A method and device for identifying a vehicle driving behavior and style

Abstract

The present invention discloses a method and device for identifying driving behaviors and styles of an automobile. The identification method includes: obtaining driving data during the driving of the vehicle; extracting first determination data from the driving data at the vehicle end, and determining the driving behavior corresponding to the first determination data according to a discrimination rule; and transmitting second determination data in the driving data to the server end to determine the driving style corresponding to the second determination data by using a driving style identification model of the server end. The present invention can reduce the occupation of computing resources of the server end, and can identify some driving behaviors with high timeliness requirements at the vehicle end, which is beneficial for the vehicle to make timely judgments on corresponding driving behaviors.

Description

Technical Field

[0001] The present invention relates to the technical field of assisted driving, and particularly relates to a method and device for identifying driving behaviors and styles of an automobile. Background Art

[0002] In recent years, the Internet of Things technology has developed rapidly and been applied in various industries. In the field of automobiles, more and more vehicles are connected to the network to form a vehicle network. Connecting an automobile to the network can provide many information services for vehicle owners, such as navigation and remote upgrade. However, many traditional driving safety problems still exist. During the driving process, a driver may have various dangerous driving behaviors, which will pose safety threats to the driver himself and other road users in terms of life, property, etc.

[0003] Therefore, in order to detect dangers in advance and avoid accidents, it is particularly necessary to analyze and identify the driving behaviors and styles of a driver. Currently, the methods for identifying driving behaviors and styles of an automobile mainly upload relevant driving data to a cloud server, and analyze and process the driving data on the cloud server to determine the corresponding driving behaviors and styles. This method needs to transmit driving data to the cloud server for processing, which cannot meet the requirements for analyzing some driving behaviors with strict timeliness requirements. Moreover, the amount of data to be transmitted to the cloud server is relatively large, which will be affected by the transmission rate. And all data is processed by the cloud server, which will impose a relatively large burden on the computing resources of the cloud server.

[0004] Therefore, it is necessary to provide a new method and device for identifying driving behaviors and styles of an automobile to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, device, electronic device and computer-readable storage medium for identifying driving behaviors and styles of an automobile, which can reduce the occupation of computing resources on the server side, and can identify some driving behaviors with high timeliness requirements at the vehicle end, which is beneficial for the vehicle to make timely judgments on corresponding driving behaviors.

[0006] To achieve the above purpose, the present invention provides a method for identifying driving behaviors and styles of an automobile, including:

[0007] Obtaining driving data during the driving process of the vehicle;

[0008] Extracting first determination data from the driving data at the vehicle end, and determining the driving behavior corresponding to the first determination data according to a discrimination rule; and

[0009] Transmit the second determination data in the driving data to the server side to use the driving style recognition model on the server side to determine the second determination data to determine the driving style corresponding to the second determination data.

[0010] Optionally, the driving data includes:

[0011] The linear speed, linear acceleration, angular speed, angular acceleration, and ACC status of the vehicle driving.

[0012] Optionally, the discrimination rules include:

[0013] Determine that the driving behavior conforms to the characteristics of sudden acceleration or sudden deceleration according to the linear acceleration;

[0014] Determine that the driving behavior conforms to the characteristics of sharp turning according to the angular acceleration;

[0015] Determine that the driving behavior conforms to the characteristics of speeding according to the linear speed;

[0016] Determine that the driving behavior conforms to the characteristics of coasting with the engine off according to the linear speed and the ACC status;

[0017] Determine that the driving behavior conforms to the characteristics of idling preheating or long-term idling according to the linear speed, the ACC status, and the duration.

[0018] Optionally, the driving style recognition model is a k-means clustering model, and a corresponding number of sampling data points are generated according to the second determination data;

[0019] The "driving style recognition model determines the second determination data" includes:

[0020] a. Set a corresponding number of clustering centers according to the preset types of driving styles;

[0021] b. Divide the several sampling data points corresponding to the second determination data into the clusters represented by the nearest clustering centers respectively;

[0022] c. Calculate the mean of the sampling data points in each divided cluster to re-determine the clustering centers;

[0023] d. Repeat steps b and c until the iteration termination condition is met, and determine the types of driving styles represented by each cluster.

