Vehicle dynamic risk scoring method and device, electronic equipment and storage medium

Abstract

The invention relates to the cross technical field of Internet of Vehicles big data analysis and financial science and technology, and discloses a vehicle dynamic risk scoring method and device, electronic equipment and a storage medium. The method comprises the steps of collecting vehicle state data and external interface data; cutting the vehicle state data and the external interface data according to the vehicle ignition signal to generate travel segments; environment characteristics in the travel segment and behavior frequency, behavior intensity and behavior duration of the driving behavior event in the travel segment are extracted; acquiring historical accident data, and constructing an environmental risk coefficient matrix based on the historical accident data; determining an initial risk score of the travel segment based on the environment characteristics, the behavior frequency, the behavior intensity, the behavior duration and the environment risk coefficient matrix; and constructing a dynamic risk baseline library, and mapping the initial risk score of the travel segment into a standard risk score based on the dynamic risk baseline library, thereby completing vehicle dynamic risk scoring. According to the invention, the accuracy of vehicle risk scoring is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Vehicles big data analysis and financial technology, and in particular to a vehicle dynamic risk scoring method and device, an electronic device and a storage medium. BACKGROUND

[0002] In the field of commercial vehicle insurance, the traditional pricing mode is mainly based on static factors such as vehicle model, vehicle age, and accident history, which cannot reflect the real driving risk of the driver, and there is an unfair pricing problem of "high risk low payment, low risk high payment". The use-based insurance (UBI) provides a direction for solving this problem by introducing vehicle terminal data. However, the existing commercial vehicle UBI model still has significant limitations.

[0003] Limitation one: insufficient consideration of environmental risk factors

[0004] Existing solutions focus on driver behaviors such as sudden acceleration and sudden braking, but ignore the significant impact of external environment on risk. For example, the same driver's sudden braking on a mountain highway at night is much riskier than the same operation on a flat road in the city during the day. Evaluating driving behavior without considering external environment cannot accurately quantify the severity of risk, resulting in distorted scores.

[0005] Limitation two: static risk scoring model

[0006] Most models either use fixed weight coefficients to integrate different indicators or use historical full data to train static models. This method is difficult to adapt to the dynamic changes of risk factors, such as sudden changes in specific road sections due to construction, or changes in day and night length due to seasonal changes. The model cannot capture these changes in a timely manner, resulting in lagging and inaccurate risk assessment. SUMMARY

[0007] The purpose of the present application is to provide a vehicle dynamic risk scoring method, device, electronic device and storage medium to at least solve the problems of insufficient consideration of environmental risk factors and static risk scoring model in existing solutions, improve the accuracy of risk scoring, and provide more scientific UBI pricing basis for insurance companies.

[0008] To solve the above technical problems, in a first aspect, the present application provides a vehicle dynamic risk scoring method, at least comprising:

[0009] Collecting vehicle state data and external interface data;

[0010] Cutting the vehicle state data and the external interface data according to the vehicle ignition signal to generate at least one trip segment;

[0011] extracting an environmental feature in each of the trip segments and a behavior frequency, a behavior intensity and a behavior duration of a driving behavior event in each of the trip segments;

[0012] obtaining historical accident data and constructing an environmental risk coefficient matrix based on the historical accident data;

[0013] determining an initial risk score of each of the trip segments based on the environmental feature, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix;

[0014] constructing a dynamic risk baseline library and mapping the initial risk score of each of the trip segments to a standard risk score based on at least the dynamic risk baseline library to complete the vehicle dynamic risk score.

[0015] Optionally, the extracting an environmental feature in each of the trip segments and a behavior frequency, a behavior intensity and a behavior duration of a driving behavior event in each of the trip segments specifically comprises:

[0016] extracting a time environmental feature, a space environmental feature and a meteorological environmental feature in each of the trip segments;

[0017] extracting a behavior frequency, a behavior intensity and a behavior duration of a driving behavior event in each of the trip segments.

