EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Multimodal Sensor Fusion for Lane-Level Traffic Speed Estimation

Kang In Choi iD, Eun Hak Lee iD, Seo Jeong Kim, Chang Ho Yu iD, Youngsun Hong iD, Sunghwan Jeong

Sensors · 06.11.2026 · 26 (21) · 7024

DOI: 10.3390/s26217024 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 06.11.2026. Verfasst von Kang In Choi und 5 weiteren Autor:innen. Beteiligte Einrichtung: AI Application Research Center, Korea Electronics Technology Institute, Jeonju 54853, Republic of Korea. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study develops a deep multimodal learning framework for lane-level traffic-speed estimation using five sensing sources: vehicle detection systems (VDS), closed-circuit television (CCTV), light detection and ranging (LiDAR), remote traffic microwave sensors (RTMS), and automatic vehicle classification (AVC). The framework integrates sensor-specific features, road attributes, modality availability, and spatial relationships. Using traffic data from Siheung, Republic of Korea, the proposed model was evaluated against five single-modal baselines. The multimodal model achieved a weighted mean absolute percentage error (WMAPE) below 10% across evaluated time steps, outperforming single-modal models in stability and cross-modal transferability. Improvements were more pronounced in selected multi-lane environments and areas with limited sensing coverage. In route guidance, lane-level multimodal information reduced mean minimum travel time by 7.31% compared with link-level information. These findings demonstrate the potential of multimodal integration for accurate lane-level estimation, recovery of missing sensor observations, and improved route guidance.

Schlagwörter

Stichwörter der Autor:innen

deep multimodal learningheterogeneous sensorslane-level traffic speedmissing observation estimationroute guidancetraffic data fusiontraffic state estimation

Autor:innen und Institutionen

  • AI Application Research Center, Korea Electronics Technology Institute, Jeonju 54853, Republic of Korea.
  • School of Railway Operation Systems, Kyungil University, Gyeongsan 38428, Republic of Korea.
  • Division of Convergence Technology Engineering, Jeonbuk National University, Jeonju 54896, Republic of Korea.

ORCID-Profile: Kang In Choi, Eun Hak Lee, Chang Ho Yu, Youngsun Hong

Zitieren

APA

Choi, K. I., Lee, E. H., Kim, S. J., Yu, C. H., Hong, Y., & Jeong, S. (2026). Deep Multimodal Sensor Fusion for Lane-Level Traffic Speed Estimation. Sensors, 26(21), 7024. https://doi.org/10.3390/s26217024

Vancouver

Choi KI, Lee EH, Kim SJ, Yu CH, Hong Y, Jeong S. Deep Multimodal Sensor Fusion for Lane-Level Traffic Speed Estimation. Sensors. 2026;26(21):7024. doi:10.3390/s26217024

BibTeX
@article{Choi2026,
  author = {Choi, Kang In and Lee, Eun Hak and Kim, Seo Jeong and Yu, Chang Ho and Hong, Youngsun and Jeong, Sunghwan},
  title = {Deep Multimodal Sensor Fusion for Lane-Level Traffic Speed Estimation},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {7024},
  doi = {10.3390/s26217024}
}
Kostenlos starten →