Driving behavior analysis and evaluation method and system

Abstract

The invention discloses a driving behavior analysis and evaluation method. The method comprises the following steps: segmenting a vehicle travel according to a preset travel segmentation logic; obtaining multi-source data generated by the vehicle in the current split journey, and performing fusion processing; a driving behavior analysis model corresponding to the current vehicle model is obtained, the driving behavior analysis model comprises a multi-dimensional calculation formula for driving behavior evaluation, the fused multi-source data is calculated, and scores of all dimensions and evaluation results of driving behaviors are obtained; and outputting the score of each dimension and the evaluation result of the driving behavior. The invention further provides a corresponding system. By implementing the method, the efficiency, comprehensiveness and accuracy of driving behavior analysis and evaluation can be improved, the cost is reduced, and the iteration speed is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle networking and big data processing, and in particular to a driving behavior analysis and evaluation method and system. Background Art

[0002] In recent years, the number of traffic accidents has remained high, and 70% of traffic accidents are caused by bad driving behavior. In order to meet the objective and microscopic cognition of drivers on their own driving behavior habits, driving behavior analysis can help drivers discover their bad driving behavior in time. In addition, as consumers' preference for green consumption increases, how to achieve higher energy saving and thus reduce energy consumption has become a topic of concern to users.

[0003] At present, the industry has some evaluation methods and product interfaces for driving behavior analysis, as well as discussions on how to use the data results for applications in insurance, user marketing and other fields. However, the existing driving behavior analysis methods have the following shortcomings:

[0004] At present, many solutions judge driving behavior based on the occurrence of preset thresholds, which cannot distinguish between car models and series, and ignore the impact of different car series and models on the thresholds.

[0005] In addition, some existing solutions evaluate driving behavior through data from a single dimension (such as driving safety-related or energy consumption-related), which cannot fully and accurately reflect driving behavior;

[0006] In the existing methods, the computing power of the vehicle itself and the ability of the cloud to modify the local model of the vehicle are not utilized. In the existing methods, the judgment of basic violent driving events such as sudden acceleration is uploaded to the cloud through the original data of the vehicle, and the cloud performs threshold judgment and calculation judgment, which increases resource consumption and cost. And the iteration cost is high. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a driving behavior analysis and evaluation method and system, which can improve the efficiency, comprehensiveness and accuracy of driving behavior analysis and evaluation, reduce costs and increase iteration speed.

[0008] As one aspect of the present invention, a driving behavior analysis and evaluation method is provided, which at least comprises the following steps:

[0009] Divide the vehicle trip according to the preset trip segmentation logic;

[0010] Obtaining multi-source data generated by the vehicle in the current segmented trip and performing fusion processing, the multi-source data including: vehicle operation status data, intelligent driving function usage data, cockpit external environment data and driver status data;

[0011] Obtaining a driving behavior analysis model corresponding to the current vehicle model, the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, performing threshold judgment and calculation on the fused multi-source data, obtaining scores of each dimension and driving behavior evaluation results, the multi-dimensional dimensions including: stability, safety, energy saving and intelligent driving dimensions;

[0012] The scores of the dimensions and the evaluation results of the driving behavior are output.

[0013] The driving behavior analysis model corresponding to the current vehicle model is obtained, and the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, and threshold judgment and calculation are performed on the fused multi-source data to obtain scores of each dimension and driving behavior evaluation results, further including:

[0014] Obtain relevant configuration information of the vehicle type, vehicle power source, vehicle series and specific intelligent driving functions according to the vehicle code;

[0015] Obtaining a corresponding driving behavior analysis model according to the configuration information of the vehicle;

[0016] According to the driving behavior analysis model, a score of each dimension of driving behavior evaluation is obtained;

[0017] The driving behavior evaluation results of the vehicle in the current segmented trip are obtained by summarizing the scores of the driving behavior evaluations of all dimensions and the primary weights corresponding to the driving behavior evaluations of each dimension.

