Vehicle control method and apparatus, and vehicle, storage medium and program product

Abstract

Disclosed in the present application are a vehicle control method and apparatus, and a vehicle, a storage medium and a program product. The method comprises: acquiring historical traveling parameters of a vehicle during the process of a user driving the vehicle within a historical time period; performing feature extraction on the historical traveling parameters to obtain a user driving feature; recognizing the user driving feature on the basis of a style recognition model, so as to obtain a user driving style of the user; and in response to detecting that the user is driving the vehicle, controlling the operation of the vehicle on the basis of the user driving style, so as to assist the user in driving the vehicle. The present application solves the technical problem in the related art of the effect of assisting a user in driving a vehicle on the basis of a user driving style being relatively poor.

Description

Vehicle control method and device, vehicle, storage medium and program product TECHNICAL FIELD

[0001] The present disclosure relates to the field of vehicle control, and in particular, to a vehicle control method and device, vehicle, storage medium and program product. BACKGROUND

[0002] Nowadays, with the development of electronic technology, the driving experience of a vehicle is widely concerned. Different users may have different driving habits when driving a vehicle, and the corresponding driving styles may also be different. In order to improve the user's driving experience of the vehicle, different vehicle-mounted devices or vehicle driving parameters on the vehicle can be controlled according to the driving style of the user when driving the vehicle. However, when determining the driving style of different users, a single classification model is usually used to classify the driving habits of the user to form the driving style of the user, which may not meet the user's expectation, resulting in an unsatisfactory effect of assisting the user to drive the vehicle according to the identified driving style.

[0003] At present, there is no effective solution to the above problems. SUMMARY

[0004] The embodiments of the present disclosure provide a vehicle control method, device, vehicle, storage medium and program product to at least solve the technical problem of poor effect of assisting a user to drive a vehicle according to the user's driving style in the related art.

[0005] According to an aspect of an embodiment of the present disclosure, a vehicle control method is provided, including: obtaining a historical driving parameter of a vehicle in a process of driving the vehicle by a user in a historical time period; performing feature extraction on the historical driving parameter to obtain a user driving feature; identifying the user driving feature based on a style recognition model to obtain a user driving style of the user, wherein the style recognition model includes a plurality of style classifiers, different style classifiers have different classification weights for classifying different preset driving styles, the preset driving style includes the user driving style, the user driving style is used to represent a style obtained by aggregating a plurality of classification results of the user driving feature classified by the style classifiers based on the classification weights, and the classification result is used to represent a matching degree between the user driving feature and different preset driving styles under different style classifiers; and in response to detecting that the user drives the vehicle, controlling the vehicle to operate based on the user driving style to assist the user to drive the vehicle.

[0006] Further, the user driving feature is identified based on the style recognition model to obtain a driving style of the user, including: performing importance evaluation on the user driving feature according to a preset evaluation manner to obtain a first importance index of the user driving feature on the historical driving parameter; selecting at least one target driving feature from the user driving feature based on the first importance index; and inputting the target driving feature into the style recognition model to obtain the driving style of the user.

[0007] Further, the importance evaluation on the user driving feature according to the preset evaluation manner to obtain the first importance index of the user driving feature on the historical driving parameter includes: performing data alignment on the user driving feature and the historical driving parameter to obtain a first observation value corresponding to the user driving feature and a second observation value corresponding to the historical driving parameter; determining a correlation coefficient between the user driving feature and the historical driving parameter based on the first observation value and the second observation value; and inputting the user driving feature into the ensemble learning model based on the correlation coefficient to obtain the first importance index.

[0008] Further, the vehicle is controlled to run based on the user driving style, including: obtaining a vehicle control parameter matched with the user driving style from a parameter library, wherein the parameter library is used to store an association relationship between the user driving style and the vehicle control parameter; outputting the vehicle control parameter in an operation interface and receiving a parameter adjustment instruction fed back by the user; adjusting the vehicle control parameter based on the parameter adjustment instruction to obtain a target control parameter; and controlling the vehicle to run based on the target control parameter.

[0009] Further, the user driving feature is identified based on the style recognition model to obtain a driving style of the user, including: dividing a plurality of classification results based on a preset driving style to obtain a plurality of classification result sets, wherein the classification results in a same classification result set correspond to a same preset driving style; merging the classification results in each classification result set based on a classification weight corresponding to the classification results in each classification result set to obtain a target classification result, wherein the target classification result is used to represent a matching degree between the user driving feature and the preset driving style corresponding to the classification result set; and determining the driving style of the user from the preset driving style based on the target classification result.

[0010] Further, the above method further includes: obtaining a sample driving parameter corresponding to at least one preset driving style; performing feature extraction on the sample driving parameter to obtain at least one sample driving feature; inputting the sample driving feature into an initial recognition model to obtain an initial recognition result corresponding to the sample driving parameter; and adjusting the initial recognition model based on the initial recognition result and the preset driving style to obtain the style recognition model.

[0011] Further, the sample driving features are input into the initial identification model to obtain an initial identification result corresponding to the sample driving parameters, including: performing importance evaluation on the sample driving features according to a preset evaluation manner to obtain a second importance index of the sample driving features for the sample driving parameters; sorting the second importance index, and selecting at least one target sample feature from the sample driving features based on a sorting result, wherein the second importance index corresponding to the target sample feature is greater than the second importance index corresponding to other sample features in the sample driving features except the target sample feature; and inputting the target sample feature into the initial identification model to obtain the initial identification result.

