Driving behavior type determination method and device, electronic equipment, vehicle and chip

Abstract

The invention provides a driving behavior type determination method and device, electronic equipment, a vehicle and a chip, and relates to the field of machine learning. The method comprises the following steps: acquiring driving behavior data of a plurality of vehicles, wherein the driving behavior data comprises vehicle data and environment data; performing cleaning and feature extraction on the driving behavior data, and determining a plurality of feature samples corresponding to the plurality of target vehicles; clustering the plurality of feature samples based on one or more feature values in the plurality of feature values to obtain a plurality of category sets; and obtaining evaluation scores of the plurality of category sets to determine the driving behavior type of each category set. According to the technical scheme provided by the invention, the portrait features of the user can be extracted and classified through a large amount of unsupervised user driving behavior data, and the evaluation score of the data sample in each category is obtained, so that the driving behavior type of each category of portrait is determined, the efficiency can be improved, and a new type of the user portrait can be obtained.

Description

Technical Field

[0001] The present disclosure relates to the field of machine learning, and in particular to a method and device for determining a driving behavior type, an electronic device, a vehicle, and a chip. Background Art

[0002] A user driving profile refers to the characteristics of a user's driving experience. This can include operational behavior characteristics, such as driving style, driving risk, and driving experience; and travel behavior characteristics, such as those associated with office workers, self-driving families, and ride-hailing services. In related technologies, algorithms or models can be theoretically established based on experience. For example, acceleration can be used to directly characterize the intensity of a user's driving. As a rule of thumb, greater acceleration indicates more aggressive driving, while smaller acceleration indicates more gentle driving. However, this approach requires a high level of experience and can lead to algorithmic limitations. Alternatively, expert evaluation or scoring can be used to label user sets. Supervised machine learning can then be performed based on the user dataset and labels to establish an algorithmic model between the user set characteristics and the labels, thereby obtaining correlations between the user set and the labels. However, this approach is inefficient and requires scoring or evaluating the behavior of a large number of users. Furthermore, the expert evaluation labels can easily limit the types of driving behaviors captured. Summary of the Invention

[0003] This disclosure provides a method and device for determining driving behavior types, as well as electronic equipment, a vehicle, and a chip. These methods address the limitations and low development efficiency of user driving profile model algorithms in related technologies. This technical solution, using large amounts of unsupervised user driving behavior big data, extracts and classifies user profile features, obtains evaluation scores for each profile feature sample, and determines the driving behavior type for each profile.

[0004] A first aspect embodiment of the present disclosure proposes a method for determining a driving behavior type, the method comprising: obtaining driving behavior data of a plurality of vehicles, the driving behavior data including vehicle data and environmental data, the vehicle data reflecting the travel data of the plurality of vehicles, and the environmental data reflecting the external environment data of the plurality of vehicles; cleaning and feature extraction of the driving behavior data to determine a plurality of feature samples corresponding to a plurality of target vehicles, each feature sample including a plurality of feature values, and the plurality of vehicles including a plurality of target vehicles; clustering the plurality of feature samples based on one or more feature values ​​among the plurality of feature values ​​to obtain a plurality of category collections, the plurality of category collections respectively including one or more feature samples among the plurality of feature samples; obtaining evaluation scores of the plurality of category collections to determine the driving behavior type of each category collection, the plurality of category collections and the driving behavior type of each category collection being data for training a first model, and the first model being used to evaluate the driving behavior type of the vehicle.

[0005] In one embodiment of the present disclosure, cleaning and feature extraction are performed on driving behavior data to determine multiple feature samples corresponding to multiple target vehicles, including: eliminating invalid data from the driving behavior data of the multiple vehicles to obtain valid driving behavior data of the multiple vehicles; screening the valid driving behavior data of the multiple vehicles to obtain training data, where the training data includes the valid driving behavior data of the multiple target vehicles; and feature extraction is performed on the training data to determine multiple feature samples corresponding to the multiple target vehicles.

[0006] In one embodiment of the present disclosure, feature extraction is performed on training data to determine multiple feature samples corresponding to multiple target vehicles, including: for each target vehicle in the training data, one or more categories of data are selected for combined calculation to determine the feature samples of each target vehicle.

[0007] In one embodiment of the present disclosure, obtaining evaluation scores of multiple category collections to determine the driving behavior type of each category collection includes: determining a data sample from each category collection; and obtaining evaluation scores of the data samples in each category collection to determine the driving behavior type of each category collection.

[0008] In one embodiment of the present disclosure, the characteristic value includes at least one of a first characteristic value, a second characteristic value, a third characteristic value, and a fourth characteristic value, wherein the first characteristic value is determined based on the speed of the target vehicle within a preset mileage, the second characteristic value is determined based on the vehicle speed and the steering wheel angle, the third characteristic value is determined based on the acceleration and the accelerator pedal opening, and the fourth characteristic value is determined based on the vehicle speed and the lateral acceleration and longitudinal acceleration.

[0009] A second embodiment of the present disclosure provides a driving behavior type determination device, which includes: an acquisition module, an extraction module, and a training module.

[0010] The acquisition module is used to obtain driving behavior data of multiple vehicles, the driving behavior data including vehicle data and environmental data, the vehicle data reflects the driving data of multiple vehicles, and the environmental data reflects the external environment data of multiple vehicles; the extraction module is used to clean and extract features from the driving behavior data, and determine multiple feature samples corresponding to multiple target vehicles respectively, each feature sample including multiple feature values, and the multiple vehicles include the multiple target vehicles.

[0011] The training module is used to perform clustering processing on multiple feature samples based on one or more feature values ​​among the multiple feature values ​​to obtain multiple category sets, and the multiple category sets respectively include one or more feature samples among the multiple feature samples.

