Driving behavior safety scoring method, device, equipment, medium and program

Abstract

The invention provides a driving behavior safety scoring method and device, equipment, a medium and a program, relates to the technical field of Internet of Vehicles big data processing, and aims to improve the accuracy of safety scoring so as to improve the driving safety. The method comprises the following steps: acquiring operation data when a driver drives a vehicle in real time; processing based on the operation data to obtain driving behavior data corresponding to the driver; and inputting the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain a safety score of the driver.

Description

Technical Field

[0001] This application relates to the technical field of big data processing in the Internet of Vehicles, and particularly to a driving behavior safety scoring method, device, equipment, medium and program. Background Art

[0002] With the increasing improvement of the Internet of Vehicles, the safety scoring of drivers' driving behaviors has attracted more and more attention, which is related to the active safety of drivers during vehicle operation. Drivers with low safety scores tend to engage in dangerous driving behaviors, such as excessive numbers of hard accelerations and lane departures; on the contrary, drivers with high safety scores will pay more attention to the safety and stability of driving behaviors and perform fewer high-risk driving operations.

[0003] However, the driving behavior scoring methods in related technologies have problems with poor scoring accuracy. Therefore, it is necessary to provide a driving behavior safety scoring method to improve the accuracy of driving behavior safety scoring and thus improve driving safety. Summary of the Invention

[0004] To solve the above technical problems, this application provides a driving behavior safety scoring method, device, equipment, medium and program to improve the accuracy of driving behavior safety scoring and thus improve driving safety.

[0005] To achieve the above technical objectives, this application provides the following technical solutions:

[0006] In a first aspect, an embodiment of this specification provides a driving behavior safety scoring method, including:

[0007] Obtain the operation data of the driver when driving the vehicle in real time;

[0008] Process the operation data to obtain the driving behavior data corresponding to the driver;

[0009] Input the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

[0010] In a second aspect, an embodiment of this specification provides a driving behavior safety scoring device, including:

[0011] An acquisition unit for obtaining the operation data of the driver when driving the vehicle in real time;

[0012] A processing unit for processing the operation data to obtain the driving behavior data corresponding to the driver;

[0013] A scoring unit for inputting the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

[0014] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the method for driving behavior safety scoring according to the first aspect or any corresponding embodiment thereof.

[0015] In a fourth aspect, an embodiment of the present specification provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the method for driving behavior safety scoring as described in any one of the above is implemented.

[0016] In a fifth aspect, an embodiment of the present specification provides a computer program product or a computer program. The computer program product includes a computer program, and the computer program is stored in a computer-readable storage medium; a processor of the computer device reads the computer program from the computer-readable storage medium, and when the processor executes the computer program, the method for driving behavior safety scoring as described in any one of the above is implemented.

[0017] As can be seen from the above technical solutions, the present application provides a method, device, equipment, medium and program for driving behavior safety scoring. The method first obtains the running data in real time when a driver drives a vehicle, then processes the running data to obtain the driving behavior data corresponding to the driver, and finally inputs the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver. By obtaining and processing multi-dimensional running data in real time, the data dimension of the safety score is improved. At the same time, using the driving behavior scoring model to score the driver's safety improves the accuracy of the safety score, thereby improving the driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0019] Figure 1 It is a schematic flowchart of a method for driving behavior safety scoring provided by an embodiment of the present specification;

[0020] Figure 2 It is a schematic flowchart of a method for training a driving behavior scoring model provided by an embodiment of the present specification;

[0021] Figure 3A structural schematic diagram of a driving behavior safety scoring device provided for the embodiments of this specification;

[0022] Figure 4 A structural schematic diagram of an electronic device provided for the embodiments of this specification. Specific embodiments

[0023] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of this specification shall have the ordinary meanings understood by those skilled in the art within the field to which this specification belongs. The "first", "second" and similar terms used in the embodiments of this specification do not denote any order, quantity or importance, but are only used to avoid confusion of components.

