Vehicle Positioning Technology Based on Stereo Vision

2025-01-7163

02/21/2025

Features
Event
2024 International Conference on Smart Transportation Interdisciplinary Studies
Authors Abstract
Content
Vehicle localization in enclosed environments, such as indoor parking lots, tunnels, and confined areas, presents significant challenges and has garnered considerable research interest. This paper proposes a localization technique based on an onboard binocular camera system, utilizing binocular ranging and spatial intersection algorithms to achieve active localization. The method involves pre-deploying reference points with known coordinates within the experimental space, using binocular ranging to measure the distance between the camera and the reference points, and applying the spatial intersection algorithm to calculate the camera’s center coordinates, thereby completing the localization process. Experimental results demonstrate that the proposed algorithm achieves sub-meter level localization accuracy. Localization accuracy is significantly influenced by the calibration precision of the binocular camera and the number of reference points. Higher calibration precision and a greater number of reference points contribute to improved localization accuracy. Through experiments, the feasibility and effectiveness of the proposed algorithm are validated, and key factors affecting localization accuracy are analyzed. This technology is an active localization technique, particularly suitable for indoor environments where GNSS signals are obstructed. Its advantages include low deployment costs and ease of implementation, providing an effective solution for indoor localization. This approach holds significant theoretical and practical value for addressing object localization challenges in obstructed environments.
Meta TagsDetails
DOI
https://doi.org/10.4271/2025-01-7163
Pages
8
Citation
Feifei, L., Haoping, Q., and Yi, W., "Vehicle Positioning Technology Based on Stereo Vision," SAE Technical Paper 2025-01-7163, 2025, https://doi.org/10.4271/2025-01-7163.
Additional Details
Publisher
Published
Feb 21
Product Code
2025-01-7163
Content Type
Technical Paper
Language
English