Highway scene-oriented low-speed vehicle grading early warning method

Abstract

The invention discloses a low-speed vehicle grading early warning method oriented to an expressway scene. The method comprises the steps that data of ETC gantries are collected and fused to obtain a to-be-processed data set; performing section passing speed feature recognition according to the driving speeds of the vehicles of different vehicle types under the condition of different sections to obtain a section passing speed feature data set; calculating the time difference when the vehicle passes through the service area and updating the passing time length data in the track data; calculating a vehicle passing speed and comparing the vehicle passing speed with a road passing speed under the same condition to judge whether a low-speed vehicle exists; constructing a subminiature membership function to calculate the membership of the current vehicle; calculating weight vectors of the driving behavior safety factors based on the scale table of the mutual proportions of the driving behavior safety factors; and calculating a vehicle section driving behavior score based on the vehicle membership degree and the weight vectors of the factors influencing the driving behavior safety so as to divide different early warning levels. According to the invention, low-speed vehicles are accurately identified and graded early warning is performed.

Description

Technical Field

[0001] The present invention relates to the technical field of highway management, and in particular to a low-speed vehicle graded warning method for highway scenarios. Background Art

[0002] The presence of low-speed vehicles on highways seriously affects the overall traffic efficiency, especially during peak traffic hours, when this impact is more significant and often causes large-scale traffic jams. What is more serious is that low-speed vehicles are very likely to cause vicious traffic accidents such as chain rear-end collisions. In order to effectively solve this problem, it is particularly important to establish a fast and accurate low-speed vehicle identification system. The establishment of this system can provide data support for the refined management of the road network, and can also significantly improve the response speed of emergency management departments, provide a basis for the formulation of scientific emergency strategies, and thus better protect the lives and property of drivers and passengers.

[0003] Existing low-speed vehicle identification methods mostly rely on technologies such as video surveillance, manual inspection, and vehicle positioning. However, these methods face a series of problems in application:

[0004] 1. Manual inspection is inefficient and time-consuming: Manual inspection requires a lot of manpower and material resources, and is limited by the experience and attention of inspectors, making it difficult to achieve real-time monitoring of all roads and all weather conditions. On long-distance highways, it takes a long time for inspection vehicles to go back and forth, making it difficult to detect and deal with low-speed vehicles in a timely manner. In addition, under severe weather conditions, the efficiency and safety of manual inspections will be greatly reduced.

[0005] 2. Video surveillance has blind spots: The existing video surveillance system is limited by the camera installation position and angle, and it is difficult to cover all lanes and sections. The monitoring effect is often not ideal on special sections such as bends and ramps. At the same time, the recognition accuracy of video surveillance is significantly reduced at night or in rainy and foggy weather. The monitoring blind spots between multiple cameras also make it difficult to continuously track low-speed vehicles.

[0006] 3. Low coverage of vehicle positioning: There are obvious limitations in relying on vehicle positioning systems for low-speed vehicle identification. First, not all vehicles are equipped with positioning equipment, especially some old vehicles and freight vehicles. Secondly, positioning signals are easily interfered with in special sections such as tunnels and mountainous areas, resulting in data loss or deviation. In addition, the data standards of positioning equipment from different manufacturers are not unified, which also makes data integration difficult.

[0007] 4. Misjudgment caused by vehicles entering and exiting service areas: When a vehicle normally enters a service area, its speed change can easily be misjudged by the system as low-speed driving. Existing recognition algorithms often have difficulty distinguishing between vehicles that normally enter a service area and vehicles that are truly driving at a low speed, resulting in a high false alarm rate. Summary of the invention

[0008] The purpose of the present invention is to provide a low-speed vehicle graded warning method for highway scenarios, based on the transaction data collected by the ETC gantry on the highway, integrated with multi-source data such as service area snapshot records, and using statistical algorithm evaluation indicators to construct a low-speed vehicle graded warning algorithm to achieve graded warning for low-speed vehicles, and provide traffic management departments and emergency rescue departments with efficient and accurate warning information for low-speed vehicles on highways.