[0024] Optionally, the driving style recognition model includes several classification models constructed based on the random forest algorithm, and each classification model is trained according to different types of sample data;

[0025] The "judgment of the driving style recognition model on the second judgment data" includes:

[0026] Input the second judgment data into the corresponding classification model according to the data type of the second judgment data;

[0027] The classification model classifies the input second judgment data according to the types of driving styles to output a classification result;

[0028] Fuse each classification result with a preset weight, and compare the weighted results of different classification results to determine the type of driving style.

[0029] Optionally, the "each classification model is trained according to different types of sample data" includes:

[0030] Obtain different types of the sample data;

[0031] Preprocess the sample data to extract feature data;

[0032] Input different types of the feature data into the corresponding classification model for training.

[0033] Optionally, the types of the sample data include angular acceleration, angular velocity, and linear acceleration;

[0034] The data types of the second judgment data are consistent with the types of the sample data;

[0035] The classification model outputs classification results according to the angular acceleration data, angular velocity data, and linear acceleration data in the second judgment data respectively.

[0036] To achieve the above object, the present invention also provides an identification device for automotive driving behaviors and styles, including:

[0037] An acquisition module, configured to acquire driving data during the driving of a vehicle;

[0038] A behavior recognition module, configured to extract first judgment data from the driving data at the vehicle end, and judge the first judgment data according to a discrimination rule to determine the driving behavior corresponding to the first judgment data;

[0039] A transmission module, configured to transmit second judgment data in the driving data to the server end;

[0040] A style recognition module, configured to judge the second judgment data based on a driving style recognition model at the server end to determine the driving style corresponding to the second judgment data.

[0041] To achieve the above object, the present invention also provides an electronic device, including:

[0042] Processor;

[0043] Memory, in which executable instructions of the processor are stored;

[0044] Wherein, the processor is configured to execute the recognition method of the above-mentioned driving behavior and style of the vehicle by executing the executable instructions.

[0045] To achieve the above object, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the recognition method of the above-mentioned driving behavior and style of the vehicle is implemented.

[0046] The present invention also provides a computer program product or a computer program, the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the recognition method of the above-mentioned driving behavior and style of the vehicle.

[0047] The present invention processes the first determination data in the driving data at the vehicle end, and processes the second determination data in the driving data at the server end. Wherein, the first determination data in the driving data is extracted at the vehicle end, and the first determination data is determined according to the discrimination rule to determine the corresponding driving behavior. At the server end, the second determination data is determined based on the driving style recognition model to determine the corresponding driving style, avoiding the situation that all recognition processes are processed at the server end, reducing the occupation of computing resources at the server end, and moreover, some driving behaviors with high timeliness requirements can be recognized at the vehicle end, so that the vehicle can make timely judgments on the corresponding driving behaviors. Description of the Drawings

[0048] Figure 1 is the data flow diagram of the recognition method of the driving behavior and style of the vehicle in the embodiment of the present invention.

[0049] Figure 2 is the method flow chart of using the k-means clustering model to recognize the driving style in the embodiment of the present invention.

[0050] Figure 3 is the method flow chart of using the classification model constructed by the random forest algorithm to recognize the driving style in the embodiment of the present invention.

[0051] Figure 4 is the schematic block diagram of the recognition device of the driving behavior and style of the vehicle in the embodiment of the present invention.

[0052] Figure 5 is the schematic block diagram of the electronic device in the embodiment of the present invention. Detailed implementation manners

[0053] In order to elaborate in detail the technical content, structural features, achieved objectives and effects of the present invention, the following is a detailed description in conjunction with the implementation manners and with reference to the accompanying drawings.

[0054] Please refer to Figure 1 , an embodiment of the present invention discloses a method for identifying driving behaviors and styles of an automobile, including:

[0055] Obtaining driving data during the driving process of the vehicle;

[0056] Extracting first determination data from the driving data at the vehicle end, and determining, according to discrimination rules, the driving behavior corresponding to the first determination data; and

[0057] Transmitting second determination data in the driving data to the server end to use the driving style recognition model at the server end to determine the driving style corresponding to the second determination data.

[0058] In the present invention, the first determination data in the driving data is processed at the vehicle end, while the second determination data in the driving data is processed at the server end. Among them, the first determination data in the driving data is extracted at the vehicle end, and the driving behavior corresponding thereto is determined according to the discrimination rules. At the server end, the second determination data is determined based on the driving style recognition model to determine the corresponding driving style, avoiding the situation where all recognition processes are processed at the server end, reducing the occupation of computing resources at the server end, and moreover, some driving behaviors with high timeliness requirements can be recognized at the vehicle end so that the vehicle can make timely judgments on the corresponding driving behaviors.

