Highway traffic state discrimination method based on vehicle level macro-micro evaluation

Abstract

The application belongs to the technical field of intelligent transportation, and particularly relates to a highway traffic state discrimination method based on vehicle-level macro and micro evaluation. First, millimeter wave radar data of a target section in a target period is acquired, and the millimeter wave radar data is preprocessed. Then, evaluation indexes are calculated according to the millimeter wave radar data, including two types of indexes, i.e., micro and macro indexes. The correlation between the indexes is tested, and some indexes with strong correlation are removed, so as to obtain micro indexes including vehicle yaw angle variance, lane changing behavior frequency, vehicle head time interval variance, acceleration variance and aggressive driving behavior variation index variance. Macro indexes include average flow, speed variation coefficient, large vehicle mixing rate and congestion index. Finally, based on the macro and micro evaluation indexes, traffic state entropy values of each section in the target period of the target section are calculated, and the traffic state is classified according to the traffic state entropy values. The method proposes the aggressive driving behavior variation index for measuring the risk degree of vehicles caused by sudden acceleration, deceleration or too fast driving speed, so as to more carefully reflect the traffic state from the micro aspect, and simultaneously consider the mutual influence between the sections to accurately quantify the traffic state.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent transportation, and particularly relates to a highway traffic state discrimination method based on vehicle-level macro-micro evaluation. BACKGROUND

[0002] Highway driving speed is fast and risk is high, real-time monitoring of traffic state can not only identify occasional traffic events, but also reflect the running state of the highway, which is crucial for precise running risk management.

[0003] In terms of traffic state representation, macro parameters such as flow, speed and density are still the main parameters for traffic state acquisition and representation. Existing researches mainly analyze traffic state based on macro parameters, but the macro expression of traffic state cannot reveal the micro characteristics of traffic flow and cannot describe the change mechanism of traffic flow. With the continuous development of intelligent transportation, the interaction and cooperation of road end sensing devices and vehicle end provide support for obtaining micro parameters of traffic state. However, the index system based on micro parameters in existing researches is not comprehensive enough. For example, the key indicators reflecting the micro traffic flow characteristics such as vehicle yaw angle and lane changing frequency are not considered, and the potential impact of driver aggressive behavior on traffic state is not considered. In addition, the existing traffic state discrimination method is mainly based on weighted entropy, which mainly quantifies the current traffic state of the road section, without considering the mutual influence between road sections. Based on the foregoing problems, the application provides a highway traffic state discrimination method based on vehicle-level macro-micro evaluation. SUMMARY

[0004] In view of the deficiencies of the prior art, the technical problem to be solved by the application is to provide a highway traffic state discrimination method based on vehicle-level macro-micro evaluation.

[0005] The application solves the technical problem by adopting the following technical scheme:

[0006] A highway traffic state discrimination method based on vehicle-level macro-micro evaluation, characterized in that the method comprises the following steps:

[0007] Step 1: Obtain the millimeter wave radar data of the target road section in the target period, and pre-process the millimeter wave radar data;

[0008] Step 2: Calculate evaluation indexes according to the millimeter wave radar data, including two types of indexes, i.e. micro and macro indexes; test the correlation between the indexes, and eliminate the indexes with strong correlation to obtain micro indexes including vehicle yaw angle variance, lane changing behavior frequency, vehicle head time interval variance, acceleration variance and aggressive driving behavior variation index variance; and macro indexes including average flow, speed variation coefficient, large vehicle mixing rate and congestion index;

[0009] Among them, the aggressive driving behavior variation index is used to measure the degree of vehicle danger, and the calculation formula is as follows:

[0010]

[0011] Where, E k is the aggressive driving behavior variation index of vehicle k, a t 、v t are the acceleration and speed of vehicle k at time t, ω1 and ω2 are weights, M represents the time period, and T represents the total time vehicle k appears on the target road section;

[0012] The variance of the aggressive driving behavior variation index is used to reflect the discrete degree of vehicle danger. The calculation formula is as follows:

[0013]

