A method, apparatus, vehicle, and storage medium for determining driving style

Abstract

This invention discloses a method, apparatus, vehicle, and storage medium for determining driving style. The method includes: acquiring kinematic segments based on vehicle driving data; acquiring a first feature parameter representing lateral driving style and a second feature parameter representing longitudinal driving style from the kinematic segments; performing dimensionality reduction processing on the first and second feature parameters respectively to obtain multiple principal component score matrices; clustering the kinematic segments based on each principal component score matrix to obtain multiple cluster centers; and determining the driver's driving style during vehicle steering, acceleration, or deceleration based on the feature parameters corresponding to each cluster center. This embodiment determines the parameters corresponding to lateral and longitudinal driving styles from the kinematic segments and uses the principal component score matrices to cluster the kinematic segments, thereby determining the driver's driving style in different dimensions and improving the intelligence of vehicle driving style judgment.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, device, vehicle, and storage medium for determining driving style. Background Technology

[0002] Driving style can be used as a holistic evaluation index to characterize a driver's inherent driving behavior. Due to differences in factors such as gender, age, and personality, there are certain regular driving behavior tendencies, resulting in significant differences in drivers' driving styles. For the same driver, their driving style will also vary depending on the working conditions or mood.

[0003] Most existing methods for identifying and judging driving styles take into account the driver's overall performance during driving, including changes in driving behavior and the degree of aggressiveness in handling, but lack analysis of different dimensions and have certain limitations. Summary of the Invention

[0004] This invention provides a method, device, vehicle, and storage medium for determining driving style, in order to address the limitations of existing technologies in determining driving style.

[0005] According to one aspect of the present invention, a method for determining driving style is provided, comprising:

[0006] Based on vehicle driving data, kinematic segments are obtained, and a first feature parameter representing lateral driving style and a second feature parameter representing longitudinal driving style are obtained from the kinematic segments.

[0007] The first feature parameter and the second feature parameter are respectively subjected to dimensionality reduction processing to obtain multiple principal component score matrices;

[0008] The kinematic segments are clustered based on the principal component score matrix to obtain multiple cluster centers. The driving style of the driver when the vehicle is turning, accelerating or decelerating is determined based on the feature parameters corresponding to each cluster center.

[0009] According to another aspect of the present invention, a driving style determination device is provided, comprising:

[0010] The feature parameter acquisition module is used to acquire kinematic segments based on vehicle driving data, and to acquire a first feature parameter representing lateral driving style and a second feature parameter representing longitudinal driving style in the kinematic segments.

[0011] The feature parameter dimensionality reduction module is used to perform dimensionality reduction processing on the first feature parameter and the second feature parameter respectively to obtain multiple principal component score matrices;

[0012] The driving style determination module is used to cluster the kinematic segments according to each principal component score matrix to obtain multiple cluster centers, and to determine the driver's driving style when the vehicle is turning, accelerating or decelerating according to the feature parameters corresponding to each cluster center.

[0013] According to another aspect of the present invention, a vehicle is provided, the vehicle comprising:

[0014] Onboard sensors are used to collect vehicle driving data;

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the driving style determination method according to any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the driving style determination method according to any embodiment of the present invention.

[0019] The technical solution of this invention determines the parameters corresponding to lateral and longitudinal driving styles from kinematic segments, and clusters the kinematic segments using the principal component score matrix, thereby determining the driver's driving style in different dimensions. This solves the problem of limitations in the determination of driving style in the prior art and improves the intelligence of vehicle driving style judgment.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a driving style determination method provided in Embodiment 1 of the present invention;

[0023] Figure 2This is a flowchart of another driving style determination method provided in Embodiment 2 of the present invention;

[0024] Figure 3 This is a flowchart of another driving style determination method provided in Embodiment 2 of the present invention.

[0025] Figure 4 This is a schematic diagram of a driving style determination device according to Embodiment 3 of the present invention;

[0026] Figure 5 This is a structural schematic diagram of a vehicle that implements the driving style determination method of this invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1 This is a flowchart of a driving style determination method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where the driver's driving style is obtained. This method can be executed by a driving style determination device, which can be implemented in hardware and / or software and can be integrated into a vehicle. Figure 1 As shown, the method includes:

[0031] S110. Obtain kinematic segments based on vehicle driving data, and obtain the first feature parameter representing the lateral driving style and the second feature parameter representing the longitudinal driving style in the kinematic segments.

[0032] Among them, vehicle driving data can be vehicle driving data collected by various sensors when the vehicle is in operation. Vehicle driving data can include, but is not limited to, various data such as time, speed, acceleration and surrounding environment information; kinematic segment can refer to a group of vehicle driving data from the start of the vehicle's idle state to the start of the next idle state. Specifically, the starting point of an idle state can be used as the starting point of a historical kinematic segment, and the starting point of the next idle state can be used as the ending point of the kinematic segment; where idle state can refer to a working condition when the engine is idling.