[0009] The technical solution adopted by the present invention is:

[0010] A low-speed vehicle graded warning method for highway scenarios comprises the following steps:

[0011] Step 1: Data collection and preprocessing: real-time data collection and fusion processing of each ETC gantry to obtain the data set to be processed. The collected data includes ETC system transaction data, ETC service area data and highway section QD;

[0012] Step 2: Mining the speed characteristics of highway sections: identifying the speed characteristics of sections according to the driving speeds of different types of vehicles on the highway under different conditions, mining the speed characteristics of vehicles to obtain the interval speed characteristic data set;

[0013] Step 3: Processing abnormal service area data: Obtaining service area data and vehicle trajectory data , calculate the time difference of vehicles passing through the service area and update the vehicle trajectory data Eliminate the impact of service area travel time on vehicle trajectory data Time interference;

[0014] Step 4: Low-speed vehicle identification: based on vehicle trajectory data Calculate the corresponding vehicle speed , which corresponds to the vehicle speed Road speed under the same conditions as in the interval speed feature dataset Compare to determine whether the current vehicle is a low-speed vehicle;

[0015] Step 5, low-speed vehicle graded scoring warning: construct a relatively small membership function as the membership function of low-speed vehicles to calculate the current vehicle membership; at the same time, the weight vector of factors affecting driving behavior safety is calculated based on the proportional scale table between factors affecting driving behavior safety; the current vehicle section driving behavior score is calculated based on the current vehicle membership and the weight vector of factors affecting driving behavior safety, so as to divide different warning levels based on the section driving behavior score.

[0016] Furthermore, step 2 specifically includes the following steps:

[0017] Step 2-1, the data set is classified by hour to obtain the overall road speed in different time periods;

[0018] Specifically, the data set can be divided into 24 categories.

[0019] Step 2-2, based on the hourly classification, the data is further refined according to the vehicle type vehclass, which can effectively solve the problem of different travel speeds of different vehicle types.

[0020] Step 2-3: According to the spatial regularity, the data set is divided into vehicle trajectory data Different segment traffic data sets are obtained through classification, so as to deeply explore the spatial regularity of vehicles.

[0021] Specifically, due to external conditions such as the number of lanes and road conditions, the speed of vehicles in different sections of highways will also be different. Therefore, according to the spatial regularity, the data set is divided into vehicle trajectory data , classified into different segment traffic data sets, so as to deeply explore the spatial regularity of vehicles.

[0022] Step 2-4, extract the vehicle's travel time in the target section through the divided section traffic data set; at the same time, extract the travel distance of the target section through the gantry distance matrix D, and use the sensorless speed measurement model to calculate the driving speed of all vehicles in the data set.

[0023] Specifically, the elements in the gantry distance matrix D are is the distance between the i-gantry and the j-gantry. If the two are connected, dis is the distance between the two gantry. If they are the same node, it is 0. The distance between unconnected nodes is inf. The elements in the gantry distance matrix D are The expression is as follows:

[0024] (1);

[0025] Step 2-5, based on the data cleaning model of the box plot, outlier abnormal data outside the upper and lower edges are deleted;

[0026] Step 2-6, calculate the average driving speed of all vehicles as the traffic speed of the road , thus approaching the true value more accurately and effectively; the speed of the road The calculation formula is as follows:

[0027] ;

[0028] in Represents vehicles traveling on this road section , Represents the total number of vehicles on this road section.

[0029] Step 2-7: Add the road speed to the highway section speed feature dataset In order to further eliminate the impact of vehicles entering the service area, the following capture technology and ETC data fusion technology are used to eliminate the impact of the service area.

[0030] Furthermore, step 4 specifically includes the following steps:

[0031] Step 4-1: Extract vehicle trajectory data and transaction data to calculate the vehicle's section speed ;

[0032] Step 4-2: Set the vehicle's section speed and section speed characteristic dataset The speed of the road under the corresponding conditions Compare and determine whether there is low-speed driving behavior; if yes, record the passing information of this low-speed driving and execute step 4-3; otherwise, execute step 4-3;

[0033] Specifically, in step 4-2, the average driving speed of all vehicles is calculated as the traffic speed of the road. ;

[0034] Step 4-3, determine whether the traversal of all vehicle trajectories is completed; if so, output the record results; otherwise, execute step 4-1.