[0018] Optionally, the determining an initial risk score of each of the trip segments based on the environmental feature, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix specifically comprises:

[0019] determining a basic risk score of a corresponding driving behavior event according to the behavior frequency, the behavior intensity and the behavior duration of each of the driving behavior events;

[0020] querying the environmental risk coefficient matrix based on the time environmental feature, the space environmental feature and the meteorological environmental feature to obtain an environmental risk coefficient of a current driving behavior event;

[0021] determining a weighted risk score of the driving behavior event according to the basic risk score and the environmental risk coefficient;

[0022] determining an initial risk score of each of the trip segments according to the weighted risk score of the driving behavior event.

[0023] Optionally, the constructing a dynamic risk baseline library and mapping the initial risk score of each of the trip segments to a standard risk score based on at least the dynamic risk baseline library to complete the vehicle dynamic risk score specifically comprises:

[0024] constructing a dynamic risk baseline library of historical risk scores according to a preset number of days of a sliding window mechanism based on a current vehicle state;

[0025] setting a preset quantile of all the historical risk scores as a dynamic risk baseline;

[0026] comparing the initial risk score corresponding to each of the trip segments with the dynamic risk baseline, and mapping the initial risk score into a standard risk score in a standard score interval by at least a setting function to complete the vehicle dynamic risk scoring.

[0027] Optionally, the vehicle state data at least includes one of a vehicle speed, a longitudinal acceleration, a lateral acceleration, a brake, an accelerator, a turn signal, GPS data, and timestamp data.

[0028] The external interface data at least includes one of map data, real-time weather data, and historical accident data.

[0029] In a second aspect, the present application further provides a vehicle dynamic risk scoring device, at least comprising:

[0030] a data collection module for collecting vehicle state data and external interface data;

[0031] a trip cutting module for cutting the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment;

[0032] a trip extraction module for extracting an environmental feature within each of the trip segments and a behavior frequency, a behavior intensity, and a behavior duration of a driving behavior event within each of the trip segments;

[0033] a matrix construction module for obtaining historical accident data and constructing an environmental risk coefficient matrix based on the historical accident data;

[0034] an initial scoring module for determining an initial risk score of each of the trip segments based on the environmental feature, the behavior frequency, the behavior intensity, the behavior duration, and the environmental risk coefficient matrix;

[0035] a standard scoring module for constructing a dynamic risk baseline library and mapping the initial risk score corresponding to each of the trip segments into a standard risk score based on at least the dynamic risk baseline library to complete the vehicle dynamic risk scoring.

[0036] Optionally, the trip extraction module is specifically configured to:

[0037] extracting time environment features, space environment features and meteorological environment features in each of the trip segments; and extracting behavior frequency, behavior intensity and behavior duration of driving behavior events in each of the trip segments.

[0038] Optionally, the initial score module is specifically configured to:

[0039] determining a basic risk score of a corresponding driving behavior event according to the behavior frequency, the behavior intensity and the behavior duration of each of the driving behavior events; and querying the environment risk coefficient matrix based on the time environment features, the space environment features and the meteorological environment features to obtain an environment risk coefficient of a current driving behavior event; and determining a weighted risk score of the driving behavior event according to the basic risk score and the environment risk coefficient; and determining an initial risk score of each of the trip segments according to the weighted risk score of the driving behavior event.

[0040] In a third aspect, the present application further provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, and the processor implements the steps in the vehicle dynamic risk scoring method according to any one of the first aspect when executing the program.

[0041] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program capable of being executed by a processor to implement the steps in the vehicle dynamic risk scoring method according to any one of the first aspect.

[0042] The technical solution provided by the embodiments of the present application first collects vehicle state data and external interface data; secondly, cuts the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment; thirdly, extracts environment features in each trip segment and behavior frequency, behavior intensity and behavior duration of driving behavior events in each trip segment; fourthly, obtains historical accident data and constructs an environment risk coefficient matrix based on the historical accident data; then, determines an initial risk score of each trip segment based on the environment features, the behavior frequency, the behavior intensity, the behavior duration and the environment risk coefficient matrix; finally, constructs a dynamic risk baseline library and maps the initial risk score corresponding to each trip segment to a standard risk score based on at least the dynamic risk baseline library to complete vehicle dynamic risk scoring.

