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

Area of Science:

• Robotics and Autonomous Systems
• Computer Vision
• Sensor Fusion

Background:

• Object occlusion is a significant challenge in roadside LiDAR-based vehicle tracking.
• Occlusion leads to inaccuracies in vehicle location and speed calculations.
• Existing methods struggle with prolonged full occlusion and sparse data.

Purpose of the Study:

• To propose a robust post-processing method for roadside LiDAR vehicle tracking under occlusion.
• To enhance the accuracy of vehicle tracking by addressing full occlusion and limited data points.
• To improve the effective detection range of roadside LiDAR systems.

Main Methods:

• A two-part post-processing approach is presented.
• Part one links disconnected vehicle trajectories caused by full occlusion.
• Part two refines the vehicle representative point for improved accuracy.

Main Results:

• The method successfully detects and reconnects trajectories of the same vehicle during full occlusion.
• Refined vehicle representative points yield more stable speed estimates with sparse data.
• The effective detection range of roadside LiDAR is significantly increased.

Conclusions:

• The proposed method effectively handles object occlusion in LiDAR-based vehicle tracking.
• Improved tracking accuracy and extended detection range are demonstrated.
• This work supports applications in emission analysis and near-crash studies using roadside LiDAR.