[0024] Unless otherwise required by the context, throughout the specification, "a plurality" means "at least two", and "including" is interpreted as open and inclusive, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples", etc., are intended to indicate that specific features, structures, materials or characteristics related to the embodiment or example are included in at least one embodiment or example of this specification. The schematic representations of the above terms do not necessarily refer to the same embodiment or example.

[0025] Next, the technical solutions in the embodiments of this specification will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this specification.

[0026] Overview

[0027] As described in the background art, with the increasing improvement of the development of the vehicle network, the problem of the safety scoring of drivers' driving behaviors has been paid more and more attention, which is related to the active safety of drivers during vehicle operation. Drivers with low safety scores tend to perform dangerous driving behaviors, such as excessive numbers of rapid accelerations, lane departure times, etc.; on the contrary, drivers with high safety scores will pay more attention to the safety and stability of driving behaviors and perform less high-risk driving operations.

[0028] However, the driving behavior scoring methods in the related art have the problem of poor scoring accuracy. Therefore, it is necessary to provide a driving behavior safety scoring method to improve the accuracy of driving behavior safety scoring and thus improve driving safety.

[0029] Therefore, it is necessary to provide a driving behavior safety scoring method to solve the above problems.

[0030] In order to solve the problem of poor scoring accuracy in the existing driving behavior scoring method, in the technical solution of this application, first, the running data of the driver when driving the vehicle is obtained in real time, then the running data is processed to obtain the driving behavior data corresponding to the driver, and finally, the driving behavior data corresponding to the driver is input into a pre-trained driving behavior scoring model to obtain the safety score of the driver. By obtaining and processing multi-dimensional running data in real time, the data dimension of the safety score is improved. At the same time, using the driving behavior scoring model to score the driver's safety improves the accuracy of the safety score, thereby improving the safety of driving.

[0031] Based on the above inventive concept, the driving behavior safety scoring method provided by the embodiments of this specification will be described exemplarily below.

[0032] Exemplary method

[0033] The embodiments of this specification provide a driving behavior safety scoring method, as Figure 1 shown, including:

[0034] S101. Obtain the running data of the driver when driving the vehicle in real time.

[0035] Specifically in implementation, according to the running data uploaded by the vehicle terminal in real time, it can be obtained through sensors set on the vehicle terminal, or obtained from the vehicle computer or other vehicle devices in other forms. Specifically, the format of the running data can be the data signal of the international standardized serial communication protocol (Controller Area Network, CAN). The acquisition interval of the data can be once per second. The running data specifically includes, for example, vehicle speed, lateral acceleration, longitudinal acceleration, total mileage, and events defined by human rules, such as the number of times of rapid acceleration per 100 kilometers, the number of times of rapid deceleration per 100 kilometers, the number of times of emergency braking per 100 kilometers, the number of times of risky cornering per 100 kilometers, the number of times of speeding per 100 kilometers, the number of times of single driving overtime per 100 kilometers, and the mileage of night driving per 100 kilometers, etc.

[0036] S102. Process the running data to obtain the driving behavior data corresponding to the driver.

[0037] Specifically in implementation, after obtaining the corresponding running data, perform statistical analysis on the running data, observe its laws and characteristics, and perform outlier determination on it to remove invalid data. At the same time, perform quantile normalization on each index to obtain the driving behavior data corresponding to the driver, thereby eliminating the sensitivity of the model to different scales and further improving the accuracy of the model.

[0038] S103. Input the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

[0039] In specific implementation, the driving behavior data obtained in step S102 is input into a trained driving behavior scoring model to obtain the safety score of the driver output by the model. Then, based on the safety score of the driver, the driving behavior of the driver is judged, so as to better perform subsequent operations such as warning the driver and providing auxiliary driving assistance, thereby improving driving safety.

[0040] The embodiment of this specification provides a specific process for a training method of a driving behavior scoring model, as Figure 2 shown, including:

[0041] S201. Obtain the historical operation data of multiple target drivers when driving vehicles.

