KR102570295B1 - Vehicle and control method thereof - Google Patents

Abstract

The present invention relates to a vehicle and a control method thereof, and in determining the driver's driving tendency, the labeling of learning data is efficiently assisted and vehicle-specific characteristic information is excluded, thereby classifying the driver's driving tendency and selecting an automatic driving mode. The purpose is to make this more accurate. To this end, a vehicle control method according to the present invention includes the steps of extracting statistical characteristics during a preset section at preset time intervals from vehicle travel information obtained through vehicle travel; extracting statistical characteristics from the converted vehicle driving information and labeling the vehicle driving information to extract a driver's driving tendency from the extracted statistical characteristics; and generating a classification logic for classifying the driver's driving tendency based on the result of the labeling.

Description

Vehicle and its control method {VEHICLE AND CONTROL METHOD THEREOF}

The present invention relates to a vehicle, and relates to a vehicle provided to set and operate a driving mode according to a driver's driving tendency, and a control method thereof.

As sensor technology and automatic control technology develop, various types of automatic driving modes are prepared in vehicles, and by selecting an automatic driving mode suitable for the driving environment and changing the system settings of the vehicle to suit the selected driving mode, driver convenience and safety are improved. I've come to try

However, such selection of an automatic driving mode entails a cumbersome process in which a driver directly selects a driving mode in consideration of his/her own driving tendency, since the automatic driving mode determined by the vehicle manufacturer is collectively applied.

According to one aspect of the present invention, a system of a vehicle selects an optimal automatic driving mode suitable for the driver's driving tendency through analysis of the driver's driving tendency, and efficiently labels learning data in determining the driver's driving tendency. Its purpose is to make the classification of the driver's driving tendency and the selection of the automatic driving mode more accurate by assisting and excluding vehicle-specific characteristic information.

A vehicle control method according to the present invention for the above object includes the steps of extracting statistical characteristics during a preset section at preset time intervals from vehicle travel information obtained through vehicle travel; extracting statistical characteristics from the converted vehicle driving information and labeling the vehicle driving information to extract a driver's driving tendency from the extracted statistical characteristics; and generating a classification logic for classifying the driver's driving tendency based on the result of the labeling.

The vehicle control method described above further includes the step of excluding the unique characteristics of the vehicle from the vehicle driving information by converting the vehicle driving information obtained through driving of the vehicle into a form of driving information of a predetermined standard vehicle. include

The vehicle control method described above distinguishes driving information generated by a driver's operation from driving information generated by a driving environment in order to exclude the unique characteristics of the vehicle from the vehicle driving information.

In the vehicle control method described above, the driving information generated by the driver's manipulation includes at least one of a damper speed, a vehicle speed, a longitudinal acceleration, a lateral acceleration, and a yaw rate.

In the vehicle control method described above, the driving information generated by the driver's manipulation includes at least one of a steering angular velocity, an accelerator pedal manipulation amount, and a brake pedal manipulation amount.

In the vehicle control method described above, by selectively taking some of the damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, yaw rate, steering angular velocity, accelerator pedal operation amount, and brake pedal operation amount, the unique characteristics of the vehicle are excluded.

In the vehicle control method described above, in the step of extracting the statistical characteristics, each of the preset sections is configured to have overlapping sections.

In the vehicle control method described above, the statistical characteristics include a feature vector, a maximum deviation, a minimum deviation, and a standard deviation.

In the above vehicle control method, machine learning-based neural network learning theory is applied for the labeling.

In the vehicle control method described above, an error matrix is used for the labeling.

In the vehicle control method described above, the labeling may include generating a classification result through a driver classification logic for the data set of the vehicle driving information; and generating the rearranged error matrix by rearranging the generated classification result through error matrix rearrangement logic.

In the vehicle control method described above, the error matrix is represented as an image in which a cluster is expressed through reordering.

The vehicle control method described above may further include acquiring the vehicle driving information through a test drive of the vehicle and building a database.

The vehicle control method described above further includes receiving the vehicle driving information stored in the database and removing noise included in the vehicle driving information.

A vehicle according to the present invention for the above object includes a feature extraction logic for each section for extracting statistical characteristics during a preset section at a preset time interval from vehicle driving information obtained through driving of the vehicle; a driving tendency labeling logic for extracting statistical characteristics from the converted vehicle driving information and labeling the vehicle driving information to extract a driver's driving tendency from the extracted statistical characteristics; and learning logic for generating classification logic for classifying the driver's driving tendency based on the result of the labeling.

