Road traffic intelligent risk early warning method, system, equipment and medium

Abstract

The invention relates to a road traffic intelligent risk early warning method, system, device and medium, and the method comprises the steps: building a vehicle driving risk prediction model and a vehicle safety grading model based on periodically collected historical trajectory data, and building a vehicle risk early warning system; before the vehicle enters the safety early warning area, factors influencing the vehicle accident risk probability are obtained, the self safety level of the current vehicle is predicted according to the vehicle safety grading model, and the safety early warning threshold value of the current vehicle is matched according to the self safety level of the current vehicle; after the vehicle enters the safety early warning area, the safety state of the current vehicle is judged according to the vehicle driving risk prediction model and the safety early warning threshold value of the current vehicle, real-time early warning is carried out, and early warning information is issued; and dynamically correcting the vehicle driving risk prediction model, the vehicle safety grading model and the threshold division in the vehicle risk early warning system every preset period. The method can be widely applied to the technical field of vehicle safety early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle safety warning, and in particular to an intelligent road traffic risk warning method, system, equipment and medium. Background Art

[0002] With the continuous development of expressways, road safety issues are becoming increasingly prominent. More and more systems are predicting vehicle risks. Since vehicles may exhibit different safety states when facing the same driving scenario, and different safety states correspond to different safety levels, risk assessment and safety grading for vehicles can personalize early warning mechanisms at different safety levels, which helps to improve the safety and efficiency of the road traffic system.

[0003] At present, the system for realizing vehicle safety warning is mainly realized in the following ways: 1. For highways, vehicle safety warnings are performed based on sensing devices such as cameras and roadside equipment to obtain vehicle position, motion parameters, and road information. However, this type of method only uses simple influencing factors such as vehicle position, speed, and road conditions, which makes it difficult to accurately and effectively determine the probability of vehicle accident risks.

[0004] 2. Predict the accident ahead based on the current vehicle and the traffic ahead information; classify and grade the accident risk of the current driving state based on the situation of the accident ahead, and generate safety warning information. This type of method mainly uses the traffic ahead information to make a dangerous judgment on the driving state of the current vehicle. It regards all vehicles as the same category, does not grade the vehicle safety status based on the vehicle type, driving characteristics, driver characteristics, etc., and does not divide the traffic conflict judgment threshold and accident probability judgment threshold applicable to different categories of vehicles. Summary of the invention

[0005] In view of the above problems, the purpose of the present invention is to provide a method, system, device and medium for intelligent road traffic risk warning, which can classify the vehicles in the warning area according to their own safety characteristics, and based on the classification results, the conflict substitution index of the vehicles of the corresponding safety level is set. Judgment threshold and vehicle accident risk probability The judgment threshold is divided and then applied to the vehicle safety warning system, which can effectively improve the accuracy of the vehicle safety warning system for dangerous events. It is suitable for smart highway control scenarios, and is helpful for active safety control in scenarios with high-frequency updates of massive data, reducing traffic accidents and ensuring traffic safety.

[0006] To achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a road traffic intelligent risk warning method, comprising the following steps: Based on the periodically collected historical trajectory data, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and a vehicle risk early warning system is established; Before a vehicle enters the safety warning area, the vehicle risk warning system obtains factors that affect the probability of vehicle accident risks, and predicts the vehicle's own safety level based on the vehicle safety grading model, and matches the vehicle's safety warning threshold based on the vehicle's own safety level; When a vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle based on the vehicle driving risk prediction model and the current vehicle safety warning threshold, issues a real-time warning, and releases warning information; At preset intervals, historical trajectory data is obtained to dynamically correct the vehicle driving risk prediction model, vehicle safety classification model and safety warning threshold division in the vehicle risk warning system.

[0007] Furthermore, based on the periodically collected historical trajectory data, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and a vehicle risk warning system is established, including: Delineate road sections or nodes as the boundaries of the safety warning area, and store the boundary information of the safety warning area in a database; Relying on roadside sensing equipment, the historical vehicle trajectory data of the preset distance in front of the safety warning area and the safety warning area are periodically collected, and the historical vehicle trajectory data is preprocessed and parameter extracted to obtain candidate feature variable parameters; Based on the candidate characteristic variable parameters, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and the traffic conflict warning threshold and the accident risk probability warning threshold are divided; Import the vehicle driving risk prediction model and vehicle safety classification model into the vehicle risk warning system to complete the construction.