[0059] Specifically, the driving data includes: linear velocity, linear acceleration, angular velocity, angular acceleration and ACC status of the vehicle. Among them, the linear acceleration, angular velocity and angular acceleration can be obtained by a linear acceleration sensor, an angular acceleration sensor and an angular velocity sensor arranged on the vehicle; the linear velocity can be obtained by a GPS module, and the GPS module can also obtain time data; the ACC status can be obtained by reading CAN bus data through an OBD interface at the vehicle end.

[0060] It can be understood that the ACC status can be 1 or 0. When the ACC status is 1, it means that the vehicle engine is in the starting state. When the ACC status is 0, it means that the vehicle engine is in the closed state.

[0061] Specifically, the first determination data includes linear velocity, linear acceleration, angular acceleration and ACC status.

[0062] Further, the discrimination rules include: determining that the driving behavior conforms to the characteristics of rapid acceleration or rapid deceleration according to the linear acceleration; determining that the driving behavior conforms to the characteristics of sharp turning according to the angular acceleration; determining that the driving behavior conforms to the characteristics of speeding according to the linear speed; determining that the driving behavior conforms to the characteristics of coasting with the engine off according to the linear speed and the ACC state; determining that the driving behavior conforms to the characteristics of idling preheating or long idling according to the linear speed, the ACC state and the duration.

[0063] When the linear acceleration exceeds the preset rapid acceleration threshold, it can be determined that the driving behavior conforms to the characteristics of rapid acceleration; when the linear acceleration exceeds the preset rapid deceleration threshold, it can be determined that the driving behavior conforms to the characteristics of rapid deceleration; when the angular acceleration exceeds the preset sharp turning threshold, it can be determined that the driving behavior conforms to the characteristics of sharp turning; when the linear speed exceeds the preset speeding threshold, it can be determined that the driving behavior conforms to the characteristics of speeding; when the linear speed exceeds the preset driving threshold and the ACC state is 0, it can be determined that the driving behavior conforms to the characteristics of coasting with the engine off; when the linear speed is lower than the preset action threshold, the ACC state is 1 and the duration exceeds the preset time, it can be determined that the driving behavior conforms to the characteristics of long idling; when the linear speed is lower than the preset action threshold, the ACC state is 1 and the duration is within the preset preheating time, it can be determined that the driving behavior conforms to the characteristics of idling preheating.

[0064] The identification of the above driving behaviors can be realized based on the determination of simple rules, which requires less computing power of the processor and can be completed by the processor set on the vehicle side. Completing the identification of the above driving behaviors on the vehicle side is beneficial to reducing the computing pressure on the server side and enabling the vehicle side to timely learn the driving behaviors with requirements for timeliness.

[0065] Specifically, the vehicle side is provided with a network transmission module, and the second determination data for driving style identification is transmitted to the server side through the network transmission module.

[0066] In some embodiments, the driving style identification model is a k-means clustering model, and a corresponding number of sampling data points are generated according to the second determination data.

[0067] Please refer to Figure 2 , "the driving style identification model determines the second determination data" includes:

[0068] a. Set the corresponding number of clustering centers according to the preset types of driving styles.

[0069] b. Divide the several sampling data points corresponding to the second determination data into the clusters represented by the nearest clustering centers respectively.

[0070] c. Calculate the mean value of the sampling data points in each divided cluster to re-determine the clustering centers.

[0071] d. Repeat steps b and c until the iteration termination condition is met, and determine the types of driving styles represented by each cluster.

[0072] Specifically, the second determination data can be the linear velocity or linear acceleration collected in units of time, or the linear velocity or linear acceleration collected in units of distance.

[0073] Furthermore, select the initial clustering centers according to the distribution of the sampling data points. To obtain a better clustering effect, the initial clustering centers can be dispersed within the distribution range of the sampling data points. For example, when the sampling data points of the linear velocity are distributed between 20 km / h and 90 km / h, and the preset types of driving styles are three, the initial clustering centers can be set to 30 km / h, 50 km / h, and 80 km / h.

[0074] It can be understood that the distance referred to in step b is the Euclidean distance; the iteration termination condition is that the clustering centers converge or the number of iterations reaches the preset number.