In the above-described vehicle, a vehicle feature standardization logic for excluding the unique characteristics of the vehicle from the vehicle driving information by converting the vehicle driving information acquired through driving of the vehicle into a form of predetermined standard vehicle driving information contains more

In the vehicle described above, the vehicle feature standardization logic selectively separates driving information generated by a driver's operation from driving information generated by a driving environment in order to exclude unique characteristics of the vehicle from the vehicle driving information. get drunk

In the vehicle described above, driving information generated by the driver's manipulation includes at least one of damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, and yaw rate.

In the vehicle described above, the driving information generated by the driver's manipulation includes at least one of a steering angular velocity, an accelerator pedal manipulation amount, and a brake pedal manipulation amount.

In the aforementioned vehicle, some of the damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, yaw rate, steering angular velocity, accelerator pedal operation amount, and brake pedal operation amount are selectively taken to exclude the inherent characteristics of the vehicle.

In the vehicle described above, the feature extraction logic for each section causes each of the preset sections to have sections overlapping each other in the step of extracting the statistical characteristics.

In the vehicle described above, the statistical characteristics include a feature vector, a maximum deviation, a minimum deviation, and a standard deviation.

In the vehicle described above, machine learning-based neural network learning theory is applied for the labeling.

In the vehicle described above, an error matrix is used for the labeling.

In the aforementioned vehicle, the driving propensity labeling logic generates a classification result through driver classification logic for the data set of the vehicle driving information; A rearranged error matrix is generated by rearranging the generated classification result through error matrix rearrangement logic.

In the vehicle described above, the error matrix is represented by an image in which clusters are expressed through reordering.

The vehicle described above further includes a database for storing the vehicle driving information obtained through test driving of the vehicle.

The vehicle described above further includes a noise canceling logic for receiving the vehicle driving information stored in the database and removing noise included in the vehicle driving information.

Another vehicle control method according to the present invention for the above object is to convert vehicle driving information obtained through driving of a test vehicle into a form of driving information of a predetermined standard vehicle, and in the vehicle driving information of the converted test vehicle allowing the unique characteristics of the test vehicle to be excluded; extracting statistical characteristics during a preset section at preset time intervals from the converted vehicle driving information; performing labeling of the vehicle driving information to extract a driver's driving tendency by extracting statistical characteristics from the converted vehicle driving information and applying machine learning-based neural network learning theory to the extracted statistical characteristics; and generating a classification logic for classifying the driver's driving tendency based on the result of the labeling.

Another vehicle control method according to the present invention for the above object is to convert vehicle driving information obtained through driving of a mass-produced vehicle into a form of driving information of a predetermined standard vehicle, and in the converted vehicle driving information of the mass-produced vehicle removing the unique characteristics of the mass-produced vehicle; extracting statistical characteristics during a preset section at preset time intervals from the converted vehicle driving information; classifying a driving tendency of a driver driving the mass-produced vehicle by using a classification logic provided through labeling of the vehicle driving information to extract the driver's driving tendency; and determining an automatic driving mode of the vehicle based on the classified driving tendency.

According to one aspect of the present invention, a system of a vehicle selects an optimal automatic driving mode suitable for the driver's driving tendency through analysis of the driver's driving tendency, and efficiently labels learning data in determining the driver's driving tendency. By assisting and excluding vehicle-specific characteristic information, the driver's driving tendency classification and automatic driving mode selection can be made more accurately.

1 is a diagram illustrating a control system for classifying a driving tendency of a vehicle according to an embodiment of the present invention.
2 is a diagram showing the flow of a learning step and a classification step for driving tendency classification according to an embodiment of the present invention.
3 is a diagram illustrating a vehicle feature standardization method according to an embodiment of the present invention.
4 is a diagram illustrating an embodiment of vehicle feature standardization according to an embodiment of the present invention.
5 is a diagram illustrating feature extraction for each section according to an embodiment of the present invention.
6 is a diagram illustrating a process of rearranging driver classification results using a mismatch matrix for driving tendency labeling according to an embodiment of the present invention.
7 is a diagram illustrating driver classification based on driving tendency according to an embodiment of the present invention.
8 is a diagram illustrating an example of classification according to a sensor input in driving tendency classification according to another embodiment of the present invention.