[0035] Furthermore, step 5 specifically includes the following steps:

[0036] Step 5-1, construct a relatively small membership function as the membership function of low-speed vehicles. The expression of the membership function of low-speed vehicles is as follows:

[0037]

[0038] in, Indicates the vehicle speed; The speed of the road in the highway section speed characteristic data set that is the same as the current situation;

[0039] Step 5-2, calculate and obtain the low-speed membership of the current vehicle ;

[0040] Step 5-3, constructing a judgment matrix R based on the scale table of the mutual proportions of the factors affecting the safety of driving behavior, wherein each element value in the judgment matrix R represents the scale of the mutual proportions of different factors affecting the safety of driving behavior;

[0041] Step 5-4, normalize the column vector of the judgment matrix R to obtain the normalized judgment matrix ; Calculate the normalized judgment matrix The mean of the row vectors gives the row vector , as the weight vector of the driving behavior safety factor of the row.

[0042] Step 5-5, calculate the current vehicle's segment driving behavior score based on the weights of different driving behavior safety factors; the expression of the specific segment driving behavior score is as follows:

[0043] ;

[0044] in, represents the driving behavior score of the vehicle in the segment; A is the set of vehicles with dangerous driving behaviors in the segment, is the membership degree of vehicle a’s driving behavior, is the driving behavior weight of vehicle a.

[0045] Step 5-6, classifying the vehicle trip into different warning levels according to the vehicle's section driving behavior score.

[0046] Furthermore, the vehicle speed classification standard corresponding to the membership function of the low-speed vehicle in step 5 is: in the first section, when the driving speed of the motor vehicle is greater than 80% of the road traffic level, it is considered to be at a normal level and the membership is 0; in the second section, when the driving speed of the motor vehicle is less than or equal to 80% of the road traffic speed and greater than 60% of the road traffic speed, the membership of this driving behavior is calculated; in the third section, when the driving speed of the motor vehicle is less than or equal to 60% of the road traffic speed and greater than 40% of the road traffic speed, the membership of this driving behavior is calculated; in the fourth section, when the vehicle's traffic speed is less than or equal to 40% of the road traffic speed, the membership is 1.

[0047] The present invention adopts the above technical solution, through intelligent data processing and algorithm, combined with ETC gantry data, road network topology and service area snapshot data, etc., to judge the stranded vehicles in real time and accurately identify them. According to factors such as different vehicle models and road sections, a flexible low-speed time threshold is designed to ensure accurate identification of low-speed vehicles in different road sections. After the stranded vehicle is identified, the system automatically generates early warning information and sends alarms to traffic management departments and emergency rescue personnel in a timely manner to provide support for accurate rescue and safety assurance.

[0048] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1) Utilize service area information to avoid a large number of false alarms: Utilize the fusion of service area snapshot records and ETC transaction records to filter out low-speed vehicles that are mistakenly identified when entering the service area, solving the problem of misidentification of low-speed vehicles due to vehicles entering the service area. 2) Low computational complexity, real-time calculation: Compared with deep learning algorithms such as neural networks, it has the advantages of small parameters and low computational complexity. In addition, due to its extremely low computational complexity, it can be used for real-time early warning of low-speed vehicles. 3) Strong interpretability: Compared with deep learning methods, this statistical method has strong interpretability and can provide managers with a clearer theoretical basis. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments;

[0050] Figure 1 It is a schematic diagram of the flow chart of the section speed feature identification of the present invention;

[0051] Figure 2 A schematic diagram of the process of low-speed driving behavior recognition of the present invention;

[0052] Figure 3 This is a schematic diagram of a relatively small S function;

[0053] Figure 4 Schematic diagram of the top 10 vehicles with low-speed risk behaviors. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0055] like Figures 1 to 4 As shown in one, the present invention discloses a low-speed vehicle graded warning method for a highway scenario, which specifically comprises the following steps:

[0056] Step 1) Data collection and preprocessing: Use big data technology to collect and integrate data from each ETC gantry in real time, including the following data:

[0057] ETC system transaction data: vehicle traffic records collected by the ETC toll collection system on highways, including vehicle entry and exit information at toll stations. It can be defined as vehicle trajectory data : The gantry trajectory formed by the vehicle passing through section QD is called , the expression is defined as follows:

[0058] ;

[0059] in Represents vehicle trajectory data The starting point, Represents vehicle trajectory data The end point.

[0060] ETC service area data: the capture rules of vehicles entering and leaving the service area, including vehicle license plate information, vehicle entry and exit time, service area latitude and longitude, and service area name.