[0033] In this embodiment, kinematic segments can be divided based on the collected vehicle driving data. The kinematic segments can include feature parameters that characterize lateral driving style and longitudinal driving style, as shown in Table 1 below:

[0034] Table 1. Kinematic segment characteristic parameters

[0035]

[0036]

[0037] Some of the parameters in Table 1 above can be used to characterize lateral driving style, as shown in Table 2 below:

[0038] Table 2. Characteristic Parameters of Lateral Driving Style

[0039] Feature parameters symbol unit average speed <![CDATA[v a ]]> km / h Maximum speed <![CDATA[v max ]]> km / h speed standard deviation <![CDATA[S v ]]> km / h Mean lateral acceleration <![CDATA[l a ]]> <![CDATA[m / s 2 ]]> Maximum lateral acceleration <![CDATA[l max ]]> <![CDATA[m / s 2 ]]> lateral acceleration standard deviation <![CDATA[S l ]]> <![CDATA[m / s 2 ]]> Mean steering wheel angle acceleration <![CDATA[ω a ]]> rad / s Steering wheel angular acceleration standard deviation <![CDATA[S ω ]]> <![CDATA[rad / s 2 ]]>

[0040] Furthermore, some parameters in Table 1 above can be used to characterize longitudinal driving style, which can be divided into driving acceleration style and driving deceleration style. Specifically, driving acceleration style can be represented by the acceleration style characteristic parameter table shown in Table 3 below:

[0041] Table 3 Acceleration Style Characteristic Parameters

[0042]

[0043]

[0044] The driving deceleration style can be represented by the deceleration style characteristic parameters shown in Table 4 below:

[0045] Table 4. Characteristic Parameters of Deceleration Style

[0046] Feature parameters symbol unit average speed <![CDATA[v a ]]> km / h Maximum speed <![CDATA[v max ]]> km / h speed standard deviation <![CDATA[S v ]]> km / h Mean longitudinal deceleration <![CDATA[d a ]]> <![CDATA[m / s 2 ]]> Maximum longitudinal deceleration <![CDATA[d max ]]> <![CDATA[m / s 2 ]]> Longitudinal deceleration standard deviation <![CDATA[S d ]]> <![CDATA[m / s 2 ]]> Average brake pedal opening <![CDATA[b a ]]> ％ Brake pedal opening standard deviation <![CDATA[S b ]]> ％

[0047] It can be seen that the first characteristic parameter used to characterize lateral driving style can be average speed, maximum speed, speed standard deviation, average lateral acceleration, maximum lateral acceleration, lateral acceleration standard deviation, average steering wheel angle acceleration, and steering wheel angle acceleration standard deviation; the first characteristic parameter used to characterize longitudinal driving style can be average speed, maximum speed, speed standard deviation, average longitudinal acceleration, maximum longitudinal acceleration, longitudinal acceleration standard deviation, average accelerator pedal opening, accelerator pedal opening standard deviation, average longitudinal deceleration, maximum longitudinal deceleration, longitudinal deceleration standard deviation, average brake pedal opening, and brake pedal opening standard deviation.

[0048] S120. Dimensionality reduction is performed on the first and second feature parameters respectively to obtain multiple principal component score matrices.

[0049] Dimensionality reduction can be achieved by using principal component analysis (PCA) to reduce the dimensionality of the first and second characteristic parameters. Specifically, PCA is a statistical method used for data dimensionality reduction. During data processing, it extracts potentially correlated variables and transforms them into a smaller, linearly uncorrelated set of variables through orthogonal transformation. This set replaces the original large number and variety of variables, while preserving as much information as possible from the original data.

[0050] In this embodiment, the principal component score matrices corresponding to the steering, acceleration, and deceleration dimensions are obtained by performing dimensionality reduction processing on the first feature parameter and the second feature parameter, respectively.

[0051] S130. Cluster the kinematic segments according to each principal component score matrix to obtain multiple cluster centers. Determine the driver's driving style when the vehicle is turning, accelerating or decelerating based on the feature parameters corresponding to each cluster center.

[0052] In this embodiment, the cluster center can be understood as a set of feature parameter data. Driving style can reflect the intensity of the driver's acceleration / deceleration / steering operations. Therefore, the driving style can be determined based on the driver's steering, acceleration and deceleration behaviors.

[0053] Clustering the principal component score matrix under each different dimension yields three cluster centers. For example, for the principal component score matrix under the turning dimension, the resulting cluster centers can be classified as aggressive, general, and robust. Similarly, for the principal component score matrices under the acceleration and deceleration dimensions, the resulting cluster centers can also be classified as aggressive, general, and robust. The feature parameters corresponding to each cluster center can be the feature parameters corresponding to the aggressive, general, and robust clusters, respectively.

[0054] Optionally, the clustering algorithm used in this embodiment may be the K-Means clustering algorithm, and this embodiment does not impose any specific restrictions on it.

[0055] This embodiment determines the parameters corresponding to lateral and longitudinal driving styles from kinematic segments, and uses the principal component score matrix to cluster the kinematic segments, thereby determining the driver's driving style in different dimensions. This solves the problem of limitations in the determination of driving style in the prior art and improves the intelligence and humanization of vehicle driving style judgment.