[0075] The following are two examples for illustration:

[0076] When the second determination data is the linear velocity collected per second, the number of clustering centers can be set to 3, corresponding to three driving styles: steady, aggressive, and fatigued. After the clustering iteration termination condition is met, several sampling data points form three clusters, each cluster corresponding to the three driving styles of steady, aggressive, and fatigued. Among them, the cluster with a relatively high linear velocity is classified as the aggressive driving style, the cluster with a relatively low linear velocity is classified as the fatigued driving style, and the cluster with a moderate linear velocity is classified as the steady driving style.

[0077] When the second determination data is the linear acceleration collected per second, the number of clustering centers can be set to 3, corresponding to three driving styles: steady, accelerating-aggressive, and decelerating-aggressive. After the clustering iteration termination condition is met, several sampling data points form three clusters, each cluster corresponding to the three driving styles of steady, accelerating-aggressive, and decelerating-aggressive. Among them, the cluster with the linear acceleration near 0 is classified as the steady driving style, the cluster with a positive and relatively high linear acceleration is classified as the accelerating-aggressive driving style, and the cluster with a moderate linear acceleration is classified as the decelerating-aggressive driving style.

[0078] In some embodiments, the driving style recognition model includes several classification models constructed based on the random forest algorithm, and each classification model is trained according to different types of sample data.

[0079] Specifically, "each classification model is trained according to different types of sample data" includes:

[0080] Obtain different types of sample data;

[0081] Preprocess the sample data to extract feature data;

[0082] Input different types of feature data into the corresponding classification models for training.

[0083] Among them, the types of sample data can include angular acceleration, angular velocity, and linear acceleration. Corresponding classification models can be trained according to different types of sample data. Making decisions on driving styles by multiple classification models can improve the decision-making accuracy.

[0084] Furthermore, preprocess the sample data through the MMA (Multiscale multifractal analysis) algorithm to generate a Hurst surface map, thereby extracting feature data, which reflects the fluctuation trend of the sample data.

[0085] Please refer to Figure 3 ,"The driving style recognition model makes a determination on the second determination data" includes:

[0086] S41. Input the second determination data into the corresponding classification model according to the data type of the second determination data.

[0087] S42. The classification model classifies the input second determination data according to the types of driving styles to output a classification result.

[0088] S43. Weight and fuse each classification result according to a preset weight, and compare the weighted results of different classification results to determine the type of driving style.

[0089] Specifically, the data type of the second determination data is the same as that of the sample data. If the data types are angular acceleration, angular velocity, and linear acceleration respectively, the components of each data on the x-axis and y-axis in the driving plane should be extracted to avoid being interfered by the z-axis component.

[0090] Furthermore, the preset weight can be set according to the accuracy of the classification model, or can be set according to the emphasis of the data type. For example, the weight of linear acceleration can be set to be higher than the weights of angular velocity and angular acceleration.

[0091] The following gives an example to illustrate:

[0092] When the second determination data is the linear acceleration, angular velocity, and angular acceleration collected per second, and the types of driving styles are divided into steady, aggressive, and fatigued styles, the linear acceleration, angular velocity, and angular acceleration are respectively input into the corresponding classification models. The classification models can respectively output classification results according to the input data, and then each classification result is weighted and fused according to a preset weight. For example, if the classification result output by the linear acceleration is the steady style, the classification result output by the angular velocity is also the steady style, and the classification result output by the angular acceleration is the aggressive style, if the determination is fused according to the equal weight method, the driving style can be determined as the steady style.

[0093] Please refer to Figure 4 , the present invention also provides an identification device for automotive driving behaviors and styles, including:

[0094] An acquisition module 100, configured to acquire driving data during the driving process of the vehicle.

[0095] A behavior identification module 200, configured to extract first determination data from the driving data at the vehicle end and determine the driving behavior corresponding to the first determination data according to the discrimination rule.

[0096] A transmission module 300, configured to transmit the second determination data in the driving data to the server end.

[0097] A style identification module 400, configured to determine the driving style corresponding to the second determination data based on the driving style identification model at the server end.

[0098] The present invention can avoid the situation where the identification processes of driving behaviors and driving styles are both processed at the server end, reduce the occupation of computing resources at the server end, and moreover, can identify some driving behaviors with high timeliness requirements at the vehicle end so that the vehicle can make timely judgments on the corresponding driving behaviors.