[0043] It can be seen that, on one hand, the embodiment of the present application fuses and analyzes the vehicle state data and the external interface data by constructing the environmental risk coefficient matrix, and then determines the initial risk score of each trip segment, so as to convert the abstract driving behavior event into a concrete and quantifiable risk value. On the other hand, the embodiment of the present application constructs a dynamic risk baseline library, continuously updates the historical accident rate and the typical risk value of each road segment under different time and weather. The baseline library serves as a dynamic calibration benchmark for the risk score, can sensitively capture the latest changes of the external environment, and makes the standard risk score always keep synchronized with the current actual risk, which is beneficial to the accuracy of the dynamic risk score. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is a flowchart of a vehicle dynamic risk scoring method provided by the embodiment of the present application;

[0045] Figure 2 is a flowchart of another vehicle dynamic risk scoring method provided by the embodiment of the present application;

[0046] Figure 3 is a structural schematic diagram of a vehicle dynamic risk scoring device provided by the embodiment of the present application;

[0047] Figure 4 is a structural schematic diagram of an electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0049] The terms used in the embodiments of the present application are merely for the purpose of describing the specific embodiments, and are not intended to limit the present application. The singular forms "a", "an" and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two.

[0050] It should be understood that the term "and / or" used herein is only to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects.

[0051] It should be understood that, although the terms first, second, third, etc. can be employed in this application embodiment, these are described not to limit the scope of the application. These terms are only used to distinguish one description from another. For example, a first can be termed a second, and, similarly, a second can be termed a first, without departing from the scope of the application.

[0052] Depending on the context, the word "if" as used herein can be interpreted to mean "when" or "while" or "in response to determining" or "in response to detecting." Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon detecting [the stated condition or event]" or "in response to detecting [the stated condition or event]."

[0053] It is also to be noted that the terms "comprising", "including", or any other variant are intended to cover non-exclusive inclusions, such that the product or process including a list of elements does not include only those elements, but can also include other elements not expressly listed, or inherent to such product or process. Without more limitations, the element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the product or process comprising the element.

[0054] It is particularly noted that the symbols and / or numbers present in the description, if not marked in the description of the drawings, are not drawing reference numbers.

[0055] Figure 1 is a flowchart of a vehicle dynamic risk scoring method provided by an embodiment of the application. The embodiment is applicable at least to the scoring of driving behavior risks of various vehicles. The vehicle dynamic risk scoring method can be executed by a vehicle dynamic risk scoring device in the embodiment of the application as an execution subject, but is not limited thereto. The execution subject can be realized in the form of software and / or hardware. As shown in Figure 1 The vehicle dynamic risk scoring method at least includes the following steps:

[0056] S1, collecting vehicle state data and external interface data.

[0057] The vehicle state data collection mode can be collecting vehicle CAN bus data through a vehicle-mounted T-Box. The external interface data collection mode can be collecting through a cloud interface. In a specific embodiment, the vehicle state data at least includes one of vehicle speed, longitudinal acceleration, lateral acceleration, brake, accelerator, turn signal, GPS data and timestamp data; the external interface data at least includes one of map data, real-time weather data and historical accident data. It can be understood that the GPS data includes longitude and latitude, speed, direction angle and the like. The map data includes road type, curve curvature, slope and the like.

[0058] S2, cutting the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment.

[0059] The vehicle ignition signal can be understood as a vehicle start signal. It can be understood that the trip between the triggering of the vehicle ignition signal and the triggering of the next vehicle ignition signal is understood as a trip segment.

[0060] S3, extracting environmental features in each trip segment and behavior frequency, behavior intensity and behavior duration of driving behavior events in each trip segment.

[0061] The environmental features can include time environmental features, space environmental features, meteorological environmental features and the like. The driving behavior events can be events such as sudden acceleration, sudden deceleration, sudden turning, overspeed, fatigue driving (based on continuous driving duration) and the like.

[0062] S4, obtaining historical accident data, and constructing an environmental risk coefficient matrix based on the historical accident data.

[0063] The historical accident data can be historical behavior events in the current driving scene, such as sudden acceleration, sudden deceleration, collision accident and the like.

[0064] S5, determining an initial risk score of each trip segment based on the environmental features, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix.

[0065] The initial risk score can be a total score of all driving event behavior risk scores in the current trip segment.

