A driving style recognition method, system, device and medium

Abstract

The application discloses a driving style recognition method, system, device and medium, wherein the method obtains natural driving data of a target driver; driving behavior recognition is performed on the natural driving data to obtain driving behavior identification of the target driver; the natural driving data is input into a driving style recognition model corresponding to the driving behavior identification for style recognition to obtain intermediate driving style information output by the driving style recognition model; and style correction is performed on the intermediate driving style information to obtain target driving style information of the target driver. The method can effectively improve the adaptive ability of driving style recognition and improve the recognition accuracy of driving style. The application relates to the technical field of driving style recognition.

Description

Technical Field

[0001] This invention relates to the field of driving style recognition technology, and in particular to a driving style recognition method, system, device and medium. Background Technology

[0002] As vehicle intelligence becomes the mainstream development trend in the global automotive industry, assisted driving is increasingly favored by consumers. Among these factors, the driver's driving style, which affects the effectiveness of assisted driving, has become one of the key areas of focus for relevant professionals.

[0003] Currently, the relevant technologies typically use offline data to train driving style recognition models and use the driving style identified by the model as the driver's inherent driving style throughout the driving process. This approach cannot adapt well to the driver's driving style under different driving behaviors, has limited adaptability, and the accuracy of driving style recognition is unsatisfactory.

[0004] Therefore, the problems with the relevant technologies still need to be solved and optimized. Summary of the Invention

[0005] The purpose of this invention is to at least partially solve one of the technical problems existing in the related art.

[0006] Therefore, one objective of this invention is to provide a driving style recognition method, system, device, and medium, wherein the method can effectively improve the adaptive capability of driving style recognition and improve the accuracy of driving style recognition.

[0007] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of this application include:

[0008] In a first aspect, embodiments of this application provide a driving style recognition method, including:

[0009] Acquire the target driver's natural driving data;

[0010] The natural driving data is used to identify driving behavior to obtain the driving behavior identifier of the target driver;

[0011] The natural driving data is input into the driving style recognition model corresponding to the driving behavior identifier for style recognition, and intermediate driving style information output by the driving style recognition model is obtained.

[0012] The intermediate driving style information is corrected to obtain the target driving style information of the target driver.

[0013] In addition, the method according to the above embodiments of this application may also have the following additional technical features:

[0014] Furthermore, in one embodiment of this application, the step of performing driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver includes:

[0015] Obtain a driving behavior recognition model;

[0016] The natural driving data is input into the driving behavior recognition model for behavior recognition, and the driving behavior identifier output by the driving behavior recognition model is obtained.

[0017] Furthermore, in one embodiment of this application, the driving behavior recognition model is constructed through the following steps:

[0018] Obtain several driving behavior rules;

[0019] Based on all the driving behavior rules, determine the behavior membership function corresponding to each driving behavior rule;

[0020] The driving behavior recognition model is constructed based on all the behavior membership functions.

[0021] Furthermore, in one embodiment of this application, the step of inputting the natural driving data into a driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtaining intermediate driving style information output by the driving style recognition model, includes:

[0022] Obtain several pre-trained driving style recognition models, each of which corresponds to a type of driving behavior;

[0023] Based on the driving behavior identifier, all the trained driving style recognition models are screened to obtain the target style recognition model;

[0024] The natural driving data is input into the target style recognition model for style recognition to obtain the intermediate driving style information.

[0025] Furthermore, in one embodiment of this application, obtaining a plurality of trained driving style recognition models includes:

[0026] Obtain the original style recognition model and driving training dataset, wherein the driving training dataset includes several sample driving data, and all the sample driving data correspond to a type of driving behavior;

[0027] Features and labels are extracted from the driving training dataset to obtain a training feature set and a feature label corresponding to each target driving feature in the training feature set. The target driving features under different types of driving behaviors are not completely the same.