[0008] Furthermore, based on the candidate characteristic variable parameters, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and traffic conflict warning thresholds and accident risk probability warning thresholds are divided, including: Based on the candidate characteristic variable parameters, the factors affecting the accident risk probability of the vehicle before and within the safety warning area are screened, and based on the screened factors affecting the accident risk probability of the vehicle, the vehicle driving risk prediction models for the vehicle before and within the safety warning area are respectively constructed; Based on the vehicle driving risk prediction model facing the safety warning area, the driving risk-related variable parameters of the vehicle before entering the safety warning area are screened, and the K-means clustering method is used to perform cluster analysis on the screened variable parameters to obtain the vehicle safety classification results and build a vehicle safety classification model; According to the vehicle safety classification results, the traffic conflict warning threshold and the accident risk probability warning threshold are divided to obtain the traffic conflict warning threshold and the accident risk probability warning threshold corresponding to each safety level.

[0009] Furthermore, the vehicle driving risk prediction model for the safety warning area is used to screen the variable parameters related to the vehicle driving risk before entering the safety warning area, and the K-means clustering method is used to perform cluster analysis on the screened variable parameters to obtain the vehicle safety classification result, and a vehicle safety classification model is constructed, including: According to the vehicle driving risk prediction model constructed for the safety warning area, the correlation between various factors and vehicle risk is obtained based on historical sample data; Based on the correlation in the above steps and the data conditions obtained before the vehicle enters the warning area, the influencing factors obtained before the vehicle enters the safety warning area are screened as vehicle safety evaluation indicators; The K-means clustering method is used to perform cluster analysis on vehicle safety evaluation indicators, obtain vehicle safety classification results, and build a vehicle safety classification model.

[0010] Furthermore, the traffic conflict warning threshold and the accident risk probability warning threshold are divided according to the vehicle safety classification result to obtain the traffic conflict warning threshold and the accident risk probability warning threshold corresponding to each safety level, including: The accident risk probability warning threshold and the traffic conflict warning threshold interval are divided into a plurality of sub-intervals according to preset interval lengths respectively; Calculate the probability and information entropy of each subinterval respectively; Find the threshold corresponding to the maximum value of information entropy as the result of dividing the traffic conflict warning threshold and the accident risk probability warning threshold.

[0011] Furthermore, when the vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle according to the vehicle driving risk prediction model and the safety warning threshold of the current vehicle, and issues warning information in real time, including: According to the real-time collected data, the factors affecting the vehicle accident risk probability of the current vehicle are obtained, and the vehicle safety classification model is used to predict the current vehicle's own safety level and the matching traffic conflict warning threshold and accident risk probability warning threshold; The vehicle risk warning system uses the roadside sensing equipment to obtain the vehicle's position in the warning area, and uses the vehicle's position information and the matching traffic conflict warning threshold and accident risk probability warning threshold to predict the current vehicle's driving risk. When the vehicle risk warning system determines that there is a collision risk in any way, the vehicle risk warning system sends risk warning information to the cloud control management platform and the vehicle terminal.

[0012] Furthermore, when the vehicle risk warning system predicts the driving risk of the current vehicle, it includes two warning modes: The first method is to determine whether the vehicle has a collision risk by calculating the conflict substitution index of the current vehicle in real time. If the conflict substitution index of the current vehicle is less than the vehicle traffic conflict warning threshold corresponding to its own safety level, it is determined that a collision may occur. The second method is to use the constructed vehicle driving risk prediction model to estimate the accident risk probability of the current vehicle. If the accident risk probability of the current vehicle is greater than the accident risk probability warning threshold corresponding to its own safety level.