[0012] Further, the initial identification model at least includes a plurality of initial classifiers, the initial identification result includes a plurality of initial classification results obtained by the plurality of initial classifiers classifying the target sample feature, and the initial identification model is adjusted based on the initial identification result and a preset driving style to obtain a style identification model, including: matching the plurality of initial classification results with the preset driving style to obtain a plurality of first matching results; adjusting the initial classifiers based on the plurality of first matching results to obtain a plurality of style classifiers; matching a plurality of style classification results with the preset driving style to obtain a plurality of second matching results, wherein the style classification result is used to represent a classification result obtained by inputting the target sample feature into the style classifier; determining a classification weight of the style classifier relative to the preset driving style based on the plurality of second matching results; and constructing the style identification model based on the classification weight and the plurality of style classifiers.

[0013] According to an aspect of an embodiment of the present disclosure, a vehicle control device is also provided, including: a parameter acquisition component configured to acquire historical driving parameters of a vehicle when a user drives the vehicle in a historical time period; a feature extraction component configured to perform feature extraction on the historical driving parameters to obtain user driving features; a style identification component configured to identify the user driving features based on a style identification model to obtain a user driving style of the user, wherein the style identification model includes a plurality of style classifiers, different classification weights of different style classifiers for classifying different preset driving styles are different, the preset driving styles include the user driving style, the user driving style is used to represent a style obtained by aggregating a plurality of classification results of the style classifiers classifying the user driving features based on the classification weights, and the classification result is used to represent a matching degree between the user driving features and different preset driving styles under different style classifiers; and a vehicle control component configured to control the vehicle to operate based on the user driving style in response to detecting that the user drives the vehicle, to assist the user in driving the vehicle.

[0014] Further, the style recognition component comprises: an importance evaluation component configured to evaluate the importance of the user driving features according to a preset evaluation manner, to obtain a first importance index of the user driving features for the historical driving parameters; a feature selection component configured to select at least one target driving feature from the user driving features based on the first importance index; and the style recognition component is configured to input the target driving feature into the style recognition model to obtain the user driving style.

[0015] Further, the importance evaluation component is further configured to: perform data alignment on the user driving features and the historical driving parameters to obtain first observation values corresponding to the user driving features and second observation values corresponding to the historical driving parameters; determine a correlation coefficient between the user driving features and the historical driving parameters based on the first observation values and the second observation values; and input the user driving features into the ensemble learning model based on the correlation coefficient to obtain the first importance index.

[0016] Further, the vehicle control component comprises: a parameter acquisition component configured to acquire vehicle control parameters matched with the user driving style from a parameter library, wherein the parameter library is used to store an association relationship between the user driving style and the vehicle control parameters; an instruction receiving component configured to output the vehicle control parameters in an operation interface and receive a parameter adjustment instruction fed back by a user; a parameter adjustment component configured to adjust the vehicle control parameters based on the parameter adjustment instruction to obtain target control parameters; and the vehicle control component is configured to control the vehicle to operate based on the target control parameters.

[0017] Further, the style recognition component comprises: a result division component configured to divide a plurality of classification results based on preset driving styles to obtain a plurality of classification result sets, wherein the classification results in a same classification result set correspond to the same preset driving style; a result merging component configured to merge the classification results in each classification result set based on classification weights corresponding to the classification results in each classification result set to obtain a target classification result, wherein the target classification result is used to represent a matching degree between the user driving features and the preset driving style corresponding to the classification result set; and a style determination component configured to determine the user driving style from the preset driving styles based on the target classification result.

[0018] Further, the above device further comprises: a sample acquisition component configured to acquire sample driving parameters corresponding to at least one preset driving style; a feature extraction component configured to extract features from the sample driving parameters to obtain at least one sample driving feature; an initial recognition component configured to input the sample driving features into an initial recognition model to obtain an initial recognition result corresponding to the sample driving parameters; and a model adjustment component configured to adjust the initial recognition model based on the initial recognition result and the preset driving style to obtain the style recognition model.

[0019] Further, the initial identification component includes: a sample feature evaluation component configured to evaluate the importance of the sample driving features according to a preset evaluation manner, to obtain a second importance index of the sample driving features on the sample driving parameters; a sample feature selection component configured to sort the second importance index, and select at least one target sample feature from the sample driving features based on the sorting result, wherein the second importance index corresponding to the target sample feature is greater than the second importance index corresponding to other sample features in the sample driving features except the target sample feature; and a sample identification component configured to input the target sample feature into the initial identification model to obtain an initial identification result.

[0020] Further, the initial identification model at least includes a plurality of initial classifiers, the initial identification result includes a plurality of initial classification results obtained by classifying the target sample feature by the plurality of initial classifiers, and the model adjustment component includes: a first matching component configured to match the plurality of initial classification results with the preset driving style to obtain a plurality of first matching results; a classifier adjustment component configured to adjust the initial classifiers based on the plurality of first matching results to obtain a plurality of style classifiers; a second matching component configured to match a plurality of style classification results with the preset driving style to obtain a plurality of second matching results, wherein the style classification result is used to represent a classification result obtained by inputting the target sample feature into the style classifier; a weight determination component configured to determine a classification weight of the style classifier with respect to the preset driving style based on the plurality of second matching results; and a model construction component configured to construct a style identification model based on the classification weight and the plurality of style classifiers.

[0021] According to another aspect of the embodiments of the present disclosure, a vehicle is further provided, including: a memory storing an executable program; and a processor configured to run the program, wherein the program runs to perform the method in various embodiments of the present disclosure.

[0022] According to another aspect of the embodiments of the present disclosure, a computer readable storage medium is further provided, including a stored executable program, wherein the executable program runs to control the device where the computer readable storage medium is located to perform the method in various embodiments of the present disclosure.

[0023] According to another aspect of the embodiments of the present disclosure, a computer program product is further provided, including a computer program, which, when executed by a processor, implements the method in various embodiments of the present disclosure.

[0024] According to a further aspect of the embodiments of the present disclosure, a computer program product is also provided, including a non-volatile computer readable storage medium storing a computer program, which, when executed by a processor, implements the method in various embodiments of the present disclosure.

