Fault prediction analysis method based on driving behavior and related device

Abstract

The invention discloses a fault prediction analysis method based on driving behaviors and a related device. The method comprises the following steps: acquiring current driving behavior data of a target driver for a target vehicle; determining a driving habit type of the target driver according to the current driving behavior data; when the driving habit type is the preset driving habit type, obtaining driver state data of the target driver; determining a target danger index according to the driving habit type and the driver state data; acquiring historical driving behavior data and historical fault data of the target vehicle, and training a preset model based on the historical driving behavior data and the historical fault data to obtain a correlation analysis model; and when the target danger index is greater than a preset danger index threshold, inputting the current driving behavior data into the correlation analysis model to obtain a target fault probability. By adopting the method and the device, the fault is accurately predicted based on the driving behavior so as to avoid the occurrence of the fault.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a fault prediction analysis method based on driving behavior and related devices. Background Art

[0002] In the existing field of fault diagnosis technology, vehicle faults are usually diagnosed only when they occur, but some faults can be predicted and avoided in advance.

[0003] Existing fault prediction technology mainly relies on vehicle sensor data, such as engine temperature, oil pressure, etc., as well as vehicle maintenance records. The data used for analysis has limitations and cannot fully reflect the vehicle condition. At the same time, the existing technology's predictive analysis of vehicle faults is only a simple analysis based on the vehicle's surrounding environment information, and does not conduct in-depth correlation analysis between driving behavior data and vehicle fault data, resulting in inaccurate and incomplete fault prediction.

[0004] Therefore, how to conduct a specific analysis of driving behavior data to accurately predict vehicle failures, thereby avoiding failures and reducing damage to vehicle components has become an urgent problem to be solved. Summary of the invention

[0005] The embodiment of the present application provides a fault prediction analysis method and related devices based on driving behavior, which analyzes the correlation between the driver's driving behavior and vehicle faults, and then accurately predicts faults based on the correlation to avoid the occurrence of faults.

[0006] In a first aspect, an embodiment of the present application provides a fault prediction analysis method based on driving behavior, which is applied to a vehicle diagnostic device, and the method includes:

[0007] Acquire current driving behavior data of a target driver for a target vehicle;

[0008] Determining the driving habit type of the target driver according to the current driving behavior data;

[0009] When the driving habit type is a preset driving habit type, obtaining driver status data of the target driver;

[0010] determining a target risk index according to the driving habit type and the driver status data;

[0011] Acquiring historical driving behavior data and historical fault data of the target vehicle, and training a preset model based on the historical driving behavior data and the historical fault data to obtain an association analysis model; the association analysis model is used to generate a fault probability based on the driving behavior data;

[0012] When the target risk index is greater than a preset risk index threshold, the current driving behavior data is input into the association analysis model to obtain a target failure probability.

[0013] In a second aspect, an embodiment of the present application provides a fault prediction and analysis device based on driving behavior, which is applied to vehicle diagnostic equipment. The fault prediction and analysis device based on driving behavior includes: a data acquisition module, a data processing module, a model training module, and a model analysis module, wherein:

[0014] The data acquisition module is used to obtain current driving behavior data of a target driver of a target vehicle;

[0015] The data processing module is used to determine the driving habit type of the target driver according to the current driving behavior data;

[0016] The data acquisition module is also used to obtain the driver status data of the target driver when the driving habit type is a preset driving habit type;

[0017] The data processing module is also used to determine a target risk index according to the driving habit type and the driver status data;

[0018] The model training module is used to obtain the historical driving behavior data and historical fault data of the target vehicle, and train the preset model based on the historical driving behavior data and the historical fault data to obtain an association analysis model; the association analysis model is used to generate a fault probability based on the driving behavior data;

[0019] The model analysis module is used to input the current driving behavior data into the association analysis model to obtain a target failure probability when the target risk index is greater than a preset risk index threshold.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in the first aspect of the embodiment of the present application.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

[0023] It can be seen that the following beneficial effects are achieved by using the embodiments of the present application:

[0024] By implementing the embodiment of the present application, the current driving behavior data of the target driver for the target vehicle is obtained; the driving habit type of the target driver is determined according to the current driving behavior data; when the driving habit type is a preset driving habit type, the driver state data of the target driver is obtained; the target risk index is determined according to the driving habit type and the driver state data; the historical driving behavior data and historical fault data of the target vehicle are obtained, and the preset model is trained based on the historical driving behavior data and the historical fault data to obtain the association analysis model; when the target risk index is greater than the preset risk index threshold, the current driving behavior data is input into the association analysis model to obtain the target failure probability. It can be seen that by analyzing the correlation between the driver's driving behavior and the vehicle failure, the vehicle failure is accurately predicted according to the correlation, thereby avoiding the occurrence of the failure and reducing the damage to the vehicle components. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the background technology, the drawings required for use in the embodiments of the present application or the background technology will be described below.

[0026] Figure 1 It is a structural diagram of a network architecture provided by an embodiment of the present application;

[0027] Figure 2 It is a flowchart of a fault prediction analysis method based on driving behavior provided in an embodiment of the present application;

[0028] Figure 3 It is a structural schematic diagram of a fault prediction and analysis device based on driving behavior provided in an embodiment of the present application;

[0029] Figure 4 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0031] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0032] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0033] The following is an explanation of the relevant contents, concepts, meanings, technical issues, technical solutions, beneficial effects, etc. involved in the embodiments of the present application.

[0034] See also Figure 1 , Figure 1 Schematic diagram of a network architecture provided by an embodiment of the present application. Figure 1 As shown, the network architecture may include a remote server 101, a vehicle diagnostic device 102 and a target vehicle 103, etc. The network architecture may include one or more remote servers. There is no limit on the number of remote servers, and it can be flexibly set according to actual application scenarios and requirements.

[0035] like Figure 1As shown, the remote server 101 can establish a network connection with the vehicle diagnostic device 102 and the target vehicle 103, so that the remote server 101 can exchange data with the vehicle diagnostic device 102 and the target vehicle 103 through the network connection. At the same time, the vehicle diagnostic device 102 and the target vehicle 103 can also be connected, and the connection method can be a network connection or a physical connection, wherein the network connection can be connected through wireless communication technology or wired network transmission, and the physical connection can be connected through data cables, interface docking, etc., so as to realize the transmission and sharing of information.

