Vehicle Trajectory Repair Under Full Occlusion and Limited Datapoints with Roadside LiDAR

Object occlusion is a common challenge in roadside LiDAR-based vehicle tracking. This issue can cause variances in vehicle location and speed calculations. This paper proposes a vehicle tracking post-processing method designed to handle full occlusion and limited datapoint conditions. The first part...

Full description

Saved in:
Bibliographic Details
Main Authors: Qiyang Luo, Zhenyu Xu, Yibin Zhang, Morris Igene, Tamer Bataineh, Mohammad Soltanirad, Keshav Jimee, Hongchao Liu
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/4/1114
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Object occlusion is a common challenge in roadside LiDAR-based vehicle tracking. This issue can cause variances in vehicle location and speed calculations. This paper proposes a vehicle tracking post-processing method designed to handle full occlusion and limited datapoint conditions. The first part of the method focuses on linking the disconnected trajectories of the same vehicle caused by full occlusion. The second part refines the vehicle representative point to enhance tracking accuracy. Performance evaluation demonstrates that the proposed method can detect and reconnect the trajectories of the same vehicle, even under prolonged full occlusion. Moreover, the refined vehicle representative point provides more stable speed estimates, even with sparse datapoints. This significantly increases the effective detection range of roadside LiDAR. This approach lays a strong foundation for the application of roadside LiDAR in emission analysis and near-crash studies.
ISSN:1424-8220