Driving behavior analysis method, server, and storage medium

Abstract

The present invention provides a driving behavior analysis method, server, and storage medium, including: obtaining driving data and first fuel consumption data of a vehicle network user; calculating correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data; using a preset algorithm to select key driving behaviors whose correlation coefficients meet a preset threshold; obtaining data on the target user's key driving behaviors and second fuel consumption data of the target user; calculating weight coefficients for the impact of each key driving behavior of the target user on the second fuel consumption data using a preset model; calculating the target user's average fuel consumption based on the second fuel consumption data; and generating a driving analysis report based on the average fuel consumption and the weight coefficients. In this embodiment of the present invention, a preliminary screening of key driving behaviors that influence fuel consumption is performed based on big data, user data is obtained, and the model is used to analyze the user's driving habits that primarily influence fuel consumption, thereby resolving the problem in existing technologies where users are unable to understand their own negative driving habits.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a driving behavior analysis method, a server, and a storage medium. Background Art

[0002] Vehicle fuel consumption is affected by many factors. In addition to the vehicle's own technical parameters, it is also closely related to the user's driving behavior. For example, sudden acceleration, sudden deceleration, and full loading will all lead to increased fuel consumption. Users are generally unaware of these factors and often simply attribute them to high engine fuel consumption or the vehicle being too heavy.

[0003] The existing solution only performs a simple evaluation of the user's fuel consumption, that is, calculating the user's average fuel consumption based on driving mileage and fuel consumption data, and displaying it on the vehicle's center console for the driver to view.

[0004] However, the current solution only calculates a user's average fuel consumption based on mileage and fuel consumption data, allowing users to understand their own fuel consumption level. However, this approach prevents users from understanding the fuel efficiency of similar users or any negative driving habits. Summary of the Invention

[0005] In view of this, the present invention aims to propose a driving behavior analysis method, server, and storage medium to solve the problem in the prior art that users cannot understand the driving fuel-saving levels of similar users and their own driving behavior.

[0006] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0007] A driving behavior analysis method, applied to a server, is characterized by comprising:

[0008] Obtain driving data and first fuel consumption data of Internet of Vehicles users;

[0009] Calculating correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and screening main driving behaviors whose correlation coefficients reach a preset threshold using a preset algorithm;

[0010] Obtaining the target user's main driving behavior data and the second fuel consumption data;

[0011] Calculating, using a preset model, a weight coefficient of the impact of each of the main driving behaviors of the target user on the second fuel consumption data;

[0012] calculating the average fuel consumption of the target user based on the second fuel consumption data;

[0013] Acquire all driving behavior data of the target user, and determine the type and frequency of instantaneous bad driving behavior based on all driving behavior data of the target user;

[0014] A driving analysis report is generated based on the average fuel consumption and the weight coefficient, wherein the driving analysis report includes:

[0015] The type and frequency of the instantaneous bad driving behaviors.

[0016] Furthermore, the obtaining of the target user's main driving behavior data includes:

[0017] Obtaining vehicle bus raw data of the main driving behavior of the target user;

[0018] Furthermore, the raw data is calculated to obtain characteristic data for evaluating the main driving behavior.

[0019] Furthermore, before using the preset model to calculate the weight coefficient of the impact of each main driving behavior of the target user on the second fuel consumption data, the method further includes:

[0020] Obtaining user driving data samples for testing, testing the model's preset parameters, obtaining test results, and adjusting the preset parameters;

[0021] When the test result meets the conditions, the preset model is determined.

[0022] Furthermore, the calculation of the average fuel consumption of the target user includes:

[0023] Acquiring engine operating status information of a target user, and determining a target trip based on the engine operating status information;

[0024] The mileage data of the target trip and the second fuel consumption data are obtained, and the average fuel consumption of the target user is calculated.

[0025] Furthermore, determining the target range according to the engine operating status information includes:

[0026] According to the engine operating state information, the vehicle state is divided into a vehicle starting state, an engine off state, and a vehicle fully powered state;

[0027] Obtain the vehicle's travel data and altitude information when it is in the starting state;

[0028] The trip whose altitude information meets the preset conditions is determined as the target trip.

[0029] Furthermore, before calculating the average fuel consumption of the target user, the method further includes:

[0030] The fuel consumption data of the warm-up phase trip in the second fuel consumption data is corrected, wherein the warm-up phase trip is a trip in the target trip in which the fuel consumption is abnormally high when the vehicle starts traveling.

