Authors:
Kohei Tokorodani
1
;
Masafumi Hashimoto
2
;
Yusuke Aihara
1
and
Kazuhiko Takahashi
2
Affiliations:
1
Graduate School of Doshisha University, Kyotanabe, Kyoto 6100321 and Japan
;
2
Faculty of Science and Engineering, Doshisha University, Kyotanabe, Kyoto 6100321 and Japan
Keyword(s):
Two-wheeled Vehicle, Lidar, Point-cloud Mapping, NDT Scan Matching, Distortion Correction, Extended Kalman Filter, Interpolation.
Related
Ontology
Subjects/Areas/Topics:
Image Processing
;
Informatics in Control, Automation and Robotics
;
Intelligent Transportation Technologies and Systems
;
Robotics and Automation
;
Sensors Fusion
;
Signal Processing, Sensors, Systems Modeling and Control
;
Vision, Recognition and Reconstruction
Abstract:
This paper presents a method for generating a 3D point-cloud map using multilayer lidar mounted on two-wheeled vehicle. The vehicle identifies its own 3D pose (position and attitude angle) in a lidar-scan period using the normal distributions transform (NDT) scan-matching method. The vehicle’s pose is updated in a period shorter than the lidar-scan period using its attitude angle and angular velocity measured by an inertial measurement unit (IMU). The pose estimation is based on the extended Kalman filter (EKF) under the assumption that the vehicle moves at nearly constant translational and angular velocities. The vehicle’s pose is further estimated in a period shorter than measurement period of the IMU using a linear interpolation method. The estimated poses of the vehicle are applied to distortion correction of lidar-scan data, and a point-cloud map is generated based on the corrected lidar-scan data. Experimental results of mapping a road environment using a 32-layer lidar mounted
on a bicycle show the efficancy of the proposed method in comparison with conventional methods of distortion correction of lidar-scan data.
(More)