[0025] According to a further aspect of the embodiments of the present disclosure, a computer program is also provided, which, when executed by a processor, implements the method in various embodiments of the present disclosure.

[0026] In the embodiments of the present disclosure, the historical driving parameters of the vehicle in the process of driving the vehicle by the user in the historical time period are acquired, the historical driving parameters are subjected to feature extraction to obtain the user driving features, the user driving features are identified based on the style recognition model to obtain the user driving style of the user, and in response to detecting that the user drives the vehicle, the vehicle is controlled based on the user driving style in the manner of assisting the user to drive the vehicle, the user driving features are classified by using the plurality of style classifiers configured in the style recognition model, and then the classification results output by the style classifiers are summarized according to the classification weights between the different style classifiers and different style types to obtain the user driving style with high matching degree of the user, which can effectively improve the accuracy of the determined user driving style and improve the effect of assisting the user to drive the vehicle by using the driving parameters corresponding to the user driving style, thereby improving the driving experience of the user and further solving the technical problem of poor effect of assisting the user to drive the vehicle according to the driving style of the user in the related art. BRIEF DESCRIPTION OF DRAWINGS

[0027] The accompanying drawings incorporated in and forming a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure. In the drawings:

[0028] FIG. 1 is a flowchart of a vehicle control method according to an embodiment of the present disclosure;

[0029] FIG. 2 is a schematic diagram of a target driving feature selection process according to an embodiment of the present disclosure;

[0030] FIG. 3 is a schematic diagram of an importance index determination process according to an embodiment of the present disclosure;

[0031] FIG. 4 is a structural block diagram of a vehicle control device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] In order to enable personnel in the technical field to better understand the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative labor should fall within the scope of protection of the present disclosure.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or components does not have to be limited to those steps or components clearly listed, but can include other steps or components that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] Embodiment 1

[0035] According to the embodiments of the present disclosure, a method embodiment of vehicle control is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that herein.

[0036] FIG. 1 is a flowchart illustrating a vehicle control method according to an embodiment of the present disclosure. As shown in FIG. 1, the method includes the following steps:

[0037] In step S102, historical driving parameters of the vehicle in the process of driving the vehicle by the user in the historical time period are acquired.

[0038] In an optional solution of the embodiment, it is considered that the driving style of the user in different periods can be different due to personal habits, living environment and other factors, and therefore, in order to accurately determine the driving style of the user, the vehicle control system can first determine whether the current driving style of the user changes, for example, the use data of the vehicle-mounted equipment commonly used by the user when driving the vehicle in the last few times, the driving data generated in the process of driving the vehicle and other parameters can be acquired, and it is determined whether these parameters change greatly, if not, it is preliminarily determined that the current driving style of the user does not change, if so, it is considered that the current driving style of the user changes. In the case that it is determined that the current driving style of the user can change, or in the case that the driving style of the user driving the vehicle is not queried in the database for storing the driving styles of different users, the vehicle control system can identify the current driving style of the user, so as to control the vehicle to run according to the driving style of the user, to assist the user to drive the vehicle. Based on this, when determining the current driving style of the user, the vehicle control system can first acquire historical driving parameters generated by the user in the process of driving the vehicle in a historical period, for example, in the last month or in the time period of the last five times of driving the vehicle by the user, such as equipment use data of historical driving parameters, such as vehicle-mounted air conditioner use time, sound volume height and other parameters, vehicle driving data of vehicle speed, acceleration, accelerator pedal opening and other parameters, so as to ensure that the historical driving parameters acquired can fully and comprehensively reflect the current driving style of the user, thereby improving the accuracy of the driving style of the user identified according to the historical driving parameters.

[0039] In step S104, the historical driving parameters are feature extracted to obtain the user driving features.

[0040] In an optional solution of the embodiment, after the historical driving parameters of the user driving the vehicle are acquired, the vehicle control system can further perform feature extraction on the historical driving parameters to obtain user driving features matched with the driving style of the user, for example, 44 feature components can be extracted from the historical driving parameters to describe the driving style.

[0041] In an optional solution of the embodiment, it is considered that there can be a lot of noise data in the acquired historical driving parameters, and the user driving features extracted from the historical driving parameters also contain a lot of noise data, and therefore, after the feature extraction on the historical parameters, the user driving features obtained by the extraction can also be de-noised, for example, the Euclidean distance between different user driving features is determined, and the discrete points are determined from the data points corresponding to the different user driving features according to the Euclidean distance, so as to remove the user driving features corresponding to the discrete points, thereby obtaining more accurate user driving features.

[0042] The formula for determining the Euclidean distance can be:

[0043] wherein d(Q, C) can refer to the Euclidean distance between the data point Q and the data point C, q i and c i are the i-th feature components of the data point Q and the data point C respectively, and n is the total number of features.

[0044] In step S106, the user driving feature is identified based on the style recognition model, and the user driving style of the user is obtained.

[0045] The style recognition model includes a plurality of style classifiers, different style classifiers have different classification weights for classifying different preset driving styles, the preset driving styles include the user driving style, the user driving style is used to represent the style obtained by aggregating a plurality of classification results obtained by classifying the user driving feature by the style classifier based on the classification weight, and the classification result is used to represent the matching degree between the user driving feature and different preset driving styles under different style classifiers.

[0046] In an optional scheme of the embodiment, after the user driving feature corresponding to the historical driving parameter is extracted, the vehicle control system can input the user driving feature into the pre-trained style recognition model to determine the current driving style of the user, i.e., to determine the user driving style.