[0013] In a second aspect, the present invention provides an intelligent road traffic risk warning system, comprising: Model building module, which is used to build a vehicle driving risk prediction model and a vehicle safety classification model based on periodically collected historical trajectory data, and to build a vehicle risk warning system; The vehicle safety classification and threshold matching module is used for the vehicle risk warning system to obtain factors that affect the vehicle accident risk probability before the vehicle enters the safety warning area, and predict the current vehicle's own safety level based on the vehicle safety classification model, and match the current vehicle's safety warning threshold according to the current vehicle's own safety level, where the safety warning threshold includes the traffic conflict warning threshold and the accident risk probability warning threshold; Warning module, used when the vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle according to the vehicle driving risk prediction model and the safety warning threshold of the current vehicle, issues a real-time warning, and releases warning information; The update module is used to obtain historical trajectory data at preset intervals to dynamically correct the vehicle driving risk prediction model, vehicle safety classification model and threshold division in the vehicle risk warning system.

[0014] In a third aspect, the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, cause the computing device to perform any method.

[0015] In a fourth aspect, the present invention provides a computing device, comprising: one or more processors and a memory, wherein the memory stores one or more programs and is configured to be executed by the one or more processors, and the one or more programs include instructions for executing any method.

[0016] The present invention adopts the above technical solution, which has the following advantages: The present invention aims to alleviate the traffic safety problem of frequent accidents on highways. In a vehicle-road collaborative environment, through the information transmission between vehicles and between vehicles and roads, the risk and safety level of vehicle driving can be accurately and effectively extracted and judged. First, based on historical data, a vehicle driving risk prediction model for the safety warning area and a vehicle safety level discrimination model before entering the warning area are established. Different warning thresholds are selected for each type of vehicle with each safety level in the safety warning area to make continuous safety status judgments and issue warnings, which can enable more accurate safety management of vehicles and effectively improve the accuracy of the vehicle safety warning system for dangerous events. Periodically updating sample data and dynamically correcting models and warning thresholds can effectively update data at a high frequency, making it easy to capture rapid changes in traffic environment and traffic flow, thereby improving prediction accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Throughout the accompanying drawings, the same reference numerals are used to represent the same components. In the accompanying drawings: Figure 1 It is a flow chart of a method for intelligent road traffic risk warning based on vehicle safety classification provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0019] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0020] The existing vehicle safety warning system mainly focuses on the prediction of vehicle accident risks in conventional road environments. There is still a lack of research on highways with complex traffic flow environments. There is a lack of differentiated warnings for vehicles based on safety classification, and there is a lack of conflict substitution indicators based on vehicles of different safety levels. Judgment threshold and vehicle accident risk probability Subdivision of judgment threshold.

[0021] Therefore, in some embodiments of the present invention, a method for intelligent risk warning of road traffic is proposed. The method comprehensively considers the vehicle type, driving characteristics, driver characteristics and surrounding driving environment characteristics, and uses historical data to establish a vehicle driving risk prediction model for the safety warning area and a vehicle safety level discrimination model before entering the warning area. When the vehicle enters the safety warning area, the vehicle safety level is judged and the appropriate safety warning threshold is matched. The safety risk prediction of the vehicle in the safety warning area is continuously performed, and timely warning is issued. At the same time, the traffic conflict threshold and accident risk probability threshold of vehicles with corresponding safety levels are regularly optimized. The method of the present invention is conducive to improving the effectiveness of the warning information of the safety warning system, reducing the misjudgment of the safety status of low-risk vehicles, and is suitable for smart highway control scenarios, which is conducive to active safety control in scenarios with high frequency updates of massive data.

[0022] Correspondingly, in other embodiments of the present invention, a road traffic intelligent risk warning system, equipment and medium are provided.

[0023] Example 1 like Figure 1 As shown, the present invention provides a road traffic intelligent risk warning method, which includes the following steps: S1. Based on the periodically collected historical trajectory data, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and a vehicle risk early warning system is established; S2. Before the vehicle enters the safety warning area, the vehicle risk warning system obtains the factors that affect the vehicle accident risk probability, and predicts the current vehicle's own safety level based on the vehicle safety classification model. , according to the current vehicle's own safety level, the current vehicle's safety warning threshold is matched, where the safety warning threshold includes the traffic conflict warning threshold and accident risk probability warning threshold ; S3. When a vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle based on the vehicle driving risk prediction model and the current vehicle safety warning threshold, issues a real-time warning, and releases warning information; S4. At each preset period, historical trajectory data is obtained to dynamically modify the vehicle driving risk prediction model, vehicle safety classification model and safety warning threshold division in the vehicle risk warning system.