[0012] The training module is used to obtain evaluation scores of multiple category collections to determine the driving behavior type of each category collection. The multiple category collections and the driving behavior type of each category collection are used to train the first model, and the first model is used to evaluate the driving behavior type of the vehicle.

[0013] The third aspect embodiment of the present disclosure proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any one of the methods in the first aspect embodiment of the present disclosure.

[0014] A fourth embodiment of the present disclosure provides a vehicle, comprising the driving behavior type determination device described in the second embodiment of the present disclosure or the electronic device described in the third embodiment of the present disclosure.

[0015] The fifth aspect embodiment of the present disclosure proposes a non-transitory computer-readable storage medium storing computer instructions, characterized in that the computer instructions are used to enable a computer to execute the method in the first aspect embodiment of the present disclosure.

[0016] The sixth aspect of the present disclosure provides a computer program product, characterized in that it includes a computer program, and when the computer program is executed by a processor, it implements any one of the methods in the first aspect of the present disclosure.

[0017] The seventh aspect embodiment of the present disclosure proposes a chip, characterized in that it includes one or more interface circuits and one or more processors; the interface circuit is used to receive signals from the memory of the electronic device and send signals to the processor, the signals including computer instructions stored in the memory, and when the processor executes the computer instructions, the electronic device executes any one of the methods in the first aspect embodiment of the present disclosure.

[0018] In summary, according to the driving behavior type determination method proposed in the present disclosure, driving behavior data of multiple vehicles is obtained, the driving behavior data including vehicle data and environmental data; the driving behavior data is cleaned and features are extracted to determine multiple feature samples corresponding to multiple target vehicles; the multiple feature samples are clustered to obtain multiple category collections, each of which includes one or more feature samples from the multiple feature samples; an evaluation score is obtained for each category collection to determine the driving behavior type of each category collection, and the multiple category collections and the driving behavior type of each category collection are used to train a first model, which is used to evaluate the driving behavior type of the vehicle. The technical solution provided by the present disclosure can extract and classify user profile features from a large amount of unsupervised user driving behavior big data, obtain sample evaluation scores for each category of portrait features, and determine the driving behavior type of each category of portrait.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0021] Figure 1 A flowchart of a method for determining a driving behavior type proposed in an embodiment of the present disclosure;

[0022] Figure 2 A flowchart of determining multiple feature samples corresponding to multiple target vehicles according to an embodiment of the present disclosure;

[0023] Figure 3 A flowchart for determining multiple feature samples proposed in an embodiment of the present disclosure;

[0024] Figure 4 A flowchart for determining the driving behavior type for each category set proposed in an embodiment of the present disclosure;

[0025] Figure 5 A flowchart of a method for determining a driving behavior type provided in an embodiment of the present disclosure;

[0026] Figure 6 A schematic diagram of the structure of a driving behavior type determination device proposed in an embodiment of the present disclosure;

[0027] Figure 7 The figure is a block diagram of an electronic device for implementing the driving behavior determination method disclosed herein, according to an exemplary embodiment. DETAILED DESCRIPTION

[0028] The following describes in detail embodiments of the present disclosure, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout identify the same or similar components or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0029] First, let’s introduce the professional terms involved in this disclosure:

[0030] Supervised machine learning: refers to the data used for machine learning that has been labeled during training.

[0031] Unsupervised machine learning: refers to the data used for machine learning that does not have corresponding labels during training.

[0032] Generally speaking, the essential conceptual difference between supervised and unsupervised learning is whether there are labels when training the machine learning model. If the dataset has labels but is not used in training, but only used in the verification stage of the model, it is also considered unsupervised machine learning.

[0033] User driving profiles refer to the characteristics of a user's driving experience, including operational and travel behavior. The specific type of profile required is determined by the modeler's needs. When creating user driving profiles, user datasets are labeled based on experience or expert evaluation. This allows for the establishment of an algorithmic model linking user set labels and the correlation between user sets and labels. However, these methods have low development efficiency and limitations.

[0034] The driving behavior determination method provided in this application is described in detail below with reference to the accompanying drawings.

[0035] Figure 1 FIG. 1 is a flow chart of a method for determining driving behavior according to an embodiment of the present disclosure. Figure 1 In the illustrated embodiment, the method includes:

[0036] Step 101: Acquire driving behavior data of multiple vehicles.

[0037] In an embodiment of the present disclosure, the driving behavior data includes vehicle data and environment data.

[0038] In an embodiment of the present disclosure, the vehicle data reflects the driving data of a plurality of vehicles, and the environment data reflects the external environment data of the plurality of vehicles.

[0039] In an embodiment of the present disclosure, vehicle data may include at least one of vehicle model information, steering wheel angle, throttle opening, brake pedal depth, vehicle speed, longitudinal acceleration, longitudinal deceleration, lateral acceleration, turn signal signal, distance between front and rear vehicles, audio working condition, and Bluetooth connection status.

[0040] In the embodiments of the present disclosure, the vehicle data may also be other data reflecting the driving data of the vehicle, which is not limited by the present disclosure.

[0041] In an embodiment of the present disclosure, the environmental data may include at least one of the grade of the vehicle driving road, the weather temperature of the vehicle driving, the slope of the vehicle driving road, the altitude of the vehicle driving road, and the number of traffic lights along the vehicle driving route.

[0042] In the embodiments of the present disclosure, the environmental data may also be other data reflecting the external environment of the vehicle, which is not limited by the present disclosure.

[0043] In an embodiment of the present disclosure, the driving behavior data of multiple vehicles may be obtained from big data uploaded to the cloud by the vehicles during operation.

[0044] In an embodiment of the present disclosure, the driving behavior data of multiple vehicles may be acquired directly through vehicle-side equipment.