[0031] Furthermore, generating a driving analysis report based on the average fuel consumption and the weight coefficient includes:

[0032] calculating, based on the driving data and the first fuel consumption data of the connected vehicle user, the standard fuel consumption of users of the same vehicle model in the connected vehicle network, and determining the fuel saving level of the target user based on the standard fuel consumption of users of the same vehicle model and the average fuel consumption of the target user;

[0033] determining a target driving behavior according to the weight coefficient and the fuel saving level;

[0034] A driving analysis report is generated according to the fuel saving level and the target driving behavior within a preset period.

[0035] Furthermore, after obtaining the driving data of the target user, the method further includes:

[0036] Obtaining real-time environmental information of the target user;

[0037] Based on the environmental information, outliers in the driving data are determined and deleted.

[0038] Another object of the present invention is to provide a server, comprising:

[0039] A selection module is configured to obtain driving data and first fuel consumption data of an Internet of Vehicles user, calculate correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and select main driving behaviors whose correlation coefficients reach a preset threshold using a preset algorithm;

[0040] Data acquisition module: used to acquire the target user's main driving behavior data and the second fuel consumption data;

[0041] A calculation module: configured to calculate, by using a preset model, a weight coefficient of the impact of each of the main driving behaviors of the target user on the second fuel consumption data;

[0042] Data Analysis Module: This module is used to calculate the average fuel consumption of the target user, obtain the data of all the target user's driving behaviors, determine the type and frequency of instantaneous bad driving behaviors based on the data of all the target user's driving behaviors, and generate a driving analysis report based on the average fuel consumption and weight coefficient. The driving analysis report includes:

[0043] Type and frequency of transient adverse driving behaviors.

[0044] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, wherein the computer program implements any of the aforementioned driving behavior analysis methods when executed by a processor.

[0045] Compared with the prior art, the driving behavior analysis method, server, and storage medium described in the present invention have the following advantages:

[0046] A driving behavior analysis method provided in an embodiment of the present invention is applied to a server, comprising: obtaining driving data and first fuel consumption data of a vehicle network user, calculating correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and screening major driving behaviors whose correlation coefficients reach a preset threshold using a preset algorithm; obtaining data on the major driving behaviors and second fuel consumption data of a target user; calculating a weight coefficient of the impact of each major driving behavior of the target user on the second fuel consumption data using a preset model; calculating the average fuel consumption of the target user based on the second fuel consumption data; and generating a driving analysis report based on the average fuel consumption and the weight coefficients.

[0047] Since it is possible to filter based on the big data of the Internet of Vehicles, it is possible to filter out the main driving behaviors that affect fuel consumption, and then obtain the driving behavior data of the target user. The preset model is used to analyze the user's current driving habits that mainly affect fuel consumption, and generate an analysis report. This solves the problem in the existing technology that users cannot understand the fuel-saving levels of similar users and the bad habits in their own driving behaviors. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which constitute part of the present invention, are provided to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are provided to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0049] Figure 1 This is a flowchart of the steps of a driving behavior analysis method according to an embodiment of the present invention;

[0050] Figure 2 This is a flowchart of another driving behavior analysis method according to an embodiment of the present invention;

[0051] Figure 3 A schematic diagram of the structure of a server provided in Embodiment 3 of the present invention; DETAILED DESCRIPTION

[0052] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0053] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0054] Example 1

[0055] Reference Figure 1 , shows a step flow chart of a driving behavior analysis method according to an embodiment of the present invention.

[0056] Step 101: Acquire driving data and first fuel consumption data of an Internet of Vehicles user.

[0057] In an embodiment of the present invention, the vehicle-side controller collects driving data during the user's driving process. The user's driving data includes various driving behaviors and corresponding original parameters during the user's driving process, and also includes the vehicle's status information. For example: vehicle longitudinal acceleration, vehicle lateral acceleration, engine speed, vehicle speed, brake pressure, accelerator pedal position, average fuel consumption, engine output torque, transmission input torque, gear signal, coolant temperature, air conditioning switch, window status, tire pressure, remaining fuel, etc. Driving data can reflect the driver's driving habits during driving. The first fuel consumption data can represent the correspondence between the vehicle speed, instantaneous fuel consumption, and mileage of the Internet of Vehicles user at each moment. The data collection method is divided into Kafka real-time collection and Sqoop and Flume offline collection, which can ensure comprehensive collection of driving data.