[0028] Based on all the aforementioned feature labels, the training feature set is input into the original style recognition model for parameter update, thereby obtaining the trained driving style recognition model.

[0029] Furthermore, in one embodiment of this application, the step of performing style correction on the intermediate driving style information to obtain the target driving style information of the target driver includes:

[0030] Obtain historical update vectors and historical driving style information. The historical update vector is the update vector of the previous time point adjacent to the current time point. The current time point is the time point of the intermediate driving style information. The historical driving style information is the target driving style information that is the same as the time point of the historical update vector.

[0031] Based on the historical update vector and the historical driving style information, the intermediate driving style information is updated to obtain the target driving style information.

[0032] Further, in one embodiment of this application, the step of updating the intermediate driving style information according to the historical update vector and the historical driving style information to obtain the target driving style information includes:

[0033] Based on the historical driving style information, the historical update vector is updated according to a time-dependent relationship to obtain the current update vector corresponding to the intermediate driving style information;

[0034] The intermediate driving style information is corrected and updated based on the current update vector to obtain the target driving style information.

[0035] Secondly, embodiments of this application provide a driving style recognition system, including:

[0036] The first processing unit is used to acquire the target driver's natural driving data;

[0037] The second processing unit is used to perform driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver;

[0038] The third processing unit is used to input the natural driving data into the driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtain the intermediate driving style information output by the driving style recognition model.

[0039] The fourth processing unit is used to perform style correction on the intermediate driving style information to obtain the target driving style information of the target driver.

[0040] Thirdly, embodiments of this application also provide an electronic device, including:

[0041] At least one processor;

[0042] At least one memory for storing at least one program;

[0043] When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.

[0044] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a processor-executable program, which, when executed by the processor, is used to implement the above-described method.

[0045] The advantages and beneficial effects of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application:

[0046] This application discloses a driving style recognition method, system, device, and medium. The method involves acquiring natural driving data of a target driver; performing driving behavior recognition on the natural driving data to obtain a driving behavior identifier for the target driver; inputting the natural driving data into a driving style recognition model corresponding to the driving behavior identifier for style recognition to obtain intermediate driving style information output by the driving style recognition model; and performing style correction on the intermediate driving style information to obtain the target driver's target driving style information. This method, by inputting natural driving data into a driving style recognition model corresponding to the driving behavior identifier for style recognition, can consider the driver's driving style under different driving behaviors such as acceleration, deceleration, and turning, thereby effectively improving the adaptive ability to recognize driving style under different driving behaviors and improving the accuracy of driving style recognition. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of this application or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions in this application. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0048] Figure 1 A flowchart illustrating a driving style recognition method provided in an embodiment of this application;

[0049] Figure 2 A schematic diagram of the framework of a driving style recognition system provided in an embodiment of this application;

[0050] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0051] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0053] Currently, relevant technologies typically utilize offline data to train driving style recognition models, and then use the driving style identified by this model as the driver's inherent driving style throughout the entire driving process. However, in real-world applications, the same driver may exhibit different driving styles throughout the entire driving process. This approach cannot adequately adapt to the driver's driving style under different driving behaviors, exhibiting limited adaptive capabilities and resulting in unsatisfactory accuracy in driving style recognition. Furthermore, this method does not consider the relationship between the driver's "long-term habits" and "instantaneous preferences" in driving, which may lead to random estimation biases, resulting in low accuracy of the inherent driving style output by the model.

[0054] It should be noted that the aforementioned related technologies are only used to assist in understanding the technical solutions of this application and do not mean that they belong to the publicly disclosed prior art.

[0055] In view of this, embodiments of the present invention provide a driving style recognition method, system, device, and medium. The method performs style recognition by inputting natural driving data into a driving style recognition model corresponding to driving behavior identifiers. It can take into account the driving style of the driver under different driving behaviors such as acceleration, deceleration, and turning, thereby effectively improving the adaptive ability to recognize the driving style under different driving behaviors and improving the accuracy of driving style recognition.

