LiDAR-based Vehicle Localization and Velocity Measurement in Long Corridors

This paper focuses on improving the roadside vehicle localization and speed measurement system for a security inspection company. First, a horizontally mounted LiDAR is added to the original system to meet the demand for whole-process vehicle detection in long corridors. A simulation environment is created using Webots, providing necessary data sources to test and validate algorithms, with ROS2(Robot Operating System 2) to enhancing real-time system performance and task prioritization management. Then, dynamic threshold is employed to identify the cab-trailer separation of truck and prevent the radiation exposure from X-ray machine to the driver during inspection. A weighted average fusion method, incorporating distance and intensity, is used to enhance the accuracy of the system’s detection scheme. Finally, experimental results show that the improved system has better accuracy, stability, real-time performance, and portability.