[0066] S6, constructing a dynamic risk baseline library, and at least based on the dynamic risk baseline library, mapping the initial risk score corresponding to each trip segment to a standard risk score to complete the vehicle dynamic risk score.

[0067] The construction method of the dynamic risk baseline library can be collecting initial risk scores corresponding to trips of similar vehicle models and similar environments (same type of road, similar weather), and then organizing them into a dynamic risk baseline library.

[0068] The technical solution provided by the embodiment first collects vehicle state data and external interface data; secondly, cuts the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment; thirdly, extracts environmental features in each trip segment and behavior frequency, behavior intensity and behavior duration of a driving behavior event in each trip segment; fourthly, obtains historical accident data and constructs an environmental risk coefficient matrix based on the historical accident data; then, determines an initial risk score of each trip segment based on the environmental features, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix; and finally, constructs a dynamic risk baseline library and maps the initial risk score corresponding to each trip segment to a standard risk score based on at least the dynamic risk baseline library to complete vehicle dynamic risk scoring.

[0069] It can be seen that, on the one hand, the embodiment fuses and analyzes the vehicle state data and the external interface data by constructing the environmental risk coefficient matrix, and then determines the initial risk score of each trip segment, so as to convert the abstract driving behavior event into a concrete and quantifiable risk value. On the other hand, the embodiment continuously updates the historical accident rate and the typical risk value of each road segment under different time and weather by constructing the dynamic risk baseline library. The baseline library serves as a dynamic calibration benchmark for risk scoring, can sensitively capture the latest changes in the external environment, and can keep the standard risk score synchronized with the current actual risk at all times, thereby improving the accuracy of risk scoring.

[0070] On the basis of the above embodiment or implementation, Figure 2 is a flowchart of another vehicle dynamic risk scoring method provided by the embodiment of the present application, which is based on the above embodiment. As shown in Figure 2 , the vehicle dynamic risk scoring method at least includes the following steps:

[0071] S1, collecting vehicle state data and external interface data.

[0072] S2, cutting the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment.

[0073] S31, extracting time environmental features, space environmental features and meteorological environmental features in each trip segment.

[0074] The time environment feature can be converting the timestamp into weekday / weekend, daytime / nighttime / sunset, holiday, etc. The space environment feature can be matching the GPS trajectory with the map to extract the road type (expressway, national highway, urban road), average lane number, tunnel and bridge proportion, historical accident frequency, etc. of each road segment in the trip. The weather environment can be matching the trip time and geographical location to obtain the weather, wind amount, etc. of the trip.

[0075] S32, extracting the behavior frequency, behavior intensity, and behavior duration of each driving behavior event in each trip segment.

[0076] S4, obtaining historical accident data and constructing an environment risk coefficient matrix based on the historical accident data.

[0077] S51, determining the basic risk score of each driving behavior event according to the behavior frequency, behavior intensity, and behavior duration of each driving behavior event.

[0078] It can be understood that the basic risk score can be a risk score given to each detected driving behavior (such as an emergency stop) according to its intensity.

[0079] S52, querying the environment risk coefficient matrix based on the time environment feature, the space environment feature, and the weather environment feature to obtain the environment risk coefficient of the current driving behavior event.

[0080] In the environment risk coefficient matrix, the time environment feature (corresponding to the time interval in Table 1), the space environment feature (corresponding to the road type in Table 1), and the weather environment feature (corresponding to the weather condition in Table 1) can uniquely determine an environment risk coefficient. The dimensions of the environment risk coefficient matrix include the time environment feature, the space environment feature, and the weather environment feature. The value of each cell in the matrix represents the basic risk coefficient under the specific environment combination (for example, the basic risk coefficient of nighttime + expressway + rainy and snowy weather is the highest). This matrix serves as a priori knowledge base to provide a basic weight for the dynamic weighting in step S53. The environment risk coefficient matrix can be as shown in Table 1 below.

[0081] Table 1

[0082]

[0083] S53, determining the weighted risk score of the driving behavior event according to the basic risk score and the environment risk coefficient.

