Method for predicting short-time traffic risk of road section by utilizing roadside observation data

Abstract

The invention discloses a method for predicting short-time traffic risk of a road section by utilizing roadside observation data. The method comprises the following steps of: 1) acquiring driving track data in a detection area by utilizing the roadside observation data; 2) counting traffic flow indexes according to continuous driving track data in the detection area, and obtaining collision safety indexes between vehicles; (3) selecting collision time and deceleration as identification indexes, and judging conflict events with collision risks in the detection area; 4) extracting traffic flow indexes and collision safety indexes within a set time before a conflict event occurs, and performing feature screening on various indexes by using a classification algorithm; 5) on the basis of the screened characteristic indexes, selecting the indexes with the top importance ranking as input, constructing a short-term traffic risk prediction model, and completing model training and testing by using training data; and 6) performing short-term traffic risk prediction based on the constructed short-term traffic risk prediction model. According to the method, the prediction accuracy can be improved.

Description

technical field [0001] The invention relates to intelligent traffic technology, in particular to a method for short-term traffic risk prediction of road sections by using roadside observation data. Background technique [0002] With the rapid development of road traffic construction in our country, the number of cars in our country is also growing continuously. Road traffic accidents are one of the important factors causing casualties in our country, and the road traffic safety situation is still grim. How to reduce road traffic risks and improve the driving safety level of road sections is an important research content of traffic safety management. [0003] Under the application of Intelligent Transportation System (Intelligent Transportation System, ITS) and Advanced Transportation Management System (Advanced Transportation Management System, ATMS), using real-time traffic flow information of road sections to predict the possibility of collision accidents in a short time ...

AI Technical Summary

Problems solved by technology

However, it only uses conventional traffic flow parameters, such as flow rate, occupancy rate, and speed, and does not consider the driving behavior of individual vehicles and the interaction behavior of vehicle groups that are closely related to collisions. It has a shortpart

Images