Vehicle driving data analysis method and device, equipment and storage medium

Abstract

The invention provides a vehicle driving data analysis method, apparatus and device, and a storage medium. The method comprises the steps of obtaining driving data when a target person drives a vehicle; determining an abnormal judgment range corresponding to the driving data based on the ID of the target person; determining an abnormal data packet containing the abnormal driving data based on the abnormal judgment range; and inputting the abnormal data packet into a pre-trained association rule model, and obtaining a target person driving abnormality occurrence reason output by the association rule model. By adopting the technical scheme provided by the invention, the driving safety is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a vehicle driving data analysis method, device, equipment and storage medium. BACKGROUND

[0002] With the rapid development of technology, people's requirements for cars are getting higher and higher. Through accurate analysis and intelligent decision-making of driving behavior, the safety and driving experience of vehicles can be enhanced. The current driving behavior analysis method is to analyze based on single driving data.

[0003] However, due to the complexity of the actual driving environment and the complexity of the actual driving scene, the existing analysis method lacks in-depth analysis of driving behavior. SUMMARY

[0004] Therefore, the embodiments of the present application provide a vehicle driving data analysis method, device, equipment and storage medium, which realizes real-time synchronization of multi-format data, provides personalized suggestions for different drivers, and improves driving safety.

[0005] The present application mainly includes the following aspects: In a first aspect, the embodiments of the present application provide a vehicle driving data analysis method, which comprises: obtaining driving data of a target person driving a vehicle; determining an abnormality discrimination range corresponding to the driving data based on the ID of the target person; determining an abnormal data packet containing abnormal driving data based on the abnormality discrimination range; inputting the abnormal data packet into a pre-trained association rule model to obtain a target person driving abnormality occurrence reason output by the association rule model.

[0006] Further, before obtaining the driving data of the target person driving the vehicle, the analysis method comprises: performing denoising processing on initial driving data of the target person driving the vehicle to obtain denoised initial driving data; wherein the initial driving data includes initial physiological data, initial vehicle data, initial scene data and initial driver feature data; aligning the denoised initial vehicle data, initial scene data and initial driver feature data in time based on the time axis of the denoised initial physiological data to obtain denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data; determining the denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data as the driving data of the target person driving the vehicle.

[0007] Furthermore, determining the anomaly detection range corresponding to the driving data based on the target person's ID includes: Based on the target person's ID, determine the target person's identity information from the feature database; Based on the identity information, the first correction coefficient, and the second correction coefficient, the anomaly detection range corresponding to the driving data is determined.

[0008] Furthermore, based on the aforementioned anomaly detection range, the method for determining abnormal data packets containing abnormal driving data includes: The moment when the driving data falls within the anomaly detection range is defined as an abnormal moment; The driving data within a preset time range from the abnormal time is packaged to obtain an abnormal data packet containing the abnormal driving data.

[0009] Furthermore, the analysis method includes: Based on the cause of the target person's abnormal driving, provide personalized suggestions.

[0010] Secondly, embodiments of this application also provide a vehicle driving data analysis device, the analysis device comprising: The acquisition module is used to acquire driving data of the target person when driving a vehicle. The range determination module is used to determine the anomaly detection range corresponding to the driving data based on the ID of the target person; The packaging module is used to determine, based on the anomaly detection range, an abnormal data packet containing the abnormal driving data; The abnormal cause determination module inputs the abnormal data packet into a pre-trained association rule model to obtain the cause of the target person's driving abnormality output by the association rule model.

[0011] Furthermore, the analysis device also includes: The denoising module is used to denoise the initial driving data of the target person driving the vehicle, and obtain the denoised initial driving data; the initial driving data includes: initial physiological data, initial vehicle data, initial scene data and initial driver feature data; The alignment module is used to align the denoised initial vehicle data, initial scene data, and initial driver feature data in time, based on the time axis of the denoised initial physiological data, to obtain the denoised and aligned initial physiological data, initial vehicle data, initial scene data, and initial driver feature data. The driving data determination module identifies the initial physiological data, initial vehicle data, initial scene data, and initial driver feature data after denoising and alignment as the driving data of the target person when driving the vehicle.

[0012] Furthermore, the range determination module is specifically used for: Based on the target person's ID, determine the target person's identity information from the feature database; Based on the identity information, the first correction coefficient, and the second correction coefficient, the anomaly detection range corresponding to the driving data is determined.

[0013] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the vehicle driving data analysis method described in the first aspect or any possible implementation of the first aspect.