[0018] Wherein, obtaining the score of each dimension of driving behavior evaluation according to the driving behavior analysis model further includes:

[0019] For the dimensions of stability, safety, and intelligent driving, multiple level events included in the driving behavior evaluation of each dimension are obtained, and the secondary weights corresponding to each level event are obtained by combining the entropy weight method with the hierarchical analysis method;

[0020] According to the scores corresponding to the multiple level events and the corresponding secondary weights, the score of each dimension of driving behavior evaluation is obtained.

[0021] The secondary weights corresponding to each level of events obtained by combining the entropy weight method with the hierarchical analysis method include:

[0022] The analytic hierarchy process (AHP) is used to assign subjective weights to each level event, which is recorded as W_a, and the entropy method is used to assign objective weights, which is recorded as W_b.

[0023] Calculate the geometric mean of the two and obtain the corresponding secondary weight: (W_a*W_b) / √(W_a^2+W_b^2).

[0024] Wherein, obtaining the score of each dimension of driving behavior evaluation according to the driving behavior analysis model further includes:

[0025] For the energy-saving dimension, select the input features that affect fuel consumption in the current segmented trip, including: accelerator pedal variation coefficient, engine operation ratio, engine average speed, air conditioning request torque, and medium-speed driving time ratio;

[0026] Linearly map each input feature to the same score interval;

[0027] The energy-saving dimension score Y is calculated using the following formula:

[0028]

[0029] Among them, M is the total number of input features, S i is the score of the i-th input feature, K i is the preset weight corresponding to the i-th input feature.

[0030] Among them, further include:

[0031] Determine whether the user authorizes the analysis of driving behavior and the output of driving behavior evaluation results; if authorized, perform threshold judgment and calculation on the fused multi-source data through the driving behavior analysis model package, and obtain the scores of each dimension and the evaluation results of the driving behavior; otherwise, end the analysis process.

[0032] As another aspect of the present invention, a driving behavior analysis and evaluation system is also provided, which at least comprises:

[0033] A segmentation processing unit, used for segmenting the vehicle trip according to a preset trip segmentation logic;

[0034] A multi-source data acquisition unit is used to obtain multi-source data generated by the vehicle in the current segmented trip and perform fusion processing, wherein the multi-source data includes: vehicle operation status data, intelligent driving function usage data, cockpit external environment data, and driver status data;

[0035] An evaluation processing unit is used to obtain a driving behavior analysis model corresponding to the current vehicle model, wherein the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, performs threshold judgment and calculation on the fused multi-source data, and obtains scores of each dimension and evaluation results of driving behavior, wherein the multi-dimensional dimensions include: stability, safety, energy saving, and intelligent driving dimensions;

[0036] The evaluation result output unit is used to output the scores of each dimension and the evaluation result of the driving behavior.

[0037] The evaluation processing unit further includes:

[0038] A vehicle configuration information acquisition unit, used to acquire relevant configuration information of the vehicle type, vehicle power source, vehicle series and specific intelligent driving functions according to the vehicle code of the vehicle;

[0039] An analysis model acquisition unit, used to obtain the driving behavior analysis model according to the configuration information of the vehicle;

[0040] A score calculation unit for each dimension, used to obtain a score for driving behavior evaluation in each dimension according to the driving behavior analysis model;

[0041] The evaluation result calculation unit is used to summarize the scores of the driving behavior evaluation in all dimensions and the primary weight corresponding to the driving behavior evaluation in each dimension to obtain the evaluation result of the driving behavior of the vehicle in the current segmented trip.

[0042] Among them, in the score calculation unit for each dimension, for the dimensions of stability, safety and intelligent driving, the score of driving behavior evaluation in each dimension is obtained by the following method:

[0043] Obtain multiple level events included in each dimension of driving behavior evaluation, and obtain the secondary weights corresponding to each level event through the entropy weight method combined with the hierarchical analysis method;

[0044] Obtaining a score for each dimension of driving behavior evaluation according to the scores corresponding to the multiple level events and the corresponding secondary weights;

[0045] Among them, the secondary weights corresponding to each level of event obtained by combining the entropy weight method with the hierarchical analysis method include:

[0046] The analytic hierarchy process (AHP) is used to assign subjective weights to each level event, which is recorded as W_a, and the entropy method is used to assign objective weights, which is recorded as W_b.