1 is a diagram illustrating a control system for classifying a driving tendency of a vehicle according to an embodiment of the present invention. The driving tendency determination unit 112 classifies and determines the driving tendency of the driver driving the vehicle and provides the classification result to the driving control unit 162 so that the driving control unit 162 can match the driver's driving tendency in the corresponding mass-produced vehicle. It enables proper driving mode selection and driving assistance control to be performed. The classification result of the driving tendency may include ECO mode/COMPORT mode/NORMAL mode/SPORT mode/SPORT+ mode.

A sensor group 150 composed of a plurality of sensors for detecting vehicle state information (driving information) is communicatively connected to the driving tendency determination unit 112 . The sensor group 150 includes a lateral acceleration sensor 132, a longitudinal acceleration sensor 134, a steering angular velocity sensor 136, a yaw rate sensor 138, a navigation system 140, a damper speed sensor 142, and various other sensors. It includes a kind of sensor 144. In addition, a database 122 is communicatively connected to the driving tendency determination unit 112. In the database 122, vehicle driving information of various types of test vehicles is stored in the form of a plurality of data sets.

The driving control unit 162 controls various elements including steering 172, engine 174, braking 176, and electric field 178 to select an appropriate driving mode suitable for the driver's driving tendency and control driving assistance in a mass-produced vehicle. Control.

2 is a diagram showing the flow of a learning step and a classification step for driving tendency classification according to an embodiment of the present invention.

In FIG. 2 , the process indicated by the dotted line arrow is the learning step for generating the classification logic 214 . The learning phase is a process of generating the classification logic 214 based on vehicle driving information acquired while driving a plurality of test vehicles by a plurality of test drivers in the vehicle development phase. In the learning step, vehicle driving information of the test vehicle obtained while the test driver drives the test vehicle and stored in the database 122 is used. In addition, in the learning step, the vehicle driving information stored in the database 122 is used by the noise cancellation logic 204, the vehicle feature standardization logic 206, the section feature extraction logic 208, the driving tendency labeling logic 210, and the learning Through the logic 212, the learned classification logic 214 based on the neural model based on machine learning is completed. The classification logic 214 thus completed is used to classify and determine the actual driving tendency of the driver driving the vehicle in the actual classification step indicated by the solid line arrow in FIG. 2 .

In FIG. 2 , the process indicated by the solid line arrow is mass production through the classification logic 214 already created in the development process based on vehicle driving information obtained during actual driving of a vehicle that has been developed, produced, and sold (hereafter, mass-produced vehicle). This is a classification step of determining and classifying the actual driving tendency of the driver of the vehicle. Unlike the learning step, in the classification step, the driving tendency is classified using vehicle driving information obtained in real time while the driver of the mass-produced vehicle, which has been produced and sold, drives the vehicle. In the classification step, the vehicle driving information obtained in real time is extracted through the noise removal logic 204, the vehicle feature standardization logic 206, and the feature extraction logic 208 for each section, and statistical characteristics are extracted. The driver's driving tendency is classified in the classification logic 214 using statistical characteristics. The classified driving tendency information is transmitted to the driving control unit 162 and used for appropriate driving mode selection and driving assistance control suitable for the driver's driving tendency in the corresponding mass-produced vehicle.

Hereinafter, the process of generating a classifier through a test driver and a test vehicle in the learning phase will be described first, and then a process of classifying and determining a driver's driving propensity in real time through the classifier that has been created for a mass-produced vehicle will be described.

<Database construction>

For the purpose of generating the machine learning-based learned classification logic 214, a plurality of test drivers drive a vehicle (hereinafter referred to as a test vehicle), and vehicle driving information is collected through driving of the test vehicle by the test driver. and accumulated and stored in the database 122.

The vehicle driving information stored in the database 122 includes vehicle driving data generated by driving of a test driver having a driving tendency to be classified. For example, when trying to classify the driving tendency for each driving mode of ECO mode / COMPORT mode / NORMAL mode / SPORT mode / SPORT+ mode, vehicle driving information for each driving tendency (by driving mode) is stored in the database. Vehicle driving information for each driving tendency includes driving speed information, lateral acceleration information, longitudinal acceleration information, and yaw rate information of the test vehicle. 'ECO Mode' is a driving mode that enables economical driving by reducing fuel consumption. 'COMPORT mode' is a comfortable and comfortable driving mode. 'NORMAL mode' is a normal driving mode. 'SPORT mode' is a dynamic and light driving mode. 'SPORT+ mode' is a driving mode with more dynamic and faster characteristics than 'SPORT mode'.