[0042] In specific implementation, obtain the historical operation data of multiple target drivers when driving vehicles, and the corresponding historical safety score of each target driver. The historical operation data can be CAN data, and the corresponding historical safety score can be the safety score obtained by manual scoring or the safety score obtained through other forms such as machine learning. This embodiment does not make any limitations in this regard. In one example, the historical operation data may include the following field information:

[0043]

[0044]

[0045] S202. Process the historical operation data to obtain the corresponding historical driving behavior data of each target driver and the corresponding historical safety score of each target driver.

[0046] In specific implementation, historical operation data is processed and analyzed to obtain input data for inputting into the driving behavior scoring model, that is, historical driving behavior data, and historical safety scores as labels of the input data. Specifically, for the historical operation data obtained in S201, preprocessing is first performed, and then safety index factor analysis is carried out to obtain safety index factors. This safety index factor is the core safety feature of driving behavior reflecting driving safety, aiming to extract the key dimensions that can best reflect driving safety by reducing the dimension and fusing the original data. Then, historical driving behavior data and historical safety scores are obtained using the safety index factors. The preprocessing includes data cleaning methods such as outlier processing and data format conversion, and normalization processing. Outlier processing and data cleaning are used to eliminate invalid and incorrect data, and normalization processing is used to eliminate the sensitivity of the model to different scales of indicators. Still using the above example, for the obtained CAN data, outlier processing is first performed, which includes the following steps: deleting vehicle data with the vehicle identification number vin not conforming to the rules, deleting row data with the daily operation mileage less than 10 kilometers, deleting invalid values with the safety index being 0 or null, and replacing safety index values exceeding the quartile with the maximum value.

[0047] After preprocessing, safety index factor analysis is carried out, that is, correlation analysis, which can effectively eliminate the redundancy between indicators. Still using the above example, performing correlation analysis on the above CAN data can show that the number of hard accelerations per 100 kilometers, the number of hard decelerations per 100 kilometers, and the number of risky corner entries per 100 kilometers are closely related, the left lane departure warning and the right lane departure warning have a strong correlation, and there is a great correlation among the number of first-level collision warnings per 100 kilometers, the number of collision alarm activations per 100 kilometers, and the number of partial braking times per 100 kilometers, which is in line with our daily driving behavior experience. Then, for the above 8 safety index factors with relatively large correlations, factor analysis is carried out, and three new independent factors factor_1, factor_2, and factoe_3 are respectively generated, so as to play the role of factor dimensionality reduction. The similarity between the newly generated safety index factors is relatively low, so the redundancy of the carried information is relatively small, to better prepare for subsequent model analysis.

[0048] It should be noted that the number of safety index factors can be set according to requirements or can change dynamically, and this embodiment does not limit this.

[0049] After obtaining the safety index factors, determine the index weights corresponding to each safety index factor. The index weights include two parts: the objective data index weight and the subjective evaluation index weight. Specifically, for the objective data index weight part, that is, through objective data, calculate the weight proportion of the safety index factor. Since there are rich data laws and characteristics in the safety index information uploaded by the vehicle terminal, in order to effectively mine the logical relationship between data, perform Gaussian clustering on the newly obtained safety index factors above, so as to divide the driving behavior data into multiple styles. Label the corresponding driving behavior data for different styles, and then use the algorithm model to train the driving behavior data, so as to obtain the objective data index weight. Specific algorithms can choose algorithms such as the CRITIC algorithm, cross-entropy algorithm, random forest algorithm, and AHP analytic hierarchy process. Compare the performance of multiple index weight calculation methods. The effect of the random forest algorithm is better. Therefore, here choose to use the random forest algorithm model to train the driving behavior data, so as to obtain the safety index factor weight determined by the random forest method.

[0050] For the subjective evaluation index weight part, that is, judge the weight proportion of the safety index factor through artificial experience. The introduction of artificial prior knowledge can effectively standardize the rationality of the index weight. Using the analytic hierarchy process, divide the safety index into three different index dimensions, and define the weights of the index factors within different dimensions respectively, so as to obtain the subjective evaluation index weight.