[0045] In an embodiment of the present disclosure, obtaining vehicle data or environmental data of multiple vehicles may be determined by modeling needs, and the type of data required for modeling is determined by the demand for driving profiling.

[0046] For example, to study the behavioral characteristics of users driving a vehicle, such as driving style, driving risk, or driving experience, it is necessary to obtain data on the vehicle's acceleration, speed, throttle and brake operations, etc.

[0047] For example, to study the driving patterns of users, such as the relationship between driving and weather, it is necessary to obtain the weather and temperature conditions along the vehicle's route.

[0048] In an embodiment of the present disclosure, driving behavior data of multiple vehicles are obtained for analyzing and extracting behavioral characteristics of vehicles driven by different users to determine user profiles, that is, the user's driving behavior type.

[0049] Step 102 : Clean and extract features from the driving behavior data to determine multiple feature samples corresponding to multiple target vehicles.

[0050] In an embodiment of the present disclosure, each feature sample includes multiple feature values, and multiple vehicles include multiple target vehicles, wherein the target vehicle is a vehicle that meets the requirements and is screened out from multiple vehicles based on modeling requirements. Each vehicle can obtain multiple feature values, thereby determining multiple feature samples corresponding to the multiple target vehicles.

[0051] In an embodiment of the present disclosure, cleaning and feature extraction of driving behavior data may be performed by cleaning driving behavior data of multiple vehicles to obtain cleaned driving behavior data, and then feature extraction is performed on these data to determine multiple feature values.

[0052] In an embodiment of the present disclosure, cleaning may be to remove erroneous data, duplicate data, or abnormal data from the driving behavior data.

[0053] In an embodiment of the present disclosure, cleaning may be data screening in driving behavior data according to modeling requirements.

[0054] In some embodiments, cleaning can include removing erroneous, duplicate, or abnormal driving behavior data and filtering the data based on modeling requirements to determine driving behavior data for target vehicles that meet the modeling requirements. Modeling requirements can include specific vehicle models, specific environments, specific roads, and the like, which are not limited in this disclosure. The target vehicle can be one or more, i.e., determining driving behavior data for multiple vehicles that meet the modeling requirements.

[0055] In an embodiment of the present disclosure, feature extraction may be performing feature analysis on the cleaned driving behavior data, and using one or more types of data in the cleaned driving behavior data for calculation to obtain multiple feature values ​​of the target vehicle.

[0056] For example, one or more categories of data in the cleaned driving behavior data are used for calculation to obtain characteristic values ​​s1, s2, s3, s4, s5, and s6. Then, each vehicle that meets the modeling requirements can obtain a feature sample, that is, the feature sample x1 of vehicle A driven by user A, the feature sample x2 of vehicle B driven by user B, and the feature sample xn of vehicle n driven by user n, where each feature sample includes the above 6 characteristic values.

[0057] In the embodiment of the present disclosure, the characteristic value may reflect the driving behavior characteristics of the user driving the vehicle, and the driving behavior characteristics reflected by each characteristic value may be the same or different.

[0058] In some embodiments, the characteristic values ​​are used to classify the driving behavior data of multiple vehicles to obtain multiple categories of vehicles. By analyzing the characteristic values ​​of the vehicles in each category, the driving behavior profile of each category can be determined.

[0059] In an embodiment of the present disclosure, multiple feature samples corresponding to multiple target vehicles can be determined by calculating multiple feature values ​​on the vehicle side and uploading the feature values ​​to the cloud without uploading the original data, thereby effectively reducing the cloud computing load and data transmission volume.

[0060] In an embodiment of the present disclosure, determining multiple feature values ​​of a target vehicle may be that after the vehicle uploads the driving behavior data to the cloud, the multiple feature values ​​are calculated in the cloud and used for subsequent clustering.

[0061] In the above embodiment, by cleaning and extracting features from the driving behavior data of multiple vehicles, different feature values ​​can be obtained based on engineering experience and non-engineering experience to reflect different driving behavior characteristics, thereby classifying the multiple vehicles.

[0062] Step 103 : performing clustering processing on the multiple feature samples based on one or more feature values ​​among the multiple feature values ​​to obtain multiple category sets.

[0063] In an embodiment of the present disclosure, the plurality of category collections respectively include one or more feature samples among the plurality of feature samples.

[0064] In an embodiment of the present disclosure, clustering of multiple feature samples may be performed based on one or more feature values ​​of a target vehicle. The multiple target vehicles may be multiple vehicles that meet the modeling requirements, and the feature values ​​used to cluster the feature samples of each vehicle should be the same or the same feature values.

[0065] In some embodiments, clustering can adopt any one of the following algorithms: k-means clustering algorithm (k-means clustering algorithm, k-means algorithm), binary k-means clustering algorithm (Binary-k-means clustering algorithm, bi-kmeans algorithm), density-based clustering method with noise (Density-Based Spatial Clustering of Applications with Noise, DBSCAN algorithm), and density-based clustering algorithm (ordering points to identify the clustering structure, OPTICS algorithm), which is not limited by the present disclosure.

[0066] For example, in step 102, six vehicle feature values ​​are determined. Clustering by each vehicle's feature value s1 can distinguish n vehicles into different small collections. Alternatively, clustering by combining each vehicle's feature values ​​s2 and s3 can distinguish n vehicles into different collections. Alternatively, clustering by all vehicle feature values ​​can distinguish n vehicles into different collections. This outputs different category collections F = {y1, y2, y3, ..., yn}. Here, y represents different categories, and each category contains different samples. For example, y1 includes vehicles driven by three users, x1, x3, and x9, or these three users belong to the same category. y2 includes two users, x2 and x7, or these two users belong to the same category. y3 includes five users, x4, x5, x6, x8, and x11, and so on.