[0056] Furthermore, this method performs style correction on the intermediate driving style information output by the driving style recognition model. Specifically, it updates the intermediate driving style information through the current update vector with temporal dependence. This fully considers the relationship between the driver's long-term driving style and instantaneous driving style, which helps to reduce the generation of random estimation bias during the recognition process (i.e., mitigate the negative impact of random fluctuations), thereby effectively improving the accuracy of driving style recognition.

[0057] Reference Figure 1 In this application embodiment, a driving style recognition method includes:

[0058] Step 110: Obtain the target driver's natural driving data;

[0059] In this embodiment, the natural driving data can be the real vehicle data under the current natural driving conditions when the target driver is driving the vehicle. Specifically, it can be obtained by collecting the vehicle's CAN signal through a multi-channel CANoe device. The natural driving data includes, but is not limited to, the vehicle's speed, longitudinal acceleration, lateral acceleration, steering wheel angle, steering wheel speed, accelerator pedal opening, and brake pedal opening.

[0060] It is understandable that due to factors such as vehicle vibration and uneven road surface, the collected natural driving data may be mixed with random noise. Therefore, after obtaining the original natural driving data, preprocessing operations such as outlier detection and removal and noise filtering can be performed on the original natural driving data to obtain preprocessed natural driving data. There are already a variety of specific preprocessing methods, which will not be elaborated here.

[0061] Step 120: Perform driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver;

[0062] In this application embodiment, the acquired natural driving data can be identified to determine the target driver's driving behavior under the current natural driving conditions, which is recorded as a driving behavior identifier. The driving behavior identifier includes left turn behavior, right turn behavior, acceleration behavior, deceleration behavior, no obvious driving behavior, etc.

[0063] In some embodiments, the step of performing driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver includes:

[0064] Obtain a driving behavior recognition model;

[0065] The natural driving data is input into the driving behavior recognition model for behavior recognition, and the driving behavior identifier output by the driving behavior recognition model is obtained.

[0066] In this embodiment of the application, a driving behavior recognition model can be constructed based on fuzzy reasoning, and natural driving data can be input into the driving behavior recognition model to identify the driving behavior corresponding to the natural driving data, thereby obtaining a driving behavior identifier.

[0067] In some embodiments, the driving behavior recognition model is constructed through the following steps:

[0068] Obtain several driving behavior rules;

[0069] Based on all the driving behavior rules, determine the behavior membership function corresponding to each driving behavior rule;

[0070] The driving behavior recognition model is constructed based on all the behavior membership functions.

[0071] In this embodiment of the application, taking driving behavior rules including acceleration rules, deceleration rules, left turn rules, and right turn rules as an example, the acceleration rule can specifically be longitudinal acceleration a. x Greater than 0.4 m / s 2 The deceleration rule can specifically be the longitudinal acceleration a. x Less than -0.4 m / s 2 The specific rules for left turns can be lateral acceleration a. y Greater than 0.4 m / s 2 Or the steering wheel angle θ is greater than 30°; while the specific rules for left turns can be lateral acceleration a. y Less than -0.4 m / s 2 Or the steering wheel angle θ is less than -30°.

[0072] It is understandable that after obtaining all driving behavior rules, the corresponding behavior membership function can be determined based on each driving behavior rule.

[0073] For example, the behavior membership function of the acceleration rule can be expressed as:

[0074]

[0075] Where, μ 加速 (a x ) is the membership function of the acceleration rule.

[0076] The behavioral membership function of the deceleration rule can be expressed as:

[0077]

[0078] Where, μ 减速 (a x ) is the behavioral membership function of the deceleration rule.

[0079] The behavioral membership function of the left turn rule can be expressed as:

[0080]

[0081] Where, μ 左横 (a y ) represents the membership function of lateral acceleration in the left-turn rule; μ 左角 (θ) is the membership function of the left-turn rule regarding the steering wheel angle θ; μ 左转弯 This is the membership function of the overall behavior of the left-turn rule.