[0014] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the vehicle driving data analysis method described in the first aspect or any possible implementation of the first aspect.

[0015] This application provides a method, apparatus, device, and storage medium for analyzing vehicle driving data. The method involves acquiring driving data of a target person driving a vehicle; determining an anomaly detection range corresponding to the driving data based on the target person's ID; determining an anomaly data packet containing the abnormal driving data based on the anomaly detection range; and inputting the anomaly data packet into a pre-trained association rule model to obtain the cause of the target person's driving anomaly, as output by the association rule model.

[0016] This enables real-time synchronization of data in multiple formats, allowing for personalized suggestions to be provided to different drivers and improving driving safety.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a method for analyzing vehicle driving data provided in an embodiment of this application is shown; Figure 2 This illustration shows one of the structural schematic diagrams of a vehicle driving data analysis device provided in an embodiment of this application; Figure 3 This is a second schematic diagram of the structure of a vehicle driving data analysis device provided in an embodiment of this application; Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0021] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] The methods, apparatus, electronic devices, or computer-readable storage media described in this application can be applied to any scenario that requires the analysis of vehicle driving data. This application does not limit the specific application scenario, and any scheme using the vehicle driving data analysis methods and apparatus provided in this application is within the protection scope of this application.

[0023] It is worth noting that with the rapid development of technology, people have increasingly higher demands for automobiles. Precise analysis and intelligent decision-making of driving behavior can enhance vehicle safety and the driving experience. Current methods for analyzing driving behavior rely on analyzing single driving data points. However, due to the complexity of real-world driving environments and scenarios, existing analysis methods lack in-depth analysis of driving behavior.

[0024] To address the aforementioned issues, this application proposes a method, apparatus, device, and storage medium for analyzing vehicle driving data, which enables real-time synchronization of multi-format data, provides personalized suggestions to different drivers, and improves driving safety.

[0025] To facilitate understanding of this application, the technical solutions provided in this application will be described in detail below with reference to specific embodiments.

[0026] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for analyzing vehicle driving data provided in an embodiment of this application.

[0027] like Figure 1 As shown in the figure, the vehicle driving data analysis method provided in this application includes the following steps: Step S101: Obtain driving data of the target person while driving the vehicle.

[0028] Here, driving data includes: physiological data, vehicle data, scene data, and driver characteristic data. Physiological data includes: electroencephalogram (EEG), heart rate, electromyography (EMG), and electrodermatology (TEG). Vehicle data includes: vehicle speed, steering wheel angle, lateral acceleration, accelerator and brake pedal travel, and air conditioning parameters. Scene data includes: external scenes (such as lane lines and obstacles), internal scenes (such as driver's line of sight and driver operations), and audio (such as human voices and prompts). Driver characteristic data includes: driver ID, age, gender, driving experience, historical driving behavior data (such as frequency of rapid acceleration and braking), operational force, and braking frequency. The driver ID is linked to driver identity information (such as age, gender, and driving experience), and driver identity information can be extracted from the characteristic database based on the driver ID. Physiological data can be collected via a head-mounted EEG device and a wristband-based physiological analyzer; vehicle data can be collected via a CAN bus interface; scene data is collected from 16 8-megapixel cameras and 8 microphones; static data in the driver characteristic data is information pre-entered into the system, while dynamic data is collected through onboard sensors. In this application, driving data can also be collected using other acquisition devices; no restrictions are placed here. Multi-dimensional data provides a foundation for subsequent data synchronization and correlation interpretation. Simultaneously, the collection of driver characteristic data addresses the deficiency of existing technologies in terms of "no human factor data."

[0029] Prior to step S101, the analysis method further includes: S01. The initial driving data of the target person driving the vehicle is denoised to obtain the denoised initial driving data.

[0030] Here, the initial driving data includes: initial physiological data, initial vehicle data, initial scene data, and initial driver feature data. The specific method for denoising the initial driving data is as follows: (1) Physiological data is highly susceptible to power frequency interference and motion artifacts caused by head shaking and muscle tension. To obtain effective physiological data that can be used to analyze the driver's psychological state, firstly, a notch filter is used to pre-process the initial physiological data (removing power frequency interference). As an example, a 50Hz notch filter is used in this application. Subsequently, wavelet transform is used for denoising. As an example, a db4 wavelet basis can be used to perform three layers of denoising on the physiological data. Specifically, the noise standard deviation σ is estimated based on the high-frequency coefficients of the initial physiological data itself, and the optimal denoising threshold λ is automatically calculated in combination with the signal length N, i.e. All wavelet coefficients below the denoising threshold (determined as noise) are removed, while coefficients above the threshold (determined as valid signals) are retained. For example, the input original signal is a 250Hz EEG signal containing δ, θ, α, β, and γ frequency band components as well as noise. After denoising using the above method, the output is an α wave reflecting attention and a β wave reflecting tension.

