CN-121982895-A - Vehicle running risk quantification method for double-decision area during yellow light of signal lamp

CN121982895ACN 121982895 ACN121982895 ACN 121982895ACN-121982895-A

Abstract

The invention provides a vehicle running risk quantification method for a double-decision area during a signal lamp yellow lamp, which is used for analyzing the vehicle running risk under different driving behavior decisions. The method comprises the steps of firstly determining the boundary of a dynamic dilemma zone according to the geometric parameters of an intersection, the duration of a yellow light, the initial speed of a vehicle, the braking performance and the like. Secondly, comprehensively considering the motion state of the vehicle, the behaviors of front and rear and lateral traffic participants and the constraint of road environment, and constructing a vehicle risk quantification evaluation system during the yellow light period of the signal lamp. On the basis, based on a risk field theory, a regional risk quantification model is established by considering the driving behavior characteristics of the self-vehicle and the running risk space attenuation of the vehicle, and the running risk quantification of the vehicle in the two difficult areas during the yellow light period of the signal lamp is realized through the risk coupling operation of each region.

Inventors

LIU MIAOMIAO
YANG HAOYI
LIU DOUDOU
ZHU MINGYUE
WANG LUYAO
JIANG DEMING
ZHANG MEIQI

Assignees

北京航空航天大学

Dates

Publication Date: 20260505
Application Date: 20260226

Claims (4)

