, EECR 2023
SP - 267
EP - 273
BT - 2023 9th International Conference on Electrical Engineering, Control and Robotics, EECR 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 9th International Conference on Electrical Engineering, Control and Robotics, EECR 2023
Y2 - 24 February 2023 through 26 February 2023
ER -
