Efficient Context-Aware Traffic Trajectory Monitoring From Roadside LiDAR for Occlusion Handling
IEEE Transactions on Intelligent Transportation Systems · 2026-06 · 27 (6) · 6509-6518
DOI: 10.1109/tits.2026.3672508 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Intelligent Transportation Systems, 2026-06. Verfasst von Zhihui Chen und 4 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben.
Abstract (Originalsprache)
Frequent occlusions in complex urban environments pose a major challenge for roadside LiDAR-based traffic monitoring, often leading to fragmented trajectories and kinematic errors. To address this, we propose a context-aware framework that shifts the sensing paradigm from individual object tracking to lane-level platoon modeling. By encoding raw LiDAR data into Lane Occupancy State, we employ an encoder–decoder LSTM architecture to capture collective traffic dynamics, enabling robust trajectory reconstruction even during prolonged occlusions. To ensure real-time scalability, a hardware-aware embedding scheme is designed to directly parse data packets from mechanical LiDAR, achieving an inference efficiency of 18.75 ms/frame. Evaluations on real-world datasets demonstrate a 95.3% occlusion reconstruction rate with low kinematic errors (0.0055 m/s speed deviation), while an ablation study confirms that integrating an attention mechanism further boosts the reconstruction rate to 97.9%. This work establishes a robust, traffic-context-centric foundation for high-fidelity digital twins and intelligent transportation systems.
Zitieren
APA
Chen, Z., Xu, H., Wang, Z., Guan, F., & Tian, M. (2026). Efficient Context-Aware Traffic Trajectory Monitoring From Roadside LiDAR for Occlusion Handling. IEEE Transactions on Intelligent Transportation Systems, 27(6), 6509-6518. https://doi.org/10.1109/tits.2026.3672508
Vancouver
Chen Z, Xu H, Wang Z, Guan F, Tian M. Efficient Context-Aware Traffic Trajectory Monitoring From Roadside LiDAR for Occlusion Handling. IEEE Transactions on Intelligent Transportation Systems. 2026;27(6):6509-6518. doi:10.1109/tits.2026.3672508
BibTeX
@article{Chen2026,
author = {Chen, Zhihui and Xu, Hao and Wang, Ziru and Guan, Fei and Tian, Muchen},
title = {Efficient Context-Aware Traffic Trajectory Monitoring From Roadside LiDAR for Occlusion Handling},
journal = {IEEE Transactions on Intelligent Transportation Systems},
year = {2026},
volume = {27},
number = {6},
pages = {6509--6518},
doi = {10.1109/tits.2026.3672508}
}
EUVIMED