[0024] Furthermore, in the above step S1, based on the periodically collected historical trajectory data, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and a vehicle risk warning system is built, including the following steps: S11. Delineate road sections or nodes as the boundaries of the safety warning area, and store the boundary information of the safety warning area in the database; S12. Relying on roadside sensing equipment, periodically collect historical vehicle trajectory data within a preset distance S (S=50 meters) in front of the safety warning area and within the safety warning area, and preprocess and extract parameters of the historical vehicle trajectory data to obtain candidate feature variable parameters; S13. Based on the candidate characteristic variable parameters, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and the traffic conflict warning threshold and the accident risk probability warning threshold are divided; S14. Import the vehicle driving risk prediction model and vehicle safety classification model into the vehicle risk warning system to complete the construction.

[0025] Furthermore, in the above step S11, the road sections or nodes are delineated as the boundaries of the safety warning area, and the boundary information of the safety warning area is stored in the database, including: S111. Use GPS or other positioning systems to accurately obtain the latitude and longitude coordinate information of the boundary points of the road section or node as the safety warning area; S112. Based on the collected latitude and longitude coordinate information of all boundary points, a closed polygonal area is formed as a safety warning area and stored in the database.

[0026] Furthermore, the present embodiment can also simultaneously collect information such as road names, road section stake numbers, etc., which helps to more comprehensively describe the safety warning area.

[0027] Furthermore, in the above step S12, relying on the roadside sensing equipment, historical vehicle trajectory data of the preset distance before the safety warning area and within the safety warning area are periodically collected, and the historical vehicle trajectory data is preprocessed and parameter extracted, including: S121. Relying on roadside sensing equipment, periodically collect vehicle trajectory data at a preset distance in front of the safety warning area and within the safety warning area.

[0028] In this embodiment, the roadside sensing device generally refers to a fixed roadside sensing device, including RSU, camera, millimeter wave radar, laser radar, radar-vision integrated device, etc. Such sensing devices can sense moving vehicles to obtain trajectory data and perform simple data processing; the historical vehicle trajectory data obtained by the roadside sensing device includes the position coordinates of the vehicle in each frame, vehicle ID information, etc.; S122. Perform parameter extraction on the vehicle trajectory data, driver information, and actual environmental conditions acquired by the roadside perception device to obtain candidate feature variable parameters.

[0029] In this embodiment, the extracted candidate feature variable parameters include: ① Vehicle motion parameters: vehicle type 、Real-time vehicle location , The lane the vehicle was in before entering the warning area , the initial speed of the vehicle before entering the warning area , Real-time vehicle speed , vehicle real-time acceleration , vehicle acceleration standard deviation 、The lane the vehicle is in in real time .

[0030] ②Driver parameters: Get the driver's gender by matching the database ,age ,Profession , Driving experience .

[0031] ③Environmental parameters: real-time traffic flow , average headway of traffic flow , average speed of traffic flow , traffic flow speed standard deviation , average acceleration of traffic flow , traffic flow acceleration standard deviation , Road width in the safety warning area , Road smoothness coefficient , Lane width , Number of lanes ,weather .

[0032] Furthermore, in the above step S13, based on the candidate characteristic variable parameters, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and the traffic conflict warning threshold and the accident risk probability warning threshold are divided, including the following steps: S131. Based on the candidate characteristic variable parameters, the factors affecting the accident risk probability of the vehicle before and within the safety warning area are screened, and based on the screened factors affecting the accident risk probability of the vehicle, a vehicle driving risk prediction model for the vehicle before and within the safety warning area is constructed respectively; S132. Based on the vehicle driving risk prediction model facing the safety warning area, the driving risk-related variable parameters of the vehicle before entering the safety warning area are screened, and the K-means clustering method is used to perform cluster analysis on the screened variable parameters to obtain the vehicle safety classification results, and a vehicle safety classification model is constructed; S133. Divide the traffic conflict warning threshold and the accident risk probability warning threshold according to the vehicle safety classification result to obtain the traffic conflict warning threshold and the accident risk probability warning threshold corresponding to each safety level.

[0033] Furthermore, in the above step S131, constructing a vehicle driving risk prediction model includes: ① Use Pearson correlation to analyze the correlation between each pair of feature variable parameters among all candidate feature variable parameters, screen the feature variable groups with strong correlation, retain one of the feature variables, and obtain the vehicle accident risk probability factors before and within the safety warning area.