[0056] Example 2

[0057] Figure 2 This is a flowchart of a driving style determination method provided in Embodiment 2 of the present invention. This embodiment further optimizes the above step of "performing dimensionality reduction on the first feature parameter and the second feature parameter respectively to obtain multiple principal component score matrices." For example... Figure 2 As shown, the method includes:

[0058] S210. Obtain kinematic segments based on vehicle driving data, and obtain the first feature parameter representing the lateral driving style and the second feature parameter representing the longitudinal driving style in the kinematic segments.

[0059] In this embodiment, the kinematic segment time T is calculated as follows:

[0060] T = N (2.1);

[0061] Where N can be used to represent the number of data points in each kinematic segment. Since the sampling frequency is 1Hz, that is, one set of data per second, the duration of each kinematic segment can be represented by the number of data points contained in the corresponding segment.

[0062] In this embodiment, for the average velocity v among the various characteristic parameters a The calculation method can be shown below:

[0063]

[0064] Among them, v i The vehicle speed, used to represent the speed at any given moment, is collected by the vehicle's sensors.

[0065] In this embodiment, for the maximum speed v among the various characteristic parameters max The calculation method can be shown below:

[0066] v max =max{v i}, where i = 1, 2, 3, ..., N (2.3);

[0067] In this embodiment, the calculation method for the lateral and longitudinal accelerations among the various characteristic parameters can be achieved by first obtaining the longitudinal acceleration 'a' of the vehicle at each moment through vehicle sensors. i and lateral acceleration l i .

[0068] In this embodiment, the velocity standard deviation S among the various characteristic parameters... v The calculation method can be shown below:

[0069]

[0070] In this embodiment, for the average longitudinal acceleration a among the various characteristic parameters a The calculation method can be shown below:

[0071]

[0072] Among them, a i1 This can be the instantaneous longitudinal acceleration value of the vehicle at that moment, T. a This can be the duration of acceleration during vehicle movement (to avoid interference from low-speed driving segment data on clustering, only the longitudinal acceleration value 'a' can be calculated). i At 0.15m / s 2 The above, and T a (Data exceeding 3 seconds).

[0073] In this embodiment, for the maximum longitudinal acceleration a among the various characteristic parameters max The calculation method can be shown below:

[0074] a max =max{a i}, where i = 1, 2, 3, ..., N (2.6);

[0075] In this embodiment, for the longitudinal acceleration standard deviation S among the various characteristic parameters a The calculation method can be shown below:

[0076]

[0077] In this embodiment, for the average longitudinal deceleration d among the various characteristic parameters a The calculation method can be shown below:

[0078]

[0079] Among them, a i2 T represents the instantaneous longitudinal deceleration of the vehicle at that moment. dThe duration of deceleration during vehicle movement (to avoid interference from low-speed road segment data on clustering, only the longitudinal deceleration value 'a' is calculated). i At -0.15m / s 2 The above, and T d (Data exceeding 3 seconds).

[0080] In this embodiment, for the maximum longitudinal deceleration d among the various characteristic parameters max The calculation method can be shown below:

[0081] d max =min{a i}, where i = 1, 2, 3, ..., N (2.9);

[0082] In this embodiment, for the longitudinal deceleration standard deviation S among the various characteristic parameters d The calculation method can be shown below:

[0083]

[0084] In this embodiment, for the average lateral acceleration l among the various characteristic parameters a The calculation method can be shown below:

[0085]

[0086] In this embodiment, for the maximum lateral acceleration l among the various characteristic parameters max The calculation method can be shown below:

[0087] l max =max{l i}, where i = 1, 2, 3, ..., N (2.12);

[0088] In this embodiment, for the lateral acceleration standard deviation S among the various characteristic parameters l The calculation method can be shown below:

[0089]

[0090] In this embodiment, for the average accelerator pedal opening t among the various characteristic parameters a The calculation method can be shown below:

[0091]

[0092] Among them, t i This can be used to determine the accelerator pedal opening of the vehicle at that moment, T t This can be the total time during which the accelerator pedal is triggered in this kinematic segment.

[0093] In this embodiment, the standard deviation S of the accelerator pedal opening among the various characteristic parameters is... t The calculation method can be shown below:

[0094]

[0095] In this embodiment, for the average brake pedal opening b among the various characteristic parameters a The calculation method can be shown below:

[0096]

[0097] Among them, b i T represents the brake pedal opening of the vehicle at that moment. b This represents the total time during which the brake pedal is triggered in this kinematic segment.

[0098] In this embodiment, the standard deviation S of the brake pedal opening among the various characteristic parameters is... b The calculation method can be shown below:

[0099]

[0100] In this embodiment, the average steering wheel angular acceleration ω among the various characteristic parameters... a The calculation method can be shown below:

[0101]

[0102] Where, ω i T represents the vehicle's steering wheel angular acceleration at that moment, collected by the vehicle's sensors. ω This represents the total time taken for the vehicle to perform a steering maneuver within this kinematic segment.