[0036] In the embodiment of the present application, the remote server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud database, cloud service, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform. The remote server 101 is used to quickly analyze and process a large amount of vehicle driving behavior data and fault data, integrate multi-source data through the big data platform, and use artificial intelligence algorithms to mine vehicle fault association patterns in multi-source data, thereby achieving accurate vehicle fault prediction and diagnosis.

[0037] The vehicle diagnostic device 102 can be a vehicle-mounted terminal device that integrates a variety of professional detection modules. It has a built-in on-board diagnostic system (On-Board Diagnostics, OBD) interface reader, which can be directly connected to the OBD interface of the target vehicle 103, and obtain detailed fault codes and various real-time operating parameters fed back by each electronic control unit (Electronic Control Unit, ECU) of the vehicle in real time, such as engine speed, coolant temperature, vehicle speed, throttle opening and other key data information. The vehicle diagnostic device 102 is also equipped with a data processing module and a communication module. The data processing module is used to analyze the correlation between the driver's driving behavior data obtained and the vehicle fault, and then accurately predict the fault according to the correlation. The communication module supports a variety of communication protocols, such as Bluetooth, Wi-Fi, 4G / 5G and other wireless communication methods and controller area network (Controller Area Network, CAN) bus and other wired communication methods, so as to ensure that the processed data can be stably and efficiently transmitted to the remote server 101, and realize data interaction between the vehicle and the remote server 101.

[0038] The target vehicle 103 may be equipped with multiple electronic control units (ECUs), which are used to accurately control and monitor different key components and systems in the target vehicle 103. The target vehicle 103 also includes an on-board diagnostic system, which is used to collect and analyze the fault codes and related operating parameter information of multiple electronic control units in real time. These multiple electronic control units exchange data through communication networks such as the CAN bus in the vehicle, coordinate the operation of various components of the vehicle in real time, and can also feed back their own operating status information to the vehicle diagnostic device 102, thereby providing a rich data foundation for vehicle fault diagnosis, driving behavior analysis, and subsequent remote maintenance management.

[0039] For example, when the driver steps on the accelerator pedal, the accelerator pedal position sensor transmits the pedal opening information to the engine control unit. The engine control unit combines the data of other related sensors (such as vehicle speed sensor, intake pressure sensor, etc.) and sends a request to adjust the gear position to the transmission control unit through the CAN bus. The transmission control unit performs reasonable gear shifting operations according to the current vehicle speed, engine speed, etc., and then feeds back the information of gear shift completion to the engine control unit and other related ECUs such as the body control unit through the CAN bus. Through the coordinated work of various electronic control units, the smooth acceleration of the vehicle is ensured. In this process, various operating status information generated by various electronic control units will be recorded in real time and can be transmitted to the vehicle diagnostic device 102, so as to perform fault diagnosis or driving behavior analysis for the vehicle. At the same time, these operating status information can also be transmitted to the remote server 101, so that the remote server 101 can perform remote maintenance management. When conducting driving behavior analysis, the driver's driving habits (such as the frequency and characteristics of sudden acceleration, sudden braking, frequent gear shifting, etc.) can be analyzed based on the operating status data of the target vehicle 103, so as to analyze the possible faults caused by these driving habits, and predict and warn of possible vehicle faults in advance to avoid the occurrence of faults and reduce damage to vehicle components.

[0040] See also Figure 2 , Figure 2 : is a flow chart of a fault prediction analysis method based on driving behavior provided in an embodiment of the present application. The method is applied to a vehicle diagnostic device. The method includes but is not limited to the following steps:

[0041] S201: Acquire current driving behavior data of a target driver of a target vehicle.

[0042] In the embodiment of the present application, the target vehicle is the vehicle driven by the target driver, which can be a car, a truck, a bus, etc., without limitation. The target vehicle is used to analyze the overall traffic travel characteristics and driving behavior patterns in specific traffic scenes (such as highways, mountain trails, urban roads, etc.).

[0043] In a specific embodiment, the target vehicle is equipped with a speed sensor, an acceleration sensor, a steering angle sensor, an accelerator pedal position sensor, a camera, a distance sensor, a global positioning system (GPS), and other devices, which are used to collect the behavior data of the target driver operating the target vehicle during the driving process of the vehicle. Specifically, the current driving behavior data of the target driver of the target vehicle can be obtained through each data collection device.

[0044] Among them, the current driving behavior data refers to the driving behavior data generated when the target driver operates the target vehicle, and the target vehicle responds to various operating behaviors of the target driver (such as changing the vehicle speed, adjusting the driving direction, controlling the power output, responding to changes in road conditions, etc.). The current driving behavior data includes: vehicle speed, acceleration, acceleration change, steering angle, steering frequency, pedal depression depth, pedal depression frequency, following distance, vehicle position, road type, etc., which are not limited here.

[0045] By analyzing the behavioral characteristics and abnormal changes that may be generated by the current driving behavior data, as well as the trend differences compared with past data, an accurate assessment of the driver's driving habits can be achieved to discover potential driving risks and issue targeted warnings to the driver.

[0046] S202: Determine the driving habit type of the target driver according to the current driving behavior data.

[0047] Among them, driving habit type refers to the general classification of relatively stable and regular driving behavior characteristics exhibited by the target driver in the long-term driving process, which includes the driver's style preferences when operating the vehicle, the way of coping with different road conditions and traffic scenarios, and driving habits in speed control, vehicle control, etc.

[0048] For example, driving habits can be divided into two categories: aggressive and conservative. The aggressive type usually shows frequent lane changes, sudden acceleration, sudden braking, and following the car in front, while the conservative type usually shows cautious speed control and gentle acceleration and deceleration. Of course, driving habits can also be subdivided into multiple other types, and different habit types can be used for different driving behavior analysis, which will not be elaborated here.

[0049] In a specific embodiment, the target driver's driving habit type can be determined based on the current driving behavior data. The current driving behavior data is analyzed to extract relevant features in the current driving behavior data, including speed, steering, pedal operation and other relevant features, and the features are evaluated and determined, so as to determine the target driver's driving habit type based on the features. Once the driver's driving habit type is determined, for driving habits with higher risks, risk reminders can be provided to the target driver in a targeted manner. Specifically, the target driver can be reminded through the vehicle voice system to adjust the driving style to avoid the occurrence of failures. For driving habits with lower risks, personalized safe driving suggestions can be provided to the target driver to ensure the good operation of the vehicle, which helps to optimize the use and maintenance of the vehicle, reduce the occurrence of failures, and extend the service life of the vehicle.