[0047] Calculate the geometric mean of the two and obtain the corresponding secondary weight: (W_a*W_b) / √(W_a^2+W_b^2).

[0048] Wherein, in the score calculation unit for each dimension, for the energy-saving dimension, the score of driving behavior evaluation in each dimension is obtained by the following method:

[0049] Obtain the input features that have a greater impact on fuel consumption in the current split trip, including: accelerator pedal variation coefficient, engine operation ratio, engine average speed, air conditioning request torque, and medium-speed driving time ratio;

[0050] Linearly map each input feature to the same score interval;

[0051] The energy-saving dimension score Y is calculated using the following formula:

[0052]

[0053] Among them, M is the total number of input features, S i is the score of the i-th input feature, K i is the preset weight corresponding to the i-th input feature.

[0054] Wherein, the driving behavior analysis model is deployed on the cloud, or at least partially deployed on the vehicle side.

[0055] Implementing the embodiments of the present invention has the following beneficial effects:

[0056] The present invention provides a driving behavior analysis and evaluation method and system, which can collect vehicle-side data and upload it to the cloud for real-time calculation and analysis, and can provide result data externally through interface services, and can be applied to different terminals. The vehicle side can upload the collected vehicle operation status, vehicle basic information, driver status data, cockpit external environment data and intelligent driving function data, and the cloud server uses its data to perform itinerary segmentation and driving behavior evaluation model analysis to evaluate the driver's driving behavior, obtain the evaluation result data and provide services externally through the data interface, which can be conveniently displayed on terminals such as mobile phones.

[0057] The present invention combines various vehicle data and various types of driver driving behavior data, and matches corresponding driving behavior analysis models according to vehicle types and power sources for analysis and calculation. The analysis model includes dimensions such as safety, energy saving, stability, and intelligent driving, which can evaluate the driver's behavior more comprehensively and make the evaluation results more accurate.

[0058] The present invention also supports the deployment of some models locally on the vehicle side, and the use of the computing power of the vehicle side can bring more efficient and accurate result data and reduce costs. In the present invention, it also supports the modification of thresholds and model feature values ​​through cloud technology and sends parameters to the vehicle side, which can quickly iterate the local model of the vehicle side. At the same time, it can provide external services through interfaces and quickly respond to the needs of different terminals. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For a person skilled in the art, other drawings obtained based on these drawings still belong to the scope of the present invention without creative labor.

[0060] Figure 1 A schematic diagram of the main process of an embodiment of a driving behavior analysis and evaluation method provided by the present invention;

[0061] Figure 2 A more detailed schematic diagram of the process provided by the present invention;

[0062] Figure 3 A flowchart of the driving behavior evaluation score calculation involved in the method provided by the present invention;

[0063] Figure 4 A schematic diagram of the structure of an embodiment of a driving behavior analysis and evaluation system provided by the present invention;

[0064] Figure 5 for Figure 4 Schematic diagram of the structure of the evaluation processing unit. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0066] like Figure 1 FIG. 1 is a schematic diagram showing a main flow chart of an embodiment of a driving behavior analysis and evaluation method provided by the present invention. Figures 2 to 3 As shown, in this embodiment, the method can be deployed on a cloud server that communicates with the vehicle end, and the method at least includes the following steps:

[0067] Step S10, dividing the vehicle trip according to a preset trip segmentation logic; in a specific example, the cloud server divides the trip according to the preset trip segmentation logic for uploading the data of the vehicle running status, and obtains the relevant data of the trip; the vehicle end can upload the data in real time, or upload the data after the vehicle stops;

[0068] In some examples, the vehicle startup status can be determined through vehicle-side data to segment the trip. For example, if a user starts the vehicle at 10 o'clock, drives for one hour, and then turns off the vehicle at 11 o'clock, the end of a trip can be determined based on the startup and shutdown times, and the trip can be segmented based on the startup and shutdown times, and the relevant data within the trip can be used as the data for the current segmented trip.

