System architecture and algorithms for detection, tracking and classification of vehicles in a free flow toll gantry using LiDARs

Abstract

One of the most important functions in free-flow systems to be performed is the detection, classification and identification of vehicles at high and low speeds. In this sense, there are several technologies that fulfill this function such as 2D LiDARs and detection cameras, but these technologies can present difficulties such as their coverage area or in the case of cameras in changing conditions of visibility and luminosity, among others. The 3D LiDAR technology has demonstrated the ability to cover these limitations, but the processing of the captured data requires specific algorithms. The objective of this research paper is to develop robust algorithms for vehicle detection and classification using 3D LiDARs technology installed in the road infrastructure.