[0067] Step 104 : Obtain evaluation scores of multiple category sets to determine the driving behavior type of each category set.

[0068] In an embodiment of the present disclosure, obtaining evaluation scores of a plurality of category collections may obtain expert evaluation or scoring results.

[0069] In an embodiment of the present disclosure, obtaining an evaluation score may be a result of obtaining one or more feature values ​​of a specific number of vehicles in each category set and performing expert evaluation or scoring, wherein the specific number may be the number of determined evaluation samples.

[0070] For example, the evaluation scores of the six feature values ​​of x1 and x3 in the y1 set are obtained to obtain the driving behavior type of the y1 set.

[0071] In an embodiment of the present disclosure, obtaining evaluation scores of multiple category collections may be obtaining evaluation scores of all data in each category collection.

[0072] For example, by performing expert evaluation or scoring on all feature values ​​of vehicles driven by five users in the y3 collection, the evaluation scores are obtained to obtain the driving behavior type of the y3 collection.

[0073] In the above embodiment, by clustering multiple vehicles based on their feature values ​​and then obtaining evaluation scores to determine the driving behavior type for each category, the evaluated objects are evenly distributed across different categories, thereby avoiding bias in the modeling results. Through unsupervised machine learning and supervised evaluation, each category collection is evaluated based on the same features to obtain a driving behavior type that is more consistent with that category collection. The multiple category collections and the driving behavior types for each category collection are used as data to train a first model, which is used to evaluate the driving behavior types of vehicles.

[0074] In summary, according to the driving behavior type determination method proposed in the present disclosure, based on unsupervised machine learning, data collection, cleaning and feature extraction are performed on multiple vehicles to classify multiple target vehicles. By obtaining the evaluation score of each category collection, the driving behavior type of each category can be determined.

[0075] Figure 2 This is a flow chart of determining multiple feature samples corresponding to multiple target vehicles according to the embodiment of the present disclosure. Figure 1 The embodiment shown, Figure 2 right Figure 1 Step 102 is further described based on Figure 2 The embodiment shown, Figure 2 The steps include:

[0076] Step 201 : Eliminate invalid data from the driving behavior data of multiple vehicles to obtain valid driving behavior data of multiple vehicles.

[0077] In the embodiment of the present disclosure, eliminating invalid data may be eliminating invalid data generated due to signal transmission errors or invalid data generated due to a short travel distance.

[0078] In the embodiment of the present disclosure, invalid data may be data loss caused by signal influence, or invalid data caused by GPS position offset.

[0079] In the above embodiment, invalid data is eliminated to obtain valid driving behavior data of multiple vehicles, which can make the subsequent determination of the driving behavior type more accurate.

[0080] Step 202 : Filter the valid driving behavior data of multiple vehicles to obtain training data.

[0081] In an embodiment of the present disclosure, the training data includes valid driving behavior data of a plurality of target vehicles.

[0082] In an embodiment of the present disclosure, the target vehicle may be a plurality of vehicles that meet the requirements obtained by screening according to modeling requirements. The modeling requirements may be a specific vehicle model, a specific road, or specific weather.

[0083] In an embodiment of the present disclosure, valid driving behavior data of multiple vehicles is filtered based on modeling requirements, and data for a specific vehicle model, a specific road, or specific weather conditions can be obtained. The filtered data includes valid driving behavior data of target vehicles that meet the modeling requirements, and the valid driving behavior data includes vehicle data and / or environmental data.

[0084] In some embodiments, the modeling requirement may be a type of requirement added separately for research purposes. According to this type of requirement, specific data of the user driving the vehicle may need to be uploaded and stored for use in the driving behavior type determination method.

[0085] Step 203 : extracting features from the training data to determine a plurality of feature samples corresponding to the plurality of target vehicles.

[0086] In an embodiment of the present disclosure, feature extraction of training data may be performed by calculating one or more types of data in the effective driving behavior data of the target vehicle to obtain multiple feature values ​​of the target vehicle.

[0087] In the embodiments of the present disclosure, feature extraction may be feature analysis and extraction based on engineering experience, or may be feature analysis and extraction not based on engineering experience.

[0088] In the above embodiment, by cleaning and extracting features from the driving behavior data of multiple vehicles, multiple feature values ​​of the target vehicle are obtained, thereby determining multiple feature samples corresponding to the multiple target vehicles, which are used to classify the feature samples.

[0089] Figure 3 This is a flow chart for determining multiple feature samples proposed in the embodiment of the present disclosure. Figure 1 、 Figure 2 The embodiment shown, Figure 3 Yes Figure 2 The specific description of step 203 is based on Figure 3 The embodiment shown includes the following steps:

[0090] Step 301 : For the effective driving behavior data of each target vehicle in the training data, one or more types of data are selected for combined calculation to determine the feature samples of each target vehicle.

[0091] In an embodiment of the present disclosure, each feature sample of a target vehicle includes multiple feature values.

[0092] In an embodiment of the present disclosure, the eigenvalue may be at least one of a first eigenvalue, a second eigenvalue, a third eigenvalue, and a fourth eigenvalue.

[0093] In an embodiment of the present disclosure, the combined calculation may be to determine the first characteristic value based on the speed of the target vehicle within a preset mileage.

[0094] In some embodiments, the first characteristic value can be the maximum acceleration, maximum deceleration, maximum acceleration change rate, top 10% percentile acceleration, etc. obtained based on engineering experience. The first characteristic value can also be other values ​​calculated by vehicle speed, which is not limited by this disclosure.

[0095] In some embodiments, the number of first eigenvalues ​​is not limited, and it may refer to any one of the above-mentioned types, or may refer to all of the above-mentioned types of eigenvalues.