[0082] The membership function of the right turn rule can be expressed as:

[0083]

[0084] Where, μ 右横 (a y ) represents the membership function of lateral acceleration in the right-turn rule; μ 右角 (θ) is the membership function of the right turn rule regarding the steering wheel angle θ; μ 右转弯 This is the membership function of the overall behavior of the right-turn rule.

[0085] It is understood that the specific values ​​related to lateral acceleration, longitudinal acceleration, and steering wheel angle in the example of this application are only optional values ​​and are set only for ease of understanding; in actual applications, the specific values ​​related to lateral acceleration, longitudinal acceleration, and steering wheel angle can be flexibly set, which will not be elaborated here.

[0086] Furthermore, after obtaining the behavior membership function of each driving behavior rule, a corresponding driving behavior recognition model can be constructed based on fuzzy reasoning based on all the obtained behavior membership functions. This driving behavior recognition model determines left-turn behavior, right-turn behavior, acceleration behavior, deceleration behavior, or no obvious driving behavior based on the input natural driving data (such as the vehicle's longitudinal acceleration, lateral acceleration, steering wheel angle, etc.). Among them, no obvious driving behavior can be the output of the driving behavior recognition model when the input natural driving data does not conform to the left-turn behavior, right-turn behavior, acceleration behavior, and deceleration behavior.

[0087] Step 130: Input the natural driving data into the driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtain the intermediate driving style information output by the driving style recognition model;

[0088] In this embodiment of the application, after determining the driving behavior of the target driver, the natural driving data of the target driver can be input into the corresponding driving style recognition model for recognition. The corresponding driving style recognition model is trained based on the training data of the corresponding driving behavior to obtain intermediate driving style information.

[0089] In some embodiments, the step of inputting the natural driving data into a driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtaining intermediate driving style information output by the driving style recognition model, includes:

[0090] Obtain several pre-trained driving style recognition models, each of which corresponds to a type of driving behavior;

[0091] Based on the driving behavior identifier, all the trained driving style recognition models are screened to obtain the target style recognition model;

[0092] The natural driving data is input into the target style recognition model for style recognition to obtain the intermediate driving style information.

[0093] In this embodiment of the application, the number of trained driving style recognition models is several. Taking four trained driving style recognition models as an example, the trained driving style recognition models can be driving style recognition models corresponding to left turn behavior, driving style recognition models corresponding to right turn behavior, acceleration style recognition models corresponding to acceleration behavior, and acceleration style recognition models corresponding to deceleration behavior.

[0094] Understandably, model selection can be based on the driving behavior indicated by the driving behavior identifier, selecting the driving style recognition model corresponding to the driving behavior, denoted as the target style recognition model. Then, natural driving data is input into the target style recognition model for style recognition, thereby obtaining the intermediate driving style information of the target driver under the current natural driving conditions.

[0095] In some embodiments, obtaining a plurality of trained driving style recognition models includes:

[0096] Obtain the original style recognition model and driving training dataset, wherein the driving training dataset includes several sample driving data, and all the sample driving data correspond to a type of driving behavior;

[0097] Features and labels are extracted from the driving training dataset to obtain a training feature set and a feature label corresponding to each target driving feature in the training feature set. The target driving features under different types of driving behaviors are not completely the same.

[0098] Based on all the aforementioned feature labels, the training feature set is input into the original style recognition model for parameter update, thereby obtaining the trained driving style recognition model.

[0099] In this embodiment, the original style recognition model can be a recurrent neural network (RNN) model, specifically a long short-term memory network (LSTM) model. Specifically, for any original style recognition model, several sample driving data points under a certain driving behavior type can be obtained. These sample driving data points can be publicly available driving data on the internet, or driving data collected with the driver's consent.