[0047] In an optional scheme of the embodiment, in order to ensure the accuracy of the determined user driving style, a plurality of pre-trained style classifiers can be configured in the style recognition model, different style classifiers have different identification focuses, for example, the style classifier A can be mainly used for identifying the preset driving style A, and the style classifier B can be mainly used for identifying the preset driving style B, i.e., the parameters configured in different style classifiers are different, different style classifiers have different classification weights for classifying different preset driving styles, and the user driving style identified by the style recognition model can be one of the preset driving styles. When the user driving feature is identified by the style recognition model, the user driving feature can be input into the above-mentioned plurality of style classifiers to correspondingly obtain a plurality of classification results, i.e., the matching degree between the above-mentioned user driving feature and different preset driving styles under different style classifiers, and then the classification weight corresponding to each style classifier is used to aggregate the plurality of classification results to determine the user driving style matching the user driving feature from the plurality of preset driving styles.

[0048] In step S108, in response to detecting that the user drives the vehicle, the vehicle operation is controlled based on the user driving style to assist the user in driving the vehicle.

[0049] After identifying the driving style of the user, the vehicle control system can store the identity information of the user and the identified driving style in a preset style database. When detecting that the user is driving the vehicle, the vehicle control system can identify the identity information of the user, obtain the driving style of the user matched with the identity information from the style database, and control the vehicle to operate according to the driving style of the user, for example, control different vehicle-mounted devices on the vehicle to operate according to specified parameters, control the driving speed of the vehicle to remain in a stable speed range, and the like, to assist the user in driving the vehicle, thereby providing the user with a driving experience.

[0050] In the embodiments of the present disclosure, the historical driving parameters of the vehicle in the process of driving the vehicle by the user in a historical time period are obtained, the user driving features are obtained by performing feature extraction on the historical driving parameters, the user driving style of the user is obtained by identifying the user driving features based on a style recognition model, and the vehicle is controlled to operate based on the user driving style in response to detecting that the user is driving the vehicle, in a manner of assisting the user in driving the vehicle, the user driving features are classified by using a plurality of style classifiers configured in the style recognition model, and then the classification results output by the style classifiers are summarized according to the classification weights between different style classifiers and different style types to obtain the user driving style with high matching degree of the user, which can effectively improve the accuracy of the determined user driving style, improve the effect of assisting the user in driving the vehicle by using the driving parameters corresponding to the user driving style, thereby improving the driving experience of the user, and further solving the technical problem that the effect of assisting the user in driving the vehicle according to the driving style of the user is poor in the related art.

[0051] Further, the user driving style is obtained by identifying the user driving features based on the style recognition model, including: performing importance evaluation on the user driving features according to a preset evaluation manner to obtain a first importance index of the user driving features for the historical driving parameters; selecting at least one target driving feature from the user driving features based on the first importance index; and inputting the target driving feature into the style recognition model to obtain the user driving style.

[0052] In an optional solution of the embodiment, considering that a large number of features can be extracted from the driving parameters, and actually only a part of the features are related to the driving style of the user, when the driving features of the user are identified by the style identification model to determine the driving style of the user, the importance of the extracted user features can be evaluated first to determine the importance of different user driving features with respect to the historical driving parameters, that is, to determine the first importance index. The greater the importance index is, the closer the relationship between the user driving feature and the historical driving parameter is, and the more accurate the user driving style determined by the user driving feature is. Based on this, after obtaining the first importance index of different user driving features with respect to the historical driving parameters, the vehicle control system can select at least one target driving feature that can reflect the driving style of the user from the user driving features according to the first importance index, and then input the target driving feature into the above-mentioned style identification model to determine the current driving style of the user.

[0053] Further, the importance of the user driving feature is evaluated according to the preset evaluation method to obtain the first importance index of the user driving feature with respect to the historical driving parameter, including: performing data alignment on the user driving feature and the historical driving parameter to obtain a first observation value corresponding to the user driving feature and a second observation value corresponding to the historical driving parameter; determining a correlation coefficient between the user driving feature and the historical driving parameter based on the first observation value and the second observation value; and inputting the user driving feature into an ensemble learning model based on the correlation coefficient to obtain the first importance index.

[0054] In an optional solution of the embodiment, in order to accurately evaluate the importance of the user driving feature with respect to the historical driving parameter, the vehicle control system can first perform data alignment on the extracted user driving feature and the historical driving parameter to obtain a first observation value corresponding to the user driving feature and a second observation value corresponding to the historical driving parameter, and then use a correlation coefficient method to determine a correlation coefficient between the user driving feature and the historical driving parameter based on the first observation value and the second observation value, and input the user driving feature into one or more preset ensemble learning models, such as a random forest model, an extreme boosting tree model, and an XGBoot model, in sequence to obtain index results output by different ensemble learning models. Finally, different index results are summarized to obtain the importance index of different user driving features with respect to the historical driving parameter. It should be noted that the above-mentioned correlation coefficient method, random forest model and the like can refer to related technologies, and will not be repeated and limited here.

[0055] For ease of understanding, FIG. 2 is a schematic diagram of a target driving feature selection process according to an embodiment of the present disclosure. As shown in FIG. 2, historical driving parameters and user driving features corresponding to the historical driving parameters can be obtained first, then the correlation coefficient between the user driving features and the historical driving parameters can be determined by using the Pearson correlation coefficient method, and the user driving features can be sequentially input into the random forest model by using the correlation coefficient to output the importance indicators corresponding to different user features, and then it can be judged whether the importance indicators meet the preset indicator conditions, for example, whether the importance indicators are greater than the preset threshold, or whether they are the maximum values among multiple importance indicators, etc. If yes, the user driving feature corresponding to the importance indicator can be determined as the target driving feature; if not, the user driving feature corresponding to the importance indicator can be discarded.

[0056] Further, the vehicle is controlled based on the user driving style, including: obtaining vehicle control parameters matched with the user driving style from a parameter library, wherein the parameter library is used to store the association relationship between the user driving style and the vehicle control parameters; outputting the vehicle control parameters in an operation interface and receiving a parameter adjustment instruction fed back by the user; adjusting the vehicle control parameters based on the parameter adjustment instruction to obtain target control parameters; and controlling the vehicle to run based on the target control parameters.