[0061] ;

[0062] ;

[0063] in, Represents the entire service area dataset, represents the captured data of service area s, is the minimum capture unit data of service area s, Indicates the number of unit data captured; Indicates a license plate; is the trip id of the vehicle; Indicates the time of entering the service area; Indicates the time of service area;

[0064] Expressway section QD: Each expressway gantries, toll station entrances and exits (including inter-provincial entrances and exits) are collectively referred to as nodes. Two adjacent nodes constitute a expressway section QD, referred to as a section, which is represented as follows

[0065] ;

[0066] Among them, QD is the highway section, Node1 is the starting point of the section, and Node2 is the end point of the section.

[0067] Step 2) Mining the speed characteristics of highway sections: Based on the driving speeds of different types of vehicles on the highway under different conditions in different sections, a section speed feature recognition algorithm is established to mine the speed characteristics of the vehicles and obtain the interval speed feature data set.

[0068] Due to the spatial regularity and time-varying nature of driving behavior, the degree of danger of vehicles driving in different sections of the highway with the same driving behavior is different, and the speed limits for different types of vehicles are also different, and the corresponding speed limit information is not static. Therefore, the present invention establishes a section speed feature recognition algorithm based on the driving speed of vehicles of different types of vehicles on the highway under conditions in different sections, mines the speed features of the vehicles, and obtains the interval speed feature data set, so as to more flexibly identify the driving behavior features of the vehicle. The specific process is as follows:

[0069] 1) As highway travel is subject to time constraints, mainly due to factors such as morning and evening rush hours and weather conditions, the data set is classified by hour to more effectively extract time features, thereby dividing the data into 24 categories. The overall road speed in different time periods.

[0070] 2) According to national regulations, vehicles currently traveling on highways are divided into 18 categories, the specific classifications are shown in Table 1.

[0071] Table 1 Vehicle classification table

[0072]

[0073] Different types of vehicles have different travel speeds on the same road. Therefore, on the basis of dividing the data set based on hours, the data is further divided according to the vehicle type vehclass. Through the refined division of vehicle types, the problem of different travel speeds of different vehicle types can be effectively solved.

[0074] 3) Due to external conditions such as the number of lanes and road conditions, the speed of vehicles in different sections of highways will also be different. Therefore, according to the spatial regularity, the data set is divided into vehicle trajectory data , classified into different segment traffic data sets, so as to deeply explore the spatial regularity of vehicles.

[0075] 4) According to the divided data set and the gantry distance matrix D, the travel time of the vehicle in the target section is extracted through the data set, and the travel distance of the target section is extracted through the distance matrix D. The sensorless speed measurement model is used to calculate the driving speed of all vehicles in the data set.

[0076] The process of calculating QD distance using Amap API is as follows: First, extract the connected portals in the adjacency matrix G ( , ), use the above mast positioning method to get the accurate mast position , The gantry’s geographic location is input into the Amap API to obtain the actual driving distance of the expressway QD.

[0077] In order to effectively store and determine the actual driving distance of QD, the present invention constructs a gantry distance matrix D, is the distance between the i-gantry and the j-gantry. If the two are connected, dis is the distance between the two gantries. If they are the same node, it is 0. The distance between unconnected nodes is inf. The calculation formula is shown in Formula 1 below, and the matrix is ​​constructed as shown in Table 2.

[0078] (1);

[0079] Table 2 Distance matrix

[0080]

[0081] 5) In order to calculate the average speed of the road, the arithmetic mean method is used to effectively remove the maximum value in the data set, so as to calculate the speed more accurately. The formula for calculating the average speed of different types of vehicles is shown below.

[0082] ;

[0083] in, For car models The average speed of vehicle p in For car models The total number of vehicles, For vehicle type Average speed;

[0084] Table 3 Average driving speed of different vehicle types

[0085]

[0086] When estimating the road speed, the average speed of vehicles passing through the section within a certain time range is generally used as the estimated value, so reducing outliers can effectively improve the estimation accuracy. However, the method mentioned above can only effectively remove large outliers, and cannot effectively remove situations such as stopping at service areas to reduce vehicle speed.

[0087] Based on the data cleaning model of the box plot, it is necessary to delete the outlier abnormal data outside the upper and lower edges, which can effectively remove the excessive or low speed caused by unexpected situations or entering the service area, and then use the following formula to calculate the average driving speed of all vehicles as the road speed , thus approaching the true value more accurately and effectively.

[0088] ;

[0089] in Represents vehicles traveling on this road section , Represents the total number of vehicles on this road section.