<Noise Reduction>

Noise included in the vehicle driving information is removed from the vehicle driving information stored in the database 122 in the noise removal logic 204 . A low pass filter may be used to remove noise included in vehicle driving information.

In addition, an exponential moving average represented by Equation 1 may be used to remove noise included in vehicle driving information. In Equation 1 below, St is performed as a result of performing the exponential moving average method using the raw data Yt.

<Equation 1>

In addition, a Gaussian filter convolution method may be used to remove noise included in vehicle driving information.

<Standardization of Vehicle Features>

A variety of different types of test vehicles may be used to collect vehicle driving information. In an embodiment of the present invention, the unique characteristics of the vehicle used for collecting the vehicle driving information are removed from the vehicle driving information. In this way, the driver's driving tendency can be determined by excluding the unique characteristics of the vehicle.

The obtained vehicle driving information is converted into the standard driving information form of the standard vehicle Z. Here, the standard driving information is driving information that can be obtained when the same type of driving is performed using the standard vehicle Z. Vehicle feature standardization is, for example, when vehicle driving information is obtained by performing a specific type of driving with different types of vehicles CAR#1, CAR#2, CAR#3, and CAR#4, vehicle CAR#1, CAR By excluding the unique characteristics of #2, CAR#3, and CAR#4, it means converting into standardized vehicle driving information that can be obtained when the same specific type of driving is performed with the standard vehicle (Z).

3 is a diagram illustrating a vehicle feature standardization method according to an embodiment of the present invention.

As shown in FIG. 3 , in an embodiment of the present invention, standardization of vehicle characteristics is performed by selectively utilizing values of vehicle sensor detection data according to circumstances. However, in FIG. 3, No. As in case 1, even if various information is used, the accuracy of driving tendency determination is not always high. Rather, no. As in the case of 2, when some vehicle driving information (damper speed, etc.) is removed, the driving tendency determination may be more accurate.

Accordingly, in this case, since the 'damper speed' is more influenced by various road environments than the driver's driving tendency, it may be more desirable to selectively remove the 'damper speed' in order to determine the driver's driving tendency. A damper refers to a damper provided to absorb vibration in a suspension system of a vehicle.

An 'accelerator pedal' refers to an operation amount of an accelerator pedal for acceleration. In the case of 'longitudinal acceleration' and 'accelerator pedal', although they are affected by the driving tendency of the driver, they are also affected by traffic lights or traffic congestion, so conditional standardization is necessary.

In the case of 'lateral acceleration', 'yaw rate', and 'steering angular velocity', lateral behavior is unavoidable according to the road pattern, so conditional standardization is required in this case as well.

4 is a diagram illustrating an embodiment of vehicle feature standardization according to an embodiment of the present invention. 4 illustrates an embodiment in which a specific sensor value is excluded from vehicle driving information under a specific driving condition.

First, the detection value of the longitudinal acceleration sensor may be excluded based on the forward inter-vehicle distance and the traffic light signal. The forward head-to-head distance can be measured through the Smart Cruise Control (SCC) function provided in the vehicle. Traffic light signals can be checked through traffic light information provided from a traffic light information providing system. When the forward head-to-head distance is greater than or equal to the preset standard head-to-head distance (Dscc) ('yes' in 412) and the signal of the traffic light is a stop signal ('yes' in 414), the longitudinal acceleration sensor (134 ) is excluded. For example, values of vehicle speed, longitudinal acceleration, brake pedal, and accelerator pedal are excluded from vehicle driving information. The detection value of the longitudinal acceleration sensor generated by the stop signal of the traffic light in a state where the distance between the vehicle in front is more than the preset distance is sufficient to indicate the external driving environment rather than the driver's driving tendency, so it is used as data for determining the driver's unique driving tendency. Since it is insufficient to be used, it is desirable to exclude it when determining driving tendency.

Also, the detection value of the lateral acceleration sensor may be excluded based on the curvature of the road on which the vehicle is driving. The curvature of the road can be checked through map information of the navigation system 140 installed in the vehicle. If the curvature of the road on which the vehicle is driving is smaller than or equal to the preset reference curvature Rnavi (Yes in step 422), the detection value of the lateral acceleration sensor is excluded (426). For example, values of lateral acceleration, yaw rate, and steering angular velocity are excluded. The detection value of the lateral acceleration sensor 132 generated when the curvature of the road is greater than the preset reference value (Rnavi) represents the external driving environment rather than the driver's driving tendency, so it is used as data to determine the driver's unique driving tendency Since it is not enough to be, it is desirable to exclude it when determining driving tendency.