[0014] Where, is the variance of the aggressive driving behavior variability index, is the average value of the aggressive driving behavior variation index, and N represents the number of vehicles on the target road section during the target period;

[0015] Step 3: Based on the macro- and micro-evaluation indicators, calculate the traffic state entropy value of each target road section during the target period, and classify the traffic state according to the traffic state entropy value;

[0016] The gray correlation analysis method is used to calculate the gray correlation coefficient of each indicator of each road section, and the weight of the indicator is calculated according to the gray correlation coefficient. The calculation formula is as follows:

[0017]

[0018] Where w ij represents the weight of the jth index of the target road segment i, δ ij is the normalized grey correlation coefficient of the jth indicator of road section i, and m is the number of indicators;

[0019] The traffic state entropy value of the road section is calculated according to the weight of each indicator. The calculation formula is:

[0020]

[0021] Where D i is the traffic state entropy value of road section i.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] 1. Based on the interactions between people, vehicles, roads, and the environment, this paper proposes a more sophisticated and comprehensive multi-dimensional macro- and micro-traffic state evaluation indicator system, characterizing traffic flow characteristics at both macro and micro levels. It also introduces the Aggressive Driving Behavior Variation Index, a novel measure of vehicle danger caused by sudden acceleration, deceleration, or excessive speed. The Aggressive Driving Behavior Variation Index variance is used as a micro-indicator to reflect the dispersion of vehicle danger levels within a specific time period or road section, providing a more detailed picture of traffic conditions.

[0024] 2. The entropy value of traffic status is calculated by combining the grey correlation analysis method with the entropy method. The traffic status is accurately quantified by considering the mutual influence between each road section. The traffic status is classified into different levels according to the entropy value, which provides an important basis for traffic management departments to make reasonable traffic planning decisions, alleviate traffic congestion, and solve traffic safety problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Fig. 1 It is the overall flow chart of the present invention;

[0026] Fig. 2 It is a correlation diagram among the various indicators of the present invention;

[0027] Fig. 3 It is a schematic diagram of the multi-dimensional macro-micro traffic status evaluation index system of the present invention. DETAILED DESCRIPTION

[0028] Specific embodiments are given below in conjunction with the accompanying drawings. The specific embodiments are only used to introduce the technical solutions of the present invention in detail and are not intended to limit the scope of protection of the present application.

[0029] The present invention provides a highway traffic status discrimination method based on vehicle-level macro-micro evaluation (hereinafter referred to as the method, see Figs. 1-3 ), including the following steps:

[0030] Step 1: Obtain millimeter-wave radar data of the target road section within the target period and pre-process the millimeter-wave radar data;

[0031] Step 2: Calculate evaluation indicators based on millimeter-wave radar data, including both micro and macro categories; micro indicators include vehicle deviation angle variance, lane change frequency, headway variance, acceleration variance, and aggressive driving behavior variation index variance, totaling five types; macro indicators include average flow rate, average density, speed variation coefficient, large vehicle mixing rate, congestion index, road saturation, and average travel delay, totaling seven types.

[0032] The vehicle deflection angle variance is used to measure the degree of dispersion of the vehicle deflection angle measurement value, that is, the stability of the vehicle deflection angle measurement value. If the vehicle deflection angle variance is large, it means that the measurement value fluctuates greatly, which may affect the stability and safety of vehicle driving. The calculation formula of vehicle deflection angle variance is as follows:

[0033]

[0034] Where, represents the vehicle deviation angle variance, AS k represents the deflection angle of vehicle k, represents the average vehicle deviation angle, and N represents the number of vehicles on the target road section during the target period.

[0035] Lane-changing frequency is an important indicator of driving behavior, reflecting how frequently a driver changes lanes while driving. Lane-changing frequency is typically defined as the number of lane changes a vehicle makes within a certain period of time or distance. The formula for calculating lane-changing frequency is as follows:

[0036]

[0037] Where, F adb Indicates the frequency of lane changing behavior, sum(event adb ) represents the total number of lane-changing behaviors, and Q represents the real-time traffic volume.