[0051] After obtaining the objective data index weight and the subjective evaluation index weight, it is necessary to perform weight fusion to obtain the index weight. In order to effectively fuse the rich information of the driving behavior data, the least squares method is introduced to combine the unsupervised clustering weight and the analytic hierarchy process weight, and finally generate a reasonable and comprehensive historical safety score. In an example, the historical safety score is in units of days, that is, the safety score value of each target driver on the same day. Specifically, for the score value, 90 - 100, 80 - 90, 70 - 80, 55 - 70, 0 - 55 can correspond to different driving behavior styles: safe type, ordinary type, aggressive type, risky type, aggressive risky type. Of course, other score segments and scoring criteria can also be selected, and this embodiment does not limit this.

[0052] S203. Input the historical driving behavior data corresponding to each target driver into the initialized driving behavior scoring model, and compare the safety score corresponding to the output of the historical driving behavior data corresponding to each target driver with the historical safety score corresponding to each target driver.

[0053] In specific implementation, the historical safety score is used as the supervision label for driving behavior data. Before model training, the complete data set is divided into a training set and a test set according to a certain proportion. A typical division method is an 8:2 allocation, where 80% of the data is used for model parameter training, and the remaining 20% is used to verify the prediction accuracy of the model. This implementation does not limit this to ensure adaptability in different application scenarios. During the training process, the model updates its parameters by continuously optimizing the difference between the predicted score and the historical true score, and finally verifies the generalization performance and score reliability of the model on the test set.

[0054] S204. Train and generate a driving behavior scoring model based on the difference between the output safety score and the historical safety score corresponding to each target driver.

[0055] In specific implementation, a driving behavior scoring model is constructed based on the difference between the current output safety score and the historical score of the target driver. Through the performance comparison and analysis of multiple machine learning models, it is found that the multi-layer perceptron (MLP) model performs outstandingly in capturing the non-linear relationship between safety index factors. As a typical representative of the feed-forward neural network, the MLP not only has high prediction accuracy but also shows excellent generalization ability, and can effectively process the feature association of complex driving behavior data. For the MLP model, first, the model parameters are initialized standardly, and the training efficiency is improved by reasonably setting the initial weights and bias values. During the iterative training process, each time the driving behavior index data and the corresponding safety score are randomly selected from the training set as the input, and the parameter gradient of the target loss function is estimated based on the current sample to accurately quantify the adjustment direction and amplitude of each parameter. The small learning rate strategy (usually set to 0.001 - 0.01) is adopted, and the network parameters are updated progressively by the gradient descent method to ensure the stable convergence of the training process. By real-time tracking the change trend of the loss function value and combining with the dynamic evaluation of the parameter update amplitude threshold, the training is terminated when the preset number of iterations is reached or the convergence condition is met (such as the loss change < 0.1% for 10 consecutive iterations).

[0056] Input the preprocessed driving behavior data into the trained MLP model, and the corresponding safety score can be output. The mean squared error (MSE) is used as the core index for model evaluation:

[0057] Mean squared error: ∑(predicted value - true value) 2 / number of samples;

[0058] The performance is continuously optimized by minimizing the mean squared error between the predicted score and the true score. Empirical studies show that this solution innovatively combines the nonlinear processing ability of the MLP with the temporal characteristics of driving behavior analysis. Its deep network structure can automatically extract complex feature interaction relationships, improving the prediction accuracy compared to traditional linear models. The model can achieve a lower error rate on the test set and maintain stable generalization ability among different driver groups, providing a reliable intelligent solution for driving safety assessment.

[0059] In this embodiment, based on the CAN data uploaded in real time by the vehicle terminal and the events defined by human rules, the data accuracy is ensured through statistical analysis and outlier processing. At the same time, quantile normalization is used to eliminate the index scale difference and enhance the model robustness. Secondly, the information closely related to driving safety is fully collected to provide comprehensive data support for safety scoring. Finally, subjectively evaluation and objective data analysis are innovatively combined. Through algorithms such as the analytic hierarchy process, factor analysis, Gaussian clustering, and random forest, multi-source data are scientifically integrated to generate reasonable driving behavior data to train the model, making the model output a more accurate safety score. It effectively solves the problems of incomplete data, strong subjectivity, and unreasonable index weights in traditional scoring methods, providing a more accurate and objective intelligent solution for driving behavior assessment and significantly improving the accuracy and reliability of driving behavior safety scoring.