[0084] The determination manner of the weighted risk score can be that the basic risk score is multiplied by the environmental risk coefficient. Exemplarily, in a specific scenario, the emergency deceleration, the emergency degree (which can be divided by the distance to the front obstacle or the slope of the speed drop) can be divided into three grades, the first grade risk score is 60, the second grade is 80, and the third grade is 100. Now according to the deceleration, it is identified as the second grade, that is, the basic risk score is 80, at this time, it is a highway-night-sunny day, and the environmental risk coefficient is 2.5, at this time, the weighted risk score of the dangerous behavior is 200.

[0085] S54, determining an initial risk score of each trip segment according to the weighted risk score of the driving behavior event.

[0086] The determination manner of the initial risk score can be that all driving behavior events in a trip segment are summed up by time-decay weighting (recent events have high weights).

[0087] S61, constructing a dynamic risk baseline library of historical risk scores based on a preset number of days of a sliding window mechanism according to a current vehicle use state.

[0088] The preset number of days can be 90 days. The current vehicle use state can be a current vehicle use scenario, that is, a vehicle model, a driving scenario, and a weather condition. The construction of the dynamic risk baseline library can be that 100 similar vehicle models, similar environments (same type of road, similar weather) and their corresponding risk scores are selected based on the current vehicle use state.

[0089] S62, setting a preset quantile of all historical risk scores as a dynamic risk baseline.

[0090] The preset quantile can be a 90th quantile. After the risk scores of 100 similar trips are sorted from small to large, the score corresponding to the 90th position (or the corresponding value calculated according to the statistical rule) is the 90th quantile. The corresponding score value can be considered as the dynamic risk baseline, which is updated daily with the window sliding, and adapts to the changes of the vehicle and the environment.

[0091] S63, comparing the initial risk score corresponding to each trip segment with the dynamic risk baseline to at least map the initial risk score to a standard risk score in a standard score interval by setting a function, to complete the dynamic risk scoring of the vehicle.

[0092] The standard score interval can be [0, 100]. The higher the standard risk score, the higher the risk level of the trip relative to its historical similar environment. The setting function can be:

[0093] S final =100 / (1+exp(-k*(S raw / B dynamic -1)));

[0094] where S final represents the standard risk score. S raw represents the initial risk score, B dynamic represents the dynamic risk baseline, and k represents an adjustment coefficient for controlling the steepness of the score curve, generally k = 4. The greater the value of k, the steeper the curve, and the more sensitive to small changes near the baseline.

[0095] For example, the S raw of the trip A is calculated to be 850 points.

[0096] According to its vehicle model, road type (highway), weather (sunny), and time (daytime holiday), the corresponding B dynamic = 1000 points.

[0097] Further, k = 4 is set;

[0098] S final = 100 / (1+exp(-4*(850 / 1000-1))) ≈ 35 points. It can be understood that although the initial risk score of trip A is 850, it is still lower compared to the historical high-risk baseline (1000 points) of the same environment, and therefore the final standard risk score is only 35 points, belonging to a "relatively safe" trip.

[0099] The technical scheme provided by the embodiment first collects vehicle state data and external interface data. Further, the vehicle state data and the external interface data are cut according to a vehicle ignition signal to generate at least one trip segment. Further, time environment features, space environment features and meteorological environment features in each trip segment are extracted. Further, behavior frequency, behavior intensity and behavior duration of each driving behavior event in each trip segment are extracted. Further, historical accident data are acquired, and an environment risk coefficient matrix is constructed based on the historical accident data. Further, a basic risk score of each driving behavior event is determined according to the behavior frequency, the behavior intensity and the behavior duration of the driving behavior event. Further, the environment risk coefficient matrix is queried based on the time environment features, the space environment features and the meteorological environment features to acquire an environment risk coefficient of the current driving behavior event. Further, a weighted risk score of the driving behavior event is determined according to the basic risk score and the environment risk coefficient. Further, an initial risk score of each trip segment is determined according to the weighted risk score of the driving behavior event. Further, a dynamic risk baseline library of historical risk scores is constructed according to a current vehicle state based on a sliding window mechanism of a preset number of days. Further, a preset quantile of all historical risk scores is set as a dynamic risk baseline. Further, the initial risk score of each trip segment is compared with the dynamic risk baseline to at least map the initial risk score to a standard risk score in a standard score interval by setting a function, so as to complete vehicle dynamic risk scoring.

