A Camera-Based Real-Time Parking Positioning Method for Air-Ground Vehicles

Lingyun Wu, Weida Wang*, Chao Yang, Xing Yue, Changle Xiang, Ying Li

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Air-ground vehicles expand the road from ground to low altitude, which can effectively improve the traffic efficiency. However, there is still a lack of research on real-time accurate autonomous parking technology for its multi-modules docking stage. We propose a camera-based real-time recognition and localization framework. Firstly, the camera internal and external reference calibration is conducted to derive the real-world coordinate from image pixels. Then, the image feature extraction is adopted to detect representative features of ArUco markers. Finally, an optimization-based edge refinement method is proposed to achieve the accurate localization of the contours of ArUco markers. Our method achieves an average computational speed of 5.23 ms/frame and a maximum localization error of 3.47 mm for the detection in video sequences under various working conditions.

Original languageEnglish
Title of host publicationProceedings of 2022 International Conference on Autonomous Unmanned Systems, ICAUS 2022
EditorsWenxing Fu, Mancang Gu, Yifeng Niu
PublisherSpringer Science and Business Media Deutschland GmbH
Pages1062-1071
Number of pages10
ISBN (Print)9789819904785
DOIs
Publication statusPublished - 2023
EventInternational Conference on Autonomous Unmanned Systems, ICAUS 2022 - Xi'an, China
Duration: 23 Sept 202225 Sept 2022

Publication series

NameLecture Notes in Electrical Engineering
Volume1010 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference on Autonomous Unmanned Systems, ICAUS 2022
Country/TerritoryChina
CityXi'an
Period23/09/2225/09/22

Keywords

  • Air-ground vehicles
  • Edge refinement
  • Parking positioning

Fingerprint

Dive into the research topics of 'A Camera-Based Real-Time Parking Positioning Method for Air-Ground Vehicles'. Together they form a unique fingerprint.

Cite this