[0038] The headway variance is used to quantify the fluctuation of headway. A larger headway variance indicates a greater degree of headway fluctuation. Specifically, it refers to the average of the squared differences between the headway of all vehicles and its average. The calculation formula is as follows:

[0039]

[0040] Where, represents the headway variance, TH k represents the headway of vehicle k, is the average headway time.

[0041] Acceleration variance reflects the degree of dispersion of vehicle acceleration within a certain time period or road section, that is, the amplitude and stability of acceleration change. The acceleration equation is calculated as follows:

[0042]

[0043] Where, is the acceleration variance, a k is the acceleration of vehicle k, is the average acceleration.

[0044] The Aggressive Driving Behavior Variation Index (ADI) measures the degree of danger caused by sudden acceleration, deceleration, or excessive speed. It is calculated using the acceleration and speed of the vehicle at various moments over a period of time. The calculation formula is as follows:

[0045]

[0046] Where, E k is the aggressive driving behavior variation index of vehicle k, a t 、v t are the acceleration and speed of vehicle k at time t, ω1 and ω2 are weights, M represents the time period, and T represents the total time that vehicle k appears on the target road section.

[0047] The variance of the aggressive driving behavior variation index reflects the discrete degree of vehicle danger within a certain time period or road section. The calculation formula is as follows:

[0048]

[0049] Where, is the average value of the aggressive driving behavior variation index.

[0050] Average traffic flow refers to the average traffic flow through a road section within a certain period of time. The calculation formula is as follows:

[0051]

[0052] Where, is the average flow rate of the target road section, q i 、l i is the flow rate and length of segment i in the target segment.

[0053] Average density refers to the density of vehicles on a certain road section, also known as vehicle density. The calculation formula is as follows:

[0054]

[0055] Where, is the average density of the target road segment, ρ i is the vehicle density of road segment i.

[0056] The speed variation coefficient is the ratio of the standard deviation of vehicle speed to the mean value. It is used to quantify the fluctuation of vehicle speed. It reflects the degree of dispersion of vehicle speed in traffic flow, that is, the difference between vehicle speeds. The calculation formula of the speed variation coefficient is as follows:

[0057]

[0058] Where C v is the speed variation coefficient, S vis the standard deviation of vehicle speed, v k is the speed of vehicle k, is the average vehicle speed;

[0059] The large vehicle mixing rate is usually defined as the proportion of large vehicles in the running traffic, and the calculation formula is as follows:

[0060]

[0061] Where R lm is the large vehicle mixing rate, Q l is the number of large vehicles, and Q is the real-time traffic volume.

[0062] The congestion index, also known as the traffic congestion index or road traffic operation index, is an indicator that comprehensively reflects the traffic operation status of the road network. The calculation formula is as follows:

[0063]

[0064] Where, CI i represents the congestion index of road section i, v i 、 are the free flow speed and the average vehicle speed of section i, respectively.

[0065] Road saturation refers to the ratio of actual traffic flow to road capacity. It reflects the traffic load of a road within a certain period of time, that is, the degree of road congestion. The calculation formula is as follows:

[0066]

[0067] Where S is the road saturation, f is the number of vehicles on a unit road section, and c is the maximum traffic capacity of the road.

[0068] Average trip delay is an important indicator reflecting road operation efficiency and service level. It refers to the travel time loss caused by factors beyond the driver's control, such as road and environmental conditions, traffic interference, and traffic management and control facilities. The calculation formula is as follows:

[0069]

[0070] Where d is the average travel delay and v0 is the free flow speed of the target road segment.

[0071] Considering the correlation between the indicators, the Pearson test method was used to test the correlation between the indicators. The test results are as follows: Fig. 2 As shown in the figure, the average flow (e8), average density (e9), road saturation (e 11 ) are strongly correlated with each other, and the congestion index (e 10 ) and average trip delay (e12 ) between them, and the indicators with strong correlation are retained, in this embodiment, the average flow and congestion index are retained, and the average density, road saturation and average travel delay are eliminated, thus the nine indicators of vehicle angle variance, lane changing behavior frequency, headway variance, acceleration variance, aggressive driving behavior variance index, average flow, speed variance coefficient, large vehicle mixing rate and congestion index form a multi-dimensional macro-micro traffic state evaluation index system, as shown in Fig. 3 .