[0100] It can be seen that the embodiment establishes an "environment risk coefficient" matrix, fuses and analyzes GPS coordinates, time stamps, weather API data and the like, matches a dynamic environment risk amplification coefficient for each driving behavior, and thus converts abstract behaviors into concrete and quantifiable risk values. In addition, the embodiment also constructs a dynamic baseline library based on a space-time sliding window, continuously updates historical accident rates and typical risk values of each section under different times and weather. The baseline library serves as a dynamic calibration benchmark for risk scoring, ensuring that the model can sensitively capture the latest changes in the external environment, so that the scoring results are always synchronized with the current actual risk. In summary, the embodiment innovatively introduces multi-dimensional environment risk factors and establishes an environment risk coefficient matrix, realizes risk quantification of driving behaviors in specific scenarios, constructs a dynamic risk baseline library based on a space-time sliding window, enables the scoring to have self-evolution ability, can respond to external environment changes in real time, significantly improves the accuracy and timeliness of risk scoring, and can complete accurate pricing and risk management of insurance based on standard risk scoring.

[0101] Figure 3is a structural schematic diagram of a vehicle dynamic risk scoring device provided by an embodiment of the present application. The embodiment is at least applicable to the scoring scene of driving behavior risk of various vehicles. The vehicle dynamic risk scoring device can be realized in the form of software and / or hardware. As shown in the figure, the vehicle dynamic risk scoring device at least comprises: Figure 3

[0102] A data acquisition module 110 is configured to acquire vehicle state data and external interface data.

[0103] A trip cutting module 120 is configured to cut the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment.

[0104] A trip extraction module 130 is configured to extract environmental features in each trip segment and behavior frequency, behavior intensity and behavior duration of driving behavior events in each trip segment.

[0105] A matrix construction module 140 is configured to acquire historical accident data and construct an environmental risk coefficient matrix based on the historical accident data.

[0106] An initial scoring module 150 is configured to determine an initial risk score of each trip segment based on the environmental features, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix.

[0107] A standard scoring module 160 is configured to construct a dynamic risk baseline library and map the initial risk score corresponding to each trip segment to a standard risk score based at least on the dynamic risk baseline library to complete the vehicle dynamic risk scoring.

[0108] Optionally, the trip extraction module 130 is specifically configured to:

[0109] extract time environmental features, space environmental features and meteorological environmental features in each trip segment; and extract behavior frequency, behavior intensity and behavior duration of driving behavior events in each trip segment.

[0110] Optionally, the initial scoring module 150 is specifically configured to:

[0111] determine a basic risk score of each driving behavior event according to the behavior frequency, the behavior intensity and the behavior duration of the driving behavior event; and query the environmental risk coefficient matrix based on the time environmental features, the space environmental features and the meteorological environmental features to acquire an environmental risk coefficient of the current driving behavior event; and determine a weighted risk score of the driving behavior event according to the basic risk score and the environmental risk coefficient; and determine the initial risk score of each trip segment according to the weighted risk score of the driving behavior event.

[0112] Optionally, the standard scoring module 160 is specifically configured to:​

[0113] constructing a dynamic risk baseline library of historical risk scores according to a preset number of days based on a sliding window mechanism of a current vehicle state; and setting a preset quantile of all historical risk scores as a dynamic risk baseline; and comparing an initial risk score corresponding to each trip segment with the dynamic risk baseline to map the initial risk score into a standard risk score in a standard score interval by at least a setting function to complete vehicle dynamic risk scoring.

[0114] Optionally, the vehicle state data at least includes one of a vehicle speed, a longitudinal acceleration, a lateral acceleration, a brake, an accelerator, a turn signal, GPS data and timestamp data.

[0115] The external interface data at least includes one of map data, real-time weather data and historical accident data.