[0034] ② Based on the vehicle accident risk probability factors in the screened safety warning area, a vehicle driving risk prediction model for the safety warning area is constructed.

[0035] In this embodiment, the vehicle driving risk prediction model constructed for the safety warning area is: (1) (2) In the formula, is a constant term, Indicates the model coefficients to be fitted, different React right Different degrees of impact. = , represents m factors that affect the vehicle driving risk; the vehicle driving risk state in the warning area ( , 1) As the dependent variable, Y is the vehicle accident risk probability P, when the predicted value , it means the vehicle is at risk. , it means the vehicle is safe.

[0036] ③ Based on the vehicle accident risk probability factors before the screened safety warning area, a vehicle driving risk prediction model for the safety warning area is constructed.

[0037] In this embodiment, the vehicle driving risk prediction model constructed for the safety warning area is: (3) (4) In the formula, is a constant term, Indicates the model coefficients to be fitted, different React right Different degrees of impact. = , represents m factors that affect the vehicle driving risk; the driving risk state of the vehicle outside the warning area ( , 1) As the dependent variable, Y' is the vehicle accident risk probability P, when the predicted value , it means the vehicle is at risk. , it means the vehicle is safe.

[0038] Furthermore, in the above step S132, constructing a vehicle safety classification model includes the following steps: ① According to the vehicle driving risk prediction model constructed for the safety warning area, the correlation between various factors and vehicle risk is obtained based on historical sample data; ② Based on the correlation in step ① and the data conditions obtained before the vehicle enters the warning area, the influencing factors obtained before the vehicle enters the safety warning area are screened as vehicle safety evaluation indicators .

[0039] ③The K-means clustering method is used to perform cluster analysis on vehicle safety evaluation indicators, obtain vehicle safety classification results, and construct a vehicle safety classification model.

[0040] Specifically, the steps include: According to previous research, the vehicle's own safety performance is usually divided into four levels: extremely low accident risk (A), normal accident risk (B), critical accident risk (C), and extremely high accident risk (D), that is, A represents the highest safety level and D represents the lowest safety level. Therefore, in this embodiment, select Vehicle samples are used as initial cluster centers = .

[0041] a. Initialize cluster centers: The eigenvector of a cluster center is = ; b. Calculate the distance: Calculate the distance between each vehicle sample and each cluster center, and assign each vehicle sample to the cluster center closest to it; No. The vehicle samples and The distance between cluster centers is: (3) In the formula, is the feature vector of a sample in the set, It is The characteristic vector of the cluster centers, is the number of features; c. Update cluster center: For each cluster, calculate the mean of all sample feature vectors assigned to the cluster as the new cluster center; The calculation formula of the mean is: (4) in, is the set of all samples assigned to the nth class, is the number of samples in the set, is the feature vector of a sample in the set; d. Repeat steps b to c above until the change of the cluster center in two consecutive iterations is less than a preset threshold, then stop the calculation, obtain the vehicle safety classification result, and establish a vehicle safety classification model.

[0042] Furthermore, in the above step S133, the traffic conflict warning threshold and the accident risk probability warning threshold are divided according to the vehicle safety classification result, including: ① Divide the accident risk probability warning threshold and traffic conflict warning threshold intervals into several sub-intervals according to the preset interval lengths.

[0043] In this embodiment, when the maximum entropy method is used, the accident risk probability P warning threshold is divided into 9 intervals of 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9, respectively; the accident risk probability P is divided into 10 sub-intervals. The vehicle traffic conflict ETTC warning threshold is 1, 2, 3, 4, 5, 6, 7, 8, and 9; the vehicle traffic conflict ETTC is divided into 10 sub-intervals.

[0044] ② Calculate the probability of each sub-interval separately and information entropy.

[0045] The probability of each subinterval is The calculation formula is: = (5) in, is the total number of data sets, represents the indicator function, when Falling in the range It takes 1 if yes, otherwise it takes 0.

[0046] According to probability Calculating Information Entropy :

[0047] ③ Find the threshold corresponding to the maximum value of information entropy as the result of dividing the traffic conflict warning threshold and the accident risk probability warning threshold.