[0090] Add the road speed to the highway section speed feature dataset In order to further eliminate the impact of vehicles entering the service area, the following capture technology and ETC data fusion technology are used to eliminate the impact of the service area.

[0091] Step 3) Service area abnormal data processing: Get service area data and vehicle trajectory data , calculate the time difference of vehicles passing through the service area and update the vehicle trajectory data Eliminate the impact of service area travel time on vehicle trajectory data Time interference;

[0092] Specifically, the service area is extracted from the service area data Snapshot records, matching through the service area Vehicles The trajectory data of the vehicle is calculated Time difference in entering and leaving the service area; updating vehicles Passing the service area The section The travel time (excluding the time spent in the service area) is updated to The vehicle trajectory travel time data avoids interference with vehicle speed calculation.

[0093] Step 4) Low-speed vehicle identification: Low-speed driving on highways has a great impact on road safety. Therefore, the present invention proposes a low-speed driving behavior feature recognition algorithm for highways.

[0094] 1) Based on vehicle trajectory data And the gantry distance matrix D established above, the segment driving distance of the vehicle is obtained and the travel time time, the segment speed of the vehicle is calculated using the sensorless speed model , the calculation formula is shown below.

[0095] ;

[0096] 2) The speed of vehicles and Under the same traffic conditions Compare and judge whether the vehicle is speeding. If 0.8 is exceeded, it is judged that there is low-speed driving behavior and the traffic information of this time is recorded; otherwise, When , it is determined that the vehicle does not have low-speed driving behavior.

[0097] 3) Determine whether the traversal is complete and the vehicle trajectory data , output the vehicle’s low speed information.

[0098] Step 5) Low-speed vehicle classification scoring warning: The present invention uses the large and small S-type membership function as a driving behavior risk degree quantification model, and its function diagram is shown as follows: Figure 2 , a relatively small membership function is used for low-speed driving.

[0099] Depend on Figure 3 As shown in the figure, the membership function is divided into four parts. The first part is relatively smooth, the second part starts to drop sharply, the third part drops slowly, and the fourth part is parallel to the x-axis. The reasons for choosing this type of membership function are summarized as follows: when the speed is within a reasonable range, it is basically not a dangerous driving behavior, and the penalty factor is small at this time. As the speed continues to decrease, the penalty factor increases, the curve becomes steeper, and the function value continues to increase, eventually reaching 1. At the same time, this membership function quantifies the degree of danger of dangerous driving to the interval of [0,1], thereby effectively realizing the quantification of the risk level of driving behavior.

[0100] After the above analysis of the driving characteristics of highway vehicles, combined with the membership function selected above, the membership function is specifically determined to calculate the low-speed membership of the vehicle:

[0101] ;

[0102] in, It is the road speed in the highway section speed characteristic dataset that is the same as the current situation.

[0103] As shown in the above formula, the present invention divides the membership function of low-speed driving behavior into four parts. Since the latest traffic law does not determine the classification of low-speed driving, but only uniformly handles vehicles below the prescribed speed limit, the present invention establishes a classification standard in accordance with the speeding rules. In the first section, when the driving speed of a motor vehicle is greater than 80% of the road traffic level, it is considered to be a normal level; in the second section, when the driving speed of a motor vehicle is less than or equal to 80% of the road traffic speed and greater than 60% of the road traffic speed, the membership of this driving behavior is calculated; in the third section, when the driving speed of a motor vehicle is less than or equal to 60% of the road traffic speed and greater than 40% of the road traffic speed, the membership of this driving behavior is calculated; in the fourth section, when the vehicle's traffic speed is less than or equal to 40% of the road traffic speed, the membership is 1.

[0104] While constructing the judgment matrix, the importance of the compared indicators and their impact level on driving safety have a great impact on the values ​​of the elements in the matrix.

[0105] Table 4 Ratio scale table

[0106]

[0107] According to the ratio scale table after the stratification of driving safety, the importance of the bottom-level elements is compared, and the present invention establishes a judgment matrix R=rij(n×n), as shown in Table 5:

[0108] Table 5 Judgment Matrix

[0109]

[0110] Among them, f1, f2, f3, ..., fn represent the various factors that affect driving safety after classification. The elements in the matrix The value of the factor is determined by experts using a proportional scale and factors The importance between them is given by the corresponding importance scale (for example, Factor Ratio If the factor is obviously important, then The scale value of is 5, The scale value is 1 / 5). The table divides the importance scale of the elements into 1-9. The larger the importance scale, the more important the element is relative to another element. In order to further determine the weight of the vehicle and facilitate calculation, the column vector of the matrix R is normalized to obtain the normalized matrix , the results are shown in Table 6. Finally, the matrix The mean of the row vectors in , we get the row vector , as shown in the following formula, as the weight vector of the elements.