[0096] For example, based on the 10,000-kilometer driving data of vehicle A driven by user A, taking a frequency of 1 Hz as an example, that is, 1 data point per second, hundreds of thousands of data points of vehicle A can be obtained. Through the vehicle speed at these 100,000 data points, the maximum acceleration or maximum deceleration, or the maximum acceleration change rate, or the top 10% acceleration can be calculated.

[0097] In an embodiment of the present disclosure, the combined calculation may be based on the vehicle speed and the steering wheel angle to determine the second characteristic value.

[0098] In some embodiments, the second characteristic value may be a value that is not obtained based on engineering experience and is used to reflect the driving behavior of the vehicle.

[0099] For example, the vehicle speed and the steering wheel angle at the same moment are multiplied together as the second characteristic value, reflecting the behavioral characteristics of the user driving the vehicle.

[0100] In some embodiments, there may be multiple second feature values, and each vehicle may determine multiple second feature values ​​based on the vehicle speed and the steering wheel angle.

[0101] In some embodiments, the combined calculation may be based on the acceleration and the accelerator pedal opening to determine the third characteristic value.

[0102] In some embodiments, the third characteristic value may be a value that is not obtained based on engineering experience and is used to reflect the user's driving behavior.

[0103] For example, the acceleration is calculated by the vehicle speed, and its rate of change is obtained by taking the derivative of the acceleration. The rate of change is also obtained by taking the derivative of the accelerator pedal opening at the same time. The average of the two is taken as the third eigenvalue to reflect the behavioral characteristics of the user driving the vehicle.

[0104] In some embodiments, there may be multiple third characteristic values, and multiple third characteristic values ​​may be determined for each vehicle.

[0105] In an embodiment of the present disclosure, the combined calculation may be based on the vehicle speed, the lateral acceleration, and the longitudinal acceleration to determine the fourth eigenvalue.

[0106] In some embodiments, the fourth characteristic value may be a value that is not obtained based on engineering experience and is used to reflect the user's driving behavior.

[0107] For example, the vehicle speed and the lateral acceleration are calculated to obtain a value a, the vehicle speed and the longitudinal acceleration are calculated to obtain a value b, and a+2*b is calculated to determine the fourth eigenvalue.

[0108] In some embodiments, there may be multiple fourth characteristic values, and multiple fourth characteristic values ​​may be determined for each vehicle.

[0109] In the above embodiment, by combining and calculating one or more types of valid driving behavior data of the target vehicle in the training data, multiple feature values ​​of the target vehicle can be determined, including but not limited to feature values ​​determined based on engineering experience and feature values ​​not determined based on engineering experience. This allows for the addition of new feature value types and the extraction of new patterns or results.

[0110] Figure 4 A flowchart for determining the driving behavior type for each category set proposed in the embodiment of the present disclosure. Figure 1 、 Figure 2 、 Figure 3 The embodiment shown, Figure 4 right Figure 1 Step 104 is further described as follows: Figure 4 As shown, the following steps are included:

[0111] Step 401: determine data samples from each category collection.

[0112] In an embodiment of the present disclosure, determining the data samples may be determining a preset number of vehicle samples from a plurality of target vehicles in each category collection, and obtaining feature samples of these vehicles.

[0113] In some embodiments, the determined data samples may be feature samples of a preset number of vehicles in each category collection, or may be feature samples of all vehicles in a certain category collection.

[0114] For example, the category set F = {y1, y2, y3}, category y3 includes 5 users, x4, x5, x6, x8, and x11. Two users are taken from these 5 users as samples, that is, x4 and x8 can be taken as user samples. x4 includes 6 feature values ​​of user D, and x8 includes 6 feature values ​​of user H.

[0115] For example, category y2 includes two users, x2 and x7. All users selected from y2 are selected as samples, and six feature values ​​of user B and user G are obtained.

[0116] Step 402: Obtain evaluation scores of data samples in each category collection and determine the driving behavior type of each category collection.

[0117] In an embodiment of the present disclosure, obtaining the evaluation score of a data sample may be obtaining the evaluation score of a feature value in the data sample, obtaining the evaluation score of one feature value, obtaining the evaluation scores of multiple feature values, or obtaining the evaluation scores of all feature values ​​to obtain the driving behavior type of each category set.

[0118] For example, among the six feature values ​​of user D and user H in the category collection of y3, the first two feature values ​​are evaluated to obtain their evaluation scores to obtain the driving behavior type of y3. For example, by evaluating the maximum acceleration and maximum deceleration, the evaluation score is obtained, and it is obtained that the driving behavior type of y3 is a mild driving type.

[0119] In the embodiments of the present disclosure, obtaining an evaluation score for a data sample in each category collection can involve evaluating all feature values ​​of vehicles driven by all users in the category collection, obtaining the evaluation scores, and then obtaining the driving behavior type for the category collection. By obtaining the evaluation scores for all data from all users, new categories can be derived.

[0120] For example, all feature values ​​of user B and user G in the category collection of y2 are evaluated to obtain evaluation scores. The evaluation score of the first feature value is used to determine it as a long-distance travel category, while the evaluation score of the second feature value is used to determine it as a short-distance travel category. Then, this category collection is neither a long-distance travel nor a short-distance travel. By obtaining the evaluation scores of all feature values, it can be concluded that this category collection belongs to the weekend travel category.

[0121] In the above embodiment, by determining the data samples and obtaining the evaluation scores of the data samples in each category collection, the evaluation workload can be reduced and efficiency can be improved. At the same time, obtaining the evaluation scores of all data in a certain category collection can also obtain new driving behavior types.