[0048] Maximizing information entropy The corresponding threshold :

[0049] Determine the vehicle traffic conflict warning threshold under four safety levels using the maximum entropy method , , , and vehicle accident risk probability warning threshold = , , , .in, , , , They represent the vehicle's own safety respectively. Vehicle traffic conflict warning threshold at level 1; , , , They represent the vehicle's own safety level. The vehicle accident risk probability warning threshold at level 1 is shown in Table 1 below.

[0050] Table 1 Vehicle traffic conflict warning thresholds and accident risk probability thresholds of vehicle safety levels

[0051] Furthermore, the above step S2 includes the following steps: S21. Obtain the factors that affect the vehicle accident risk probability of the current vehicle based on the real-time collected data, and use the vehicle safety classification model to predict the current vehicle's own safety level and the matching traffic conflict warning threshold and accident risk probability warning threshold; S22. The vehicle risk warning system obtains the position of the vehicle in the warning area through the roadside sensing device, and uses the vehicle position information and the matching traffic conflict warning threshold and accident risk probability warning threshold to predict the driving risk of the current vehicle; S23. When the vehicle risk warning system determines that there is a collision risk through any method, the vehicle risk warning system sends risk warning information to the cloud control management platform and the vehicle terminal.

[0052] Furthermore, in the above step S22, when the vehicle risk warning system predicts the driving risk of the current vehicle, it includes two warning modes: The first is to calculate the vehicle conflict alternative index in real time Determine whether the vehicle has a collision risk, that is, for the current vehicle with a safety level of A, if , it is determined that a collision may occur; for vehicles with a safety level of B, when When , it is determined that a collision may occur; for vehicles with a safety level of C, when When , it is determined that a collision may occur; for a vehicle with a safety level of D, when When , it is determined that a collision may occur.

[0053] The second method is to use the constructed vehicle driving risk prediction model to estimate the vehicle accident risk probability , that is, for the current vehicle with a safety level of A, when When , it is judged that there may be an accident risk; for the current vehicle with its own safety level B, when When , it is judged that there may be an accident risk; for vehicles with a safety level of C, when When , it is determined that a collision may occur; for a vehicle with a safety level of D, when When , it is determined that a collision may occur.

[0054] Furthermore, in the above step S4, historical sample data is obtained at every preset period to dynamically modify the vehicle driving risk prediction model, vehicle safety classification model, and threshold division in the vehicle risk warning system. Specifically, it includes: S41. Periodically collect historical vehicle trajectory data within the safety warning area and S (S = 50) meters before the safety warning area, where the cycle is consistent with the cycle in step S12; S42. Perform cyclic extraction with a detection period of T days, and select five data segments in one day: morning peak, evening peak, morning flat peak, afternoon flat peak and night, and the time length of each data segment is 45 minutes.

[0055] S43. Recalibrate the vehicle driving risk prediction model and vehicle safety classification model based on the historical sample data, and reclassify the traffic conflict warning threshold and accident risk probability warning threshold. Ensure that the above models constructed use the latest historical data.

[0056] Example 2 The above-mentioned embodiment 1 provides an intelligent risk warning method for road traffic, and correspondingly, this embodiment provides an intelligent risk warning system for road traffic. The system provided by this embodiment can implement the intelligent risk warning method for road traffic of embodiment 1, and the system can be implemented by software, hardware, or a combination of software and hardware. For example, the system may include integrated or separate functional modules or functional units to execute the corresponding steps in each method of embodiment 1. Since the system of this embodiment is basically similar to the method embodiment, the description process of this embodiment is relatively simple, and the relevant parts can refer to the partial description of embodiment 1. The embodiment of the system provided in this embodiment is only illustrative.

[0057] The intelligent road traffic risk warning system provided in this embodiment includes: Model building module, which is used to build a vehicle driving risk prediction model and a vehicle safety classification model based on periodically collected historical trajectory data, and to build a vehicle risk warning system; The vehicle safety classification and threshold matching module is used before the vehicle enters the safety warning area. The vehicle risk warning system obtains the factors that affect the probability of vehicle accident risks, and predicts the current vehicle's own safety level based on the vehicle safety classification model. , according to the current vehicle's own safety level, the current vehicle's safety warning threshold is matched, where the safety warning threshold includes the traffic conflict warning threshold and accident risk probability warning threshold ; Warning module, used when the vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle according to the vehicle driving risk prediction model and the safety warning threshold of the current vehicle, issues a real-time warning, and releases warning information; The update module is used to obtain historical trajectory data at preset intervals to dynamically correct the vehicle driving risk prediction model, vehicle safety classification model and threshold division in the vehicle risk warning system.