[0069] Step S11, obtaining multi-source data generated by the vehicle in the current segmented trip and performing fusion processing, wherein the multi-source data includes: vehicle operation status data, intelligent driving function usage data, cockpit external environment data, and driver status data;

[0070] Among them, the vehicle operation status data can be data such as vehicle speed, acceleration, mileage, fuel consumption of this trip, etc.; the intelligent driving function usage data can be data such as front collision warning, automatic emergency braking, lane departure, speeding recognition, etc.; the data of the external environment of the cabin can be data such as the average temperature outside the cabin, which can be uploaded to the cloud server through TBOX; and the driver status data can be result data such as smoking, making phone calls, distraction and fatigue triggers, which can be uploaded to the cloud server through the vehicle computer embedding method.

[0071] The data uploaded by the vehicle can be stored and analyzed on the cloud server. Before analysis, the data generally needs to be preprocessed, including but not limited to data cleaning, filtering, processing of abnormal values ​​such as null values, and merging data from different sources according to travel time to obtain all the data for this section of the journey.

[0072] Step S12, obtaining a driving behavior analysis model corresponding to the current vehicle model, the driving behavior analysis model including a multi-dimensional calculation formula for driving behavior evaluation, performing threshold judgment and calculation on the fused multi-source data, obtaining scores of each dimension and driving behavior evaluation results, the multi-dimensional dimensions including: stability, safety, energy saving and intelligent driving dimensions;

[0073] Step S13: output the scores of the dimensions and the evaluation results of the driving behavior.

[0074] In a specific example, the step S13 further includes:

[0075] Step S130, obtaining the relevant configuration information of the vehicle type, vehicle power source, vehicle series and specific intelligent driving function according to the vehicle code of the vehicle;

[0076] Step S131, according to the configuration information of the vehicle, obtain a corresponding driving behavior analysis model; it is understandable that corresponding driving behavior analysis models are established in advance for vehicles with different configuration information, such as the model development can be carried out in advance using a Web terminal; each driving behavior analysis model must include multiple dimensions such as stability, safety, energy saving and intelligent driving; among which stability is used to reflect the comfort level in a journey; safety is used to reflect dangerous driving and behaviors that endanger road safety in a journey; energy saving is used to reflect behaviors that affect energy consumption in a journey; and intelligent driving is used to reflect the degree to which intelligent driving functions are applied in a journey;

[0077] Step S132, obtaining a score for each dimension of driving behavior evaluation according to the driving behavior analysis model;

[0078] More specifically, in one example, step S132 further includes:

[0079] For the dimensions of stability, safety, and intelligent driving, multiple level events included in the driving behavior evaluation of each dimension are obtained, and the secondary weights corresponding to each level event are obtained by combining the entropy weight method with the hierarchical analysis method;

[0080] According to the scores corresponding to the multiple level events and the corresponding secondary weights, the score of each dimension of driving behavior evaluation is obtained. In some examples, the score of each level event can be obtained by converting the number of times the event occurs.

[0081] It is understandable that in this step, the severity level of the entire event needs to be distinguished according to the degree to which the relevant indicators exceed the threshold range. For example, in one example, it can be divided into the following levels of events: Level 1, which means that the event occurs but to a weaker degree; Level 2, which means that the event occurs to a moderate degree; Level 3, which means that the event occurs to a serious degree. The level of the event is defined based on the historical data distribution of the indicator and the degree of exceeding the threshold. Each score consists of several events, and each event is assigned a corresponding secondary weight to obtain the final score. For example, in the safety dimension: if the speeding event occurs, the first-level event is converted to 1 point, while the second-level event is converted to 2 points, and the third-level event is converted to 3 points; if the turn signal event is not turned on, the first-level event is converted to 0.5 points, while the second-level event is converted to 1 point, and the third-level event is converted to 1.5 points; and so on. In one dimension, it is necessary to calculate the scores of all level events and their corresponding secondary weights, and summarize them to obtain the final score.