[0050] Optionally, the step of determining the driving habit type of the target driver according to the current driving behavior data may specifically include the following steps:

[0051] Extracting operation frequency data from the current driving behavior data; the operation frequency data including at least one of the following: acceleration frequency, braking frequency, steering frequency, deceleration frequency, and lane change frequency; acquiring driving environment data corresponding to the current driving behavior data; determining target driving behavior characteristics based on the operation frequency data and the driving environment data; determining the driving habit type based on the target driving behavior characteristics.

[0052] In a specific embodiment, operation frequency data can be extracted from the current driving behavior data, wherein the operation frequency data includes at least one of the following: acceleration frequency, braking frequency, steering frequency, deceleration frequency, lane change frequency, etc., which are not limited here. Specifically, the operation frequency data refers to the number of times the target driver performs the corresponding operation per unit time or per unit mileage, and the target driver's driving habit type can be preliminarily determined based on the operation frequency data. However, due to different traffic conditions or traffic scenes, the environment will also affect the target driver's habit of operating the vehicle. Therefore, the driving environment data corresponding to the current driving behavior data can be obtained, wherein the driving environment data can include: road type, traffic flow conditions, weather conditions, etc., which are not limited here.

[0053] For example, a target driver drives on a mountain road for a long time. The mountain road is winding and rugged with many bends, so the target driver needs to frequently perform steering operations, which makes the steering frequency relatively high. When determining the target driver's driving habit type based on the current driving behavior data, if the driving habit type is evaluated by the steering frequency, the characteristic environmental factors of the mountain road will have a very large impact on the evaluation criteria, which cannot reflect the target driver's driving habits. Therefore, it is necessary to conduct a closer analysis based on the specific operation frequency data and the driving environment data corresponding to the current driving behavior data to accurately judge the target driver's driving habits.

[0054] Furthermore, after extracting the operation frequency data and obtaining the driving environment data corresponding to the current driving behavior data, the target driving behavior characteristics can be determined based on the operation frequency data and the driving environment data, wherein the target driving behavior characteristics refer to the representative characteristics of the driver's operation frequency under different driving conditions that can reflect his driving behavior pattern. For example, in a crowded urban road section with large traffic flow and narrow roads, each vehicle needs to pass smoothly and orderly, but the braking frequency and lane change frequency of a certain target driver are high, so it can be determined that the target driver has driving behavior characteristics such as frequent braking and frequent lane changes. By quantitatively evaluating the target driving behavior characteristics, the target driver's driving habit type can be determined by setting threshold scores, model judgments, etc.

[0055] S203: When the driving habit type is a preset driving habit type, obtaining driver status data of the target driver.

[0056] In the embodiment of the present application, the preset driving habit type refers to the type of driver assessed to have aggressive driving habits, which means that the target driver's driving operations are relatively frequent.

[0057] In a specific embodiment, when the driving habit type is a preset driving habit type, the driver status data of the target driver can be obtained, and the driver status data can be analyzed to determine the danger index of the current driving habits. Since the driving habit type only reflects long-term behavioral preferences and cannot reflect the real-time conditions during the current driving process, in order to comprehensively assess driving risks and conduct targeted intervention reminders, the driver status data of the target driver can be obtained to provide timely warnings when unfavorable driving conditions are discovered to reduce accident risks.

[0058] Among them, the driver status data is relevant data that reflects the real-time physical, mental and attention status of the target driver during the driving process. The driver status data may include: fatigue status data, attention status data, emotional status data, physiological indicator data, etc., which are not limited here. According to the driver status data, it is possible to evaluate whether the target driver's status is suitable for driving, and analyze the driving risk under the driver's status. When it is found that the target driver is not suitable for driving the target vehicle or has a high driving risk, an early warning prompt can be issued immediately to prompt the target driver to adjust his status.

[0059] When the driving habit type is not a preset driving habit type, the target driver's driving behavior data and driver status data can be continuously monitored and recorded, and the risk change trend can be determined by analyzing the changes in driving behavior and driving status over time, road conditions, and other factors. Driving suggestions can also be provided to the target driver based on the characteristics reflected in the driving behavior data (such as reasonably controlling the speed and continuing to maintain a safe distance) to help the target driver optimize driving behavior and reduce the occurrence of faults.

[0060] Optionally, the step of obtaining the driver status data of the target driver may specifically include the following steps:

[0061] Collect target image data of the target driver; perform demeanor analysis on the target image data to obtain driving demeanor data of the target driver; and determine the driver state data of the target driver based on the driving demeanor data.

[0062] In a specific embodiment, the vehicle diagnostic device can be equipped with a camera module and an image analysis and processing module. The camera module can instruct the camera on the target vehicle to collect the target image data of the target driver in real time. The target image data includes the face and part of the upper body area of ​​the target driver. The collected target image data can be transmitted to the image analysis and processing module of the vehicle diagnostic device in real time for image analysis. By performing demeanor analysis on the target image data, the driving demeanor data of the target driver can be obtained, and the driver status data of the target driver can be determined based on the driving demeanor data.

[0063] For example, when it is detected that the target driver closes his eyes for a long time and frequently while driving, exceeding the reasonable range of normal blinking, and the head droops unconsciously, the driver can be judged to be in a fatigued state according to the preset judgment rules, and the corresponding fatigue driving demeanor data can be generated. Or if it is found that the driver's eyes frequently wander, his sight deviates from the road ahead for a long time, and he looks at other irrelevant areas in the car, such as frequently looking at the mobile phone or the in-car entertainment device, it is judged as inattentive driving demeanor data.

[0064] By collecting the target image data of the target driver and analyzing and determining the driver status data, it is possible to ensure that the target driver's driver status data is accurate. The driver status data is a quantitative value to evaluate the different driving states of the target driver. When the target driver is identified as being in a bad state, such as fatigue driving or distraction, the vehicle can immediately remind the driver through voice prompts, flashing dashboard warning lights, and seat vibrations, so that the driver can adjust his state and concentrate in time, thereby effectively reducing the risk of traffic accidents and vehicle failures caused by driver status problems.

