Pavement characteristic parameter automatic extraction method and system

Abstract

The invention provides a pavement characteristic parameter automatic extraction method and systems. The method comprises the following steps: scanning the ground in front of a vehicle by using an intelligent networked vehicle-mounted laser radar, and extracting a point cloud initial region of interest by using a single-frame point cloud grid height difference feature; partitioning the point cloudin front of the vehicle based on the density degree of the point cloud, performing final extraction of a region of interest to obtain ground point cloud, and finally performing plane fitting on groundpoints to form a ground reference plane; based on a ground reference plane, calculating an included angle between front and rear blocked ground areas, and completing extraction of longitudinal gradient and transverse gradient characteristic parameters of the front ground; and calculating the distance between the ground point cloud and the reference plane, and calculating the front ground unevenness parameter characteristics by using the extracted unevenness evaluation indexes. According to the method, the characteristic parameters of the road surface in front of the intelligent networked vehicle can be automatically extracted, the perception information of the vehicle to the front environment is enriched, and the safety and stability of the vehicle in the driving process are improved.

Description

technical field [0001] The invention relates to the field of environment perception of intelligent networked vehicles, in particular to a method and system for automatically extracting road surface characteristic parameters. Background technique [0002] In recent years, as issues such as vehicle safety, environmental protection and energy saving have become increasingly prominent, intelligent and connected vehicles (ICVs, Intelligent and connected vehicles) have been researched worldwide. By integrating on-board perception, positioning, communication, computing and control modules, intelligent networked vehicles can drive autonomously and cooperatively without manual driving, which is conducive to reducing driving burden, improving driving safety, improving vehicle operating efficiency, and improving the entire transportation system. performance. [0003] However, intelligent networked vehicles often encounter roads with large slopes or large undulations during driving, wh...

AI Technical Summary

Problems solved by technology

This method has the following disadvantages: (1) It is impossible to extract the road surface information in front of the vehicle in real time; (2) It needs to collect road information in advance, and the workload is heavy; (3) The neural network model of the trained road surface characteristics cannot adapt to different roads and different scenarios condition

[0005] Therefore, the existing road surface feature parameter extraction methods can only provide information on the current road surface where the vehicle is located, and cannot provide real-time information on the road ahead and surrounding roads for the vehicle.

Moreover, the existing methods have poor adaptability to different roads and scenarios, and are not suitable for extracting road surface feature parameters of roads without historical data and roads with variable characteristics.

Images