In addition, the detection value of the damper speed sensor 142 may be excluded based on the presence or absence of a speed bump on the road on which the vehicle is driving. Whether there are speed bumps on the road can be checked through map information of the navigation system 140 installed in the vehicle. If there is no speed bump on the road on which the vehicle is traveling ('No' in 432), the detection value of the damper speed sensor 142 is excluded (436). For example, the value of the vertical direction acceleration of the damper is excluded. The vertical acceleration value of the damper caused by the speed bump on the road can be seen as representing the driver's driving tendency when passing the speed bump, but the change in the damper's vertical acceleration when the speed bump does not exist represents only the road surface roughness (external driving environment). Therefore, it is desirable to exclude the vertical direction acceleration value of the damper at a place other than a speed bump ('No' in 432) when determining the driver's driving tendency.

In order to standardize vehicle features, it is necessary to convert vehicle driving information into a form necessary for learning the conversion model. An encoder-decoder structure can be used for this conversion. In order to learn the conversion model, a large amount of data sets acquired from the standard vehicle (Z) in various environments (various driving patterns by multiple drivers with different driving tendencies, various road environments, various weather conditions, etc.) are required.

Learning of the internal parameters of the encoder-decoder of the neural network structure is performed using a plurality of data sets. At this time, an objective function that receives each data value of the vehicle driving information that has passed through the encoder-decoder as an input is used. A discriminator of a neural network structure is used for the objective function, and the discriminator enables the conversion model to repeatedly learn the unique characteristics of the standard vehicle Z among a plurality of data sets. The conversion model does not extract only the characteristics of the standard vehicle Z, but enables data conversion in which the characteristics of the standard vehicle Z are reflected. A conversion model of an encoder-decoder structure may be created through iterative learning.

Through such vehicle feature standardization, it is possible to solve a data bias problem that may occur due to differences in vehicle driving information acquisition means (i.e., vehicle model) and vehicle driving environment (road, weather, etc.).

<Extraction of statistical characteristics by section>

5 is a diagram illustrating feature extraction for each section according to an embodiment of the present invention.

As shown in FIG. 5 , in driving tendency classification based on vehicle driving information according to an embodiment of the present invention, statistical characteristics are extracted from vehicle driving information from which noise and vehicle characteristics are removed during a preset interval at a preset time interval. Execute feature extraction. Each section partitioned during feature extraction for each section may have overlapping sections.

For example, as shown in FIG. 5 , statistical characteristics are extracted at 4-second intervals for a 4-second interval, but an overlapping interval of 2 seconds may exist in each interval. Statistical properties may include feature vectors, maximum deviations, minimum deviations, and standard deviations. Statistical characteristics are extracted for all sensor detection values to be input, but different types of statistical characteristics may be provided for each sensor detection value, if necessary.

<Driving tendency labeling>

6 is a diagram illustrating a process of rearranging driver classification results using a mismatch matrix for driving tendency labeling according to an embodiment of the present invention. In the driving propensity labeling step, a clustering algorithm is applied to the extracted statistical characteristics (feature vectors) to assist labeling for extracting driving propensities.

As shown in FIG. 6 , in driving tendency classification according to an embodiment of the present invention, a data set 602 of vehicle driving information including a driver ID is classified as a driver in order to label learning data for machine learning-based neural network learning. Logic 604 to generate a classification result 606, and generated classification result 606 to be rearranged through a confusion matrix rearrangement logic 608 to obtain a rearranged confusion matrix 610. The error matrix is represented by an image in which clusters are expressed through reordering. The rearrangement may use Maciejewski's method or the method of the Cuthill-McKee algorithm. In addition, the number of classes to be labeled may be determined using an image in which clusters are expressed, and a driver's own driving tendency may be used as a labeling criterion.

7 is a diagram illustrating driver classification based on driving tendency according to an embodiment of the present invention.

7(A) is a diagram showing the results of driver classification performed with one driver (A) having a driving tendency of 'COMPORT mode' and two drivers (B, C) having a driving tendency of 'SPORT mode'. . In FIG. 7(A), numbers without parentheses are test results, and numbers in parentheses are normalized results for true values.