[0103] In this embodiment, the standard deviation S of steering wheel angular acceleration among the various characteristic parameters is... ω The calculation method can be shown below:

[0104]

[0105] S220. The first feature parameter is reduced in dimensionality using the principal component analysis algorithm to obtain the first principal component score matrix.

[0106] This embodiment uses principal component analysis (PCA) to reduce the dimensionality of the first characteristic parameter representing lateral driving style and the second characteristic parameter representing longitudinal driving style in the aforementioned kinematic segment. During the calculation, the new variable with the largest variance in all linear combinations of steering, acceleration, and deceleration can be defined as the first principal component M1. If the first principal component cannot adequately reflect the information contained in the original data, the remaining new variables with the largest variance in the linear combinations are selected as the second principal component M2, and so on. When the cumulative contribution rate of the selected principal components exceeds 85%, the new variables can be considered to have relatively completely reflected the information in the original data.

[0107] The steps of the principal component analysis algorithm are as follows:

[0108] First, the original data needs to be standardized. Since the original data have different units, if the original data is analyzed directly without processing, the results will have a large degree of dispersion, which obviously has a certain impact on classification. Therefore, the original data needs to be standardized to eliminate this impact. Specifically, the original data matrix can be converted into a standardized matrix according to a certain method, so that the mean of each column of the matrix is ​​0 and the variance is 1.

[0109] Suppose the original data matrix is ​​X (the elements in the matrix are feature parameters), where p is the number of kinematic segments and n is the number of feature parameters;

[0110]

[0111] Standardize matrix X to obtain standardized matrix Y. The standardization process is shown in the following formula:

[0112]

[0113]

[0114]

[0115] Where r = 1, 2, 3...p, j = 1, 2, 3...n;

[0116]

[0117] Next, calculate the correlation coefficient and its corresponding correlation coefficient matrix:

[0118] This embodiment uses the following formula to calculate the correlation coefficient:

[0119]

[0120] The corresponding correlation coefficient matrix can be:

[0121]

[0122] Third, determine the eigenvectors and eigenvalues ​​of the correlation coefficient matrix based on the correlation coefficient matrix. Specifically, find the eigenvalues ​​λ of the correlation coefficient matrix, arrange the eigenvalues ​​in descending order as λ1>λ2>λ3>……>λn, and find the eigenvectors corresponding to the eigenvalues ​​as ξr and ξrj, where ξrj represents the j-th component of the eigenvector ξr.

[0123] Fourth, calculate the contribution rate of the principal components. Specifically, the formula for calculating the contribution rate of the j-th principal component is shown in Formula 2.24:

[0124]

[0125] Calculate the contribution rate of each principal component according to Formula 3.24 above, and find the cumulative contribution rate of each principal component. When the cumulative contribution rate of the current m principal components reaches more than 85%, it can be considered that the information of the original data has been preserved relatively completely, and these principal components are selected as the basis for classification.

[0126] Fifth, calculate the principal component loading matrix Z. Specifically, calculate it using the following formula:

[0127]

[0128] Where j = 1, 2, 3...n, q = 1, 2, 3...m;

[0129]

[0130] Sixth, multiply the standardized feature parameter matrix by the principal component loading matrix to obtain the principal component score matrix. Specifically, the principal component score matrix is ​​shown below:

[0131]

[0132] In this embodiment, the first feature parameter can be reduced in dimensionality using the principal component analysis algorithm described above to obtain the first principal component score matrix.

[0133] S230. The second feature parameter representing vehicle acceleration is reduced in dimensionality using the principal component analysis algorithm to obtain the second principal component score matrix.

[0134] In this embodiment, the second feature parameter representing vehicle acceleration can be reduced in dimensionality using the principal component analysis algorithm described above, to obtain the second principal component score matrix.

[0135] S240. The principal component analysis algorithm is used to reduce the dimensionality of the second characteristic parameter representing vehicle deceleration, and the third principal component score matrix is ​​obtained.

[0136] In this embodiment, the principal component analysis algorithm described above can be used to reduce the dimensionality of the second feature parameter representing vehicle deceleration, thereby obtaining the third principal component score matrix.

[0137] S250. Cluster the kinematic segments according to each principal component score matrix to obtain multiple cluster centers. Determine the driver's driving style when the vehicle is turning, accelerating or decelerating based on the feature parameters corresponding to each cluster center.

[0138] In this embodiment, the K-means clustering algorithm can be used to cluster the processed feature parameters to obtain multiple cluster centers.

[0139] In this embodiment, the steps for clustering kinematic segments based on each principal component score matrix to obtain multiple cluster centers are as follows:

[0140] 1. Depending on the actual situation, we can assume that the number of clusters K = 2 or 3, and randomly set an initial cluster center for each category.