[0065] Optionally, the driving demeanor data includes: facial expression data, eye gaze direction data and head posture data; the step of determining the driver status data of the target driver according to the driving demeanor data may specifically include the following steps:

[0066] S2301, obtaining reference facial expression data, reference eye gaze direction data, and reference head posture data;

[0067] S2302, determining first state offset data according to the facial expression data and the reference facial expression data;

[0068] S2303, determining second state offset data according to the eye gaze direction data and the reference eye gaze direction data;

[0069] S2304, determining third state offset data according to the head posture data and the reference head posture data;

[0070] S2305, obtaining weight information corresponding to the facial expression data, the eye gaze direction data, and the head posture data, and obtaining first weight information, second weight information, and third weight information respectively;

[0071] S2306. Determine the driver state data according to the first state offset data, the second state offset data, the third state offset data, the first weight information, the second weight information and the third weight information.

[0072] Among them, the reference facial expression data, reference eye gaze direction data and reference head posture data are standard driving data. For example, the reference facial expression data include the calm expression characteristics when driving normally, awake and focused, the reference eye gaze direction data include data such as the angle range of both eyes looking directly at the road ahead, and the reference head posture data include standard posture data of keeping the head upright and slightly tilted forward to facilitate observation of road conditions.

[0073] In a specific embodiment, the driver state data of the target driver is determined based on the facial expression data, the eye gaze direction data and the head posture data. The facial expression data is compared with the reference facial expression data to determine the degree of difference between the facial expression of the target driver and the normal state, and the first state deviation data is obtained. For example, the yawning frequency is higher than the normal reference value, and the deviation data shows that the deviation degree is large. The eye gaze direction data is compared with the reference eye gaze direction data to determine whether the eye gaze direction of the target driver is within the set eye gaze range, and the second state deviation data is obtained to determine the deviation of the eye gaze direction of the target driver. The head posture data is compared with the reference head posture data to determine the abnormality of the head posture. For example, if the head of the target driver droops from time to time, the third state deviation data is obtained.

[0074] Furthermore, weight information corresponding to the facial expression data, the eye gaze direction data and the head posture data is obtained to obtain first weight information, second weight information and third weight information respectively, the weight information is determined by analyzing the historical data, and different weight information represents the relative importance of judging the driver state data. The driver state data is obtained by weighted calculation based on the first state offset data, the second state offset data, the third state offset data, the first weight information, the second weight information and the third weight information, and the driving state of the target driver can be determined by analyzing the driver state data.

[0075] S204: Determine a target risk index according to the driving habit type and the driver status data.

[0076] In a specific embodiment, the driving habit type and the driver status data can be quantified to obtain a target danger index, which can be used to characterize the driver's current comprehensive danger level. When the target danger index exceeds a preset danger threshold, it indicates that the target vehicle may fail. Therefore, it is necessary to predict possible failures based on the target driver's driving habit type and his current driver status. The preset danger threshold can be a critical value used to evaluate whether the target driver's current driving condition is at a higher risk level. When the current danger index exceeds the preset danger threshold, it indicates that the possibility of a failure has increased significantly, prompting the target driver to take corresponding measures.

[0077] It should be noted that when the target danger index exceeds the preset danger threshold, it means that the current driving condition of the target driver is at a higher risk. Although it does not directly indicate that the target vehicle has broken down, it is likely to cause vehicle failure due to certain specific driving habits and poor driver conditions.

[0078] For example, if the target driver frequently accelerates and brakes suddenly, such high-intensity operation will put great pressure on the target vehicle's brake system, transmission system and other components, which will in turn aggravate the wear of the brake pads, and then there will be hidden dangers of longer braking distance or even brake failure. At the same time, frequent sudden acceleration will also cause great damage to the engine and transmission, which may easily cause engine shaking, abnormal transmission shifting and other problems. In this case, if the target driver is in a state of fatigue and has a slow reaction speed, when responding to sudden road conditions (such as passing speed bumps, potholes, etc.), and when driving, he cannot detect the abnormal warning signals issued by the target vehicle in time (such as the fault indicator light on the dashboard is on, etc.), and cannot make correct operations in time, so that small faults that can be handled in time develop into more serious fault problems.

[0079] Optionally, the step of determining a target risk index according to the driving habit type and the driver status data may specifically include the following steps:

[0080] The driving habit type and the driver status data are scored based on preset scoring rules to obtain a target score; a mapping relationship between the score and the danger index is obtained to obtain a target mapping relationship; and the target danger index is determined based on the target score and the target mapping relationship.

[0081] Among them, the preset scoring rules refer to the pre-set standards and methods for quantitative scoring of driving habit types and driver status data, which can convert data such as driving habit types and driver status data into specific, comparable and analyzable values, and thus accurately assess the driver's current comprehensive level of danger.

[0082] In a specific embodiment, the driving habit type and the driver status data can be scored based on a preset scoring rule to obtain a target score, and then the mapping relationship between the score and the danger index is obtained to obtain a target mapping relationship, and the target danger index is determined according to the target score and the target mapping relationship. In the actual driving process, the score and the danger index are not in the same proportional growth relationship, so it is necessary to obtain the target mapping relationship and determine the danger index corresponding to the current score according to the target mapping relationship. For example, although a driver is slightly aggressive in driving habits, his driving state is good, while another driver is also slightly aggressive in driving habits, but is in an extremely fatigued state. Due to the interaction between the fatigue state and the aggressive driving habits, the latter's driving behavior will increase exponentially. The mapping relationship between the score and the danger index is usually analyzed based on a large number of vehicle failure cases, risk research, etc., so as to construct a corresponding relationship that conforms to this nonlinear risk growth logic and more accurately reflect the real danger situation.

[0083] S205: Acquire historical driving behavior data and historical fault data of the target vehicle, and train a preset model based on the historical driving behavior data and the historical fault data to obtain a correlation analysis model.

[0084] Among them, the preset model refers to a basic model framework pre-constructed before performing tasks such as data analysis and machine learning, which may include a neural network model, a support vector machine (SVM) model, and a naive Bayes model. In the embodiment of the present application, by training the preset model, an association analysis model can be obtained to analyze the correlation between driving behavior data and fault data, and then predict the probability of fault occurrence.

[0085] In the embodiment of the present application, the association analysis model refers to a model constructed for analyzing the intrinsic connection between driving behavior data and vehicle fault data, which is used to generate fault probability based on driving behavior data. The association analysis model can learn and analyze a large amount of historical driving behavior data reflecting the driver's operating habits and vehicle driving status information (such as the vehicle's acceleration frequency, braking force, steering operation, driving speed fluctuation, etc.) and the corresponding historical fault data (such as engine failure, brake system failure, electrical system failure, etc.), wherein the historical fault data includes data such as fault type, occurrence time, and vehicle operating conditions at the time.