[0057] The above parameter library can be used to store the association relationship between different user driving styles and corresponding vehicle control parameters. The vehicle control parameters can be control parameters summarized by the vehicle control system according to historical driving parameters, or control parameters set by the user, which are not limited herein. Different vehicle control parameters can also be configured for different users in the same driving style.

[0058] In an optional solution of the present embodiment, when the vehicle is controlled based on the user driving style, the vehicle control system can first obtain vehicle control parameters matched with the user driving style from a preset parameter library, and then output the vehicle control parameters in a preset operation interface, such as a vehicle screen, to show the user. The user can view and adjust the vehicle control parameters in the operation interface. When the vehicle control system detects the adjustment of the vehicle control parameters by the user, it can generate a corresponding parameter adjustment instruction and use the parameter adjustment instruction to adjust the obtained vehicle control parameters to obtain the target control parameters described above. Finally, the vehicle is controlled to run by using the target control parameters, so that the running condition of the vehicle can be more in line with the user's expectation. While adjusting to obtain the target control parameters, the target control parameters can also be selectively updated to the above parameter library for easy use next time.

[0059] Further, the user driving feature is identified based on the style recognition model to obtain a user driving style of the user, including: dividing the plurality of classification results based on the preset driving styles to obtain a plurality of classification result sets, wherein the classification results in the same classification result set correspond to the same preset driving style; merging the classification results in each classification result set based on the classification weights corresponding to the classification results in the classification result set to obtain a target classification result, wherein the target classification result is used to represent a matching degree between the user driving feature and the preset driving style corresponding to the classification result set; and determining the user driving style from the preset driving styles based on the target classification result.

[0060] In an optional solution of the embodiment, when the user driving feature is classified and identified by using the plurality of style classifiers in the style recognition model, the plurality of classification results can be first divided according to different preset driving styles to obtain a plurality of classification result sets, the classification results in the same classification result set correspond to the same preset driving style, and the difference lies in the classification weights of the different style classifiers for the preset driving style. After the classification result sets are divided, the classification results in the classification result set can be merged according to the classification weights corresponding to the different classification results in the classification result set to obtain a target classification result corresponding to the preset driving style, that is, to obtain a matching degree between the user driving feature and the preset driving style corresponding to the classification result set. Finally, according to the target classification results of the different classification result sets, that is, the matching degrees between the user driving feature and the different preset driving styles, the vehicle control system can determine the driving style with the highest matching degree from the different preset driving styles as the user driving style.

[0061] Further, the above method further includes: obtaining sample driving parameters corresponding to at least one preset driving style; extracting features of the sample driving parameters to obtain at least one sample driving feature; inputting the sample driving feature into the initial recognition model to obtain an initial recognition result corresponding to the sample driving parameter; and adjusting the initial recognition model based on the initial recognition result and the preset driving style to obtain the style recognition model.

[0062] In an optional solution of the embodiment, when the style recognition model is trained, the sample driving parameters corresponding to different preset driving styles can be first obtained, then features of the sample driving parameters are extracted to obtain at least one sample driving feature, and the sample driving feature is input into the initial recognition model to obtain an initial recognition result corresponding to the sample driving parameter. Finally, the initial recognition model is adjusted based on the initial recognition result and the corresponding preset driving style to obtain a style recognition model with higher recognition accuracy.

[0063] Further, the sample driving features are input into the initial identification model to obtain an initial identification result corresponding to the sample driving parameters, including: performing importance evaluation on the sample driving features according to a preset evaluation manner to obtain a second importance index of the sample driving features for the sample driving parameters; sorting the second importance index, and selecting at least one target sample feature from the sample driving features based on the sorting result, wherein the second importance index corresponding to the target sample feature is greater than the second importance index corresponding to other sample features in the sample driving features except the target sample feature; and inputting the target sample feature into the initial identification model to obtain the initial identification result.

[0064] Similar to the foregoing process of identifying the user driving features, when the sample driving features are input into the initial identification model for identification, the sample driving features can be evaluated in the manner of evaluating the user driving features to obtain a second importance index of the sample driving features for the sample driving parameters, and then the sample driving features are sorted according to the second importance index, and one or more target sample features are selected from the sample driving features according to the sorting result, the second importance index corresponding to the target sample feature is greater than the second importance index corresponding to other sample features in the sample driving features except the target sample feature, and finally the target sample feature is input into the initial identification model, so that the initial identification result described above can be obtained.

[0065] Further, the initial identification model includes at least a plurality of initial classifiers, the initial identification result includes a plurality of initial classification results obtained by classifying the target sample feature by the plurality of initial classifiers, and the initial identification model is adjusted based on the initial identification result and the preset driving style to obtain a style identification model, including: matching the plurality of initial classification results with the preset driving style to obtain a plurality of first matching results; adjusting the initial classifiers based on the plurality of first matching results to obtain a plurality of style classifiers; matching a plurality of style classification results with the preset driving style to obtain a plurality of second matching results, wherein the style classification result is used to represent a classification result obtained by inputting the target sample feature into the style classifier; determining a classification weight of the style classifier with respect to the preset driving style based on the plurality of second matching results; and constructing the style identification model based on the classification weight and the plurality of style classifiers.