[0082] The secondary weights corresponding to each level of events obtained by combining the entropy weight method with the hierarchical analysis method include:

[0083] The analytic hierarchy process (AHP) is used to assign subjective weights to each level event, which is recorded as W_a, and the entropy method is used to assign objective weights, which is recorded as W_b.

[0084] Calculate the geometric mean of the two and obtain the corresponding secondary weight: (W_a*W_b) / √(W_a^2+W_b^2).

[0085] In other examples, custom configurable ways to assign weights may be supported.

[0086] More specifically, in one example, step S132 further includes:

[0087] For the energy-saving dimension, a machine learning method is used to score. Specifically, the input features that have a greater impact on fuel consumption in the current segmented trip are selected, including: accelerator pedal variation coefficient, engine operation ratio, engine average speed, air conditioning request torque, and medium-speed driving time ratio;

[0088] Linearly map each input feature to the same scoring interval. Specifically, in one example, the fuel consumption per 100 kilometers can be linearly mapped to 60-100 points as a model label. It is understandable that different models determine the upper / lower limits based on the fuel consumption distribution per 100 kilometers.

[0089] In a specific example, a binned Lasso regression model can be used to train the regression model using input features and labels as data, and SHAP Value calculation is used to quantify the impact of features: SHAP is used to interpret the model constructed in the above steps to quantify the specific behaviors that affect economy (fuel consumption). Finally, the following formula can be used to calculate the score Y of the energy-saving dimension:

[0090]

[0091] Among them, M is the total number of input features, S i is the score of the i-th input feature, K i is the preset weight corresponding to the i-th input feature.

[0092] For example, in one example, the formula may be: Y (economy score) = 0.1*acceleration pedal fluctuation + 0.22 low-speed driving proportion + correlation coefficient 3*feature 3 + correlation coefficient 4*feature 4….

[0093] Step S133, summarizing the scores of the driving behavior evaluations in all dimensions and the primary weights corresponding to the driving behavior evaluations in each dimension to obtain the evaluation results of the driving behavior of the vehicle in the current segmented trip.

[0094] In a specific example, the method further comprises:

[0095] Determine whether the user authorizes the analysis of driving behavior and the output of driving behavior evaluation results; if authorized, perform threshold judgment and calculation on the fused multi-source data through the driving behavior analysis model package, and obtain the scores of each dimension and the evaluation results of the driving behavior; otherwise, end the analysis process.

[0096] In step S13, the driving behavior analysis data may be output. Specifically, after obtaining the user's authorization, the driving behavior analysis data service may be provided to the outside in the form of an interface. Figure 2As shown, these data provided to the outside can be processed and displayed on the APP side, displayed on the vehicle side, and analyzed and displayed on the enterprise side (B side).

[0097] In an example of the present invention, the selection of the judgment threshold and model feature value of the model supports rapid online code modification, publishing and application to the vehicle-side model in the cloud.

[0098] It can be understood that the present invention fully considers the differences in vehicle types and power sources, and matches the corresponding models for analysis and calculation; the present invention can more comprehensively evaluate the driver's behavior from the aspects of safety, energy saving, stability and intelligent driving:

[0099] The present invention can calculate more accurate data results by performing partial model calculations on the vehicle side, modifying thresholds and model feature values ​​on the cloud side, and sending parameters to the vehicle side, so as to quickly iterate the local model on the vehicle side. Compared with implementing partial algorithms on the cloud side, such as judging the driver's fatigue state, deploying partial models on the vehicle side can obtain more efficient and high-quality data results. The present invention collects vehicle-side trigger and result data for summary calculation, achieving more efficient calculation speed and saving cloud resource costs;

[0100] The present invention refines the granularity to the comparison of the same car series, providing a more comprehensive and accurate evaluation method and suggestions; at the same time, external services are provided through interfaces, which can quickly respond to the needs of different terminals.

