Fusion of Airborne and Terrestrial Sensed Data for Real-time Monitoring of Traffic Networks

Abstract

Expanding the deployment of traffic monitoring systems and information transmission is a crucial step toward increasing the efficiency, reliability and safety of vehicular transportation. In this project, we present a real-time data analysis system for aerial LiDAR roadway vehicle detection as an option for expanding traffic monitoring. The presented solution is able to perform typical post-flight processing in real time with minimal computational and power requirements, which allows its implementation on light-weight unmanned aircraft systems (UAS). The solution utilizes adaptive segmentation and 3D convolutions, which take advantage of the structure of the LiDAR point cloud, to identify vehicles and their respective positions within 3D point cloud segments that may include background clutter. All the necessary positioning information required to run the algorithm is introduced, along with a detailed description of the computational steps for extracting the desired features from the raw data. We provide the timing constraints for the system and evaluate its performance while considering different optimization variables and computation capabilities. The proposed convolution algorithm can achieve a detection ratio of greater than 85 percent and is able to detect the presence of vehicles with less than 0.75 seconds of computation time.