1. A method for quantifying the risk of vehicle operation in a critical decision area during the yellow light of a signal lamp, comprising the steps of: step 1, comprehensively considering the geometry and signal parameters of an intersection by analyzing the known initial speed, braking performance and yellow light duration of a vehicle so as to determine the boundary of a dynamic dilemma area; Step 2, constructing a vehicle risk quantification evaluation system during the period of yellow lights of the multi-area signal lamps, and specifically calculating and quantifying risk sources of all areas, wherein the system specifically comprises front risk intensity Intensity of risk at rear Intensity of lateral risk Coefficient of environmental risk ; And3, based on a risk field theory, establishing a regional risk quantification model by considering the driving behavior characteristics of the self-vehicle and the running risk space attenuation of the vehicle, and quantifying the running risk of the vehicle in the two-difficulty region during the yellow lamp period of the signal lamp through the risk coupling operation of each region.
2. The method of claim 1, wherein specific quantization operations are performed on the regional risk sources and a running risk model of vehicles currently in the behinderance region during the yellow light is built: Wherein, the A risk of operation for the vehicle currently in the hard choice region during the yellow light; the front risk intensity is the risk intensity generated by a vehicle positioned in front of the current vehicle in a lane where the current vehicle is positioned; the risk intensity is the risk intensity at the rear, namely the risk intensity generated by the vehicle positioned at the rear of the current vehicle in the lane where the current vehicle is positioned; The lateral risk intensity is the risk intensity generated by vehicles, pedestrians and non-motor vehicles in the lateral area of the current vehicle; 、、 And (3) calibrating by means of traffic flow actual measurement data, driving behavior actual measurement data and historical risk event statistical data during the yellow lights of intersections as weight coefficients.
3. The method of claim 1, wherein the driving behavior coefficients corresponding to driving behaviors in class 3 influencing front-to-rear risk intensity in the double choice zone during yellow light are simulated And (3) with : The two driving behavior coefficients respectively quantify the front risk intensity of the current driving behavior Intensity of risk with rear The values of the action effects of the driving direction and the risk direction are divided into 3 grades, namely 0.5, 1 and 2, respectively corresponding to 3 action modes of negative influence, no influence and positive influence of the driving behavior on the risk intensity, wherein when the value of the coefficient is 0.5, the current driving behavior has a weakening effect on the risk of the corresponding direction and can reduce the risk intensity of the direction, when the value of the coefficient is 1, the current driving behavior is not obviously related to the risk intensity of the corresponding direction and cannot influence the risk intensity of the corresponding direction, and when the value of the coefficient is 2, the current driving behavior has an amplifying effect on the risk of the corresponding direction and can improve the risk intensity of the direction; The driving behavior characteristics of the actual intersection yellow light scene are combined, drivers are divided into cautious, dangerous and conventional 3 types, and the coefficient values and action mechanisms corresponding to various driving behaviors are as follows: For cautious driving behavior, the driver usually adopts an operation strategy of early deceleration and stable braking after the yellow light is turned on, and in the behavior mode, the risk faced by the vehicle mainly comes from the following vehicle, the risk is mainly caused by insufficient following distance, the collision probability in front of the vehicle is obviously reduced, and the front driving behavior coefficient of the driving behavior is greatly reduced Rear driving behavior coefficient ; For conventional driving behavior, after the yellow light is turned on, a driver usually adopts an operation strategy of keeping the original speed or slightly decelerating and safely passing during the yellow light, in the behavior mode, the motion state of the vehicle is basically stable, the driving behavior does not influence front and rear risks, the only situation with possible risk is that the transverse vehicle breaks out of rules and robs, the external occasional risk is caused, and the front driving behavior coefficient of the driving behavior is irrelevant to the behavior of the driver Rear driving behavior coefficient ; For dangerous driving behavior, after a yellow light is started, a driver usually adopts an active acceleration operation strategy to rob the driver to pass in front of a red light, and in the behavior mode, the risks of the vehicle mainly come from rear-end collision risks uncoordinated with the driving state of the vehicle in front and risks of insufficient braking redundancy caused by too fast speed and incapability of coping with intersection emergency conditions are mainly sourced from rear-end collision risks and the front driving behavior coefficients of the driving behavior are obviously reduced Rear driving behavior coefficient 。
4. The method according to claim 1, wherein the specific calculation and quantification of each regional risk source is performed taking into account the characteristics of the self-driving behavior and the attenuation of the vehicle running risk space; For the front risk intensity It is necessary to consider the relative distance of the preceding vehicle, the speed of the preceding vehicle, the length of the preceding vehicle, the speed of the current vehicle, and incorporate the forward driving behavior coefficient Quantification was performed using an inverse proportional decay model: Wherein, for a front vehicle, an x-axis and a y-axis in a coordinate system are respectively parallel to the longitudinal direction and the transverse direction of the front vehicle, an origin of the coordinate system is positioned at the center of the front vehicle, and the current vehicle is positioned in the coordinate system At the position of the first part, At the coordinates for the current vehicle The attenuation factor of the position, because the current vehicle and the front vehicle are positioned in the same lane, the coordinates in the front vehicle coordinate system The attenuation factor at can be regarded as being equal to only In relation to the use of a liquid crystal display device, For the current speed of the vehicle, Is a minimum value for avoiding abnormal calculation with a 0 denominator when the speed is 0, 、、 Respectively, the speed of the preceding vehicle, the relative distance from the preceding vehicle, the extent to which the speed of the current vehicle affects the risk ahead, and the type of vehicle, For the length of the vehicle in front of the vehicle, Is the speed of the vehicle in front; for rear risk intensity It is necessary to consider the relative distance of the rear vehicle, the speed of the rear vehicle, the length of the rear vehicle, the speed of the current vehicle, and incorporate the rear driving behavior coefficient Quantification was performed using an inverse proportional decay model: Wherein for a rear vehicle, the x-axis and the y-axis in the coordinate system are respectively parallel to the longitudinal and transverse directions of the rear vehicle, the origin of the coordinate system is located at the center of the rear vehicle, and the current vehicle is located in the coordinate system At the position of the first part, At the coordinates for the current vehicle The attenuation factor of the position, because the current vehicle and the rear vehicle are positioned on the same lane, the coordinates in the coordinate system of the rear vehicle The attenuation factor at can be regarded as being equal to only In relation to the use of a liquid crystal display device, For the current speed of the vehicle, 、、 The speed of the rear vehicle, the relative distance from the rear vehicle, the extent to which the speed of the current vehicle affects the rear risk, respectively, are related to the vehicle type, For the length of the vehicle behind it, Is the speed of the rear vehicle; Lateral risk intensity The method comprises the steps that relative distances of a lateral vehicle, a pedestrian and a non-motor vehicle, speeds of the lateral vehicle, the pedestrian and the non-motor vehicle, the length of the lateral vehicle, the width of the vehicle and the speed of a current vehicle are considered, the method adopts an inverse proportion attenuation model to quantify, firstly, the boundary range of a lateral region is defined, the lateral dimension defines the adjacent left and right lanes on two sides of the current driving lane as the effective coverage range of the lateral region, the longitudinal dimension extends to a distance equivalent to the length of the two current vehicles by taking the tail of the current vehicle as an initial reference, the longitudinal boundary of the lateral region is defined, and after the definition of the lateral region is completed, quantitative analysis and evaluation are further carried out on the lateral driving risk in the region: Vehicle k in the lateral region: Wherein for a vehicle k in the lateral region, the x-axis and the y-axis in the coordinate system are parallel to the longitudinal and transverse directions of travel of k, respectively, the origin of the coordinate system is located at the geometric center of k, For the current vehicle in a coordinate system established with k as the center A risk attenuation factor at the location of the risk attenuation factor, For the current speed of the vehicle, And The extent of influence of the longitudinal speed of k and the relative spacing in the longitudinal direction on the lateral risk is expressed respectively, At the longitudinal velocity of k, For a length of k, And The degree of influence of the transverse velocity of k and the relative spacing of the transverse direction on the lateral risk is expressed respectively, Indicating the extent to which the current vehicle speed affects the risk of the motor vehicle, At the transverse velocity of k, A width of k; pedestrians and non-motor vehicles p in the lateral region: wherein for pedestrians and non-vehicles p in the lateral region, the x-axis and the y-axis in the coordinate system are respectively parallel to the longitudinal and transverse directions of p, the origin of the coordinate system is located at the geometric center of p, For the current vehicle in a coordinate system built with p as the center A risk attenuation factor at the location of the risk attenuation factor, For the current speed of the vehicle, And The extent of influence of the longitudinal speed of p and the relative spacing in the longitudinal direction on the lateral risk is expressed respectively, At the longitudinal speed of p, And The degree of influence of the lateral velocity of p and the relative spacing of the lateral direction on the lateral risk is expressed respectively, Indicating the extent to which the speed of the current vehicle affects the risk of lateral non-motor vehicles and pedestrians, A lateral velocity of p; finally, counting all traffic participants in the lateral areas, and taking the lateral risk generated by the traffic participant generating the maximum risk at the current vehicle as the lateral risk intensity : Wherein, the Representing vehicles i in lateral regions The lateral risk intensity created at the current vehicle, Representing pedestrians or non-motor vehicles j in lateral regions The lateral risk intensity created at the current vehicle.