[0101] like Figure 4 FIG. 1 is a schematic diagram showing a structure of an embodiment of a driving behavior analysis and evaluation system 1 provided by the present invention. Figure 5 As shown, in this embodiment, the system 1 at least includes:

[0102] A segmentation processing unit 10 is used to segment the vehicle trip according to a preset trip segmentation logic;

[0103] The multi-source data acquisition unit 11 is used to obtain the multi-source data generated by the vehicle in the current segmented trip and perform fusion processing, wherein the multi-source data includes: vehicle operation status data, intelligent driving function usage data, cockpit external environment data and driver status data;

[0104] The evaluation processing unit 12 is used to obtain a driving behavior analysis model corresponding to the current vehicle model, wherein the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, performs threshold judgment and calculation on the fused multi-source data, and obtains scores of each dimension and evaluation results of the driving behavior, wherein the multi-dimensional dimensions include: stability, safety, energy saving, and intelligent driving dimensions;

[0105] The evaluation result output unit 13 is used to output the scores of each dimension and the evaluation result of the driving behavior.

[0106] Furthermore, if Figure 5 As shown, in the example of the present invention, the evaluation processing unit 12 further includes:

[0107] The vehicle configuration information acquisition unit 120 is used to acquire the relevant configuration information of the vehicle type, vehicle power source, vehicle series and specific intelligent driving functions according to the vehicle code of the vehicle;

[0108] The analysis model acquisition unit 121 is used to obtain the driving behavior analysis model according to the configuration information of the vehicle;

[0109] The score calculation unit 122 for each dimension is used to obtain the score of each dimension of driving behavior evaluation according to the driving behavior analysis model;

[0110] The evaluation result calculation unit 123 is used to summarize the scores of the driving behavior evaluation in all dimensions and the primary weight corresponding to the driving behavior evaluation in each dimension to obtain the evaluation result of the driving behavior of the vehicle in the current segmented trip.

[0111] In a specific example, in the dimension score calculation unit 122, for the dimensions of stability, safety, and intelligent driving, the scores of driving behavior evaluation in each dimension are obtained in the following manner:

[0112] Obtain multiple level events included in each dimension of driving behavior evaluation, and obtain the secondary weights corresponding to each level event through the entropy weight method combined with the hierarchical analysis method;

[0113] Obtaining a score for each dimension of driving behavior evaluation according to the scores corresponding to the multiple level events and the corresponding secondary weights;

[0114] Among them, the secondary weights corresponding to each level of event obtained by combining the entropy weight method with the hierarchical analysis method include:

[0115] The analytic hierarchy process (AHP) is used to assign subjective weights to each level event, which is recorded as W_a, and the entropy method is used to assign objective weights, which is recorded as W_b.

[0116] Calculate the geometric mean of the two and obtain the corresponding secondary weight: (W_a*W_b) / √(W_a^2+W_b^2).

[0117] In a specific example, in the dimension score calculation unit 122, for the energy saving dimension, the score of the driving behavior evaluation of each dimension is obtained in the following manner:

[0118] Obtain the input features that have a greater impact on fuel consumption in the current split trip, including: accelerator pedal variation coefficient, engine operation ratio, engine average speed, air conditioning request torque, and medium-speed driving time ratio;

[0119] Linearly map each input feature to the same score interval;

[0120] The energy-saving dimension score Y is calculated using the following formula:

[0121]

[0122] Among them, M is the total number of input features, S i is the score of the i-th input feature, K i is the preset weight corresponding to the i-th input feature.

[0123] It is understandable that the driving behavior analysis model involved in the present invention can be deployed entirely on the cloud, or at least partially on the vehicle side.

[0124] For more details, please refer to the above Figures 1 to 3 The description is not repeated here.

[0125] As another aspect of the present invention, there is also provided a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the above-mentioned Figures 1 to 3 For more details, please refer to the above Figures 1 to 3 The description will not be repeated here.

[0126] As another aspect of the present invention, a vehicle is provided, comprising a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the above-mentioned Figures 1 to 3 For more details, please refer to the above Figures 1 to 3 The description will not be repeated here.