[0122] In summary, the method for determining the driving behavior type proposed in the present disclosure obtains driving behavior data of multiple vehicles; cleans and extracts features from the driving behavior data to determine multiple feature samples corresponding to multiple target vehicles; clusters the multiple feature samples based on one or more feature values ​​among the multiple feature values ​​to obtain multiple category collections, each of which includes one or more feature samples among the multiple feature samples; obtains evaluation scores for the multiple category collections to determine the driving behavior type of each category collection. User profiling is performed based on unsupervised machine learning to obtain a category collection, and then the evaluation score of each category collection is obtained through supervised learning to determine the driving behavior type of each category. This can not only improve efficiency but also obtain new types of user profiles.

[0123] Figure 5A flowchart of a method for determining a driving behavior type provided by an embodiment of the present disclosure is shown in FIG. Figure 5 As shown, the method includes the following steps:

[0124] Step 1: Collect a large amount of user driving behavior data.

[0125] It can generally be obtained from the big data uploaded to the cloud when the vehicle is running. The signals needed for modeling are determined by the demand for driving profiling.

[0126] For example, to study the behavioral characteristics of users driving vehicles (driving style, driving risks, driving experience, etc.), it is generally necessary to obtain data on acceleration, vehicle speed, throttle and brake operations, etc.

[0127] For example, it can be the vehicle's own signals (vehicle model information, steering wheel angle, throttle opening, brake pedal depth, vehicle speed, longitudinal acceleration, deceleration, lateral acceleration, turn signal signals, smart driving sensor-related signals, such as the distance between the front and rear vehicles, the working status of the in-car entertainment facilities, etc.), as well as the vehicle's external environment signals obtained through the network and sensors (the grade of the vehicle's road, the weather and temperature conditions of the vehicle's driving, the number of traffic lights in the journey, the slope of the road, the altitude of the road, etc.).

[0128] Optionally, driving behavior data of multiple vehicles is acquired, where the driving behavior data includes vehicle data and environment data.

[0129] Step 2: Data cleaning.

[0130] Poor-quality data is eliminated, such as invalid data caused by signal transmission errors or short trips. At the same time, data is filtered according to modeling needs, such as user data for specific models, user data on specific roads, or road data under specific weather conditions.

[0131] Optionally, cleaning the data may be cleaning the driving behavior data, including eliminating invalid data and screening the data according to the first modeling requirement to obtain training data that meets the first modeling requirement.

[0132] Step 3: Feature analysis.

[0133] Perform feature analysis and extraction on user data to obtain user feature samples.

[0134] For example, if user A has 10,000 kilometers of driving data and the data frequency is 1 Hz, that is, one data point per second, then user A's acceleration data corresponds to hundreds of thousands or even millions of data points, and acceleration features need to be extracted for model training.

[0135] This requires both engineering and non-engineering experience. For example, engineering experience involves extracting features such as maximum acceleration, maximum deceleration, maximum acceleration change rate, and the top 10% acceleration, all of which represent corresponding physical meanings. Non-engineering experience involves extracting features such as the product of vehicle speed and steering wheel angle, the average of the derivative of acceleration and the derivative of accelerator pedal opening, and the sum of the product of vehicle speed and lateral acceleration and twice the product of vehicle speed and longitudinal acceleration.

[0136] Optionally, performing feature analysis on the data may be performing feature extraction on the training data to determine a plurality of feature values ​​of the first vehicle.

[0137] Step 4: Obtain a collection of user samples.

[0138] Through step 3, we obtain feature samples x1 for user A and x2 for user B, where x1 may include multiple features extracted above, such as maximum acceleration and maximum deceleration. The set of samples for all users is D = {x1, x2, x3, …, xn}, where each x variable represents a feature sample for a user.

[0139] Step 5: Perform machine learning cluster analysis to obtain different category collections.

[0140] Performing cluster analysis on the above user sample collection D can divide the user samples into different small collections. Input the user sample collection D, that is, the feature sample x of each user, and output different category collections F = {y1, y2, y3, ..., yn}, where y represents different categories. Each category contains different user samples x n For example, y1 = {x1, x3, x9}, that is, category y1 includes 3 users x1, x3, and x9, or these three users belong to the same category.

[0141] When performing cluster analysis, the feature quantities in the user feature sample x can be used in batches. Clustering can be performed feature by feature, or several or all features can be selected for clustering. However, it is necessary to ensure that the features used for clustering each user are consistent.

[0142] Optionally, obtaining the user sample collection may be performing clustering processing on the driving behavior data of the first vehicle, for example, determining multiple category collections based on one or more feature values ​​among multiple feature values ​​of the first vehicle.

[0143] Step 6: Label evaluation and verification.

[0144] Through the above cluster analysis, we obtain a collection of multiple different categories. From each category, we select a portion of samples for expert evaluation or scoring to obtain their labels. In addition to using sample averaging for expert evaluation, we can also conduct in-depth analysis of the entire data set of a specific user category to obtain new labels.

[0145] Optionally, label evaluation and verification may be to obtain evaluation scores of multiple category sets and determine the driving behavior type of each category set.

[0146] Step 7: Iterative optimization of the model.

[0147] The above modeling method can be iteratively optimized after a period of data accumulation, and a new model can also be re-modeled according to new needs.

[0148] Regarding the aforementioned method for determining driving behavior types, steps 1-3 can be performed either in the cloud or on the vehicle. Specifically, features are collected and calculated on the vehicle and then uploaded, rather than the raw data. This on-vehicle computation followed by uploading effectively reduces cloud computing load and data transmission. The aforementioned model training requires big data in the cloud. After model training is complete, it can be deployed on the vehicle of a new user, enabling the identification of the new user's profile type there, also effectively reducing cloud computing load and data transmission.