[0086] In a specific embodiment, the historical driving behavior data and historical fault data of the target vehicle are obtained, and the preset model is trained based on the historical driving behavior data and historical fault data to obtain an association analysis model. Through the association analysis model, after obtaining the real-time driving behavior data of the target vehicle, the vehicle's subsequent possible faults can be predicted. For example, when the model finds that there is a strong correlation between the driving behavior of frequent emergency braking and the wear and tear of a component in the brake system, if it is monitored that a certain vehicle's current driving behavior has frequent emergency braking, the model can be used to infer that the possibility of the corresponding brake system failure of the vehicle has increased, and then measures such as reminding the driver to pay attention to the driving method and arranging the vehicle for targeted inspection and maintenance can be taken in advance to improve the safety of the driving vehicle.

[0087] Optionally, the step of training a preset model based on the historical driving behavior data and the historical fault data to obtain a correlation analysis model may specifically include the following steps:

[0088] S2501, preprocessing the historical driving behavior data and the historical fault data to obtain target driving behavior data and target fault data respectively; the preprocessing includes: standardization, removal of outliers, and deduplication processing;

[0089] S2502, extracting features of the target driving behavior data and the target fault data to obtain a first feature vector and a second feature vector respectively;

[0090] S2503, performing data division on the first feature vector and the second feature vector to obtain a first training set and a first validation set;

[0091] S2504, constructing a first model according to the preset model and the attention mechanism;

[0092] S2505: Train the first model according to the first training set to obtain a second model;

[0093] S2506. Validate the second model according to the first validation set to obtain a performance indicator value of the second model;

[0094] S2507, determining adjustment parameters of the second model according to the performance indicator value to obtain target adjustment parameters;

[0095] S2508. Adjust the model parameters of the second model according to the target adjustment parameters to obtain the association analysis model.

[0096] In a specific embodiment, the historical driving behavior data and the historical fault data are preprocessed to obtain target driving behavior data and target fault data, respectively, wherein the preprocessing includes: standardization, removal of outliers, deduplication and other operations. Features such as average acceleration frequency and braking force change trend are extracted from the target driving behavior data and combined into a first feature vector, and features such as fault type and first occurrence time of the fault are extracted from the target fault data to form a second feature vector. The first feature vector and the second feature vector are divided into data to obtain a first training set and a first verification set, wherein the first training set is used for model training, and the first verification set is used for subsequent verification of the effect.

[0097] In an embodiment of the present application, a long short-term memory network is selected as a preset model, and the preset model is combined with an attention mechanism to obtain a first model, wherein the attention mechanism allows the model to focus on more critical driving behaviors and fault-related features.

[0098] Furthermore, the first model is trained using the first training set, and the parameters within the model are adjusted through multiple iterations to obtain the second model. Specifically, the second model is verified using the first validation set to obtain performance indicators such as accuracy and recall. By analyzing these performance indicator values, if the accuracy does not meet expectations, the target adjustment parameters such as increasing the number of network layers and adjusting the attention weight are determined. Next, the model parameters of the second model are adjusted according to the target adjustment parameters. Through multiple iterations of adjustment and verification, a correlation analysis model that can accurately analyze the correlation between driving behavior and faults can be obtained.

[0099] S206: When the target risk index is greater than a preset risk index threshold, input the current driving behavior data into the association analysis model to obtain a target failure probability.

[0100] The preset danger index threshold refers to a pre-set critical numerical standard used to measure the degree of driving danger of the driver, which is determined based on a comprehensive assessment of risk analysis and vehicle failure statistics in a large number of actual driving scenarios. When the target danger index is higher than the preset danger index threshold, it means that the current driver's driving situation is already in a high-risk state and is very likely to cause vehicle failure.

[0101] In a specific embodiment, when the target risk index is greater than the preset risk index threshold, the operation of inputting the current driving behavior data into the association analysis model is performed to obtain the target fault probability. For example, the current driving behavior data including the current vehicle speed, acceleration, braking force, steering angle and other data are input into the association analysis model, and the association analysis model outputs the target fault probability of the target vehicle having an engine overheating fault under the current driving behavior, which is 30%, the target fault probability of a brake system wear fault is 20%, and other probabilities corresponding to different fault types.

[0102] It should be noted that the association analysis model is mainly used to analyze the correlation between driving behavior and specific vehicle failures, and outputs relevant information such as the probability of various failures by analyzing driving behavior. However, if the association analysis model is relied upon for fault prediction throughout the entire process, and various operating behaviors are generated at all times during driving, the association analysis model will continue to output a large number of fault possibility results of varying degrees, including some situations with low probabilities and little actual impact in the short term. This may lead to frequent warning information being sent to the driver or vehicle system, and may also consume more system resources, such as computing resources, storage resources and other resources to process these large numbers of prediction results with low impact. Therefore, by setting a preset danger index threshold, different risk level intervals are divided. Only when the danger index exceeds the preset danger index threshold, further fault prediction is performed, thereby reducing unnecessary warnings and concentrating resources on more critical scenarios where serious problems are more likely to occur, thereby improving overall management efficiency and the accuracy and effectiveness of warnings.

[0103] Optionally, the vehicle diagnostic device is communicatively connected to the remote server, and the following steps may also be included:

[0104] S2601: when the target failure probability is less than or equal to a preset first probability, generating a first text prompt message; the first text prompt message is used to prompt the target driver to pay attention to driving behavior;

[0105] S2602, when the target failure probability is greater than the preset first probability and less than or equal to the preset second probability, generating a first voice prompt message; the first voice prompt message is used to remind the target driver to drive safely; the preset first probability is less than the preset second probability;

[0106] S2603, generating a first vehicle failure risk report according to the current driving behavior data and the target failure probability;

[0107] S2604, storing the first vehicle failure risk report in a local storage system of the target vehicle;

[0108] S2605: when the target failure probability is greater than the preset second probability, generating a first alarm message; the first alarm message is used to warn the target driver;

[0109] S2606, generating a second vehicle failure risk report according to the current driving behavior data and the target failure probability;

[0110] S2607. Send the second vehicle fault risk report to the remote server, diagnose the target vehicle based on the second vehicle fault risk report through the remote server, obtain a target fault diagnosis plan, and return the target fault diagnosis plan to the vehicle diagnostic device.

[0111] Among them, the preset first probability is a critical probability value set for distinguishing between lower risk and medium risk intervals, the preset second probability is a critical probability value set for distinguishing between medium risk and higher risk intervals, and the preset first probability is less than the preset second probability.