[0060] Exemplary device

[0061] In an exemplary embodiment of this specification, a driving behavior safety scoring device 300 is further provided, as Figure 3 shown, including:

[0062] An acquisition unit 301, configured to acquire the operation data of a driver when driving a vehicle in real time;

[0063] A processing unit 302, configured to process based on the operation data to obtain the driving behavior data corresponding to the driver;

[0064] A scoring unit 303, configured to input the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

[0065] In an implementation manner, the scoring unit 303 specifically trains the driving behavior scoring model through the following method:

[0066] Acquire the historical operation data of multiple target drivers when driving a vehicle;

[0067] Process the historical operation data to obtain the historical driving behavior data corresponding to each target driver and the historical safety score corresponding to each target driver;

[0068] Input the historical driving behavior data corresponding to each target driver into the initialized driving behavior scoring model, and compare the safety scores output corresponding to the historical driving behavior data of each target driver with the historical safety scores corresponding to each target driver;

[0069] Train and generate a driving behavior scoring model based on the difference between the output safety score and the historical safety score corresponding to each target driver.

[0070] In one implementation, the scoring unit 303 is specifically further configured to:

[0071] Preprocess the historical operation data to obtain the preprocessed historical operation data. The preprocessing includes cleaning, outlier processing, and quantile normalization;

[0072] Perform factor analysis on the preprocessed historical operation data to obtain safety index factors for characterizing driving safety correlation, and the index weights corresponding to each safety index factor;

[0073] Determine the historical driving behavior data corresponding to each target driver and the historical safety score corresponding to each target driver according to the safety index factors and index weights.

[0074] In one implementation, the scoring unit 303 is specifically further configured to:

[0075] For each target driver, use the preprocessed historical operation data to obtain safety index factors;

[0076] Based on the safety index factors, obtain the objective data index weights corresponding to each safety index factor. The objective data index weights are used to characterize the weights corresponding to each safety index factor determined according to objective data;

[0077] Based on the safety index factors, obtain the subjective evaluation index weights corresponding to each safety index factor. The subjective evaluation index weights are used to characterize the weights corresponding to each safety index factor determined based on human experience;

[0078] Determine the objective data index weights and the subjective evaluation index weights as the index weights.

[0079] In one implementation, the scoring unit 303 is specifically further configured to:

[0080] Use the analytic hierarchy process to determine the subjective evaluation index weights based on the safety index factors.

[0081] In one implementation, the scoring unit 303 is specifically further configured to:

[0082] Use the random forest algorithm to determine the objective data index weights based on the safety index factors.

[0083] The driving behavior safety scoring device provided in this embodiment belongs to the same inventive concept as the driving behavior safety scoring method provided in the above embodiments of the present application. It can execute the driving behavior safety scoring method provided in any of the above embodiments of the present application and has corresponding functional modules and beneficial effects for executing the driving behavior safety scoring method. For technical details not described in detail in this embodiment, reference may be made to the specific processing content of the driving behavior safety scoring method provided in the above embodiments of the present application, which will not be elaborated here.

[0084] Exemplary equipment

[0085] In an exemplary embodiment of this specification, an electronic device is further provided. As Figure 4 shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communications interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call the logical instructions in the memory 430 to execute the driving behavior safety scoring method, which includes:

[0086] Obtain the running data of the vehicle when the driver is driving in real time;

[0087] Process the running data to obtain the driving behavior data corresponding to the driver;

[0088] Input the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

[0089] In addition, when the logical instructions in the above-mentioned memory 430 are implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0090] Exemplary computer program product and storage medium

[0091] In addition to the above methods and devices, the driving behavior safety scoring method provided in the embodiments of this specification may also be a computer program product, which includes computer program instructions that, when run by a processor, cause the processor to execute the steps in the driving behavior safety scoring method according to various embodiments of this specification described in the "Exemplary Method" section above of this specification.