[0111] Table 6 Normalized judgment matrix

[0112] in, Represents the value of the element in the second row and first column of the judgment matrix, which is the value in Table 5 value; Represents the judgment matrix Line The value of the element in the column is the value in Table 5 value;

[0113] (2)

[0114] in, Represents the value of the element in the first row and first column of the normalized judgment matrix; The normalized judgment matrix The value of the element in the row and column 1;

[0115] After determining the weights of different driving behaviors, it is necessary to calculate the driving safety of a certain trip of the vehicle according to the calculated weights. The present invention proposes a segment driving behavior scoring model, and the calculation formula is shown in the following formula.

[0116] ;

[0117] in, represents the driving behavior score of the vehicle in the segment; A is the set of vehicles with dangerous driving behaviors in the segment, is the membership degree of vehicle a’s driving behavior, is the driving behavior weight of vehicle a in formula (2).

[0118] The present invention divides the low-speed travel into 4 categories, and the classification meanings are shown in Table 7 below.

[0119] Table 7 Definition of vehicle driving behavior

[0120] Logo meaning 1 The vehicle's current trip is a normal trip 2 The vehicle's current trip is a light low-speed trip 3 The vehicle's current trip is a moderate low-speed risk trip 4 The vehicle's current trip is a severe low-speed risk trip

[0121] In order to effectively verify the validity and reliability of the model, the present invention uses the following formula to convert the above scoring results into the 4 types in the table:

[0122] ;

[0123] Here, x represents the score obtained by the evaluation model above, and C(x) represents the classification result.

[0124] Different scoring results are used to achieve graded and classified warnings for low-speed vehicles.

[0125] We selected some sections of roads from September to December 2024 to evaluate the effectiveness of the algorithm, and detected 2,862 low-speed behaviors of 124 vehicles. The top 10 vehicles with serious low-speed risks are: Figure 4 shown.

[0126] The present invention is particularly suitable for automatically identifying low-speed risk vehicles and issuing safety warnings in severe weather scenarios (such as typhoons, heavy rains, etc.).

[0127] The present invention adopts the above technical solution, through intelligent data processing and algorithm, combined with ETC gantry data, road network topology and service area snapshot data, etc., to judge the stranded vehicles in real time and accurately identify them. According to factors such as different vehicle models and road sections, a flexible low-speed time threshold is designed to ensure accurate identification of low-speed vehicles in different road sections. After the stranded vehicle is identified, the system automatically generates early warning information and sends alarms to traffic management departments and emergency rescue personnel in a timely manner to provide support for accurate rescue and safety assurance.

[0128] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1) Utilize service area information to avoid a large number of false alarms: Utilize the fusion of service area snapshot records and ETC transaction records to filter out low-speed vehicles that are mistakenly identified when entering the service area, solving the problem of misidentification of low-speed vehicles due to vehicles entering the service area. 2) Low computational complexity, real-time calculation: Compared with deep learning algorithms such as neural networks, it has the advantages of small parameters and low computational complexity. In addition, due to its extremely low computational complexity, it can be used for real-time early warning of low-speed vehicles. 3) Strong interpretability: Compared with deep learning methods, this statistical method has strong interpretability and can provide managers with a clearer theoretical basis.

[0129] Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians of the art without making creative work are within the scope of protection of the present application.

Claims

1. A low-speed vehicle graded warning method for highway scenarios, characterized by: The following steps are involved: Step 1, data collection and preprocessing: collect data from each ETC gantry in real time and perform fusion processing to obtain the data set to be processed. The collected data includes ETC system transaction data, ETC service area data and highway section QD; Step 2: Mining the speed characteristics of highway sections: identifying the speed characteristics of sections according to the driving speeds of different types of vehicles on the highway under different conditions, mining the speed characteristics of vehicles to obtain the interval speed characteristic data set; Step 3: Processing abnormal service area data: Obtaining service area data and vehicle trajectory data , calculate the time difference of vehicles passing through the service area and update the vehicle trajectory data Eliminate the impact of service area travel time on vehicle trajectory data Time interference; Step 4: Low-speed vehicle identification: based on vehicle trajectory data Calculate the corresponding vehicle speed , which corresponds to the vehicle speed Road speed under the same conditions as in the interval speed feature dataset Compare to determine whether the current vehicle is a low-speed vehicle; Step 5, low-speed vehicle classification score warning: construct a small-sized membership function as the membership function of low-speed vehicles to calculate the current vehicle membership; at the same time, calculate the weight vector of the factors affecting driving behavior safety based on the mutual proportion scale table of factors affecting driving behavior safety; The current vehicle section driving behavior score is calculated based on the current vehicle membership and the weight vector of the factors affecting the driving behavior safety, so as to divide different warning levels based on the section driving behavior score.