[0112] In an embodiment of the present application, the preset first probability may be 20%, and the preset second probability may be 50%. Specifically, when the target failure probability is less than or equal to the preset first probability, a first text prompt message is generated, wherein the first text prompt message is used to prompt the target driver to pay attention to driving behavior. For example, through the association analysis model analysis, it is concluded that the current target failure probability is 15%, which is less than the preset first probability, so the first text prompt message "Please pay attention to driving behavior, standardized operation can reduce potential risks" is generated, which can be displayed on the central control display screen of the target vehicle and presented to the driver in a clear and eye-catching text form, so that the driver can see the prompt content at a glance during driving.

[0113] When the target failure probability is greater than the preset first probability and less than or equal to the preset second probability, a first voice prompt message is generated, wherein the first voice prompt message is used to remind the target driver to drive safely. For example, when the target failure probability is 30%, which is greater than the preset first probability and less than or equal to the preset second probability, the vehicle's voice assistant will issue a first voice prompt message of "Please drive safely, the current vehicle has certain risks, please operate with caution" to remind the target driver. Then, a first vehicle failure risk report is generated based on the current driving behavior data and the target failure probability, and the first vehicle failure risk report is stored in the local storage system of the target vehicle, so as to facilitate the subsequent review of vehicle risk changes.

[0114] When the target failure probability is greater than the preset second probability, a first alarm message is generated, wherein the first alarm message is used to warn the target driver. For example, when the target failure probability reaches 60%, the target vehicle immediately emits a sharp alarm sound, and at the same time flashes a warning light on the dashboard. The generated first alarm message strongly warns the target driver that the current driving situation is very dangerous. Then, a second vehicle failure risk report is generated based on the current driving behavior data and the target failure probability, and the second vehicle failure risk report is sent to a remote server. The target vehicle is diagnosed based on the second vehicle failure risk report by the remote server to obtain a target failure diagnosis plan, and the target failure diagnosis plan is returned to the vehicle diagnosis device.

[0115] The mechanism of taking different countermeasures based on different target failure probabilities allows drivers to know the risk level of the vehicle in an appropriate way. Text prompts and voice reminders allow them to adjust their driving habits and enhance safety awareness in a timely manner at low-risk and medium-risk stages, while alarm information can prompt them to pay attention and take action immediately at high risks to ensure driving safety. The fault risk report stored locally in the vehicle facilitates tracing the trajectory of vehicle risk changes, and the diagnosis and returned diagnostic solutions involving the remote server help to accurately locate the fault, improve maintenance efficiency, and reduce the impact of the vehicle on the fault. At the same time, a large amount of vehicle fault data is collected for analysis, which can improve the safety and reliability of vehicle travel.

[0116] In summary, by implementing the embodiments of the present application, the current driving behavior data of the target driver for the target vehicle can be obtained; the driving habit type of the target driver can be determined based on the current driving behavior data; when the driving habit type is a preset driving habit type, the driver state data of the target driver can be obtained; the target risk index can be determined based on the driving habit type and the driver state data; the historical driving behavior data and historical fault data of the target vehicle can be obtained, and the preset model can be trained based on the historical driving behavior data and the historical fault data to obtain the association analysis model; when the target risk index is greater than the preset risk index threshold, the current driving behavior data is input into the association analysis model to obtain the target failure probability. It can be seen that by analyzing the correlation between the driver's driving behavior and the vehicle failure, the vehicle failure can be accurately predicted based on the correlation, thereby avoiding the occurrence of the failure and reducing the damage to the vehicle components.

[0117] See also Figure 3 , Figure 3 300 is a structural diagram of a fault prediction and analysis device based on driving behavior provided in an embodiment of the present application. The fault prediction and analysis device based on driving behavior 300 is applied to vehicle diagnostic equipment. The device includes: a data acquisition module 301, a data processing module 302, a model training module 303, and a model analysis module 304, wherein:

[0118] The data acquisition module 301 is used to obtain the current driving behavior data of the target driver of the target vehicle;

[0119] The data processing module 302 is used to determine the driving habit type of the target driver according to the current driving behavior data;

[0120] The data acquisition module 301 is also used to obtain the driver status data of the target driver when the driving habit type is a preset driving habit type;

[0121] The data processing module 302 is also used to determine a target risk index according to the driving habit type and the driver status data;

[0122] The model training module 303 is used to obtain the historical driving behavior data and historical fault data of the target vehicle, and train the preset model based on the historical driving behavior data and the historical fault data to obtain an association analysis model; the association analysis model is used to generate a fault probability based on the driving behavior data;

[0123] The model analysis module 304 is used to input the current driving behavior data into the association analysis model to obtain a target failure probability when the target risk index is greater than a preset risk index threshold.

[0124] Optionally, in determining the driving habit type of the target driver according to the current driving behavior data, the data processing module 302 is further specifically configured to:

[0125] Extracting operation frequency data from the current driving behavior data; the operation frequency data includes at least one of the following: acceleration frequency, braking frequency, steering frequency, deceleration frequency, and lane change frequency;

[0126] Acquire driving environment data corresponding to the current driving behavior data;

[0127] determining a target driving behavior characteristic according to the operation frequency data and the driving environment data;

[0128] The driving habit type is determined according to the target driving behavior characteristics.

[0129] Optionally, in the aspect of acquiring the driver status data of the target driver, the data acquisition module 301 is further specifically used for:

[0130] Collecting target image data of the target driver;

[0131] Performing demeanor analysis on the target image data to obtain driving demeanor data of the target driver;

[0132] The driver state data of the target driver is determined according to the driving demeanor data.

[0133] Optionally, the driving demeanor data includes: facial expression data, eye gaze direction data and head posture data; in determining the driver status data of the target driver according to the driving demeanor data, the data acquisition module 301 is further specifically used for:

[0134] Acquire reference facial expression data, reference eye gaze direction data, and reference head posture data;

[0135] Determine first state offset data according to the facial expression data and the reference facial expression data;

[0136] Determining second state offset data according to the eye gaze direction data and the reference eye gaze direction data;

[0137] Determine third state offset data according to the head posture data and the reference head posture data;

[0138] Obtain weight information corresponding to the facial expression data, the eye gaze direction data, and the head posture data, and obtain first weight information, second weight information, and third weight information respectively;

[0139] The driver state data is determined according to the first state offset data, the second state offset data, the third state offset data, the first weight information, the second weight information and the third weight information.

