Vehicle collision risk assessment and prediction method for plane intersection

Abstract

The invention relates to a vehicle collision risk assessment and prediction method for plane intersection. The method comprises the following steps of 1, determining risk factors influencing vehicle collision; 2, building and training a vehicle collision risk assessment and prediction model based on an LSTM network; 3, evaluating and predicting the vehicle collision risk of the plane intersectionin real time. According to the method, the vehicle collision risk can be evaluated in real time, the vehicle collision risk prediction of different time scales can be realized, the early warning of different time scales is carried out on the related vehicles with the collision risk, the traffic conflict between the vehicles at the plane intersection can be obviously reduced, and the safety of thekey areas of the road is effectively improved.

Description

technical field [0001] The invention relates to a prediction method, in particular to a vehicle collision risk assessment and prediction method facing level intersections, and belongs to the technical field of traffic safety assessment. Background technique [0002] As a key area of ​​the road, vehicle collision accidents occur frequently at level intersections. Therefore, vehicle collision risk assessment for level intersections is particularly important. At present, the "central cloud" processing technology is adopted at the macro level, that is, the basic road traffic parameters and traffic flow information are gathered through the traffic command and control platform to evaluate whether there is vehicle congestion or collision at the level intersection. Since there is a transmission delay in the transfer of data from the level intersection to the traffic command and control platform, and there is also a delay in the centralized processing of the "central cloud", this typ...

AI Technical Summary

Problems solved by technology

In short, the current assessment of vehicle collision accidents at level intersections is a centralized "post-event" assessment, which cannot evaluate the vehicle collision risk at level intersections in real time, and lacks the ability to "pre-event" predict vehicle collision risks on different time scales.

Images