[0141] 2. Calculate the Euclidean distance between each kinematic segment and the cluster center using the following formula (2.26):

[0142]

[0143] Where xik represents the k-th variable of the i-th segment, each segment has p variables (determined by the principal component analysis algorithm), and y represents the selected cluster center, which is a randomly selected set of data in the first calculation, also containing p variables.

[0144] 3. Group the data that are close together into one class, calculate the center of each class, and set it as the new cluster center.

[0145] 4. Based on the new cluster centers, repeat steps 2-3 above. By iteratively calculating, continuously determine the new cluster centers until the cluster centers no longer change.

[0146] 5. Based on the value of the cluster center, the clusters are classified into three types from largest to smallest: aggressive, general, and robust.

[0147] In this embodiment, the driver's driving style is determined based on the characteristic parameters corresponding to each cluster center, namely, aggressive, normal, and conservative, when the vehicle is turning, accelerating, or decelerating.

[0148] For example, determining the driver's driving style when the vehicle is turning based on the feature parameters corresponding to each cluster center can be done by determining the driving style of each first cluster center when the vehicle is turning based on the value of each first cluster center corresponding to the first principal component score matrix; determining the number of kinematic segments corresponding to each driving style when the vehicle is turning based on the driving style of each first cluster center when the vehicle is turning; and determining the driver's driving style when the vehicle is turning based on the proportion of the number of kinematic segments corresponding to each driving style when the vehicle is turning in the number of kinematic segments corresponding to the first cluster center.

[0149] The first cluster center can be any kinematic segment in the cluster corresponding to the first cluster center, and these kinematic segments can be distributed in the vicinity of the first cluster center.

[0150] In this embodiment, after determining the value of the first cluster center, the specific style of the driver when performing steering behavior can be determined based on the proportion of the number of kinematic segments corresponding to each driving style in the steering dimension of the kinematic segments generated by the driver.

[0151] In this embodiment, if we assume that the proportion of aggressive kinematic segments is 40%, the proportion of normal kinematic segments is 30%, and the proportion of conservative kinematic segments is 30%, then we can determine that the driver's driving style when the vehicle is turning is aggressive. This embodiment does not impose specific limitations on this.

[0152] For example, determining the driving style of each first cluster center when the vehicle is turning, based on the value of each first cluster center corresponding to the first principal component score matrix, may include: determining the value of each first cluster center based on the average of the mean lateral acceleration, maximum lateral acceleration, and standard deviation of lateral acceleration of each first cluster center corresponding to the first principal component score matrix; sorting the first cluster centers in descending order based on the values; and determining the driving style of each first cluster center when the vehicle is turning based on the sorting result.

[0153] In this embodiment, since the selected indicators reflect the intensity of the driver's acceleration / deceleration / steering operations, the rule for each characteristic parameter should be: aggressive type > general type > robust type. That is, the value of the aggressive type is greater than that of the general type, and the value of the general type is greater than that of the robust type.

[0154] Specifically, for the driving style when the vehicle is turning, the average value of the average lateral acceleration, the maximum lateral acceleration, and the standard deviation of the lateral acceleration can be selected. The average value is sorted in descending order from large to small, and the corresponding first cluster centers are divided into aggressive, general, and robust types according to the pattern of the feature parameters.

[0155] For example, based on the value of each second cluster center corresponding to the second principal component score matrix, the driving style of each second cluster center during vehicle acceleration is determined; based on the driving style of each second cluster center during vehicle acceleration, the number of kinematic segments corresponding to each driving style during vehicle acceleration is determined; based on the proportion of the number of kinematic segments corresponding to each driving style during vehicle acceleration in the number of kinematic segments corresponding to the second cluster center, the driver's driving style during vehicle acceleration is determined.

[0156] In this embodiment, it can be assumed that when the vehicle accelerates, the proportion of aggressive kinematic segments is 30%, the proportion of normal kinematic segments is 30%, and the proportion of conservative kinematic segments is 40%. Then it can be determined that the driver's driving style when the vehicle accelerates is conservative.

[0157] For example, determining the driving style of each second cluster center during vehicle acceleration based on the value of each second cluster center corresponding to the second principal component score matrix may include: determining the value of each second cluster center based on the average of the mean longitudinal acceleration, maximum longitudinal acceleration, and standard deviation of longitudinal acceleration of each second cluster center corresponding to the second principal component score matrix; sorting the second cluster centers in descending order based on the values; and determining the driving style of each second cluster center during vehicle acceleration based on the sorting result.

[0158] For the acceleration driving style, the mean of the average longitudinal acceleration, maximum longitudinal acceleration, and standard deviation of longitudinal acceleration of each data cluster center can be calculated. According to the mean from large to small, the corresponding second cluster centers are divided into aggressive, general, and robust types according to the pattern of the feature parameters.

[0159] For example, determining the driver's driving style during vehicle deceleration based on the feature parameters corresponding to each cluster center may include: determining the driving style of each third cluster center during vehicle deceleration based on the value of each third cluster center corresponding to the third principal component score matrix; determining the number of kinematic segments corresponding to each driving style during vehicle deceleration based on the driving style of each third cluster center during vehicle deceleration; and determining the driver's driving style during vehicle deceleration based on the proportion of the number of kinematic segments corresponding to each driving style during vehicle deceleration to the number of kinematic segments corresponding to the third cluster center.