[0149] Figure 6 FIG. 6 is a schematic diagram of a driving behavior type determination device 600 according to an embodiment of the present disclosure. Figure 6 As shown, the device includes: an acquisition module 610, an extraction module 620, and a training module 630.

[0150] The acquisition module 610 is used to acquire driving behavior data of multiple vehicles. The driving behavior data includes vehicle data and environment data. The vehicle data reflects the travel data of the multiple vehicles, and the environment data reflects the external environment data of the multiple vehicles.

[0151] The extraction module 620 is used to clean and extract features from the driving behavior data, and determine multiple feature samples corresponding to multiple target vehicles. Each feature sample includes multiple feature values, and the multiple vehicles include multiple target vehicles.

[0152] The training module 630 is configured to perform clustering processing on the plurality of feature samples based on one or more feature values ​​among the plurality of feature values ​​to obtain a plurality of category sets, wherein the plurality of category sets respectively include one or more feature samples among the plurality of feature samples.

[0153] The training module 630 is used to obtain evaluation scores of multiple category sets to determine the driving behavior type of each category set. The multiple category sets and the driving behavior type of each category set are used to train the first model, and the first model is used to evaluate the driving behavior type of the vehicle.

[0154] In some embodiments, the extraction module is also used to eliminate invalid data from the driving behavior data of multiple vehicles to obtain valid driving behavior data of multiple vehicles; screen the valid driving behavior data of multiple vehicles to obtain training data, which includes valid driving behavior data of multiple target vehicles; and perform feature extraction on the training data to determine multiple feature samples corresponding to the multiple target vehicles.

[0155] In some embodiments, the extraction module is further configured to select one or more types of data for combined calculation based on the effective driving behavior data of each target vehicle in the training data to determine a feature sample of each target vehicle.

[0156] In some embodiments, the training module is further configured to determine data samples from each category set; and obtain evaluation scores of the data samples in each category set to determine the driving behavior type of each category set.

[0157] In some embodiments, the characteristic value includes at least one of a first characteristic value, a second characteristic value, a third characteristic value, and a fourth characteristic value, wherein the first characteristic value is determined based on the speed of the target vehicle within a preset mileage, the second characteristic value is determined based on the speed and the steering wheel angle, the third characteristic value is determined based on the acceleration and the accelerator pedal opening, and the fourth characteristic value is determined based on the speed and the lateral acceleration and longitudinal acceleration.

[0158] In summary, a driving behavior type determination device is used to obtain driving behavior data of multiple vehicles; clean and extract features from the driving behavior data to determine multiple feature samples corresponding to multiple target vehicles; cluster the multiple feature samples based on one or more feature values ​​among the multiple feature values ​​to obtain multiple category collections, each of which includes one or more feature samples among the multiple feature samples; and obtain evaluation scores for the multiple category collections to determine the driving behavior type of each category collection. User profiling is performed based on unsupervised machine learning to obtain a category collection, and then an evaluation score for each category collection is obtained through supervised learning to determine the driving behavior type of each category collection. This can not only improve efficiency but also obtain new types of user profiling. The above-mentioned multiple category collections and the driving behavior type of each category collection are used to train a first model, and the first model can be used to evaluate the driving behavior type of a new user driving a vehicle.

[0159] In the embodiments provided above, the methods and devices provided in the embodiments of the present application are introduced. In order to implement the various functions of the methods provided in the embodiments of the present application, the electronic device may include a hardware structure and a software module, and implement the aforementioned functions in the form of a hardware structure, a software module, or a hardware structure plus a software module. One of the aforementioned functions may be executed in the form of a hardware structure, a software module, or a hardware structure plus a software module.

[0160] Figure 7 is a block diagram of an electronic device 700 for implementing the above-mentioned method for determining the driving behavior type according to an exemplary embodiment.

[0161] For example, the electronic device 700 may be a mobile phone, a computer, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, or the like.

[0162] Reference Figure 7 , electronic device 700 may include one or more of the following components: a processing component 702 , a memory 704 , a power component 706 , a multimedia component 708 , an audio component 710 , an input / output (I / O) interface 712 , a sensor component 714 , and a communication component 716 .

[0163] The processing component 702 generally controls the overall operation of the electronic device 700, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 702 may include one or more modules to facilitate interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate interaction between the multimedia component 708 and the processing component 702.

[0164] The memory 704 is configured to store various types of data to support operations on the electronic device 700. Examples of such data include instructions for any application or method operating on the electronic device 700, contact data, phone book data, messages, pictures, videos, etc. The memory 704 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0165] The power supply component 706 provides power to the various components of the electronic device 700. The power supply component 706 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device 700.

[0166] The multimedia component 708 includes a screen that provides an output interface between the electronic device 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensor can not only sense the boundaries of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 708 includes a front camera and / or a rear camera. When the electronic device 700 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capabilities.

[0167] The audio component 710 is configured to output and / or input audio signals. For example, the audio component 710 includes a microphone (MIC), which is configured to receive external audio signals when the electronic device 700 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 704 or transmitted via the communication component 716. In some embodiments, the audio component 710 also includes a speaker for outputting audio signals.

[0168] I / O interface 712 provides an interface between processing component 702 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0169] The sensor assembly 714 includes one or more sensors for providing various aspects of status assessment for the electronic device 700. For example, the sensor assembly 714 can detect the open / closed state of the electronic device 700, the relative positioning of components, such as the display and keypad of the electronic device 700. The sensor assembly 714 can also detect changes in the position of the electronic device 700 or a component of the electronic device 700, the presence or absence of user contact with the electronic device 700, the orientation or acceleration / deceleration of the electronic device 700, and temperature changes of the electronic device 700. The sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 714 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 714 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0170] The communication component 716 is configured to facilitate wired or wireless communication between the electronic device 700 and other devices. The electronic device 700 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, 4G LTE, 5G NR (NewRadio) or a combination thereof. In an exemplary embodiment, the communication component 716 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 716 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0171] In an exemplary embodiment, the electronic device 700 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0172] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions, and the instructions can be executed by the processor 720 of the electronic device 700 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0173] An embodiment of the present disclosure further provides a vehicle, comprising the driving behavior type determination device or electronic device described in the above embodiment of the present disclosure.