[0031] (2) The acquisition equipment used to collect the initial vehicle data is affected by various noises, such as uneven road surface or equipment error. This application uses Kalman filtering to remove noise from the initial vehicle data. As an example, the input original signal is lateral acceleration (collected by the vehicle-mounted accelerometer), which contains random noise (such as high-frequency fluctuations caused by road bumps). Through Kalman filtering, high-frequency noise is removed, and the processed lateral acceleration data is output.

[0032] (3) Initial scene data is first denoised by Gaussian filtering on video frames; then, machine learning algorithms are used to remove false alarm signals. As an example, 3×3 Gaussian filtering can be used to denoise video frames. Then, the DBSCAN algorithm can be used to remove outliers, i.e., false targets in the scene (such as false alarm signals of obstacles caused by ground reflection, weather factors or stationary objects).

[0033] (4) The dynamic data in the initial driver characteristic data are noise removed using Kalman filtering.

[0034] S02. Using the time axis of the denoised initial physiological data as a reference, the denoised initial vehicle data, initial scene data and initial driver feature data are time-aligned to obtain the denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data.

[0035] Specifically, in this embodiment, to ensure data synchronization and accuracy, all acquisition devices use a unified clock (such as GPS time and the device's local clock) to ensure that each frame or each data entry has a timestamp in a unified format. For example, EEG signal jumps are correlated with steering wheel angle changes at the same time. A differentiated alignment strategy is adopted for data with different sampling rates, using the denoised initial vehicle data as the synchronization reference axis, maintaining the original sampling rate, and aligning other data. Specifically, physiological data has the highest sampling rate. By setting a sliding window, the sampled data under each sliding window is averaged. For example, in this application, the window size is 5, assuming the original sampling rate of the denoised initial physiological data is 250Hz, and the sampling rate after time alignment is 50Hz. If the scene data is video data, the denoised initial scene data uses a frame-level alignment method, that is, extracting the timestamp for each frame of video and aligning it with the time of the denoised initial vehicle data using an interpolation algorithm. Dynamic data in the denoised initial driver characteristic data is aligned with the time of the denoised initial vehicle data using an interpolation algorithm. This achieves millisecond-level time alignment and noise filtering for multi-format heterogeneous data, ensuring data availability.

[0036] S03. The initial physiological data, initial vehicle data, initial scene data and initial driver feature data after noise reduction and alignment are identified as the driving data of the target person when driving the vehicle.

[0037] See again Figure 1 Step S102: Based on the ID of the target person, determine the anomaly detection range corresponding to the driving data.

[0038] The following explains in detail how to determine the anomaly detection range corresponding to the driving data based on the target person's ID.

[0039] Regarding step S102, as an example in specific implementation, it may include the following steps: Step S1021: Determine the identity information of the target person from the feature database based on the target person's ID.

[0040] Step S1022: Based on the identity information, the first correction coefficient, and the second correction coefficient, determine the anomaly detection range corresponding to the driving data.

[0041] Here, based on the target person's identity information, the corresponding anomaly detection threshold is automatically loaded. The anomaly detection range can be determined based on the anomaly detection threshold without manual intervention. As an example, the anomaly detection threshold can be obtained by correcting the base threshold using driving experience and age. The anomaly detection threshold can be represented by formula (1).

[0042] (1).

[0043] in, The threshold for anomaly detection. This is the first correction factor. This is the second correction factor. For the driver's driving experience, For age, The basic threshold is used. Formula (1) is a statistical regression model based on a preset number of driver samples. and The least squares method was used for fitting. As an example, formula (1) was obtained by statistical regression of the age, driving experience, and actual threshold of 1000 groups of drivers. As an example, it is assumed that the anomaly detection range of driving data in the high-speed scenario is shown in Table 1.

[0044] Table 1. Examples of anomaly detection ranges for vehicle driving data

[0045] Step S103: Based on the anomaly detection range, determine the abnormal data packet containing the abnormal driving data.