[0066] In an optional solution of the embodiment, multiple initial classifiers can be included in the initial identification model, multiple initial classification results obtained by different initial classifiers classifying the sample driving features are included in the initial identification result, and when the initial identification model is adjusted using the initial identification result and the preset driving style, the different initial classification results can be matched with the preset driving style respectively to obtain multiple first matching results, and then the initial classifiers are adjusted using the multiple first matching results to obtain the multiple style classifiers. After the style classifiers are adjusted, the target sample features screened out can be input into the style classifiers to obtain the style classification results, and after the style classification results are matched with the preset style types to obtain second matching results, the classification weights of the different style classifiers for the preset style types can be determined according to the second matching results. For example, if the result output by the style classifier A shows that there is a 50% probability that the target sample features belong to the corresponding preset driving style, it can be determined that the initial weight of the style classifier for the preset driving style is 0.5, if the result output by the style classifier B shows that there is a 60% probability that the target sample features belong to the corresponding preset driving style, it can be determined that the initial weight of the style classifier for the preset driving style is 0.6, and if the result output by the style classifier shows that there is a 90% probability that the target sample features belong to the corresponding preset driving style, it can be determined that the initial weight of the style classifier for the preset driving style is 0.9. The initial weights of the different style classifiers are integrated to obtain the classification weights of the different style classifiers for the preset driving style. For example, the classification weight of the style classifier A for the preset style type can be 0.25, the classification weight of the style classifier B for the preset style type can be 0.3, and the classification weight of the style classifier A for the preset style type can be 0.45.

[0067] After the classification weights of the different style classifiers for the different preset driving styles are determined, the classification weights and the multiple style classifiers can be used to construct the style identification model, so as to ensure the accuracy of the style identification model.

[0068] For ease of understanding, FIG. 3 is a schematic diagram illustrating a user driving style determination process according to an embodiment of the present disclosure. As shown in FIG. 3, in determining the user driving style, the historical driving parameters can be obtained first, and the historical driving parameters are preprocessed, for example, denoising processing is performed. Then, the user driving features are extracted from the preprocessed historical driving parameters, and the importance of the user driving features is evaluated to select at least one target driving feature from the user driving features. Finally, the target driving feature is input into the style recognition model, and the target driving feature is recognized by multiple different style classifiers to obtain multiple style classification results. The classification weights of different style classifiers for different driving styles are used to summarize the multiple style classification results to determine the user driving style matching the target driving feature.

[0069] Embodiment 2

[0070] According to an aspect of an embodiment of the present disclosure, corresponding to the above vehicle control method, a vehicle control device is also provided. FIG. 4 is a structural block diagram of a vehicle control device according to an embodiment of the present disclosure. As shown in FIG. 4, the device includes a parameter acquisition component 402, a feature extraction component 404, a style recognition component 406, and a vehicle control component 408.

[0071] The parameter acquisition component 402 is configured to obtain historical driving parameters of the vehicle when the user drives the vehicle in a historical time period. The feature extraction component 404 is configured to extract features from the historical driving parameters to obtain user driving features. The style recognition component 406 is configured to recognize the user driving features based on a style recognition model to obtain the user driving style of the user, wherein the style recognition model includes multiple style classifiers, different style classifiers have different classification weights for classifying different preset driving styles, the preset driving styles include the user driving style, the user driving style is used to represent a style obtained by summarizing multiple classification results of the user driving features classified by the style classifiers based on the classification weights, and the classification results are used to represent the matching degree between the user driving features and different preset driving styles under different style classifiers. The vehicle control component 408 is configured to control the vehicle to operate based on the user driving style in response to detecting that the user drives the vehicle, to assist the user in driving the vehicle.

[0072] Further, the style recognition component includes an importance evaluation component configured to evaluate the importance of the user driving features according to a preset evaluation method to obtain a first importance indicator of the user driving features for the historical driving parameters; a feature selection component configured to select at least one target driving feature from the user driving features based on the first importance indicator; and a style recognition component configured to input the target driving feature into the style recognition model to obtain the user driving style.

[0073] Further, the importance evaluation component is further configured to: perform data alignment on the user driving feature and the historical driving parameter to obtain a first observation value corresponding to the user driving feature and a second observation value corresponding to the historical driving parameter; determine a correlation coefficient between the user driving feature and the historical driving parameter based on the first observation value and the second observation value; and input the user driving feature into the ensemble learning model based on the correlation coefficient to obtain a first importance index.

[0074] Further, the vehicle control component includes: a parameter acquisition component configured to acquire a vehicle control parameter matched with the user driving style from a parameter library, wherein the parameter library is configured to store an association relationship between the user driving style and the vehicle control parameter; an instruction receiving component configured to output the vehicle control parameter in an operation interface and receive a parameter adjustment instruction fed back by the user; a parameter adjustment component configured to adjust the vehicle control parameter based on the parameter adjustment instruction to obtain a target control parameter; and a vehicle control component configured to control the vehicle to operate based on the target control parameter.

[0075] Further, the style recognition component includes: a result division component configured to divide the plurality of classification results based on the preset driving styles to obtain a plurality of classification result sets, wherein the classification results included in a same classification result set correspond to the same preset driving style; a result merging component configured to merge the classification results in each classification result set based on the classification weights corresponding to the classification results in each classification result set to obtain a target classification result, wherein the target classification result is configured to represent a matching degree between the user driving feature and the preset driving style corresponding to the classification result set; and a style determination component configured to determine the user driving style from the preset driving styles based on the target classification result.

[0076] Further, the above device further includes: a sample acquisition component configured to acquire sample driving parameters corresponding to at least one preset driving style; a feature extraction component configured to extract features from the sample driving parameters to obtain at least one sample driving feature; an initial recognition component configured to input the sample driving feature into an initial recognition model to obtain an initial recognition result corresponding to the sample driving parameter; and a model adjustment component configured to adjust the initial recognition model based on the initial recognition result and the preset driving style to obtain the style recognition model.

[0077] Further, the initial identification component includes: a sample feature evaluation component configured to evaluate the importance of the sample driving features according to a preset evaluation manner to obtain a second importance index of the sample driving features on the sample driving parameters; a sample feature selection component configured to sort the second importance index and select at least one target sample feature from the sample driving features based on a sorting result, wherein the second importance index corresponding to the target sample feature is greater than the second importance index corresponding to other sample features in the sample driving features except the target sample feature; and a sample identification component configured to input the target sample feature into the initial identification model to obtain an initial identification result.