[0140] Optionally, in determining the target risk index according to the driving habit type and the driver status data, the data acquisition module 301 is further specifically used for:

[0141] Scoring the driving habit type and the driver status data based on a preset scoring rule to obtain a target score;

[0142] Obtain the mapping relationship between the score and the risk index, and obtain the target mapping relationship;

[0143] The target risk index is determined according to the target score and the target mapping relationship.

[0144] Optionally, in the aspect of training the preset model based on the historical driving behavior data and the historical fault data to obtain the association analysis model, the model training module 303 is further specifically used for:

[0145] Preprocessing the historical driving behavior data and the historical fault data to obtain target driving behavior data and target fault data, respectively; the preprocessing includes: standardization, removal of outliers, and deduplication processing;

[0146] Extracting features of the target driving behavior data and the target fault data to obtain a first feature vector and a second feature vector respectively;

[0147] Performing data division on the first eigenvector and the second eigenvector to obtain a first training set and a first validation set;

[0148] A first model is constructed according to the preset model and the attention mechanism;

[0149] Training the first model according to the first training set to obtain a second model;

[0150] Verify the second model according to the first verification set to obtain a performance indicator value of the second model;

[0151] Determine an adjustment parameter of the second model according to the performance indicator value to obtain a target adjustment parameter;

[0152] The model parameters of the second model are adjusted according to the target adjustment parameters to obtain the association analysis model.

[0153] Optionally, the vehicle diagnostic device is in communication connection with a remote server; after inputting the current driving behavior data into the association analysis model to obtain a target fault probability, the device is further specifically used to:

[0154] When the target failure probability is less than or equal to a preset first probability, generating a first text prompt message; the first text prompt message is used to prompt the target driver to pay attention to driving behavior;

[0155] When the target failure probability is greater than the preset first probability and less than or equal to the preset second probability, a first voice prompt message is generated; the first voice prompt message is used to remind the target driver to drive safely; the preset first probability is less than the preset second probability;

[0156] generating a first vehicle failure risk report according to the current driving behavior data and the target failure probability;

[0157] storing the first vehicle failure risk report in a local storage system of the target vehicle;

[0158] When the target failure probability is greater than the preset second probability, generating a first alarm message; the first alarm message is used to warn the target driver;

[0159] generating a second vehicle failure risk report according to the current driving behavior data and the target failure probability;

[0160] The second vehicle fault risk report is sent to the remote server, the target vehicle is diagnosed by the remote server based on the second vehicle fault risk report, a target fault diagnosis plan is obtained, and the target fault diagnosis plan is returned to the vehicle diagnostic device.

[0161] The driving behavior-based fault prediction and analysis device 300 described in the present application can obtain the current driving behavior data of the target driver of the target vehicle; determine the driving habit type of the target driver according to the current driving behavior data; when the driving habit type is a preset driving habit type, obtain the driver state data of the target driver; determine the target risk index according to the driving habit type and the driver state data; obtain the historical driving behavior data and historical failure data of the target vehicle, and train the preset model based on the historical driving behavior data and the historical failure data to obtain the association analysis model; when the target risk index is greater than the preset risk index threshold, input the current driving behavior data into the association analysis model to obtain the target failure probability. It can be seen that by analyzing the correlation between the driver's driving behavior and the vehicle failure, the vehicle failure is accurately predicted according to the correlation, thereby avoiding the occurrence of the failure and reducing the damage to the vehicle components.

[0162] See also Figure 4 , Figure 4 : is a structural diagram of an electronic device provided in an embodiment of the present application, the electronic device may include a processor, a memory, a communication interface and one or more programs, the processor, the memory and the communication interface may be interconnected through a bus; the one or more programs are stored in the memory and are configured to be executed by the processor; in an embodiment of the present application, the program is applied to a vehicle diagnostic device, and the program includes instructions for executing the following steps:

[0163] Acquire current driving behavior data of a target driver for a target vehicle;

[0164] Determining the driving habit type of the target driver according to the current driving behavior data;

[0165] When the driving habit type is a preset driving habit type, obtaining driver status data of the target driver;

[0166] determining a target risk index according to the driving habit type and the driver status data;

[0167] Acquiring historical driving behavior data and historical fault data of the target vehicle, and training a preset model based on the historical driving behavior data and the historical fault data to obtain an association analysis model; the association analysis model is used to generate a fault probability based on the driving behavior data;

[0168] When the target risk index is greater than a preset risk index threshold, the current driving behavior data is input into the association analysis model to obtain a target failure probability.

[0169] The electronic device described in the present application can obtain the current driving behavior data of the target driver of the target vehicle; determine the driving habit type of the target driver according to the current driving behavior data; obtain the driver status data of the target driver when the driving habit type is a preset driving habit type; determine the target risk index according to the driving habit type and the driver status data; obtain the historical driving behavior data and historical fault data of the target vehicle, and train the preset model based on the historical driving behavior data and the historical fault data to obtain the association analysis model; when the target risk index is greater than the preset risk index threshold, input the current driving behavior data into the association analysis model to obtain the target failure probability. It can be seen that by analyzing the correlation between the driver's driving behavior and the vehicle failure, the vehicle failure is accurately predicted according to the correlation, thereby avoiding the occurrence of the failure and reducing the damage to the vehicle components.

[0170] An embodiment of the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiment, and the above computer includes an electronic device.

[0171] The embodiment of the present application also provides a computer program product, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, the computer program is operable to cause a computer to execute some or all of the steps of any method described in the method embodiment. The computer program product may be a software installation package, and the computer includes an electronic device.

[0172] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

[0173] The steps of the method or algorithm described in the embodiments of the present application can be implemented in hardware or by executing software instructions by a processor. The software instructions can be composed of corresponding software modules, and the software modules can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (electrically EPROM, EEPROM), registers, hard disks, mobile hard disks, read-only compact disks (CD-ROMs) or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a terminal device or a management device. Of course, the processor and the storage medium can also be present in a terminal device or a management device as discrete components.

[0174] Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server, or data center to another website site, computer, server, or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0175] The modules / units included in the devices and products described in the above embodiments may be software modules / units or hardware modules / units, or may be partially software modules / units and partially hardware modules / units. For example, for the devices and products applied to or integrated in the chip, the modules / units included therein may all be implemented in the form of hardware such as circuits, or at least some of the modules / units may be implemented in the form of software programs, which run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for the devices and products applied to or integrated in the chip module, the modules / units included therein may all be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as a chip, circuit module, etc.) or in different components of the chip module, or at least some of the modules / units may be implemented in the form of software programs. The software programs run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits. It is implemented in the form of a software program, which runs on a processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in hardware such as circuits; for various devices and products applied to or integrated in the terminal equipment, the various modules / units contained therein can be implemented in hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or in different components in the terminal equipment, or, at least some modules / units can be implemented in the form of a software program, which runs on a processor integrated inside the terminal equipment, and the remaining (if any) modules / units can be implemented in hardware such as circuits.