[0058] Example 3 This embodiment provides a processing device corresponding to the intelligent road traffic risk warning method provided in this embodiment 1. The processing device can be a processing device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the method of embodiment 1.

[0059] The processing device includes a processor, a memory, a communication interface and a bus, and the processor, the memory and the communication interface are connected through the bus to complete mutual communication. The memory stores a computer program that can be run on the processor, and the processor executes the intelligent road traffic risk warning method provided in this embodiment 1 when running the computer program.

[0060] Preferably, the memory may be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0061] Preferably, the processor may be a central processing unit (CPU), a digital signal processor (DSP) or other general-purpose processors, which are not limited here.

[0062] Example 4 The intelligent road traffic risk warning method of this embodiment 1 can be specifically implemented as a computer program product, which may include a computer-readable storage medium carrying computer-readable program instructions for executing the intelligent road traffic risk warning method described in this embodiment 1.

[0063] Computer readable storage media can be tangible devices that hold and store instructions used by instruction execution devices. Computer readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any combination thereof.

[0064] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks. These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 One or more processes and / or boxes Figure 1 A function specified in one or more boxes. These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A road traffic intelligent risk warning method, characterized in that: The following steps are involved: Based on the periodically collected historical trajectory data, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and a vehicle risk early warning system is established; Before a vehicle enters the safety warning area, the vehicle risk warning system obtains factors that affect the probability of vehicle accident risks, and predicts the vehicle's own safety level based on the vehicle safety classification model, and matches the vehicle's safety warning threshold based on the vehicle's own safety level; When a vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle based on the vehicle driving risk prediction model and the current vehicle safety warning threshold, issues a real-time warning, and releases warning information; At preset intervals, historical trajectory data is obtained to dynamically correct the vehicle driving risk prediction model, vehicle safety classification model and safety warning threshold division in the vehicle risk warning system.

2. A road traffic intelligent risk early warning method as claimed in claim 1, characterized in that: The vehicle driving risk prediction model and vehicle safety classification model are constructed based on the periodically collected historical trajectory data, and a vehicle risk early warning system is established, including: Delineate road sections or nodes as the boundaries of the safety warning area, and store the boundary information of the safety warning area in a database; Relying on roadside sensing equipment, the historical vehicle trajectory data of the preset distance in front of the safety warning area and the safety warning area are periodically collected, and the historical vehicle trajectory data is preprocessed and parameter extracted to obtain candidate feature variable parameters; Based on the candidate characteristic variable parameters, a vehicle driving risk prediction model and a vehicle safety classification model are constructed, and the traffic conflict warning threshold and the accident risk probability warning threshold are divided; Import the vehicle driving risk prediction model and vehicle safety classification model into the vehicle risk warning system to complete the construction.

3. A road traffic intelligent risk early warning method as claimed in claim 2, characterized in that: The vehicle driving risk prediction model and the vehicle safety classification model are constructed based on the candidate characteristic variable parameters, and the traffic conflict warning threshold and the accident risk probability warning threshold are divided, including: Based on the candidate characteristic variable parameters, the factors affecting the accident risk probability of the vehicle before and within the safety warning area are screened, and based on the screened factors affecting the accident risk probability of the vehicle, the vehicle driving risk prediction models for the vehicle before and within the safety warning area are respectively constructed; Based on the vehicle driving risk prediction model facing the safety warning area, the driving risk-related variable parameters of the vehicle before entering the safety warning area are screened, and the K-means clustering method is used to perform cluster analysis on the screened variable parameters to obtain the vehicle safety classification results and build a vehicle safety classification model; According to the vehicle safety classification results, the traffic conflict warning threshold and the accident risk probability warning threshold are divided to obtain the traffic conflict warning threshold and the accident risk probability warning threshold corresponding to each safety level.