[0046] The following explains in detail how to determine the abnormal data packet containing the abnormal driving data based on the aforementioned anomaly discrimination range.

[0047] Regarding step S103, as an example in specific implementation, it may include the following steps: Step S1031: The moment when the driving data is within the anomaly detection range is determined as an abnormal moment.

[0048] Here, the abnormal moment in this application is accurate to the millisecond level.

[0049] Step S1032: Package the driving data within a preset time range of the abnormal time interval to obtain an abnormal data packet containing the abnormal driving data.

[0050] Here, as an example, driving data from 5 seconds before and after the abnormal moment is packaged, that is, physiological signal data, vehicle data, scene data, and driver characteristic data are linked and integrated to form structured data (such as JSON format). As an example, the linked and integrated driving data at the abnormal moment is as follows: {"Abnormal moment": "10:23:45.123", "Physiological characteristics": {"Heart rate change": "+30 beats / min", "EEG alpha wave": "reduced by 50%"}, "Vehicle characteristics": {"Lateral acceleration": "0.45g", "Steering wheel angle": "20°", "Vehicle speed": "60km / h"}, "Scene characteristics": {"Outside the vehicle": "Pedestrian crossing", "Inside the vehicle": "Driver not looking ahead"}.

[0051] Step S104: Input the abnormal data packet into the pre-trained association rule model to obtain the cause of the abnormal driving of the target person output by the association rule model.

[0052] In this embodiment of the application, the Apriori association rule algorithm is used to construct an association rule model, and the association relationship between "scene-physiology-vehicle-driver" is mined from multi-dimensional data to form abnormal causal judgment rules (such as "pedestrian crossing - sudden increase in heart rate - sudden turn of steering wheel"). Specifically, the input to the pre-training association rule model consists of a pre-defined number of anomaly records, each associated with driving data, and pre-defined minimum support and minimum confidence thresholds. The output is a set of association rules in the format "condition-conclusion (confidence)". For example, "pedestrian crossing outside the vehicle - sudden increase in heart rate of 20 beats / min - abnormal operation (confidence 92%)". The output rules are sorted by a combination of confidence and support, prioritizing rules with higher confidence. The pre-training association rule model processes the input as follows: iteratively generating a candidate set and performing pruning to select multimodal feature combinations with support no less than the minimum support threshold (e.g., "pedestrian + sudden increase in heart rate"). It extracts "condition-conclusion" relationships from the frequent itemset, calculates the confidence score, and filters out rules with confidence scores below the threshold. Finally, it verifies the generalization ability of the rules using a test set and removes overfitting rules. The confidence score is calculated as: (Number of samples meeting the condition and the conclusion) ÷ (Total number of samples meeting the condition) × 100%.

[0053] As an example, as shown in Table 2, the abnormal scenario sample library contains 1000 abnormal records. Assuming the minimum support threshold is 5% and the minimum confidence threshold is 80%, when the three features "pedestrian crossing," "heart rate spike of 20 beats / min," and "steering wheel angle 15° / 0.5s" are found to frequently appear in the abnormal samples (support > 5%, i.e., occurrence frequency exceeds 5%), and 92 out of 100 simultaneously occurring abnormal samples lead to vehicle abnormalities, then the rule "pedestrian crossing - heart rate spike - abnormal operation (confidence 92%)" is generated. When real-time driving data triggers this rule, it will be immediately marked as a valid abnormality, and the output "abnormal operation caused by scenario trigger" will be displayed.

[0054] Table 2. Example Table of Causal Association Rules for Driving Abnormalities

[0055] In one possible implementation, the analysis method further includes providing personalized suggestions to the target person based on the cause of the driving abnormality. As an example, this specifically includes the following steps: S1. Convert the association rules related to the abnormal driving of the target person into natural language descriptions to obtain the scene reconstruction text of the abnormal driving of the target person.

[0056] S2. Extract key abnormal information from the scene reconstruction text using a preset extraction model.

[0057] In this embodiment, the BERT-NER model is used to extract key anomaly information (e.g., problem type "distraction", associated object "pedestrian", driver characteristic "novice") from the scene reconstruction text. The input to the BERT-NER model before training consists of a preset number of manually annotated driving scene texts, with annotations including entity labels such as "problem type (distraction, accidental touch, etc.), associated object (call function, steering wheel, etc.), driver characteristic (novice, experienced driver, etc.)". The output of the BERT-NER model before training is the key anomaly information. The processing procedure for the input content by the BERT-NER model before training is as follows: converting the manually annotated driving text into BERT input format (word segmentation, adding special symbols [CLS] and [SEP]), training the model, optimizing parameters using the cross-entropy loss function, and outputting entity labels for newly input driving scene texts.