[0176] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the embodiments of the present application. It should be understood that the above description is only the specific implementation method of the embodiments of the present application and is not intended to limit the protection scope of the embodiments of the present application. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application should be included in the protection scope of the embodiments of the present application.

Claims

1. A fault prediction analysis method based on driving behavior, characterized in that: Applied to vehicle diagnostic equipment, the method comprises: Acquire current driving behavior data of a target driver for a target vehicle; Determining the driving habit type of the target driver according to the current driving behavior data; When the driving habit type is a preset driving habit type, obtaining driver status data of the target driver; determining a target risk index according to the driving habit type and the driver status data; Acquiring historical driving behavior data and historical fault data of the target vehicle, and training a preset model based on the historical driving behavior data and the historical fault data to obtain an association analysis model; the association analysis model is used to generate a fault probability based on the driving behavior data; When the target risk index is greater than a preset risk index threshold, the current driving behavior data is input into the association analysis model to obtain a target failure probability.

2. The method according to claim 1, characterized in that The determining the driving habit type of the target driver according to the current driving behavior data includes: Extracting operation frequency data from the current driving behavior data; the operation frequency data includes at least one of the following: acceleration frequency, braking frequency, steering frequency, deceleration frequency, and lane change frequency; Acquire driving environment data corresponding to the current driving behavior data; determining a target driving behavior characteristic according to the operation frequency data and the driving environment data; The driving habit type is determined according to the target driving behavior characteristics.

3. The method according to claim 1, characterized in that The obtaining of the driver status data of the target driver includes: Collecting target image data of the target driver; Performing demeanor analysis on the target image data to obtain driving demeanor data of the target driver; The driver state data of the target driver is determined according to the driving demeanor data.

4. The method according to claim 3, characterized in that The driving demeanor data includes: facial expression data, eye gaze direction data and head posture data; the driver status data of the target driver determined according to the driving demeanor data includes: Acquire reference facial expression data, reference eye gaze direction data, and reference head posture data; Determine first state offset data according to the facial expression data and the reference facial expression data; Determining second state offset data according to the eye gaze direction data and the reference eye gaze direction data; Determine third state offset data according to the head posture data and the reference head posture data; Obtain weight information corresponding to the facial expression data, the eye gaze direction data, and the head posture data, and obtain first weight information, second weight information, and third weight information respectively; The driver state data is determined according to the first state offset data, the second state offset data, the third state offset data, the first weight information, the second weight information and the third weight information.

5. The method according to any one of claims 1 to 4, characterized in that: The determining of the target risk index according to the driving habit type and the driver status data comprises: Scoring the driving habit type and the driver status data based on a preset scoring rule to obtain a target score; Obtain the mapping relationship between the score and the risk index, and obtain the target mapping relationship; The target risk index is determined according to the target score and the target mapping relationship.

6. The method according to any one of claims 1 to 4, characterized in that: The training of a preset model based on the historical driving behavior data and the historical fault data to obtain a correlation analysis model includes: Preprocessing the historical driving behavior data and the historical fault data to obtain target driving behavior data and target fault data, respectively; the preprocessing includes: standardization, removal of outliers, and deduplication processing; Extracting features of the target driving behavior data and the target fault data to obtain a first feature vector and a second feature vector respectively; Performing data division on the first eigenvector and the second eigenvector to obtain a first training set and a first validation set; A first model is constructed according to the preset model and the attention mechanism; Training the first model according to the first training set to obtain a second model; Verify the second model according to the first verification set to obtain a performance indicator value of the second model; Determine an adjustment parameter of the second model according to the performance indicator value to obtain a target adjustment parameter; The model parameters of the second model are adjusted according to the target adjustment parameters to obtain the association analysis model.

7. The method according to any one of claims 1 to 4, characterized in that: The vehicle diagnostic device is communicatively connected to a remote server; after inputting the current driving behavior data into the association analysis model to obtain a target failure probability, the method further includes: When the target failure probability is less than or equal to a preset first probability, generating a first text prompt message; the first text prompt message is used to prompt the target driver to pay attention to driving behavior; When the target failure probability is greater than the preset first probability and less than or equal to the preset second probability, a first voice prompt message is generated; the first voice prompt message is used to remind the target driver to drive safely; the preset first probability is less than the preset second probability; generating a first vehicle failure risk report according to the current driving behavior data and the target failure probability; storing the first vehicle failure risk report in a local storage system of the target vehicle; When the target failure probability is greater than the preset second probability, generating a first alarm message; the first alarm message is used to warn the target driver; generating a second vehicle failure risk report according to the current driving behavior data and the target failure probability; The second vehicle fault risk report is sent to the remote server, the target vehicle is diagnosed by the remote server based on the second vehicle fault risk report, a target fault diagnosis plan is obtained, and the target fault diagnosis plan is returned to the vehicle diagnostic device.

8. A fault prediction and analysis device based on driving behavior, characterized in that: Applied to vehicle diagnostic equipment, the fault prediction and analysis device based on driving behavior includes: a data acquisition module, a data processing module, a model training module, and a model analysis module, wherein: The data acquisition module is used to obtain current driving behavior data of a target driver of a target vehicle; The data processing module is used to determine the driving habit type of the target driver according to the current driving behavior data; The data acquisition module is also used to obtain the driver status data of the target driver when the driving habit type is a preset driving habit type; The data processing module is also used to determine a target risk index according to the driving habit type and the driver status data; The model training module is used to obtain the historical driving behavior data and historical fault data of the target vehicle, and train the preset model based on the historical driving behavior data and the historical fault data to obtain an association analysis model; the association analysis model is used to generate a fault probability based on the driving behavior data; The model analysis module is used to input the current driving behavior data into the association analysis model to obtain a target failure probability when the target risk index is greater than a preset risk index threshold.

9. An electronic device, characterized in that: include: A processor, a memory, a communication interface, and one or more programs; The one or more programs are stored in the memory and configured to be executed by the processor, the programs comprising instructions for executing the steps in the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.

Images