4. A road traffic intelligent risk warning method as claimed in claim 3, characterized in that: The vehicle driving risk prediction model for the safety warning area is used to screen the driving risk-related variable parameters of the vehicle before entering the safety warning area, and the K-means clustering method is used to perform cluster analysis on the screened variable parameters to obtain the vehicle safety classification result, and a vehicle safety classification model is constructed, including: According to the vehicle driving risk prediction model constructed for the safety warning area, the correlation between various factors and vehicle risk is obtained based on historical sample data; Based on the correlation in the above steps and the data conditions obtained before the vehicle enters the warning area, the influencing factors obtained before the vehicle enters the safety warning area are screened as vehicle safety evaluation indicators; The K-means clustering method is used to perform cluster analysis on vehicle safety evaluation indicators, obtain vehicle safety classification results, and build a vehicle safety classification model.

5. The intelligent road traffic risk warning method according to claim 3, characterized in that: The traffic conflict warning threshold and the accident risk probability warning threshold are divided according to the vehicle safety classification result to obtain the traffic conflict warning threshold and the accident risk probability warning threshold corresponding to each safety level, including: The accident risk probability warning threshold and the traffic conflict warning threshold interval are divided into a plurality of sub-intervals according to preset interval lengths respectively; Calculate the probability and information entropy of each subinterval respectively; Find the threshold corresponding to the maximum value of information entropy as the result of dividing the traffic conflict warning threshold and the accident risk probability warning threshold.

6. A road traffic intelligent risk warning method as claimed in claim 3, characterized in that: When the vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle according to the vehicle driving risk prediction model and the safety warning threshold of the current vehicle, and issues warning information in real time, including: According to the real-time collected data, the factors affecting the vehicle accident risk probability of the current vehicle are obtained, and the vehicle safety classification model is used to predict the current vehicle's own safety level and the matching traffic conflict warning threshold and accident risk probability warning threshold; The vehicle risk warning system uses the roadside sensing equipment to obtain the vehicle's position in the warning area, and uses the vehicle's position information and the matching traffic conflict warning threshold and accident risk probability warning threshold to predict the current vehicle's driving risk. When the vehicle risk warning system determines that there is a collision risk in any way, the vehicle risk warning system sends risk warning information to the cloud control management platform and the vehicle terminal.

7. A road traffic intelligent risk warning method as claimed in claim 6, characterized in that: The vehicle risk warning system includes two warning modes when predicting the driving risk of the current vehicle: The first method is to determine whether the vehicle has a collision risk by calculating the conflict substitution index of the current vehicle in real time. If the conflict substitution index of the current vehicle is less than the vehicle traffic conflict warning threshold corresponding to its own safety level, it is determined that a collision may occur. The second method is to use the constructed vehicle driving risk prediction model to estimate the accident risk probability of the current vehicle. If the accident risk probability of the current vehicle is greater than the accident risk probability warning threshold corresponding to its own safety level.

8. An intelligent road traffic risk warning system, characterized in that: include: Model building module, which is used to build a vehicle driving risk prediction model and a vehicle safety classification model based on periodically collected historical trajectory data, and to build a vehicle risk warning system; The vehicle safety classification and threshold matching module is used for the vehicle risk warning system to obtain factors that affect the vehicle accident risk probability before the vehicle enters the safety warning area, and predict the current vehicle's own safety level based on the vehicle safety classification model, and match the current vehicle's safety warning threshold according to the current vehicle's own safety level, where the safety warning threshold includes the traffic conflict warning threshold and the accident risk probability warning threshold; Warning module, used when the vehicle enters the safety warning area, the vehicle risk warning system determines the safety status of the current vehicle according to the vehicle driving risk prediction model and the safety warning threshold of the current vehicle, issues a real-time warning, and releases warning information; The update module is used to obtain historical trajectory data at preset intervals to dynamically correct the vehicle driving risk prediction model, vehicle safety classification model and threshold division in the vehicle risk warning system.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions, which, when executed by a computing device, cause the computing device to perform any one of the methods of claims 1 to 7.

10. A computing device, characterized in that include: One or more processors and a memory, wherein the memory stores one or more programs and is configured to be executed by the one or more processors, wherein the one or more programs include instructions for executing any one of the methods described in claims 1 to 7.

Images