[0058] The acquired driving data and first fuel consumption data of the Internet of Vehicles user are uploaded to the server via the gateway by the Internet of Vehicles terminal T-BOX on the vehicle side.

[0059] Step 102 : Calculate the correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and use a preset algorithm to select the main driving behaviors whose correlation coefficients reach a preset threshold.

[0060] Statistical algorithms and random forest algorithms are used to analyze large amounts of driving data, calculating correlation coefficients between various driving behaviors and fuel consumption data. These correlation coefficients are then filtered out to identify driving behaviors that meet a threshold. This threshold can be determined based on the specific vehicle model and server computing power. For example, key driving behaviors selected include vehicle speed, accelerator pedal position, air conditioning switch, window status, and tire pressure. Subsequently, only key driving behaviors are analyzed to reduce server computing pressure. Data calculation methods primarily include Flink real-time computing and Spark offline computing.

[0061] Step 103: Obtain the target user's main driving behavior data and the target user's second fuel consumption data.

[0062] Obtain the above-mentioned main driving behaviors of the target user and the fuel consumption data corresponding to the driving behaviors. The vehicle terminal adopts a continuous sampling method during the obtaining process. The data collection time at the vehicle terminal each time is less than or equal to 1 second, and the sampled data is transmitted in binary form. The upload interval time can be verified according to different vehicle models and configured on the in-vehicle terminal, the user's vehicle networking APP or the web terminal.

[0063] Optionally, obtaining data on the main driving behaviors of the target user includes: obtaining the original vehicle bus data of the main driving behaviors of the target user; and calculating the original data to obtain characteristic data for evaluating the main driving behaviors.

[0064] Since the data obtained after the secondary calculation of the original data can reflect the change process of the original data. It can analyze the original data from more dimensions, thereby reflecting what kind of driving behavior habits the user has.

[0065] In practical applications, not only the original vehicle bus (CAN) data during the actual driving process is used, but also the characteristic data obtained after calculating the data. For example, the obtained original data includes the throttle pedal opening. Then calculate the standard deviation of the throttle pedal opening. The standard deviation of the throttle pedal opening can reflect the difference in the use of the throttle by the driver during driving and can be used as a characteristic of the throttle pedal smoothness to evaluate the user's usage habit of the throttle pedal. Another example: the original vehicle bus data includes vehicle speed data. Then calculate the vehicle speed data. The time taken to accelerate the vehicle speed from 0 kph to 30 kph is used as new characteristic data. The time taken to accelerate the vehicle speed from 0 kph to 30 kph is used as the starting duration and used as the evaluation data for the user's acceleration habit.

[0066] The calculation method can refer to the sliding window logic to calculate the time taken for the vehicle speed to continuously rise from 0 kph to 30 kph within the continuously rising segment of the vehicle speed. It is also possible to convert the vehicle speed data: when the vehicle speed (spd == 0), it is recorded as 0; when the vehicle speed (0 < spd < 30), it is recorded as 1; when the vehicle speed (spd ≈ 30), it is recorded as 3; when the vehicle speed (spd > 30), it is recorded as 6. Differentiate the converted data, screen the data with differential values of 1 and 2, and differentiate again to screen the data with a differential value of 1 as the data of starting successfully (successfully accelerating to 30 kph). Subtract the starting moment from the moment of starting successfully as the starting duration. If the starting duration is too short, it is counted as one case of starting too fast.

[0067] In addition to the aforementioned feature data, in embodiments of the present invention, feature data can also be combined to evaluate a user's driving behavior. For example, this includes simultaneously turning on the air conditioner and windows, or simultaneously opening windows at excessive speed. For example, based on the air conditioner switch signal data, the driver's window position status, the passenger window position command signal, the left rear window position command, and the right rear window position command data (all four window states are combined and counted as a single window state), the duration and number of windows opened with the air conditioner on during each trip segment are calculated. For example, based on the vehicle speed data, the driver's window position status, the passenger window position command signal, the left rear window position command, and the right rear window position command data (all four window states are combined and counted as a single window state), the duration and number of windows opened simultaneously at speeds greater than 60 kph during each trip segment are calculated. This combination can reflect issues with a user's driving habits from multiple perspectives, thereby providing more accurate driving advice to the user in subsequent behavior analysis reports.