[0116] The technical scheme provided by the embodiment first acquires vehicle state data and external interface data through a data acquisition module; further, cuts the vehicle state data and the external interface data according to a vehicle ignition signal through a trip cutting module to generate at least one trip segment; further, extracts environmental features in each trip segment and behavior frequency, behavior intensity and behavior duration of a driving behavior event in each trip segment through a trip extraction module; further, acquires historical accident data and constructs an environmental risk coefficient matrix based on the historical accident data through a matrix construction module; further, determines an initial risk score of each trip segment based on the environmental features, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix through an initial scoring module; and finally, constructs a dynamic risk baseline library through a standard scoring module, and maps the initial risk score corresponding to each trip segment into a standard risk score based on at least the dynamic risk baseline library to complete vehicle dynamic risk scoring.

[0117] It can be seen that, on the one hand, the embodiment fuses and analyzes the vehicle state data and the external interface data by constructing the environmental risk coefficient matrix, and then determines the initial risk score of each trip segment, so as to convert the abstract driving behavior event into a concrete and quantifiable risk value. On the other hand, the embodiment continuously updates the historical accident rate and the typical risk value of each road segment under different times and weather by constructing the dynamic risk baseline library. The baseline library serves as a dynamic calibration benchmark for risk scoring, can sensitively capture the latest changes of the external environment, makes the standard risk score always keep synchronized with the actual risk, and is beneficial to improving the accuracy of the dynamic risk scoring.

[0118] The embodiment provides an electronic device, Figure 4 is a structural schematic diagram of an electronic device provided by the embodiment of the application, referring to Figure 4The electronic device 1000 comprises a processor 1001 and a memory 1002, and the memory 1002 stores computer readable instructions, when the computer readable instructions are executed by the processor 1001, the steps in any one of the vehicle dynamic risk scoring methods described above are executed. Through the above technical solution, the processor 1001 and the memory 1002 are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanism (not marked), the memory 1002 stores a computer program executable by the processor, when the electronic device 1000 is running, the processor 1001 executes the computer program to execute the vehicle dynamic risk scoring method in any one of the optional implementation manners of the above embodiments, to at least realize the following functions: collecting vehicle state data and external interface data; cutting the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment; extracting environmental features in each trip segment and behavior frequency, behavior intensity and behavior duration of driving behavior events in each trip segment; obtaining historical accident data, and constructing an environmental risk coefficient matrix based on the historical accident data; determining an initial risk score of each trip segment based on the environmental features, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix; constructing a dynamic risk baseline library, and mapping the initial risk score corresponding to each trip segment to a standard risk score based on at least the dynamic risk baseline library, to complete the vehicle dynamic risk scoring.

[0119] The embodiment provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the vehicle dynamic risk scoring method provided by all the embodiments of the application: collecting vehicle state data and external interface data; cutting the vehicle state data and the external interface data according to a vehicle ignition signal to generate at least one trip segment; extracting environmental features in each trip segment and behavior frequency, behavior intensity and behavior duration of driving behavior events in each trip segment; obtaining historical accident data, and constructing an environmental risk coefficient matrix based on the historical accident data; determining an initial risk score of each trip segment based on the environmental features, the behavior frequency, the behavior intensity, the behavior duration and the environmental risk coefficient matrix; constructing a dynamic risk baseline library, and mapping the initial risk score corresponding to each trip segment to a standard risk score based on at least the dynamic risk baseline library, to complete the vehicle dynamic risk scoring.

[0120] Any combination of one or more computer readable medium can be utilized. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium can be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0121] A computer readable signal medium can include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal can take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium can be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0122] Program code embodied on a computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0123] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In an embodiment, the present application is directed to computer program products comprising machine-readable media for carrying or having machine-executable instructions or programs

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions recorded in the above embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for vehicle dynamic risk scoring, characterized in that, At least including: Collect vehicle status data and external interface data; The vehicle status data and the external interface data are segmented based on the vehicle ignition signal to generate at least one travel segment; Extract the environmental features within each trip segment and the frequency, intensity, and duration of driving behavior events within each trip segment; Acquire historical accident data and construct an environmental risk coefficient matrix based on the historical accident data; An initial risk score is determined for each of the following trip segments based on the environmental characteristics, the frequency of the behavior, the intensity of the behavior, the duration of the behavior, and the environmental risk coefficient matrix. A dynamic risk baseline library is constructed, and the initial risk score corresponding to each trip segment is mapped to a standard risk score based at least on the dynamic risk baseline library to complete the vehicle dynamic risk score.