[0072] Third step: based on the multi-dimensional macro-micro traffic state evaluation index system, calculate the traffic state entropy value, and divide the traffic state categories according to the traffic state entropy value;

[0073] Firstly, the grey correlation analysis method is used to calculate the grey correlation coefficient of each index of each road section; the target road section is divided into multiple road sections, and all the indexes of each section in each period form an index sequence, and each road section corresponds to an optimal index sequence at the best traffic state; according to the grey system theory, let:

[0074]

[0075] In the formula, x i * is the index sequence of road section i, is the jth index value of road section i, y i * is the optimal index sequence of road section i, is the optimal value of the jth index of road section i, and m is the number of indexes;

[0076] The index sequence and the optimal index sequence of each road section are normalized to obtain the normalized index sequence and the optimal index sequence; the normalization formula is:

[0077]

[0078] In the formula, is the jth index value of road section i after normalization, is the optimal value of the jth index of road section i after normalization;

[0079] According to the normalized index sequence and the optimal index sequence, the maximum and minimum absolute differences of each index of the road section are calculated, and the grey correlation coefficient of each index of the road section is calculated therefrom;

[0080]

[0081] In the formula, Δ maxij , Δ minij are the maximum and minimum absolute differences of the jth index of road section i, respectively, and ξij is the grey correlation coefficient of the jth index of the road section i, is the deviation of the jth index of the road section i in the current period from the best traffic state, and μ is a resolution coefficient, generally 0.5;

[0082] The grey correlation coefficient is normalized, and the normalization formula is as follows:

[0083]

[0084] In the formula, δ ij is the normalized grey correlation coefficient of the jth index of the road section i;

[0085] The weight of each index is calculated according to formula (20):

[0086]

[0087] In the formula, w ij represents the weight of the jth index of the road section i;

[0088] The traffic state entropy value of the road section is calculated according to the weight of each index, and the calculation formula is:

[0089]

[0090] In the formula, D i is the traffic state entropy value of the road section i.

[0091] Similarly, the traffic state entropy values of each road section in the target road section are calculated, and the traffic state is determined according to the traffic state entropy values.

[0092] Embodiment

[0093] In this embodiment, a certain expressway is taken as an example, the length of the target road section is 400 meters, the millimeter wave radar data of the target road section in the target period (from 0 o'clock in the morning to 11 o'clock at night, with an interval of 15 minutes) is obtained, and the traffic state entropy values of each period are calculated by using the millimeter wave radar data, as shown in Table 1.

[0094] Table 1 Traffic state entropy values of each period of the target road section

[0095]

[0096]

[0097] In this embodiment, the traffic state is divided into four categories, and the value range of the traffic state entropy value of each category is shown in Table 2.

[0098] Table 2 Value range of traffic state entropy value

[0099]

[0100] Therefore, the traffic state can be determined based on the traffic state entropy value. The road network system can issue warning information to drivers, traffic police departments, etc. based on the traffic state through warning information release facilities. Warning information release facilities include variable message boards, variable speed limit signs, traffic broadcasts, signal light systems, public information query systems, and vehicle information terminals. Variable message boards are used to provide dynamic traffic information, and the layout information can be adjusted in real time according to road traffic safety conditions; variable speed limit signs are used to remind drivers of their driving speed at any time, and the speed limit value can be displayed through the monitoring center according to the actual road conditions; traffic broadcasts are used to broadcast authoritative information in a timely manner and deploy emergency measures according to emergency needs in the event of major emergencies, playing a role in early warning, communication, and information notification; the signal light system is used to visualize early warnings. For example, extreme chaos corresponds to a red first-level warning signal light, indicating that the possibility of a traffic accident is very high and safety control measures must be taken immediately; chaos corresponds to an orange second-level warning signal light, indicating that a traffic accident may occur and safety control measures must be taken in a timely manner; general chaos corresponds to a yellow third-level warning signal light, indicating that a traffic accident is unlikely and no safety control measures are required; normal corresponds to a green fourth-level warning signal light, indicating that traffic accidents are unlikely and no measures are required; the public information inquiry system is used to inquire about congestion conditions, the shortest path, route information, weather conditions, etc.; and the on-board information terminal is used to receive early warning information and remind drivers to drive carefully.