[0058] As an example, suppose the scene reconstruction text is: Driver A (23 years old, 0.5 years of driving experience, novice driver) experienced a sudden increase in heart rate to 30 beats per minute when a pedestrian was crossing outside the car. Due to distraction, he failed to observe in time, causing the steering wheel to turn sharply by 20 degrees, resulting in a lateral acceleration of 0.45g (within the anomaly detection range of 0.4g). The above scene reconstruction text is input into the BERT-NER model. The BERT-NER model outputs {"Problem Type": "Distraction + Overreaction", "Related Objects": "Pedestrian crossing scene, novice driver lacks driving experience", "Key Anomaly Features": "Excessive lateral acceleration, sudden increase in heart rate"}.

[0059] S3. Calculate the similarity between the key anomaly information and the matching information in the optimization solution library, and select the optimization solutions corresponding to the matching information with a similarity greater than the preset similarity threshold.

[0060] In this embodiment, the matching information is the key anomaly information corresponding to the optimization scheme. As an example, the GNN matching algorithm can be used for filtering, and the preset similarity threshold can be 0.7. An example of the optimization scheme library content is shown in Table 3.

[0061] Table 3. Examples of Optimization Solution Library Contents

[0062] S4. Prioritize the selected optimization solutions and determine the ranked optimization solutions as personalized suggestions for the target individuals.

[0063] Here, for each selected optimization scheme, its proximity score is determined based on its feasibility, effectiveness, and cost. Schemes are then prioritized according to their proximity scores. The proximity score is obtained by multiplying the feasibility, effectiveness, and cost of each optimization scheme by their respective weights and then summing the results. For example, the feasibility weight could be 0.4, the effectiveness weight could be 0.35, and the cost weight could be 0.25.

[0064] As an example, the personalized suggestions are: Short-term solution: When the pedestrian detection module is triggered, activate the HUD visual reminder simultaneously (proximity 1.0); Medium-term solution: Add steering wheel vibration feedback (proximity 0.65); Long-term solution: Optimize the AEB system to trigger automatic deceleration 0.5 seconds in advance in novice driving mode (proximity 0.42).

[0065] This application provides a method for analyzing vehicle driving data. This method enables real-time synchronization of multi-format data, provides personalized suggestions to different drivers, and improves driving safety.

[0066] Based on the same application concept, this application also provides a vehicle driving data analysis device corresponding to the vehicle driving data analysis method provided in the above embodiments. Since the principle of the device in this application to solve the problem is similar to the vehicle driving data analysis method in the above embodiments of this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0067] Please see Figures 2 to 3 , Figure 2 This is one of the structural schematic diagrams of a vehicle driving data analysis device provided in an embodiment of this application. Figure 3 This is a second schematic diagram of a vehicle driving data analysis device provided in an embodiment of this application.

[0068] like Figure 2 As shown in the figure, the vehicle driving data analysis device 210 provided in this application embodiment includes: Module 211 is used to acquire driving data of the target person when driving a vehicle; The range determination module 212 is used to determine the anomaly detection range corresponding to the driving data based on the ID of the target person; Packaging module 213 is used to determine, based on the anomaly discrimination range, an abnormal data packet containing the abnormal driving data; The anomaly cause determination module 214 inputs the anomaly data packet into a pre-trained association rule model to obtain the cause of the target person's driving anomaly output by the association rule model.

[0069] like Figure 3As shown, the analysis device 210 further includes: The denoising module 215 is used to denoise the initial driving data of the target person driving the vehicle to obtain denoised initial driving data; wherein, the initial driving data includes: initial physiological data, initial vehicle data, initial scene data and initial driver feature data; Alignment module 216 is used to perform time alignment of the denoised initial vehicle data, initial scene data and initial driver feature data based on the time axis of the denoised initial physiological data, so as to obtain the denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data. The driving data determination module 217 determines the initial physiological data, initial vehicle data, initial scene data and initial driver feature data after noise reduction and alignment as the driving data when the target person is driving the vehicle.

[0070] Furthermore, the range determination module 212 is specifically used for: Based on the target person's ID, determine the target person's identity information from the feature database; Based on the identity information, the first correction coefficient, and the second correction coefficient, the anomaly detection range corresponding to the driving data is determined.