[0068] Step 104 : Calculate the weight coefficient of the impact of each main driving behavior of the target user on the second fuel consumption data using a preset model.

[0069] In an embodiment of the present invention, the driving data of the main driving behaviors and the fuel consumption data of the target user can be analyzed by a statistical model, and a fitting formula can be used to analyze the weight coefficient of the target user's influence on the vehicle fuel consumption under different driving behaviors.

[0070] Step 105: Calculate the average fuel consumption of the target user based on the second fuel consumption data.

[0071] In the embodiment of the present invention, the average fuel consumption of the target user may be calculated based on the fuel consumption data of the target user and the mileage of the target user.

[0072] Optionally, the average fuel consumption of the target user is calculated based on the second fuel consumption data, including: obtaining the target user's engine operating status information, and dividing the vehicle status into a started state, an off state, and a fully powered state based on the engine operating status information; obtaining the vehicle's travel data and altitude information when in the started state; determining a trip whose altitude information meets preset conditions as a target trip; obtaining the mileage data and the second fuel consumption data of the target trip, and calculating the target user's average fuel consumption.

[0073] In an embodiment of the present invention, the vehicle operating state can be determined based on the operating state of the engine. By obtaining the real-time operating state of the engine, the vehicle operating state can be divided into the vehicle starting state, the flameout state, and the vehicle power-on state. In the flameout state and the vehicle power-on state, the engine is not started and there is no fuel consumption. Therefore, only the travel data of the vehicle starting state is considered, wherein the travel parameters include the vehicle's mileage, vehicle positioning information, various driving behaviors and corresponding parameters, etc. The altitude change threshold is set according to the altitude information of the vehicle, and the process of the vehicle being in a clear uphill or downhill process is screened out. The data in the clear uphill or downhill process is deleted, and the corresponding travel does not participate in the calculation of the user's average fuel consumption.

[0074] Because the present invention analyzes the relationship between different driving behaviors and fuel consumption, even if the vehicle maintains the same driving behavior, significant changes in fuel consumption can occur when the vehicle is traveling uphill or downhill, significantly affecting fuel consumption calculations. Therefore, in this embodiment of the present invention, trips where the altitude change exceeds a threshold are deleted.

[0075] Step 106 : Obtain all driving behavior data of the target user, and determine the type and frequency of instantaneous bad driving behaviors based on all driving behavior data of the target user.

[0076] The data of all driving behaviors can also include statistics on the number of sudden accelerations, sudden decelerations, and sharp turns. It should be noted that the four bad driving behavior indicators of starting too fast, sudden acceleration, sudden deceleration, and sharp turns will affect instantaneous fuel consumption, but will not necessarily affect the overall average fuel consumption. The above-mentioned bad driving behavior indicators do not participate in the correlation calculation of the monthly average fuel consumption, and do not require additional quantification or evaluation. They simply count the number of bad driving behaviors and push them to users to remind them to avoid these bad driving behaviors.

[0077] Step 107 : Generate a driving analysis report based on the average fuel consumption and the weight coefficient.

[0078] Optionally, generate a driving analysis report based on average fuel consumption and weighting factors, including:

[0079] Based on the driving data and first fuel consumption data of the Internet of Vehicles users, the standard fuel consumption of users of the same car model in the Internet of Vehicles is calculated. Based on the standard fuel consumption of users of the same car model and the average fuel consumption of the target users, the fuel-saving level of the target user is determined; the target driving behavior is determined based on the weight coefficient and the fuel-saving level; and a driving analysis report is generated based on the fuel-saving level and target driving behavior within a preset period.

[0080] In actual applications, the average fuel consumption of users of the same model over a period of time can be obtained through the Internet of Vehicles. This time period can be set to 1 day, 1 week or 1 month, etc.

[0081] In the analysis report, users with the same characteristics are grouped together. Criteria for classifying users with the same characteristics include mileage, average speed, time of day, weather and traffic events, and vehicle condition. PCA dimensionality reduction analysis revealed that mileage has the highest impact on average fuel consumption, significantly higher than other indicators. Therefore, mileage alone is sufficient as a user characteristic for classification.

[0082] During the classification process, users within 10km (such as daily commuting) are classified into one category; users between 10km and 50km (such as round-trip travel between the city and the suburbs) are classified into one category; and users over 50km (such as long-distance travel) are classified into one category.