[0101] Any matters not described in the present invention are applicable to the prior art.

Claims

1. A highway traffic status identification method based on vehicle-level macro-micro evaluation, characterized in that: The method comprises the following steps: Step 1: Obtain millimeter-wave radar data of the target road section within the target period and pre-process the millimeter-wave radar data; Step 2: Calculate evaluation indicators based on millimeter-wave radar data, including both micro and macro indicators. Correlations between indicators are tested, and those with strong correlations are eliminated. Micro indicators include vehicle deviation angle variance, lane change frequency, headway variance, acceleration variance, and aggressive driving behavior variance. Macro indicators include average flow rate, speed coefficient of variation, large vehicle mixing rate, and congestion index. Among them, the aggressive driving behavior variation index is used to measure the degree of vehicle danger, and the calculation formula is as follows: Where, E k is the aggressive driving behavior variation index of vehicle k, a t 、v t are the acceleration and speed of vehicle k at time t, ω1 and ω2 are weights, M represents the time period, and T represents the total time vehicle k appears on the target road section; The variance of the aggressive driving behavior variation index is used to reflect the discrete degree of vehicle danger. The calculation formula is as follows: Where, is the variance of the aggressive driving behavior variation index, E is the average value of the aggressive driving behavior variation index, and N represents the number of vehicles on the target road section during the target period; Step 3: Based on the macro- and micro-evaluation indicators, calculate the traffic state entropy value of each target road section during the target period, and classify the traffic state according to the traffic state entropy value; The gray correlation analysis method is used to calculate the gray correlation coefficient of each indicator of each road section, and the weight of the indicator is calculated according to the gray correlation coefficient. The calculation formula is as follows: Where w ij represents the weight of the jth index of the target road segment i, δ ij is the normalized grey correlation coefficient of the jth indicator of road section i, and m is the number of indicators; The traffic state entropy value of the road section is calculated according to the weight of each indicator. The calculation formula is: Where D i is the traffic state entropy value of road section i.

2. The highway traffic status identification method based on vehicle-level macro-micro evaluation according to claim 1 is characterized in that: The calculation formula of vehicle deflection angle variance is as follows: Where, represents the vehicle deviation angle variance, AS k represents the deflection angle of vehicle k, Indicates the average value of vehicle deviation angle; The calculation formula for lane changing behavior frequency is as follows: Where, F adb Indicates the frequency of lane changing behavior, sum(event adb ) represents the sum of the number of lane-changing behaviors, and Q represents the real-time traffic volume; The calculation formula of headway variance is as follows: Where, represents the headway variance, TH k represents the headway of vehicle k, is the average headway time; The calculation formula for average flow is as follows: Where, is the average flow rate of the target road section, q i 、l i is the flow and length of road segment i; The calculation formula of speed variation coefficient is as follows: Where C v is the speed variation coefficient, S v is the standard deviation of vehicle speed, v k is the speed of vehicle k, v is the average speed of the vehicles; The formula for calculating the large vehicle mixing rate is as follows: Where R lm is the large vehicle mixing rate, Q l is the number of large vehicles; The calculation formula of the congestion index is as follows: Where, CI i represents the congestion index of road section i, v i 、 are the free flow speed and the average vehicle speed of section i, respectively.

3. The highway traffic status identification method based on vehicle-level macro-micro evaluation according to claim 1 or 2 is characterized in that: According to the traffic state entropy value, the traffic state is divided into four categories: normal, general chaos, chaos and extreme chaos; extreme chaos means that the possibility of a traffic accident is very high and safety control measures must be taken immediately; chaos means that a traffic accident may occur and safety control measures must be taken in time; general chaos means that a traffic accident is unlikely to occur and no safety control measures are necessary; normal means that traffic accidents are basically unlikely to occur and no measures need to be taken.

Images