2. The method for hierarchical warning of low-speed vehicles on highways according to claim 1 is characterized by: Step 2 specifically includes the following steps: Step 2-1, the data set is classified by hour to obtain the overall road speed in different time periods; Step 2-2: Based on the hourly classification, the data is further divided according to the vehicle type; Step 2-3: sort the data set into vehicle trajectory data according to spatial regularity Different segment traffic data sets are obtained through classification; Step 2-4: extract the travel time of vehicles in the target section through the divided section traffic data set; at the same time, extract the travel distance of the target section through the gantry distance matrix D, and use the sensorless speed measurement model to calculate the driving speed of all vehicles in the data set; Step 2-5, deleting abnormal data of the vehicle's driving speed based on the data cleaning model of the box plot; Step 2-6, calculate the average driving speed of all vehicles as the traffic speed of the road , the speed of the road The calculation formula is as follows: ； in, Represents vehicles traveling on the road segment , represents the total number of vehicles on the road section; Step 2-7: Add the road speed to the highway section speed feature dataset middle.

3. The method for hierarchical warning of low-speed vehicles on highways according to claim 2 is characterized in that: In step 2-1, the data set is divided into 24 categories.

4. The method for hierarchical warning of low-speed vehicles on highways according to claim 2 is characterized in that: The elements in the gantry distance matrix D in step 2-4 The expression is as follows: (1)； Among them, dis is the distance between the two connected portals, and inf means that the two portal nodes are not connected.

5. The method for hierarchical warning of low-speed vehicles on highways according to claim 1 is characterized by: Step 4 specifically includes the following steps: Step 4-1: Extract vehicle trajectory data and transaction data to calculate the vehicle's section speed ; Step 4-2: Set the vehicle's section speed and section speed characteristic dataset The speed of the road under the corresponding conditions Compare and determine whether there is low-speed driving behavior; if yes, record the passing information of this low-speed driving and execute step 4-3; otherwise, execute step 4-3; Step 4-3, determine whether the traversal of all vehicle trajectories is completed; if so, output the record results; otherwise, execute step 4-1.

6. The method for hierarchical warning of low-speed vehicles on highways according to claim 5 is characterized by: In step 4-2, the average driving speed of all vehicles is calculated as the traffic speed of the road. .

7. The method for hierarchical warning of low-speed vehicles on highways according to claim 1 is characterized by: Step 5 specifically includes the following steps: Step 5-1, construct a relatively small membership function as the membership function of low-speed vehicles. The expression of the membership function of low-speed vehicles is as follows: ； in, Indicates the vehicle speed; The speed of the road in the highway section speed characteristic data set that is the same as the current situation; Step 5-2, calculate and obtain the low-speed membership of the current vehicle ; Step 5-3, constructing a judgment matrix R based on the scale table of the mutual proportions of the factors affecting the safety of driving behavior, wherein each element value in the judgment matrix R represents the scale of the mutual proportions of different factors affecting the safety of driving behavior; Step 5-4, normalize the column vector of the judgment matrix R to obtain the normalized judgment matrix ; Calculate the normalized judgment matrix The mean of the row vectors gives the row vector , as the weight vector of the driving behavior safety factor of the line; Step 5-5, calculate the segment driving behavior score of the current vehicle according to the weights of different driving behavior safety factors; the expression of the specific segment driving behavior score is as follows: ; in, represents the driving behavior score of the vehicle in the segment; A is the set of vehicles with dangerous driving behaviors in the segment, is the membership degree of vehicle a’s driving behavior, is the driving behavior weight vector of vehicle a; Step 5-6, classifying the vehicle trip into different warning levels according to the vehicle's section driving behavior score.

Images