Research about LiDAR pointcloud processing method and its application for autonomous driving = 자율 주행을 위한 라이다 포인트 클라우드 처리 기법과 응용에 관한 연구

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 40
  • Download : 0
LiDAR sensors are used as core sensors in autonomous vehicles. The range of use varies from positioning to recognition. In this paper, we introduce a method that can utilize the LiDAR point cloud for recognition and positioning. In the positioning section, after detecting the lane with a LiDAR, the positioning information is corrected using this information. First of all, lane detection using LiDAR has the advantage of being able to instantly determine the three-dimensional position of the lane, and the detection performance is not affected by the illuminance. Lane detection using LiDAR detects all road markings on the ground and preprocesses them to provide them as input to a multi-layer perceptron-based network to instantly obtain the three-dimensional position of points corresponding to lanes. In this way, IMU is corrected by fusing lane information detected in real time and lane information obtained using the yaw rate from the vehicle's IMU. Finally, we propose a Kalman Filter-based sensor fusion complex positioning system that corrects the absolute position estimated by the GPS absolute position and the corrected yaw rate in conjunction with high definition map.
Advisors
Kong, Seung-Hyunresearcher공승현researcher
Description
한국과학기술원 :로봇공학학제전공,
Publisher
한국과학기술원
Issue Date
2020
Identifier
325007
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 로봇공학학제전공, 2020.8,[iv, 48 p. :]

Keywords

LiDAR▼aAutonomous Vehicle▼aLane Detection▼aMulti Layer Perceptron▼aLocalization▼aSensor Fusion; 라이다▼a자율주행▼a차선 인식▼a측위▼a센서 융합

URI
http://hdl.handle.net/10203/284961
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=925120&flag=dissertation
Appears in Collection
RE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0