[0083] In the analysis report, fuel consumption rankings are displayed based on different user categories. The average fuel consumption of a large number of users with the same vehicle model is randomly sampled, and after filtering out extreme values, the standard fuel consumption is used as the standard fuel consumption for users with the same vehicle model. For example, based on the monthly average fuel consumption of users with the same vehicle model, the top 5% and bottom 5% of users in terms of fuel consumption are removed, and the average fuel consumption of the middle 80% of users is taken. Therefore, the standard fuel consumption can represent the majority of users, comprehensively consider road conditions and vehicle conditions, and select the optimal fuel consumption that balances vehicle power and economy. By comparing the average fuel consumption of the target user with the standard fuel consumption, the user's fuel-saving level can be determined. In the analysis report, a driving behavior report is generated and displayed based on the average fuel consumption of the user's vehicle, the above-mentioned general indicators, and the counts of various bad driving behaviors.

[0084] In addition, the analysis report provides comparisons based on the average fuel consumption of each user's vehicle. This includes horizontal comparisons: using average fuel consumption as the dimension, a horizontal comparison of each user's driving behavior is made. Vertical comparisons: using average fuel consumption as the dimension, a vertical comparison of each user's driving behavior this month and last month is made.

[0085] According to the weight coefficients of the main driving behaviors, a certain number of driving behaviors that best reflect the user's driving habits are screened out, and the corresponding driving behavior guidance information is output in turn.

[0086] In summary, embodiments of the present invention provide a driving behavior analysis method, including: obtaining driving data and first fuel consumption data of a target user in an Internet of Vehicles (IoV), calculating correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, using a preset algorithm to select key driving behaviors whose correlation coefficients reach a preset threshold, obtaining data on the target user's key driving behaviors and the target user's second fuel consumption data, and using a preset model to calculate a weight coefficient for the impact of each of the target user's key driving behaviors on the second fuel consumption data; calculating the target user's average fuel consumption based on the second fuel consumption data, and generating a driving analysis report based on the average fuel consumption and the weight coefficients. Because this method can be filtered based on IoV big data, it is possible to identify key driving behaviors that affect fuel consumption. Then, the driving behavior data of a single user is obtained, and the preset model is used to analyze the user's current driving habits that primarily affect fuel consumption, generating an analysis report. This method addresses the prior art issue where users are unable to understand the fuel-efficiency levels of similar users and any undesirable driving habits within their own driving behavior.

[0087] Example 2

[0088] Reference Figure 2 , shows a step flow chart of another driving behavior analysis method described in an embodiment of the present invention.

[0089] An embodiment of the present invention provides a driving behavior analysis method.

[0090] Step 201: Acquire driving data and first fuel consumption data of an Internet of Vehicles user.

[0091] This step is the same as step 101 and will not be described in detail here.

[0092] Step 202 : Calculate the correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and use a preset algorithm to select the main driving behaviors whose correlation coefficients reach a preset threshold.

[0093] This step is the same as step 102 and will not be described in detail here.

[0094] Step 203: Obtain the target user's main driving behavior data and the target user's second fuel consumption data.

[0095] This step is the same as step 103 and will not be described in detail here.

[0096] Step 204: Acquire real-time environmental information of the target user; based on the environmental information, determine and delete abnormal values ​​in the driving data.

[0097] In this embodiment of the present invention, due to factors such as the vehicle's driving environment, poor communication quality may cause missing or abnormal data on the target user's primary driving behavior and secondary fuel consumption data. Using the data analysis model, when a sudden change or missing data is detected, the abnormal data is deleted to avoid affecting the reliability of the overall data.

[0098] Step 205: Obtain a user driving data sample for testing, test the model preset parameters, obtain the test results and adjust the preset parameters; when the test results meet the conditions, determine the preset model.

[0099] In practical applications, test samples can be pre-set to test model parameters, and the model parameters can be adjusted according to the test results until the test accuracy reaches a preset threshold, confirming that the model can be used.

[0100] Step 206 : Calculate the weight coefficient of the impact of each main driving behavior of the target user on the second fuel consumption data using a preset model.

[0101] This step is the same as step 104 and will not be described in detail here.

[0102] Step 207 , correcting the fuel consumption data of the warm-up phase trip in the second fuel consumption data, wherein the warm-up phase trip is a trip in the target trip in which the fuel consumption is abnormally high when the vehicle starts traveling.