[0127] Implementing the embodiments of the present invention has the following beneficial effects:

[0128] The present invention provides a driving behavior analysis and evaluation method and system, which can collect vehicle-side data and upload it to the cloud for real-time calculation and analysis, and can provide result data externally through interface services, and can be applied to different terminals. The vehicle side can upload the collected vehicle operation status, vehicle basic information, driver status data, cockpit external environment data and intelligent driving function data, and the cloud server uses its data to perform itinerary segmentation and driving behavior evaluation model analysis to evaluate the driver's driving behavior, obtain the evaluation result data and provide services externally through the data interface, which can be conveniently displayed on terminals such as mobile phones.

[0129] The present invention combines various vehicle data and various types of driver driving behavior data, and matches corresponding driving behavior analysis models according to vehicle types and power sources for analysis and calculation. The analysis model includes dimensions such as safety, energy saving, stability, and intelligent driving, which can evaluate the driver's behavior more comprehensively and make the evaluation results more accurate.

[0130] The present invention also supports the deployment of some models locally on the vehicle side, and the use of the computing power of the vehicle side can bring more efficient and accurate result data and reduce costs. In the present invention, it also supports the modification of thresholds and model feature values ​​through cloud technology and sends parameters to the vehicle side, which can quickly iterate the local model of the vehicle side. At the same time, it can provide external services through interfaces and quickly respond to the needs of different terminals.

[0131] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0132] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] The above disclosure is only a preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A driving behavior analysis and evaluation method, characterized in that: At least the following steps are included: Divide the vehicle trip according to the preset trip segmentation logic; Obtaining multi-source data generated by the vehicle in the current segmented trip and performing fusion processing, the multi-source data including: vehicle operation status data, intelligent driving function usage data, cockpit external environment data and driver status data; Obtaining a driving behavior analysis model corresponding to the current vehicle model, the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, calculating the fused multi-source data, obtaining scores for each dimension and driving behavior evaluation results, the multi-dimensional dimensions including: stability, safety, energy saving, and intelligent driving dimensions; The scores of the dimensions and the evaluation results of the driving behavior are output.

2. The method according to claim 1, characterized in that The driving behavior analysis model corresponding to the current vehicle model is obtained, the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, threshold judgment and calculation are performed on the fused multi-source data to obtain scores of each dimension and driving behavior evaluation results, further including: Obtain relevant configuration information of the vehicle type, vehicle power source, vehicle series and intelligent driving function according to the vehicle code; Obtaining a corresponding driving behavior analysis model according to the configuration information of the vehicle; According to the driving behavior analysis model, a score of each dimension of driving behavior evaluation is obtained; The driving behavior evaluation results of the vehicle in the current segmented trip are obtained by summarizing the scores of the driving behavior evaluations of all dimensions and the primary weights corresponding to the driving behavior evaluations of each dimension.

3. The method according to claim 2, characterized in that Obtaining the score of each dimension of driving behavior evaluation according to the driving behavior analysis model further includes: For the dimensions of stability, safety, and intelligent driving, multiple level events included in the driving behavior evaluation of each dimension are obtained, and the secondary weights corresponding to each level event are obtained by combining the entropy weight method with the hierarchical analysis method; According to the scores corresponding to the multiple level events and the corresponding secondary weights, the score of each dimension of driving behavior evaluation is obtained.

4. The method according to claim 3, characterized in that: The secondary weights corresponding to each level of events obtained by combining the entropy weight method with the hierarchical analysis method include: The analytic hierarchy process (AHP) is used to assign subjective weights to each level event, which is recorded as W_a, and the entropy method is used to assign objective weights, which is recorded as W_b. Calculate the geometric mean of the two and obtain the corresponding secondary weight: (W_a*W_b) / √(W_a^2+W_b^2).