[0078] Further, the initial identification model includes at least a plurality of initial classifiers, the initial identification result includes a plurality of initial classification results obtained by classifying the target sample feature by the plurality of initial classifiers, and the model adjustment component includes: a first matching component configured to match the plurality of initial classification results with the preset driving style to obtain a plurality of first matching results; a classifier adjustment component configured to adjust the initial classifiers based on the plurality of first matching results to obtain a plurality of style classifiers; a second matching component configured to match a plurality of style classification results with the preset driving style to obtain a plurality of second matching results, wherein the style classification result is used to represent a classification result obtained by inputting the target sample feature into the style classifier; a weight determination component configured to determine a classification weight of the style classifier relative to the preset driving style based on the plurality of second matching results; and a model construction component configured to construct a style identification model based on the classification weight and the plurality of style classifiers.

[0079] Embodiment 3

[0080] Embodiments of the present disclosure also provide a vehicle, including: a memory storing an executable program; and a processor configured to run the program, wherein the program performs the method in various embodiments of the present disclosure when running.

[0081] Embodiment 4

[0082] Embodiments of the present disclosure also provide a computer-readable storage medium, which includes a stored executable program, wherein the computer-readable storage medium controls a device where the computer-readable storage medium is located to perform the method in various embodiments of the present disclosure when the executable program runs.

[0083] Embodiment 5

[0084] Embodiments of the present disclosure also provide a computer program product, which includes a computer program, and the computer program implements the method in various embodiments of the present disclosure when executed by a processor.

[0085] Embodiment 6

[0086] The embodiment of the present disclosure further provides a computer program product comprising a non-volatile computer readable storage medium for storing a computer program, which, when executed by a processor, implements the method in each of the embodiments of the present disclosure.

[0087] Embodiment 7

[0088] The embodiment of the present disclosure further provides a computer program, which, when executed by a processor, implements the method in each of the embodiments of the present disclosure.

[0089] The above-mentioned serial numbers of the embodiments of the present disclosure are only for description, and do not represent the advantages or disadvantages of the embodiments.

[0090] In the above-mentioned embodiments of the present disclosure, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0091] In the several embodiments of the present disclosure, it should be understood that the disclosed technology can be implemented in other ways. Of course, the embodiment described above is only a schematic, for example, the division of components can be a logical function division, and there can be another division manner in actual implementation, for example, a plurality of components or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed components can be through some interface, indirect coupling or communication connection between components or components, and can be electrical or other forms.

[0092] The components described as separate components can or can not be physically separate, and the components shown as components can or can not be physical components, that is, they can be located in one place, or they can be distributed on multiple components. According to actual needs, part or all of the components can be selected to achieve the purpose of the present embodiment.

[0093] In addition, each functional component in each embodiment of the present disclosure can be integrated in one processing component, or each component can exist physically, or two or more components can be integrated in one component. The above integrated component can be realized in the form of hardware or software functional component.

[0094] The integrated components, if implemented in the form of software function components and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present disclosure, essentially or in the part that contributes to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of the present disclosure. The aforementioned storage medium includes various media that can store program codes, such as a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and the like.

[0095] The above are only preferred embodiments of the present disclosure, and it should be pointed out that, for those skilled in the art, without departing from the principles of the present disclosure, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present disclosure. Industrial applicability

[0096] The scheme provided by the embodiments of the present disclosure can be applied to determining the driving style of a user, to assist the user in driving a vehicle. By using multiple style classifiers configured in the style recognition model to classify the driving features of the user respectively, and then according to the classification weights between different style classifiers and different style types, the classification results output by the style classifiers are summarized to obtain a user driving style with high matching degree for the user, the accuracy of the determined user driving style can be effectively improved, the effect of using the driving parameters corresponding to the user driving style to assist the user in driving the vehicle can be improved, thereby improving the driving experience of the user, and further solving the technical problem that the effect of assisting the user in driving the vehicle according to the driving style of the user is poor in the related art.

Claims

1. A vehicle control method, comprising: obtaining historical driving parameters of a vehicle driven by a user in a historical time period; extracting features of the historical driving parameters to obtain user driving features; identifying the user driving features based on a style recognition model to obtain a user driving style of the user, wherein the style recognition model comprises a plurality of style classifiers, different style classifiers have different classification weights for classifying different preset driving styles, the preset driving styles include the user driving style, the user driving style is used to represent a style obtained by aggregating a plurality of classification results of the user driving features classified by the style classifiers based on the classification weights, and the classification results are used to represent matching degrees between the user driving features and different preset driving styles under different style classifiers; in response to detecting that the user drives the vehicle, controlling the vehicle to operate based on the user driving style to assist the user in driving the vehicle.

2. The method of claim 1, wherein, The identification of the user driving features based on the style recognition model to obtain the user driving style comprises: performing importance evaluation on the user driving features according to a preset evaluation manner to obtain a first importance indicator of the user driving features for the historical driving parameters; selecting at least one target driving feature from the user driving features based on the first importance indicator; inputting the target driving feature into the style recognition model to obtain the user driving style.

3. The method of claim 2, wherein, The importance evaluation on the user driving features according to the preset evaluation manner to obtain the first importance indicator of the user driving features for the historical driving parameters comprises: performing data alignment on the user driving features and the historical driving parameters to obtain a first observation value corresponding to the user driving features and a second observation value corresponding to the historical driving parameters; determining a correlation coefficient between the user driving features and the historical driving parameters based on the first observation value and the second observation value; inputting the user driving features into an ensemble learning model based on the correlation coefficient to obtain the first importance indicator.