[0071] Furthermore, the packaging module 213 is specifically used for: The moment when the driving data falls within the anomaly detection range is defined as an abnormal moment; The driving data within a preset time range from the abnormal time is packaged to obtain an abnormal data packet containing the abnormal driving data.

[0072] Furthermore, the analysis device 210 also includes: The suggestion module 218 is used to determine personalized suggestions for the target person based on the cause of the target person's abnormal driving.

[0073] This application provides a vehicle driving data analysis device that enables real-time synchronization of multi-format data, provides personalized suggestions to different drivers, and improves driving safety.

[0074] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0075] like Figure 4 As shown, the electronic device 400 includes a processor 410, a memory 420, and a bus 430.

[0076] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, they can perform the operations described above. Figure 1 The steps of the vehicle driving data analysis method in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0077] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 The steps of the vehicle driving data analysis method in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0078] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0079] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0080] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0081] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the 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 to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0082] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of analyzing vehicle driving data, characterized by, The analysis method comprises: obtaining driving data of a target person driving a vehicle; determining an abnormality judgment range corresponding to the driving data based on the ID of the target person; determining an abnormal data packet containing the abnormal driving data based on the abnormality judgment range; inputting the abnormal data packet into a pre-trained association rule model to obtain a driving abnormality occurrence reason of the target person output by the association rule model.

2. The analysis method of vehicle driving data according to claim 1, characterized in that, Before obtaining the driving data of the target person driving the vehicle, the analysis method further comprises: performing denoising processing on initial driving data of the target person driving the vehicle to obtain denoised initial driving data; wherein the initial driving data comprises initial physiological data, initial vehicle data, initial scene data and initial driver feature data; performing time alignment on the denoised initial vehicle data, initial scene data and initial driver feature data based on the time axis of the denoised initial physiological data to obtain denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data; determining the denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data as the driving data of the target person driving the vehicle.

3. The analysis method of vehicle driving data according to claim 1, characterized in that, The determination of the abnormality judgment range corresponding to the driving data based on the ID of the target person comprises: determining the identity information of the target person from a feature database according to the ID of the target person; determining the abnormality judgment range corresponding to the driving data based on the identity information, a first correction coefficient and a second correction coefficient.

4. The analysis method of vehicle driving data according to claim 1, characterized in that, The determination of the abnormal data packet containing the abnormal driving data based on the abnormality judgment range comprises: determining an abnormal time point as a time point at which the driving data is in the abnormality judgment range; packing the driving data within a preset time range from the abnormal time point to obtain the abnormal data packet containing the abnormal driving data.

5. The analysis method of vehicle driving data according to claim 1, characterized in that, The analysis method comprises: determining a personalized suggestion for the target person based on the driving abnormality occurrence reason of the target person.

6. An analysis device of vehicle driving data, characterized by, The analysis device comprises: an obtaining module configured to obtain driving data of a target person driving a vehicle; a range determining module configured to determine an abnormality judgment range corresponding to the driving data based on the ID of the target person; a packing module configured to determine an abnormal data packet containing the abnormal driving data based on the abnormality judgment range; an abnormality reason determining module configured to input the abnormal data packet into a pre-trained association rule model to obtain a driving abnormality occurrence reason of the target person output by the association rule model.

7. The device for analyzing vehicle driving data according to claim 6, characterized by, The analysis device further comprises: a denoising module configured to perform denoising processing on initial driving data of the target person driving the vehicle to obtain denoised initial driving data; wherein the initial driving data comprises initial physiological data, initial vehicle data, initial scene data and initial driver feature data; An alignment module is configured to perform time alignment on the denoised initial vehicle data, the initial scene data and the initial driver feature data based on the time axis of the denoised initial physiological data, to obtain denoised and aligned initial physiological data, initial vehicle data, initial scene data and initial driver feature data. A driving data determination module is configured to determine the denoised and aligned initial physiological data, the initial vehicle data, the initial scene data and the initial driver feature data as driving data of the target person driving the vehicle.

8. The device for analyzing vehicle driving data according to claim 6, characterized by, The range determination module is specifically configured to: determine identity information of the target person from the feature database according to an ID of the target person; and determine an abnormality judgment range corresponding to the driving data based on the identity information, a first correction coefficient and a second correction coefficient.

9. An electronic device, comprising: The electronic device comprises: a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the vehicle driving data analysis method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to perform the steps of the vehicle driving data analysis method according to any one of claims 1 to 5.

Images