[0100] It is understandable that, since the data features corresponding to different driving styles under different driving behaviors are different, the feature extraction in this embodiment can be based on the driving behavior type corresponding to the sample driving data, extracting the data features of each sample driving data in the driving training dataset. Specifically, for acceleration and deceleration behaviors, statistical values ​​(such as mean, standard deviation, and maximum value) of vehicle speed, longitudinal acceleration, accelerator pedal opening, and brake pedal opening can be extracted from the sample driving data as feature parameters characterizing their driving style, denoted as target driving features; or, for left-turn and right-turn behaviors, statistical values ​​(mean, standard deviation, and maximum value) of vehicle speed, longitudinal acceleration, lateral acceleration, steering wheel angle, steering wheel speed, accelerator pedal opening, and brake pedal opening can be extracted from the sample driving data as feature parameters characterizing their driving style, also denoted as target driving features.

[0101] It should be noted that after extracting the target driving features for each sample of driving data, a training feature set can be constructed based on all target driving features. Furthermore, label extraction can begin by using Principal Component Analysis (PCA) to reduce the dimensionality of all target driving features, resulting in several dimensionality-reduced target driving features. Then, a clustering algorithm (such as K-means clustering) is used to cluster all the dimensionality-reduced target driving features, resulting in several clusters, each with a corresponding cluster label. Finally, based on the cluster labels, each dimensionality-reduced target driving feature within a cluster is labeled, thus obtaining the feature label corresponding to each target driving feature.

[0102] Step 140: Perform style correction on the intermediate driving style information to obtain the target driving style information of the target driver.

[0103] In this embodiment of the application, after the driving style recognition model outputs intermediate driving style information, the intermediate driving style information can be corrected to obtain the final target driving style information.

[0104] In some embodiments, the step of performing style correction on the intermediate driving style information to obtain the target driving style information of the target driver includes:

[0105] Obtain historical update vectors and historical driving style information. The historical update vector is the update vector of the previous time point adjacent to the current time point. The current time point is the time point of the intermediate driving style information. The historical driving style information is the target driving style information that is the same as the time point of the historical update vector.

[0106] In this embodiment, the intermediate driving style information time point is the predicted time point output by the driving style recognition model. The update vector can be the weighted probability of the target driving style information at previous time points, and the historical driving style information can be the target driving style information at previous time points, such as the target driving style information at the previous time point adjacent to the current time point.

[0107] Based on the historical update vector and the historical driving style information, the intermediate driving style information is updated to obtain the target driving style information.

[0108] Further, the step of updating the intermediate driving style information based on the historical update vector and the historical driving style information to obtain the target driving style information includes:

[0109] Based on the historical driving style information, the historical update vector is updated according to a time-dependent relationship to obtain the current update vector corresponding to the intermediate driving style information;

[0110] Based on the current update vector, the intermediate driving style information is corrected and updated to obtain the target driving style information.

[0111] In this embodiment, the intermediate driving style information can record the initial probability distribution of several driving styles corresponding to the target driver's natural driving data. Taking driving styles including aggressive, normal, and mild as an example, the functional expression of the intermediate driving style information can be:

[0112] P initial =[p0,p1,p2]

[0113] Among them, P initial p0 represents the initial probability distribution of the aggressive driving style; p1 represents the initial probability distribution of the normal driving style; and p2 represents the initial probability distribution of the mild driving style.

[0114] Understandably, time-dependent updates can be achieved by fusing historical information into the update vector to suppress random noise in real-time data. The current update vector has time-dependent properties, which can be achieved by weighting the probabilities of historical driving styles and historical update vectors.

[0115] Specifically, the intermediate driving style information at the initial moment is updated using the initial update vector to obtain the target driving style information at the initial moment. The initial update vector can be specifically represented as:

[0116] M initial =[p ′ 0,p ′ 1,p ′ 2]

[0117] Among them, M initial p is the initial update vector; ′ 0 represents the initial update weight for an aggressive driving style; p ′ 1 represents the initial update weight for the normal driving style; p ′ 2 is the initial update weight for the mild driving style.