4. The method of claim 1, wherein, The control of the vehicle to operate based on the user driving style comprises: obtaining vehicle control parameters matched with the user driving style from a parameter library, wherein the parameter library is used to store an association relationship between the user driving style and the vehicle control parameters; outputting the vehicle control parameters in an operation interface and receiving a parameter adjustment instruction fed back by the user; adjusting the vehicle control parameters based on the parameter adjustment instruction to obtain target control parameters; controlling the vehicle to operate based on the target control parameters.

5. The method of claim 1, wherein, The identification of the user driving features based on the style recognition model to obtain the user driving style of the user comprises: dividing the plurality of classification results based on the preset driving styles to obtain a plurality of classification result sets, wherein the classification results included in a same classification result set correspond to the same preset driving style. merge the classification results in each of the classification result sets based on classification weights corresponding to the classification results in the classification result sets, to obtain a target classification result, wherein the target classification result is used to represent a matching degree between the user driving feature and the preset driving style corresponding to the classification result sets; determine the user driving style from the preset driving styles based on the target classification result.

6. The method of claim 1, wherein, The method further comprises: obtaining sample driving parameters corresponding to at least one preset driving style; extracting features of the sample driving parameters to obtain at least one sample driving feature; inputting the sample driving features into an initial identification model to obtain an initial identification result corresponding to the sample driving parameters; adjusting the initial identification model based on the initial identification result and the preset driving style to obtain the style identification model.

7. The method of claim 6, wherein, inputting the sample driving features into an initial identification model to obtain an initial identification result corresponding to the sample driving parameters, comprising: performing importance evaluation on the sample driving features according to a preset evaluation method to obtain a second importance index of the sample driving features for the sample driving parameters; sorting the second importance index and selecting at least one target sample feature from the sample driving features based on a sorting result, wherein the second importance index corresponding to the target sample feature is greater than the second importance index corresponding to other sample features in the sample driving features except the target sample feature; inputting the target sample feature into the initial identification model to obtain the initial identification result.

8. The method of claim 6, wherein, The initial identification model at least includes a plurality of initial classifiers, and the initial identification result includes a plurality of initial classification results obtained by classifying the target sample feature by the plurality of initial classifiers. Adjusting the initial identification model based on the initial identification result and the preset driving style to obtain the style identification model, comprising: matching the plurality of initial classification results with the preset driving style to obtain a plurality of first matching results; adjusting the initial classifiers based on the plurality of first matching results to obtain a plurality of style classifiers; matching a plurality of style classification results with the preset driving style to obtain a plurality of second matching results, wherein the style classification result is used to represent a classification result obtained by inputting the target sample feature into the style classifier; determining classification weights of the style classifiers relative to the preset driving style based on the plurality of second matching results; constructing the style identification model based on the classification weights and the plurality of style classifiers.

9. A vehicle control device, comprising: a parameter acquisition component configured to obtain historical driving parameters of a vehicle driven by a user in a historical time period; a feature extraction component configured to extract features of the historical driving parameters to obtain a user driving feature; The style recognition component is configured to identify the user driving feature based on a style recognition model to obtain a user driving style of the user, wherein the style recognition model comprises a plurality of style classifiers, different style classifiers have different classification weights for classifying different preset driving styles, the preset driving styles include the user driving style, the user driving style is used to represent a style obtained by aggregating a plurality of classification results of the user driving feature classified by the style classifiers based on the classification weights, and the classification results are used to represent matching degrees between the user driving feature and different preset driving styles under different style classifiers. The vehicle control component is configured to control the vehicle to operate based on the user driving style to assist the user in driving the vehicle in response to detecting that the user drives the vehicle.

10. A vehicle comprising: a memory storing an executable program; a processor configured to run the program, wherein the program performs the method of any one of claims 1 to 8 when running.

11. A computer readable storage medium, the computer readable storage medium comprising a stored program, wherein, The device is controlled to perform the following method when the program is running: obtain historical driving parameters of a vehicle driven by a user during driving in a historical time period; extract features from the historical driving parameters to obtain user driving features; identify the user driving features based on a style recognition model to obtain a user driving style of the user, wherein the style recognition model comprises a plurality of style classifiers, different style classifiers have different classification weights for classifying different preset driving styles, the preset driving styles include the user driving style, the user driving style is used to represent a style obtained by aggregating a plurality of classification results of the user driving feature classified by the style classifiers based on the classification weights, and the classification results are used to represent matching degrees between the user driving feature and different preset driving styles under different style classifiers; control the vehicle to operate based on the user driving style to assist the user in driving the vehicle in response to detecting that the user drives the vehicle.

12. The storage medium of claim 11, wherein, The device is further controlled to perform the following method when the program is running: evaluate the importance of the user driving features according to a preset evaluation method to obtain a first importance indicator of the user driving features for the historical driving parameters; select at least one target driving feature from the user driving features based on the first importance indicator; input the target driving feature into the style recognition model to obtain the user driving style.

13. The storage medium of claim 12, wherein, The device is further controlled to perform the following method when the program is running: align the user driving features and the historical driving parameters to obtain first observation values corresponding to the user driving features and second observation values corresponding to the historical driving parameters; determine a correlation coefficient between the user driving features and the historical driving parameters based on the first observation values and the second observation values; input the user driving feature into an ensemble learning model based on the correlation coefficient to obtain the first importance index.

14. The storage medium of claim 11, wherein, The device where the storage medium is located also performs the following method when the program is running: obtain vehicle control parameters matched with the user driving style from a parameter library, wherein the parameter library is used to store an association between the user driving style and the vehicle control parameters; output the vehicle control parameters in an operation interface and receive a parameter adjustment instruction fed back by the user; adjust the vehicle control parameters based on the parameter adjustment instruction to obtain target control parameters; control the vehicle to operate based on the target control parameters.

15. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images