2. The vehicle dynamic risk scoring method according to claim 1, characterized in that, The extraction of environmental features within each trip segment and the frequency, intensity, and duration of driving behavior events within each trip segment specifically includes: Extract the temporal, spatial, and meteorological environmental features within each of the aforementioned travel segments; Extract the frequency, intensity, and duration of driving behavior events within each trip segment.

3. The vehicle dynamic risk scoring method according to claim 2, characterized in that, The determination of the initial risk score for each trip segment based on the environmental characteristics, the frequency of the behavior, the intensity of the behavior, the duration of the behavior, and the environmental risk coefficient matrix specifically includes: A basic risk score for each driving behavior event is determined based on the frequency, intensity, and duration of the behavior. Based on the time environment characteristics, the spatial environment characteristics, and the meteorological environment characteristics, the environmental risk coefficient matrix is ​​queried to obtain the environmental risk coefficient of the current driving behavior event; The weighted risk score of the driving behavior event is determined based on the basic risk score and the environmental risk coefficient. An initial risk score is determined for each segment of the trip based on the weighted risk score of the driving behavior event.

4. The vehicle dynamic risk scoring method according to claim 2, characterized in that, The process of constructing a dynamic risk baseline library and mapping the initial risk score corresponding to each trip segment to a standard risk score based at least on the dynamic risk baseline library to complete the vehicle dynamic risk scoring specifically includes: A dynamic risk baseline library of historical risk scores is constructed based on a sliding window mechanism with a preset number of days, according to the current vehicle usage status. Set the preset quantiles of all the historical risk scores as the dynamic risk baseline; The initial risk score corresponding to each trip segment is compared with the dynamic risk baseline to map the initial risk score to a standard risk score within a standard score range, at least by setting a function, in order to complete the vehicle dynamic risk score.

5. The vehicle dynamic risk scoring method according to claim 1, characterized in that, The vehicle status data includes at least one of the following: vehicle speed, longitudinal acceleration, lateral acceleration, braking, accelerator, turn signal, GPS data, and timestamp data; The external interface data includes at least one of the following: map data, real-time weather data, and historical accident data.

6. A vehicle dynamic risk scoring device, characterized in that, At least including: The data acquisition module is used to collect vehicle status data and external interface data; The stroke cutting module is used to cut the vehicle status data and the external interface data according to the vehicle ignition signal to generate at least one stroke segment; The trip extraction module is used to extract environmental features within each trip segment and the frequency, intensity, and duration of driving behavior events within each trip segment. A matrix construction module is used to acquire historical accident data and construct an environmental risk coefficient matrix based on the historical accident data. An initial scoring module is used to determine an initial risk score for each of the trip segments based on the environmental characteristics, the frequency of the behavior, the intensity of the behavior, the duration of the behavior, and the environmental risk coefficient matrix. The standard scoring module is used to construct a dynamic risk baseline library and, at least based on the dynamic risk baseline library, map the initial risk score corresponding to each trip segment to a standard risk score to complete the vehicle dynamic risk scoring.

7. The vehicle dynamic risk scoring device according to claim 6, characterized in that, The trip extraction module is specifically used for: Extract the temporal, spatial, and meteorological environmental features within each trip segment; and extract the frequency, intensity, and duration of driving behavior events within each trip segment.

8. The vehicle dynamic risk scoring device according to claim 7, characterized in that, The initial scoring module is specifically used for: A basic risk score for each driving behavior event is determined based on the frequency, intensity, and duration of the behavior. Furthermore, based on the temporal environmental characteristics, the spatial environmental characteristics, and the meteorological environmental characteristics, the environmental risk coefficient matrix is ​​queried to obtain the environmental risk coefficient of the current driving behavior event; Furthermore, a weighted risk score for the driving behavior event is determined based on the basic risk score and the environmental risk coefficient; Furthermore, an initial risk score is determined for each of the aforementioned trip segments based on the weighted risk score of the driving behavior events.

9. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the vehicle dynamic risk scoring method according to any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps in the vehicle dynamic risk scoring method according to any one of claims 1 to 5.

Images