Description

Vehicle running risk quantification method for double-decision area during yellow light of signal lamp Vehicle running risk quantification method for double-decision area during yellow light of signal lamp Technical Field The invention relates to the field of intersection safety assessment in traffic engineering, in particular to a vehicle running risk quantification method and a risk field modeling technology for a double-decision area during a signal lamp yellow lamp, which are suitable for risk perception of vehicles and intersection safety management and control optimization. Background The urban road intersection is a core node for traffic flow intersection, a transition stage of green light to red light conversion is used during a yellow light period, vehicles face a stop or pass decision, a formed decision area is a key risk point for high occurrence of traffic accidents, and accurate risk quantification and dynamic risk perception become core requirements for supporting vehicle safety decision. The formed double-decision area has extremely strong dynamic uncertainty and risk coupling as a transition stage of green light to red light conversion during the yellow light, vehicles in the area need to face binary decision dilemma of 'parking' and 'passing', the complexity of constraint conditions is increased in a large scale, and the solving difficulty of accurate quantification and dynamic characterization is increased. Therefore, risk assessment of the dilemma area during yellow lights becomes one of the most challenging problems in intersection risk assessment and security decisions. Since the concept of the two-difficult-area in the 60 th century of the 20 th century was formally proposed, a great deal of research was conducted in the academic world around three core directions of basic properties, driver behaviors, management and control, but the quantification of the risks of vehicles in the two-difficult-area is still difficult to adapt to the current accurate risk perception requirements, the accurate risk pre-judgment is difficult to support, and meanwhile, the accurate quantification of the risk distribution in the two-difficult-area cannot be realized due to the lack of a dynamic risk characterization method of a system. Therefore, the invention provides a vehicle running risk quantification method for a double-decision area during a signal lamp yellow lamp, which aims to solve the problems of one-sided risk assessment, inaccurate quantification and insufficient dynamic property in the prior art, so that space-time resources of an intersection are more reasonably distributed, the overall passing efficiency of the intersection is improved, the collision risk caused by decision uncertainty is effectively avoided, and the running safety of vehicles and traffic participants is improved. Disclosure of Invention The invention is based on the above situation of the prior art, the invention aims to provide a precise and comprehensive vehicle running risk quantification method for a signal lamp yellow light period double-decision area, and the vehicle running risk quantification in the double-decision area is realized by determining the boundary of a dynamic double-decision area, constructing a multi-area risk quantification evaluation system and establishing a regional risk quantification model, wherein the research scene of the invention is an urban road intersection entrance way, and is suitable for risk assessment during the yellow light period, and the specific realization steps are as follows: step 1, comprehensively considering the geometry and signal parameters of an intersection by analyzing the known initial speed, braking performance and yellow light duration of a vehicle so as to determine the boundary of a dynamic dilemma area; Step 2, constructing a vehicle risk quantification evaluation system during the period of yellow lights of the multi-area signal lamps, and specifically calculating and quantifying risk sources of all areas, wherein the system specifically comprises front risk intensity Intensity of risk at rearIntensity of lateral riskCoefficient of environmental risk; And3, based on a risk field theory, establishing a regional risk quantification model by considering the driving behavior characteristics of the self-vehicle and the running risk space attenuation of the vehicle, and quantifying the running risk of the vehicle in the two-difficulty region during the yellow lamp period of the signal lamp through the risk coupling operation of each region. In step 3, a regional risk quantization model is established by considering the driving behavior characteristics of the self-vehicle and the running risk space attenuation of the self-vehicle, and the steps of the risk coupling operation of each region are as follows: a. Performing specific quantification operation on risk sources of all areas and establishing an operation risk model of vehicles in the area which is currently in the