5. The method according to claim 4, characterized in that Obtaining the score of each dimension of driving behavior evaluation according to the driving behavior analysis model further includes: For the energy-saving dimension, select the input features that affect fuel consumption in the current segmented trip, including: accelerator pedal variation coefficient, engine operation ratio, engine average speed, air conditioning request torque, and medium-speed driving time ratio; Linearly map each input feature to the same score interval; The energy-saving dimension score Y is calculated using the following formula: Among them, M is the total number of input features, S i is the score of the i-th input feature, K i is the preset weight corresponding to the i-th input feature.

6. The method according to claim 5, characterized in that Further including: Determine whether the user authorizes the analysis of driving behavior and the output of driving behavior evaluation results; If authorized, the driving behavior analysis model package performs threshold judgment and calculation on the fused multi-source data, and obtains the scores of each dimension and the evaluation results of the driving behavior; otherwise, the analysis process ends.

7. A driving behavior analysis and evaluation system, characterized in that: At least: A segmentation processing unit, used for segmenting the vehicle trip according to a preset trip segmentation logic; A multi-source data acquisition unit is used to obtain multi-source data generated by the vehicle in the current segmented trip and perform fusion processing, wherein the multi-source data includes: vehicle operation status data, intelligent driving function usage data, cockpit external environment data, and driver status data; An evaluation processing unit is used to obtain a driving behavior analysis model corresponding to the current vehicle model, wherein the driving behavior analysis model includes a multi-dimensional calculation formula for driving behavior evaluation, calculates the fused multi-source data, obtains scores of each dimension and evaluation results of driving behavior, and the multi-dimensional dimensions include: stability, safety, energy saving, and intelligent driving dimensions; The evaluation result output unit is used to output the scores of each dimension and the evaluation result of the driving behavior.

8. The system according to claim 7, characterized in that The evaluation processing unit further comprises: A vehicle configuration information acquisition unit, used to acquire relevant configuration information of the vehicle type, vehicle power source, vehicle series and intelligent driving function according to the vehicle code of the vehicle; An analysis model acquisition unit, used to obtain the driving behavior analysis model according to the configuration information of the vehicle; A score calculation unit for each dimension, used to obtain a score for driving behavior evaluation in each dimension according to the driving behavior analysis model; The evaluation result calculation unit is used to summarize the scores of the driving behavior evaluation in all dimensions and the primary weight corresponding to the driving behavior evaluation in each dimension to obtain the evaluation result of the driving behavior of the vehicle in the current segmented trip.

9. The system according to claim 8, characterized in that In the dimension score calculation unit, for the dimensions of stability, safety and intelligent driving, the scores of driving behavior evaluation in each dimension are obtained in the following manner: Obtain multiple level events included in each dimension of driving behavior evaluation, and obtain the secondary weights corresponding to each level event through the entropy weight method combined with the hierarchical analysis method; Obtaining a score for each dimension of driving behavior evaluation according to the scores corresponding to the multiple level events and the corresponding secondary weights; Among them, the secondary weights corresponding to each level of event obtained by combining the entropy weight method with the hierarchical analysis method include: The analytic hierarchy process (AHP) is used to assign subjective weights to each level event, which is recorded as W_a, and the entropy method is used to assign objective weights, which is recorded as W_b. Calculate the geometric mean of the two and obtain the corresponding secondary weight: (W_a*W_b) / √(W_a^2+W_b^2).

10. The system according to claim 9, characterized in that In the dimension score calculation unit, for the energy saving dimension, the score of driving behavior evaluation in each dimension is obtained by the following method: Obtain the input features that have a greater impact on fuel consumption in the current split trip, including: accelerator pedal variation coefficient, engine operation ratio, engine average speed, air conditioning request torque, and medium-speed driving time ratio; Linearly map each input feature to the same score interval; The energy-saving dimension score Y is calculated using the following formula: Among them, M is the total number of input features, S i is the score of the i-th input feature, K i is the preset weight corresponding to the i-th input feature.

11. The system according to any one of claims 7 to 8, characterized in that: The driving behavior analysis model is deployed in the cloud, or at least partially deployed in the vehicle.

Images