[0092] The computer program product can be written in any combination of one or more programming languages to write program code for performing the operations of the embodiments of this specification. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages.

[0093] In addition, the embodiments of this specification also provide a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to perform the steps in the driving behavior safety scoring method according to various embodiments of this specification described in the "Exemplary Method" section above of this specification.

[0094] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium, and when the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in this specification can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0095] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0096] The above-described embodiments merely represent several implementation manners of this specification. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the solutions provided by the embodiments of this specification. It should be noted that for those of ordinary skill in the art, without departing from the concept of this specification, several modifications and improvements can still be made, and these all belong to the protection scope of this specification. Therefore, the protection scope of the patent of this specification shall be subject to the appended claims.

Claims

1. A driving behavior safety scoring method, characterized in that including: Obtaining the running data of the driver when driving a vehicle in real time; Processing the running data to obtain the driving behavior data corresponding to the driver; Inputting the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

2. The method according to claim 1, characterized in that The driving behavior scoring model is trained by the following method: Obtaining the historical running data of multiple target drivers when driving a vehicle; Processing the historical running data to obtain the historical driving behavior data corresponding to each target driver and the historical safety score corresponding to each target driver; Inputting the historical driving behavior data corresponding to each target driver into an initialized driving behavior scoring model, and comparing the safety score output corresponding to the historical driving behavior data of each target driver with the historical safety score corresponding to each target driver; Training and generating the driving behavior scoring model based on the difference between the output safety score and the historical safety score corresponding to each target driver.

3. The method according to claim 2, wherein The processing of the historical running data to obtain the historical driving behavior data corresponding to each target driver and the historical safety score corresponding to each target driver includes: Preprocessing the historical running data to obtain preprocessed historical running data, and the preprocessing includes cleaning, outlier processing, and quantile normalization; Performing factor analysis on the preprocessed historical running data to obtain safety index factors for characterizing driving safety correlation and the index weights corresponding to each safety index factor; Determining the historical driving behavior data corresponding to each target driver and the historical safety score corresponding to each target driver according to the safety index factors and the index weights.

4. The method according to claim 3, wherein The analysis of the preprocessed historical running data to obtain the safety index factors for characterizing driving safety correlation and the index weights corresponding to each safety index factor includes: For each target driver, using the preprocessed historical running data to obtain the safety index factors; Based on the safety index factors, obtaining the objective data index weights corresponding to each safety index factor, and the objective data index weights are used to characterize the weights corresponding to each safety index factor determined according to objective data; Based on the safety index factors, obtaining the subjective evaluation index weights corresponding to each safety index factor, and the subjective evaluation index weights are used to characterize the weights corresponding to each safety index factor determined based on human experience; Determining the objective data index weights and the subjective evaluation index weights as the index weights.

5. The method according to claim 4, wherein The obtaining of the objective data index weights corresponding to each safety index factor based on the safety index factors includes: Using the analytic hierarchy process to determine the subjective evaluation index weights based on the safety index factors.

6. The method according to claim 4, wherein The obtaining of the subjective evaluation index weights corresponding to each safety index factor based on the safety index factors includes: Using the random forest algorithm to determine the objective data index weights based on the safety index factors.

7. A driving behavior safety scoring device, characterized in that, including: An acquisition unit for acquiring in real time the operation data of a vehicle when a driver is driving the vehicle; A processing unit for processing based on the operation data to obtain the driving behavior data corresponding to the driver; A scoring unit for inputting the driving behavior data corresponding to the driver into a pre-trained driving behavior scoring model to obtain the safety score of the driver.

8. A computer device, characterized in that, Comprising: A memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the driving behavior safety scoring method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the driving behavior safety scoring method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the driving behavior safety scoring method according to any one of claims 1 to 6.

Images