[0099] Please refer to Figure 5 , the present invention also provides an electronic device, including:

[0100] A processor 40;

[0101] A memory 50, which stores executable instructions of the processor 40;

[0102] Wherein, the processor 40 is configured to execute the above-mentioned method for identifying automotive driving behaviors and styles by executing the executable instructions.

[0103] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned method for identifying automotive driving behaviors and styles is implemented.

[0104] An embodiment of the present invention also discloses a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the method for identifying the driving behavior and style of an automobile as described above.

[0105] It should be understood that in the embodiment of the present invention, the so-called processor may be a central processing module (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0106] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by hardware related to computer program instructions. The program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), etc.

[0107] The above-disclosed are only the preferred examples of the present invention, and the scope of the rights of the present invention cannot be limited thereby. Therefore, all equivalent changes made according to the claims of the present invention fall within the scope covered by the present invention.

Claims

1. A method for identifying driving behaviors and styles of a vehicle, characterized in that, it includes: Obtaining driving data of the vehicle during driving; Extracting first determination data from the driving data at the vehicle end, and determining the driving behavior corresponding to the first determination data according to discrimination rules; and Transmitting second determination data in the driving data to the server end to use the driving style recognition model at the server end to determine the second determination data to determine the driving style corresponding to the second determination data; The driving style recognition model is a k-means clustering model, and generating corresponding several sampling data points according to the second determination data; The "determining the second determination data by the driving style recognition model" includes: a. Setting corresponding numbers of clustering centers according to the preset types of driving styles, and dispersedly setting the initial clustering centers within the distribution range of the sampling data points; b. Dividing the several sampling data points corresponding to the second determination data into the clusters represented by the nearest clustering centers respectively; c. Calculating the mean value of the sampling data points within each divided cluster to re-determine the clustering centers; d. Repeating steps b and c until the iteration termination condition is met, and determining the types of driving styles represented by each cluster; The driving style recognition model includes several classification models constructed based on the random forest algorithm, and each classification model is trained according to different types of sample data; The "determining the second determination data by the driving style recognition model" includes: Inputting the second determination data into the corresponding classification model according to the data type of the second determination data; The classification model classifies the input second determination data according to the types of driving styles to output a classification result; Performing weighted fusion on each classification result according to preset weights, and comparing the weighted results of different classification results to determine the type of driving style.

2. The method for identifying driving behaviors and styles of a vehicle according to claim 1, characterized in that, the driving data includes: Linear speed, linear acceleration, angular speed, angular acceleration and ACC status of the vehicle during driving.

3. The method for identifying driving behaviors and styles of a vehicle according to claim 2, characterized in that, the discrimination rules include: Determining that the driving behavior conforms to the characteristics of sudden acceleration or sudden deceleration according to the linear acceleration; Determining that the driving behavior conforms to the characteristics of sharp turning according to the angular acceleration; Determining that the driving behavior conforms to the characteristics of speeding according to the linear speed; Determining that the driving behavior conforms to the characteristics of coasting with the engine off according to the linear speed and the ACC status; Determining that the driving behavior conforms to the characteristics of idling preheating or long-term idling according to the linear speed, the ACC status and the duration.

4. The method for identifying driving behaviors and styles of a vehicle according to claim 1, characterized in that, the "each classification model is trained according to different types of sample data" includes: Obtaining different types of the sample data; Preprocessing the sample data to extract feature data; Input different types of the feature data into the corresponding classification model for training.

5. The method for identifying a driving behavior and style of a vehicle according to claim 4, wherein, the types of the sample data include angular acceleration, angular velocity, and linear acceleration; the data types of the second determination data are the same as those of the sample data; the classification model outputs classification results respectively according to the angular acceleration data, angular velocity data, and linear acceleration data in the second determination data.

6. An apparatus for identifying a driving behavior and style of a vehicle, wherein, the identification apparatus operates based on the method for identifying a driving behavior and style of a vehicle according to any one of claims 1 to 5.

7. An electronic device, wherein, comprising: a processor; a memory storing executable instructions of the processor; wherein, the processor is configured to execute the method for identifying a driving behavior and style of a vehicle according to any one of claims 1 - 5 by executing the executable instructions.

8. A computer-readable storage medium having a computer program stored thereon, wherein, the computer program, when executed by a processor, implements the method for identifying a driving behavior and style of a vehicle according to any one of claims 1 - 5.

Images