[0103] Step 208: Calculate the average fuel consumption of the target user based on the second fuel consumption data.

[0104] In this embodiment of the present invention, the calculation of the displayed average fuel consumption involves an initialization process when the vehicle is first started from a cold engine. Furthermore, the initial engine water temperature is relatively low, resulting in low engine thermal efficiency. For example, fuel consumption may be abnormally high during the first 2 kilometers of a trip. However, after a period of driving, initialization is completed, and as the water temperature rises, overall fuel consumption returns to a reasonable value. Therefore, the average fuel consumption for the first 2 kilometers of a trip requires processing before use. Average fuel consumption values ​​after 2 kilometers can be calculated directly based on the displayed average fuel consumption value.

[0105] In actual applications, the calculation of average fuel consumption uses the average fuel consumption value displayed in the car network data. For example, the data before the mileage of 2km is (y1, y2, y3......y n ), the data after 2km is (y n+1 ,y n+2 ,y n+3 ......y m ), the average fuel consumption is corrected in step 207 to be:

[0106]

[0107] The subscript i represents the intermediate variable in the accumulation process, and λ represents the adjustment coefficient, which adjusts the fuel consumption during trips with abnormally high fuel consumption.

[0108] Step 209 : Generate a driving analysis report based on the average fuel consumption and the weight coefficient.

[0109] This step is the same as step 107 and will not be described in detail here.

[0110] In summary, embodiments of the present invention provide a driving behavior analysis method, comprising: obtaining driving data and first fuel consumption data of a target user in an Internet of Vehicles (IoV) network; calculating correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data; using a preset algorithm to select key driving behaviors whose correlation coefficients reach a preset threshold; obtaining data on the target user's primary driving behaviors and second fuel consumption data; and obtaining real-time environmental information about the target user; identifying and deleting outliers in the driving data based on the environmental information; calculating weight coefficients for the impact of each of the target user's primary driving behaviors on the second fuel consumption data using a preset model; and correcting the fuel consumption data for warm-up phase trips in the second fuel consumption data, where warm-up phase trips are trips in the target trip where fuel consumption is abnormally high at the start of the vehicle's journey; calculating the target user's average fuel consumption based on the second fuel consumption data; and generating a driving analysis report based on the average fuel consumption and the weight coefficients. Because this method can be filtered based on IoV big data, it is possible to identify key driving behaviors that affect fuel consumption. Then, driving behavior data for individual users is obtained and filtered based on environmental information, thereby improving data reliability. The preset model is used to analyze the user's current driving habits that mainly affect fuel consumption, and an analysis report is generated. The analysis report can display the user's fuel-saving level and corresponding driving suggestions, which can help users understand their own driving conditions. This solves the problem in the existing technology that users cannot understand the fuel-saving level of similar users and the bad habits in their own driving behavior.

[0111] Example 3

[0112] Reference Figure 3 , shows a server 30 according to an embodiment of the present invention, including:

[0113] The Internet of Vehicles data acquisition module 301 is used to acquire driving data and first fuel consumption data of an Internet of Vehicles user.

[0114] Screening module 302: Calculates correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and uses a preset algorithm to screen out major driving behaviors whose correlation coefficients reach a preset threshold;

[0115] Target data acquisition module 303: used to acquire the target user's main driving behavior data and the second fuel consumption data;

[0116] Calculation module 304: used to calculate the weight coefficient of the impact of each main driving behavior of the target user on the second fuel consumption data using a preset model;

[0117] Data analysis module 305 is used to calculate the average fuel consumption of the target user, obtain all driving behavior data of the target user, determine the type and frequency of instantaneous bad driving behavior based on all driving behavior data of the target user, and generate a driving analysis report based on the average fuel consumption and weight coefficient, wherein the driving analysis report includes:

[0118] Type and frequency of transient adverse driving behaviors.

[0119] The specific implementation of the driving behavior analysis method implemented by each module in the embodiment of the present invention has been introduced in detail in the method side, so it will not be repeated here.