[0160] In this embodiment, it can be assumed that when the vehicle decelerates, the proportion of aggressive kinematic segments is 30%, the proportion of normal kinematic segments is 30%, and the proportion of conservative kinematic segments is 40%. Then it can be determined that the driver's driving style when the vehicle decelerates is conservative.

[0161] For example, determining the driving style of each third cluster center during vehicle deceleration based on the values ​​of each third cluster center corresponding to the third principal component score matrix may include: determining the value of each third cluster center based on the average of the mean longitudinal deceleration, the maximum longitudinal deceleration, and the standard deviation of the longitudinal deceleration of each third cluster center corresponding to the third principal component score matrix; sorting the third cluster centers in descending order based on the values; and determining the driving style of each third cluster center during vehicle deceleration based on the sorting result.

[0162] For deceleration driving styles, the average of the mean longitudinal deceleration, maximum longitudinal deceleration, and standard deviation of longitudinal deceleration (all absolute values) can be calculated. Based on the mean from largest to smallest, the corresponding third cluster centers are divided into aggressive, general, and robust types according to the pattern of the characteristic parameters.

[0163] For example, in this embodiment, the driving style when the vehicle is turning can be input into the steering system, and the steering system can adjust the driving mode and / or driving parameters based on the driving style when the vehicle is turning; the driving style when the vehicle is accelerating can be input into the power system, and the power system can adjust the driving mode and / or driving parameters based on the driving style when the vehicle is accelerating; the driving style when the vehicle is decelerating can be input into the braking system, and the braking system can adjust the driving mode and / or driving parameters based on the driving style when the vehicle is decelerating.

[0164] In this embodiment, the determination result of driving style can be used as an important basis for adaptive adjustment of driving mode and / or driving parameters as well as related parameters of power, braking and steering systems.

[0165] One method for determining driving style in this embodiment can be as follows: Figure 3 As shown, this demonstrates how kinematic segments are obtained from vehicle driving data, and how eight feature parameters are selected based on these kinematic segments to correspond to different driving styles: lateral driving style, longitudinal acceleration driving style, and longitudinal deceleration driving style. Data dimensionality reduction and K-means clustering are then performed on these three types of feature parameters to determine each driving style type. The corresponding driving styles are then input into the steering system, powertrain system, and braking system, thereby achieving adaptive adjustment of driving modes and / or driving parameters.

[0166] It should be emphasized that this embodiment can set driving styles in three dimensions: steering, acceleration, and deceleration, and continuously accumulate new driving data as the driver's driving time increases, and iterate repeatedly to ensure the timeliness and long-term nature of the classification results.

[0167] This embodiment can determine the parameters corresponding to lateral and longitudinal driving styles from kinematic segments, and use the principal component score matrix to cluster the kinematic segments, thereby determining the driver's driving style in different dimensions, improving the intelligence and humanization of vehicle driving style judgment.

[0168] Example 3

[0169] Figure 4 This is a schematic diagram of a driving style determination device provided in Embodiment 3 of the present invention. Figure 4 As shown, the device includes: a feature parameter acquisition module 401, a feature parameter dimensionality reduction module 402, and a driving style determination module 403;

[0170] Among them, the feature parameter acquisition module 401 is used to acquire kinematic segments based on vehicle driving data, and acquire a first feature parameter representing lateral driving style and a second feature parameter representing longitudinal driving style in the kinematic segments;

[0171] The feature parameter dimensionality reduction module 402 is used to perform dimensionality reduction processing on the first feature parameter and the second feature parameter respectively to obtain multiple principal component score matrices.

[0172] The driving style determination module 403 is used to cluster the kinematic segments according to each principal component score matrix to obtain multiple cluster centers, and to determine the driver's driving style when the vehicle is turning, accelerating or decelerating according to the feature parameters corresponding to each cluster center.

[0173] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0174] The first feature parameter is reduced in dimensionality using principal component analysis algorithm to obtain the first principal component score matrix;

[0175] Principal component analysis algorithm is used to reduce the dimensionality of the second feature parameter representing vehicle acceleration to obtain the second principal component score matrix.

[0176] Principal component analysis (PCA) algorithm is used to reduce the dimensionality of the second feature parameter representing vehicle deceleration, resulting in the third principal component score matrix.

[0177] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0178] Based on the value of each first cluster center corresponding to the first principal component score matrix, determine the driving style of each first cluster center when the vehicle is turning;

[0179] Based on the driving style of each first cluster center when the vehicle is turning, determine the number of kinematic segments corresponding to each driving style when the vehicle is turning.

[0180] The driver's driving style during vehicle turning is determined by the proportion of the number of kinematic segments corresponding to each driving style in the total number of kinematic segments corresponding to the first cluster center.