[0118] It should be noted that the initial update weights for the aggressive driving style, the normal driving style, and the mild driving style in the initial update vector are set to have equal probability and their sum is 1.

[0119] The current update vector can incorporate historical information and be used to update the current intermediate driving style information, which can be specifically represented as:

[0120]

[0121] Among them, M (t) M is the current update vector, i.e., the update vector at time t; λ is the forgetting factor, λ∈(0,1); (t-1) This is the historical update vector, that is, the update vector at time point t-1; This refers to historical driving style information, specifically the target driving style information at time t-1.

[0122] It should be noted that the correction update can use the current update vector as the update weight parameter to adjust the intermediate driving style, thereby obtaining the target driving style information. This target driving style information can include the target probability distribution of aggressive driving style, the target probability distribution of normal driving style, and the target probability distribution of mild driving style.

[0123] For example, the target driving style information at time point t can be specifically represented as:

[0124]

[0125] Among them, P correct The target probability distribution of target driving style information at time point t; The target probability for an aggressive driving style; The target probability for a normal driving style; The target probability for a mild driving style.

[0126] For example, for time point t, the target probability distribution of an aggressive driving style can be expressed as:

[0127]

[0128] in, The update weights for aggressive driving styles at time point t.

[0129] It is worth mentioning that the target probability distributions for normal driving style and mild driving style are similar to those for the aforementioned aggressive driving style and can be easily deduced, so this application will not elaborate further here. Furthermore, after obtaining the target probability distributions for all types of driving styles, these distributions can be integrated to obtain target driving style information. The expression of this target driving style information is similar to that of the aforementioned intermediate driving style information and can be easily deduced.

[0130] It should be noted that in practical applications, after obtaining the target driving style information, the driving style type with the highest probability distribution is often determined as the final driving style recognition result. Furthermore, the driving style types in the examples of this application are for illustrative purposes only and are not intended to limit the application; the actual driving style types and their number can be flexibly set according to the actual situation.

[0131] A driving style recognition system according to an embodiment of this application is described in detail below with reference to the accompanying drawings.

[0132] Reference Figure 2 The driving style recognition system proposed in this application includes:

[0133] The first processing unit 101 is used to acquire the natural driving data of the target driver;

[0134] The second processing unit 102 is used to perform driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver.

[0135] The third processing unit 103 is used to input the natural driving data into the driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtain the intermediate driving style information output by the driving style recognition model.

[0136] The fourth processing unit 104 is used to perform style correction on the intermediate driving style information to obtain the target driving style information of the target driver.

[0137] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0138] Reference Figure 3 This application also provides an electronic device, including:

[0139] At least one processor 201;

[0140] At least one memory 202 is used to store at least one program;

[0141] When the at least one program is executed by the at least one processor 201, the at least one processor 201 implements the above-described method embodiments.

[0142] Similarly, it can be understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0143] This application also provides a computer-readable storage medium storing a program executable by a processor 201, which, when executed by the processor 201, is used to implement the above-described method embodiments.

[0144] Similarly, the content of the above method embodiments is applicable to the present computer-readable storage medium embodiments. The specific functions implemented by the present computer-readable storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0145] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0146] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods.