[0120] An embodiment of the present invention further provides a computer-readable storage medium, including a computer program stored therein, for executing any one of the driving behavior analysis methods in the embodiments.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0122] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0123] In addition, the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in the embodiments of the present invention are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

Claims

1. A driving behavior analysis method, applied to a server, characterized in that: include: Obtain driving data and first fuel consumption data of Internet of Vehicles users; Calculating correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and screening main driving behaviors whose correlation coefficients reach a preset threshold using a preset algorithm; Obtaining the target user's main driving behavior data and the second fuel consumption data; Calculating, using a preset model, a weight coefficient of the impact of each of the main driving behaviors of the target user on the second fuel consumption data; calculating the average fuel consumption of the target user based on the second fuel consumption data; Acquire all driving behavior data of the target user, and determine the type and frequency of instantaneous bad driving behavior based on all driving behavior data of the target user; A driving analysis report is generated based on the average fuel consumption and the weight coefficient, wherein the driving analysis report includes: The type and number of instantaneous bad driving behaviors, the horizontal and vertical comparison results of the user's driving behaviors based on the average fuel consumption of the user's vehicle, and the main driving behavior guidance information generated according to the weight coefficient.

2. The method according to claim 1, characterized in that The acquiring of the target user's main driving behavior data includes: Obtaining vehicle bus raw data of the main driving behavior of the target user; Furthermore, the raw data is calculated to obtain characteristic data for evaluating the main driving behavior.

3. The method according to claim 1, characterized in that Before calculating the weight coefficient of the impact of each main driving behavior of the target user on the second fuel consumption data using a preset model, the method further includes: Obtaining user driving data samples for testing, testing the model's preset parameters, obtaining test results, and adjusting the preset parameters; When the test result meets the conditions, the preset model is determined.

4. The method according to claim 1, wherein The calculating the average fuel consumption of the target user according to the second fuel consumption data includes: Acquiring engine operating status information of a target user, and determining a target trip based on the engine operating status information; The mileage data of the target trip and the second fuel consumption data are obtained, and the average fuel consumption of the target user is calculated.

5. The method according to claim 4, characterized in that The determining of the target range according to the engine operating state information includes: According to the engine operating state information, the vehicle state is divided into a vehicle starting state, an engine off state, and a vehicle fully powered state; Obtain the vehicle's travel data and altitude information when it is in the starting state; The trip whose altitude information meets the preset conditions is determined as the target trip.

6. The method according to claim 4, characterized in that Before calculating the average fuel consumption of the target user according to the second fuel consumption data, the method further includes: The fuel consumption data of the warm-up phase trip in the second fuel consumption data is corrected, wherein the warm-up phase trip is a trip in the target trip in which the fuel consumption is abnormally high when the vehicle starts traveling.

7. The method according to claim 1, characterized in that The generating of the driving analysis report based on the average fuel consumption and the weight coefficient includes: calculating, based on the driving data and the first fuel consumption data of the connected vehicle user, the standard fuel consumption of users of the same vehicle model in the connected vehicle network, and determining the fuel saving level of the target user based on the standard fuel consumption of users of the same vehicle model and the average fuel consumption of the target user; determining a target driving behavior according to the weight coefficient and the fuel saving level; A driving analysis report is generated according to the fuel saving level and the target driving behavior within a preset period.

8. The method according to claim 1, characterized in that After obtaining the target user's driving data, the method further includes: Obtaining real-time environmental information of the target user; Based on the environmental information, outliers in the driving data are determined and deleted.

9. A server, characterized in that: include: IoV data acquisition module: used to acquire IoV user's driving data and first fuel consumption data; A screening module is configured to calculate the correlation coefficients between all driving behaviors in the driving data and the first fuel consumption data, and to screen the main driving behaviors whose correlation coefficients reach a preset threshold using a preset algorithm; Target data acquisition module: used to acquire the target user's main driving behavior data and the second fuel consumption data; A calculation module: configured to calculate, by using a preset model, a weight coefficient of the impact of each of the main driving behaviors of the target user on the second fuel consumption data; The data analysis module is configured to calculate the average fuel consumption of a target user, obtain data on all driving behaviors of the target user, determine the type and frequency of instantaneous bad driving behaviors based on the data on all driving behaviors of the target user, and generate a driving analysis report based on the average fuel consumption and the weight coefficient, wherein the driving analysis report includes: The type and number of instantaneous bad driving behaviors, the horizontal and vertical comparison results of the user's driving behaviors based on the average fuel consumption of the user's vehicle, and the main driving behavior guidance information generated according to the weight coefficient.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the driving behavior analysis method according to any one of claims 1 to 8 is implemented.

Images