Scholarlink

<p>The connected vehicle (CV) environment represents a transformative step for future road networks, with real-time data acquisition serving as a critical foundation. High-resolution traffic data (HRTD) encompassing all road users, both connected and unconnected, is essential for CV systems, traffic safety assessments, performance evaluations, and fuel efficiency analyses. Current CV implementations primarily focus on gathering real-time data from connected vehicles, creating a significant gap in the observation of unconnected road users. This research investigates the role of roadside LiDAR (Light Detection and Ranging) sensors in addressing this gap by enabling the real-time collection of 3D point clouds to deliver HRTD for all road users. The study introduces an innovative background filtering algorithm designed to improve the precision and efficiency of HRTD extraction from raw LiDAR data. By leveraging the spatial arrangement of laser points and applying distance-dependent point density thresholds, the algorithm effectively filters out static and dynamic background elements. A detailed case study demonstrates scalability by successfully applying LiDAR-based analysis at moderate to high-volume intersections reflecting real urban conditions. it checks the algorithm’s effectiveness across a range of scenarios, including varying road layouts, traffic volumes, lighting conditions, and speed environments. The results show that vehicle and pedestrian contours remain intact after filtering, and the algorithm’s low computational demands make it well-suited for real-time applications such as vehicle tracking and pedestrian monitoring. This research emphasizes the importance of integrating roadside LiDAR sensors for comprehensive data acquisition in mixed traffic settings, advancing CV systems, and enhancing overall traffic safety and operational efficiency.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Real-time traffic data collection using roadside LiDAR: a novel background filtering approach

  • 期刊论文

  • 发表日期：

    2025年9月25日

查看全文
  • Adil Ansari,
  • B. Raghuram Kadali

摘要

The connected vehicle (CV) environment represents a transformative step for future road networks, with real-time data acquisition serving as a critical foundation. High-resolution traffic data (HRTD) encompassing all road users, both connected and unconnected, is essential for CV systems, traffic safety assessments, performance evaluations, and fuel efficiency analyses. Current CV implementations primarily focus on gathering real-time data from connected vehicles, creating a significant gap in the observation of unconnected road users. This research investigates the role of roadside LiDAR (Light Detection and Ranging) sensors in addressing this gap by enabling the real-time collection of 3D point clouds to deliver HRTD for all road users. The study introduces an innovative background filtering algorithm designed to improve the precision and efficiency of HRTD extraction from raw LiDAR data. By leveraging the spatial arrangement of laser points and applying distance-dependent point density thresholds, the algorithm effectively filters out static and dynamic background elements. A detailed case study demonstrates scalability by successfully applying LiDAR-based analysis at moderate to high-volume intersections reflecting real urban conditions. it checks the algorithm’s effectiveness across a range of scenarios, including varying road layouts, traffic volumes, lighting conditions, and speed environments. The results show that vehicle and pedestrian contours remain intact after filtering, and the algorithm’s low computational demands make it well-suited for real-time applications such as vehicle tracking and pedestrian monitoring. This research emphasizes the importance of integrating roadside LiDAR sensors for comprehensive data acquisition in mixed traffic settings, advancing CV systems, and enhancing overall traffic safety and operational efficiency.