[0181] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0182] The value of each first cluster center is determined based on the average of the mean lateral acceleration, the maximum lateral acceleration, and the standard deviation of the lateral acceleration for each first cluster center corresponding to the first principal component score matrix.

[0183] The first cluster centers are sorted in descending order based on the values, and the driving style of each first cluster center when the vehicle is turning is determined based on the sorting results.

[0184] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0185] Based on the value of each second cluster center corresponding to the second principal component score matrix, determine the driving style of each second cluster center during vehicle acceleration;

[0186] Based on the driving style of each second cluster center during vehicle acceleration, determine the number of kinematic segments corresponding to each driving style during vehicle acceleration.

[0187] The driver's driving style during vehicle acceleration is determined by the proportion of the number of kinematic segments corresponding to each driving style in the number of kinematic segments corresponding to the second cluster center.

[0188] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0189] The value of each second cluster center is determined based on the average of the mean longitudinal acceleration, the maximum longitudinal acceleration, and the average of the standard deviation of the longitudinal acceleration for each second cluster center corresponding to the second principal component score matrix.

[0190] The second cluster centers are sorted in descending order based on the values, and the driving style of each second cluster center during vehicle acceleration is determined based on the sorting results.

[0191] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0192] Based on the values ​​of each third cluster center corresponding to the third principal component score matrix, the driving style of each third cluster center during vehicle deceleration is determined;

[0193] Based on the driving style of each third cluster center during vehicle deceleration, determine the number of kinematic segments corresponding to each driving style during vehicle deceleration.

[0194] The driver's driving style during vehicle deceleration is determined by the proportion of the number of kinematic segments corresponding to each driving style in the number of kinematic segments corresponding to the third cluster center.

[0195] Optionally, the feature parameter dimensionality reduction module 402 is specifically used for:

[0196] The value of each third cluster center is determined based on the average of the mean longitudinal deceleration, the maximum longitudinal deceleration, and the average of the standard deviation of the longitudinal deceleration for each third cluster center corresponding to the third principal component score matrix.

[0197] The third cluster centers are sorted in descending order based on the values, and the driving style of each third cluster center during vehicle deceleration is determined based on the sorting results.

[0198] Optionally, the above-mentioned device further includes:

[0199] The steering module is used to input the driving style of the vehicle when turning into the steering system, and the steering system adjusts the driving mode and / or driving parameters based on the driving style of the vehicle when turning.

[0200] The power module is used to input the driving style of the vehicle during acceleration into the power system, and the power system adjusts the driving mode and / or driving parameters based on the driving style of the vehicle during acceleration.

[0201] The braking module is used to input the driving style of the vehicle during deceleration into the braking system, and the braking system adjusts the driving mode and / or driving parameters based on the driving style of the vehicle during deceleration.

[0202] The driving style determination device provided in this embodiment of the invention can execute a driving style determination method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0203] Example 4

[0204] Figure 5 A schematic diagram of the structure of a vehicle 10 that can be used to implement embodiments of the present invention is shown. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0205] like Figure 5As shown, vehicle 10 includes at least one processor 11 and a memory, such as read-only memory (ROM) 12 and random access memory (RAM) 13, communicatively connected to at least one processor 11. The memory stores computer programs executable by at least one processor. Processor 11 can perform various appropriate actions and processes based on the computer program stored in ROM 12 or loaded from storage unit 18 into RAM 13. RAM 13 can also store various programs and data required for the operation of vehicle 10. Processor 11, ROM 12, and RAM 13 are interconnected via bus 14. Input / output (I / O) interface 15 is also connected to bus 14.

[0206] Multiple components in vehicle 10 are connected to I / O interface 15, including: input unit 16, such as buttons and on-board sensors; wherein the on-board sensors are used to collect vehicle driving data; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disks, optical discs, etc.; and communication unit 19, such as network cards, modems, wireless transceivers, etc. The communication unit 19 allows vehicle 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0207] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a driving style determination method.

[0208] In some embodiments, a driving style determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on vehicle 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the driving style determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform a driving style determination method by any other suitable means (e.g., by means of firmware).

[0209] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0210] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0211] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0212] To provide interaction with the user, the systems and technologies described herein can be implemented in a vehicle having: a display device (e.g., a touchscreen) for displaying information to the user; and buttons through which the user can provide input to the vehicle. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including voice input, speech input, or tactile input).

[0213] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0214] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining driving style, characterized in that, include: Based on vehicle driving data, kinematic segments are obtained, and a first feature parameter representing lateral driving style and a second feature parameter representing longitudinal driving style are obtained from the kinematic segments. The first feature parameter and the second feature parameter are respectively subjected to dimensionality reduction processing to obtain multiple principal component score matrices; The kinematic segments are clustered based on the principal component score matrix to obtain multiple cluster centers. The driving style of the driver when the vehicle is turning, accelerating or decelerating is determined based on the feature parameters corresponding to each cluster center.

