Summary of Dataset Scenario and Performance.
<div><p>Roadside LiDAR systems can generate real-time microscopic vehicle trajectories applicable to develop intelligent transportation systems and aid the operations of connected and autonomous vehicles. Tracking is the most crucial data processing step to generate accurate and reliable...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , , , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <div><p>Roadside LiDAR systems can generate real-time microscopic vehicle trajectories applicable to develop intelligent transportation systems and aid the operations of connected and autonomous vehicles. Tracking is the most crucial data processing step to generate accurate and reliable trajectories of road users from raw point clouds collected from LiDAR sensors. In this paper, a new tracking mechanism is proposed for real-time tracking, which is based on the 2D LiDAR data structure with the Simple Online and Real-Time Tracking (SORT) algorithm. The traditional method of using bounding boxes to identify vehicles is replaced by center points of vehicles inspired by track-by-point approach. The developed index that integrates the distance between the LiDAR serves as a more accurate way of defining the spatial location of vehicles. The proposed methodology was evaluated using data collected by a 32-channel portable LiDAR at three signalized intersections. The results showed that the proposed method has a higher tracking accuracy and a faster computation speed compared to the traditional bounding box approach, indicating improvement toward real-time applications. <i>Index Terms</i>—Real-time tracking; Roadside LiDAR; Trajectory; Track by point; SORT.</p></div> |
|---|