[0147] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0148] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional technology for an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0149] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0151] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0152] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0153] In the foregoing description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0154] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

[0155] The above is a detailed description of the preferred embodiments of this application, but this application is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A driving style recognition method, characterized in that, include: Acquire the target driver's natural driving data; The natural driving data is used to identify driving behavior to obtain the driving behavior identifier of the target driver; The natural driving data is input into the driving style recognition model corresponding to the driving behavior identifier for style recognition, and intermediate driving style information output by the driving style recognition model is obtained. The intermediate driving style information is corrected to obtain the target driving style information of the target driver. The step of inputting the natural driving data into a driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtaining intermediate driving style information output by the driving style recognition model, includes: Obtain several pre-trained driving style recognition models, each of which corresponds to a type of driving behavior; Based on the driving behavior identifier, all the trained driving style recognition models are screened to obtain the target style recognition model; The natural driving data is input into the target style recognition model for style recognition to obtain the intermediate driving style information; The step of performing style correction on the intermediate driving style information to obtain the target driving style information of the target driver includes: Obtain historical update vectors and historical driving style information. The historical update vector is the update vector of the previous time point adjacent to the current time point. The current time point is the time point of the intermediate driving style information. The historical driving style information is the target driving style information that is the same as the time point of the historical update vector. Based on the historical update vector and the historical driving style information, the intermediate driving style information is updated to obtain the target driving style information.

2. The method according to claim 1, characterized in that, The step of performing driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver includes: Obtain a driving behavior recognition model; The natural driving data is input into the driving behavior recognition model for behavior recognition, and the driving behavior identifier output by the driving behavior recognition model is obtained.

3. The method according to claim 2, characterized in that, The driving behavior recognition model is constructed through the following steps: Obtain several driving behavior rules; Based on all the driving behavior rules, determine the behavior membership function corresponding to each driving behavior rule; The driving behavior recognition model is constructed based on all the behavior membership functions.

4. The method according to claim 1, characterized in that, The acquisition of several trained driving style recognition models includes: Obtain the original style recognition model and driving training dataset, wherein the driving training dataset includes several sample driving data, and all the sample driving data correspond to a type of driving behavior; Features and labels are extracted from the driving training dataset to obtain a training feature set and a feature label corresponding to each target driving feature in the training feature set. The target driving features under different types of driving behaviors are not completely the same. Based on all the aforementioned feature labels, the training feature set is input into the original style recognition model for parameter update, thereby obtaining the trained driving style recognition model.

5. The method according to claim 1, characterized in that, The step of updating the intermediate driving style information based on the historical update vector and the historical driving style information to obtain the target driving style information includes: Based on the historical driving style information, the historical update vector is updated according to a time-dependent relationship to obtain the current update vector corresponding to the intermediate driving style information; Based on the current update vector, the intermediate driving style information is corrected and updated to obtain the target driving style information.

6. A driving style recognition system, characterized in that, include: The first processing unit is used to acquire the target driver's natural driving data; The second processing unit is used to perform driving behavior recognition on the natural driving data to obtain the driving behavior identifier of the target driver; The third processing unit is used to input the natural driving data into the driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtain the intermediate driving style information output by the driving style recognition model. The fourth processing unit is used to perform style correction on the intermediate driving style information to obtain the target driving style information of the target driver; The step of inputting the natural driving data into a driving style recognition model corresponding to the driving behavior identifier for style recognition, and obtaining intermediate driving style information output by the driving style recognition model, includes: Obtain several pre-trained driving style recognition models, each of which corresponds to a type of driving behavior; Based on the driving behavior identifier, all the trained driving style recognition models are screened to obtain the target style recognition model; The natural driving data is input into the target style recognition model for style recognition to obtain the intermediate driving style information; The step of performing style correction on the intermediate driving style information to obtain the target driving style information of the target driver includes: Obtain historical update vectors and historical driving style information. The historical update vector is the update vector of the previous time point adjacent to the current time point. The current time point is the time point of the intermediate driving style information. The historical driving style information is the target driving style information that is the same as the time point of the historical update vector. Based on the historical update vector and the historical driving style information, the intermediate driving style information is updated to obtain the target driving style information.

7. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method as described in any one of claims 1-5.

8. A computer-readable storage medium storing a processor-executable program, characterized in that, The processor-executable program, when executed by the processor, is used to implement the method as described in any one of claims 1-5.

Images