[0174] An embodiment of the present disclosure further provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the driving behavior type determination method described in the above embodiment of the present disclosure.

[0175] An embodiment of the present disclosure further provides a computer program product, including a computer program, which is used by a processor to execute the driving behavior type determination method described in the above embodiment of the present disclosure.

[0176] An embodiment of the present disclosure also proposes a chip, which includes one or more interface circuits and one or more processors; the interface circuit is used to receive signals from the memory of the electronic device and send signals to the processor, the signals including computer instructions stored in the memory, and when the processor executes the computer instructions, the electronic device executes the driving behavior type determination method described in the above embodiment of the present disclosure.

[0177] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0178] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with an embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, the illustrative use of the above terms does 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 any one or more embodiments or examples.

[0179] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present disclosure belong.

[0180] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the 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 (e.g., a computer-based system, a system including a processing module, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection having one or more wires (control method), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing it in a suitable manner if necessary, and then storing it in a computer memory.

[0181] It should be understood that the various parts of the embodiments of the present disclosure 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 a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0182] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0183] In addition, the functional units in the various embodiments of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium. The above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disk, etc.

[0184] Although the embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are exemplary and are not to be construed as limitations on the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. A method for determining a driving behavior type, characterized in that: The method comprises: Acquiring driving behavior data of a plurality of vehicles, the driving behavior data including vehicle data and environmental data, the vehicle data reflecting travel data of the plurality of vehicles, and the environmental data reflecting external environment data of the plurality of vehicles; Cleaning and feature extracting the driving behavior data to determine a plurality of feature samples corresponding to a plurality of target vehicles, each feature sample including a plurality of feature values, the plurality of vehicles including the plurality of target vehicles; Based on one or more feature values ​​among the multiple feature values, clustering the multiple feature samples to obtain multiple category sets, where the multiple category sets respectively include one or more feature samples among the multiple feature samples; Obtain evaluation scores of the multiple category collections to determine the driving behavior type of each category collection, the multiple category collections and the driving behavior type of each category collection are data for training a first model, and the first model is used to evaluate the driving behavior type of a vehicle.

2. The method according to claim 1, characterized in that The cleaning and feature extraction of the driving behavior data to determine a plurality of feature samples corresponding to a plurality of target vehicles includes: Eliminating invalid data from the driving behavior data of the multiple vehicles to obtain valid driving behavior data of the multiple vehicles; Screening the valid driving behavior data of the plurality of vehicles to obtain training data, wherein the training data includes the valid driving behavior data of a plurality of target vehicles; Feature extraction is performed on the training data to determine a plurality of feature samples corresponding to the plurality of target vehicles respectively.

3. The method according to claim 2, characterized in that The extracting features from the training data to determine a plurality of feature samples corresponding to a plurality of target vehicles comprises: For the effective driving behavior data of each target vehicle in the training data, one or more types of data are selected for combined calculation to determine the feature samples of each target vehicle.

4. The method according to claim 3, characterized in that The characteristic value includes at least one of a first characteristic value, a second characteristic value, a third characteristic value, and a fourth characteristic value, wherein the first characteristic value is determined based on the speed of the target vehicle within a preset mileage, the second characteristic value is determined based on the vehicle speed and the steering wheel angle, the third characteristic value is determined based on the acceleration and the accelerator pedal opening, and the fourth characteristic value is determined based on the vehicle speed and the lateral acceleration and longitudinal acceleration.

5. The method according to claim 1, wherein Obtaining the evaluation scores of the plurality of category sets to determine the driving behavior type of each category set includes: determining a data sample from each of the category collections; Obtain evaluation scores of the data samples in each category set to determine the driving behavior type of each category set.

6. A driving behavior type determination device, characterized in that: Including acquisition module, extraction module and training module, The acquisition module is used to acquire driving behavior data of multiple vehicles, wherein the driving behavior data includes vehicle data and environmental data, wherein the vehicle data reflects the travel data of the multiple vehicles, and the environmental data reflects the external environment data of the multiple vehicles; The extraction module is used to clean and extract features from the driving behavior data, determine a plurality of feature samples corresponding to a plurality of target vehicles, each feature sample including a plurality of feature values, and the plurality of vehicles including the plurality of target vehicles; The training module is configured to perform clustering processing on the plurality of feature samples based on one or more feature values ​​among the plurality of feature values ​​to obtain a plurality of category collections, wherein the plurality of category collections respectively include one or more feature samples among the plurality of feature samples; The training module is used to obtain evaluation scores of the multiple category collections to determine the driving behavior type of each category collection. The multiple category collections and the driving behavior type of each category collection are data used to train a first model, and the first model is used to evaluate the driving behavior type of a vehicle.

7. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 5.

8. A vehicle, characterized in that: Comprising the apparatus according to claim 6 or the electronic device according to claim 7.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 5.

10. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 5.

11. A chip, characterized in that: The electronic device comprises one or more interface circuits and one or more processors; the interface circuit is used to receive a signal from a memory of the electronic device and send the signal to the processor, wherein the signal includes a computer instruction stored in the memory, and when the processor executes the computer instruction, the electronic device executes the method according to any one of claims 1 to 5.

Images