2. The method according to claim 1, characterized in that, The dimensionality reduction processing of the first feature parameter and the second feature parameter respectively yields multiple principal component score matrices, including: The first feature parameter is reduced in dimensionality using principal component analysis algorithm to obtain the first principal component score matrix; Principal component analysis algorithm is used to reduce the dimensionality of the second feature parameter representing vehicle acceleration to obtain the second principal component score matrix. Principal component analysis (PCA) algorithm is used to reduce the dimensionality of the second feature parameter representing vehicle deceleration, resulting in the third principal component score matrix.

3. The method according to claim 2, characterized in that, The driver's driving style during vehicle steering is determined based on the feature parameters corresponding to each cluster center, including: Based on the value of each first cluster center corresponding to the first principal component score matrix, determine the driving style of each first cluster center when the vehicle is turning; Based on the driving style of each first cluster center when the vehicle is turning, determine the number of kinematic segments corresponding to each driving style when the vehicle is turning. The driver's driving style during vehicle turning is determined by the proportion of the number of kinematic segments corresponding to each driving style in the total number of kinematic segments corresponding to the first cluster center.

4. The method according to claim 3, characterized in that, The step of determining the driving style of each first cluster center during vehicle steering based on the value of each first cluster center corresponding to the first principal component score matrix includes: The value of each first cluster center is determined based on the average of the mean lateral acceleration, the maximum lateral acceleration, and the standard deviation of the lateral acceleration for each first cluster center corresponding to the first principal component score matrix. The first cluster centers are sorted in descending order based on the values, and the driving style of each first cluster center when the vehicle is turning is determined based on the sorting results.

5. The method according to claim 2, characterized in that, The driver's driving style during vehicle acceleration is determined based on the feature parameters corresponding to each cluster center, including: Based on the value of each second cluster center corresponding to the second principal component score matrix, determine the driving style of each second cluster center during vehicle acceleration; Based on the driving style of each second cluster center during vehicle acceleration, determine the number of kinematic segments corresponding to each driving style during vehicle acceleration. The driver's driving style during vehicle acceleration is determined by the proportion of the number of kinematic segments corresponding to each driving style in the number of kinematic segments corresponding to the second cluster center.

6. The method according to claim 5, characterized in that, The step of determining the driving style of each second cluster center during vehicle acceleration based on the value of each second cluster center corresponding to the second principal component score matrix includes: The value of each second cluster center is determined based on the average of the mean longitudinal acceleration, the maximum longitudinal acceleration, and the average of the standard deviation of the longitudinal acceleration for each second cluster center corresponding to the second principal component score matrix. The second cluster centers are sorted in descending order based on the values, and the driving style of each second cluster center during vehicle acceleration is determined based on the sorting results.

7. The method according to claim 2, characterized in that, The driver's driving style during vehicle deceleration is determined based on the feature parameters corresponding to each cluster center, including: Based on the value of each third cluster center corresponding to the third principal component score matrix, the driving style of each third cluster center during vehicle deceleration is determined; Based on the driving style of each third cluster center during vehicle deceleration, determine the number of kinematic segments corresponding to each driving style during vehicle deceleration. The driver's driving style during vehicle deceleration is determined by the proportion of the number of kinematic segments corresponding to each driving style in the number of kinematic segments corresponding to the third cluster center.

8. The method according to claim 7, characterized in that, The step of determining the driving style of each third cluster center during vehicle deceleration based on the value of each third cluster center corresponding to the third principal component score matrix includes: The value of each third cluster center is determined based on the average of the mean longitudinal deceleration, the maximum longitudinal deceleration, and the average of the standard deviation of the longitudinal deceleration for each third cluster center corresponding to the third principal component score matrix. The third cluster centers are sorted in descending order based on the values, and the driving style of each third cluster center during vehicle deceleration is determined based on the sorting results.

9. The method according to claim 1, characterized in that, Also includes: The driving style of the vehicle when turning is input into the steering system, and the steering system adjusts the driving mode and / or driving parameters based on the driving style of the vehicle when turning. The driving style during vehicle acceleration is input into the power system, and the power system adjusts the driving mode and / or driving parameters based on the driving style during vehicle acceleration. The driving style during vehicle deceleration is input into the braking system, which then adjusts the driving mode and / or driving parameters based on the driving style during deceleration.

10. A driving style determining device, characterized in that, include: The feature parameter acquisition module is used to acquire kinematic segments based on vehicle driving data, and to acquire a first feature parameter representing lateral driving style and a second feature parameter representing longitudinal driving style in the kinematic segments. The feature parameter dimensionality reduction module is used to perform dimensionality reduction processing on the first feature parameter and the second feature parameter respectively to obtain multiple principal component score matrices; The driving style determination module is used to cluster the kinematic segments according to each principal component score matrix to obtain multiple cluster centers, and to determine the driver's driving style when the vehicle is turning, accelerating or decelerating according to the feature parameters corresponding to each cluster center.

11. A vehicle, characterized in that, The vehicles include: Onboard sensors are used to collect vehicle driving data; At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the driving style determination method according to any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the driving style determination method according to any one of claims 1-9.

Images