EUVIMEDEUVIMED entdecken
Crossref

Visual-Inertial-GNSS Fusion Positioning for Vehicles With Deep-Learning-Based Feature Extraction and Outlier Detection

Shengjun Hu, Genyou Liu, Minghui Lyu, Run Wang, Wenhao Zhao, Bo Zhang

IEEE Internet of Things Journal · 01.08.2025 · 12 (15) · 29830-29843

DOI: 10.1109/jiot.2025.3569166 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 01.08.2025. Verfasst von Shengjun Hu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: National Key Research Program of China Collaborative Precision Positioning Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reliable and continuous high-precision positioning is a fundamental requirement for navigation, guidance, and control of intelligent mobile platforms, such as autonomous vehicles, drones, and robots. However, achieving such positioning in complex urban environments remains a significant challenge due to Global Navigation Satellite System (GNSS) signal obstructions and multipath effects. Moreover, traditional visual feature extraction algorithms struggle with variations in lighting, perspectives, and noise, limiting their adaptability. Consequently, relying solely on a single sensor often fails to provide a stable and accurate positioning solution. This article addresses the challenges encountered in typical urban environments and proposes a Visual-Inertial-GNSS fusion positioning framework, in which deep-learning-based feature extraction is applied to the visual front-end process. The deep-learning-based feature extraction approach fully exploits image information to derive more accurate and robust features, thereby enabling reliable feature matching. Additionally, we introduce a K-means clustering-based outlier detection method to remove dynamic features and mismatched features, thereby enhancing optical flow tracking and overall positioning accuracy. Experimental results in urban environments demonstrate that the proposed approach achieves high-precision positioning results under challenging GNSS conditions and complex visual environments, with improvements of (29.6%, 10.0%, and 7.3%) over SIFT and (47.8%, 7.8%, and 15.5%) over ORB in the North-East-Down (N-E-D) directions. Moreover, in dynamic road environments, the proposed outlier detection method achieves accuracy improvements of 19.6%, 3.0%, and 4.0% in the N-E-D directions compared to the RANSAC-based method.

Förderung

  • National Key Research Program of China Collaborative Precision Positioning Project

Zitieren

APA

Hu, S., Liu, G., Lyu, M., Wang, R., Zhao, W., & Zhang, B. (2025). Visual-Inertial-GNSS Fusion Positioning for Vehicles With Deep-Learning-Based Feature Extraction and Outlier Detection. IEEE Internet of Things Journal, 12(15), 29830-29843. https://doi.org/10.1109/jiot.2025.3569166

Vancouver

Hu S, Liu G, Lyu M, Wang R, Zhao W, Zhang B. Visual-Inertial-GNSS Fusion Positioning for Vehicles With Deep-Learning-Based Feature Extraction and Outlier Detection. IEEE Internet of Things Journal. 2025;12(15):29830-29843. doi:10.1109/jiot.2025.3569166

BibTeX
@article{Hu2025,
  author = {Hu, Shengjun and Liu, Genyou and Lyu, Minghui and Wang, Run and Zhao, Wenhao and Zhang, Bo},
  title = {Visual-Inertial-GNSS Fusion Positioning for Vehicles With Deep-Learning-Based Feature Extraction and Outlier Detection},
  journal = {IEEE Internet of Things Journal},
  year = {2025},
  volume = {12},
  number = {15},
  pages = {29830--29843},
  doi = {10.1109/jiot.2025.3569166}
}
Kostenlos starten →