As shown in FIG. 7(A), it can be seen that a block occurs between B and C due to a relatively high ratio of mismatch between the actual driving tendency and the classification result. For example, a block was formed between B and C because the ratio of actually classifying drivers with B tendencies as C and drivers with C tendencies as B was relatively high.

In this case, the driver's driving tendency is classified as 'COMPORT - SPORT', and the data of A can be used as COMPORT learning data, and the data of B and C can be used as SPORT learning data.

7(B) shows one driver (a) with a driving tendency in 'COMPORT mode', three drivers (b, c, d) with a driving tendency in 'SPORT mode', and one driver with a driving tendency in SPORT+ mode. This is the result of performing driver classification with one driver (e). In FIG. 7(B), numbers without parentheses represent test results, and numbers in parentheses are results normalized to true values.

As shown in FIG. 7(B), it can be seen that a block occurs between b-c-d where the ratio of the actual driving tendency and the classification result is relatively low.

In this case, the driver's driving tendency is classified as 'COMPORT - SPORT - SPORT+', and the data of a can be used as COMPORT learning data, the data of b, c, and d can be used as SPORT learning data, and the data of e Can be used as SPORT+ training data.

<Learning>

In the learning logic, machine learning-based neural network learning is performed using the statistical characteristics (feature vectors) extracted in the statistical feature extraction step and the driving propensity extracted in the driving propensity labeling step, and a driving propensity classifier is generated as a result of the neural network learning. .

The classification logic 214 for classifying the driving tendency may use convolutional neural networks, recurrent neural networks, and the like. The classification logic 214 may be designed by learning a feature vector obtained through noise removal, vehicle feature removal, and feature extraction for each section using a neural network model. In the classification logic 214, sufficient learning is repeatedly performed until the weights converge. The value of the weight inside the classification logic 214 is updated according to the training data, and finally has a weight suitable for the training data.

<Driving Tendency Classification>

A driver's driving tendency classification result may be obtained by inputting the feature vector extracted from the feature extraction logic for each section to the classification logic 214 . A machine learning-based neural network model can be built using a Long Short-Term Memory (LSTM) model. In this way, the long-term dependency problem of RNN can be solved by building a machine learning-based neural network model using the LSTM model. In this case, the internal cell state is used to determine what information to leave.

8 is a diagram illustrating an example of classification according to a sensor input in driving tendency classification according to another embodiment of the present invention. 8 shows graphs of detection curves of the lateral acceleration sensor 132, the longitudinal acceleration sensor 134, the steering angular velocity sensor 136, and the yaw rate sensor 138 of the vehicle. As a model for the classification logic 214, the previously described recurrent neural network-based Long Short-Term Memory (LSTM) may be used.

The task of classifying the driver's driving tendency from the detection values of various sensors is performed every 60 seconds, and the classification result observed for the past 120 seconds can be used.

As shown in FIG. 8, most of the sections were predicted as 'COMPORT', but a large change was detected in all sensors around 1750 seconds (802) in FIG. 8, and the corresponding section 804 could be classified as 'SPORT'. there is.

Returning to FIG. 2, once the classification logic is completed through a series of learning steps (the process of the dotted line arrow in FIG. ) may be performed (solid line arrow process in FIG. 2 ) for classifying and determining the driving tendency. In the classification step, the vehicle driving information obtained in real time is extracted through the noise removal logic 204, the vehicle feature standardization logic 206, and the feature extraction logic 208 for each section, and statistical characteristics are extracted. Classification of driving propensity is performed in the classification logic 214 of statistical characteristics. The classified driving tendency information is transmitted to the driving control unit 162 and used for appropriate driving mode selection and driving assistance control suitable for the driver's driving tendency in the corresponding mass-produced vehicle.

The above description is only an illustrative example of the technical idea, and those skilled in the art will be able to make various modifications, changes, and substitutions without departing from the essential characteristics. Therefore, the embodiments disclosed above and the accompanying drawings are intended to explain rather than limit the technical idea, and the scope of the technical idea is not limited by these embodiments and the accompanying drawings. The scope of protection should be construed according to the scope of claims below, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of rights.

112: driving tendency determination unit
122: database
132: lateral acceleration sensor
134: longitudinal acceleration sensor
136: steering angular speed sensor
138: yaw rate sensor
140: navigation
142: damper speed sensor
150: sensor group
162: driving control unit
172: steering
174: engine
176: braking
178: battlefield
204: noise cancellation logic
206: vehicle feature standardization logic
208: feature extraction logic for each section
210: Driving tendency labeling logic
212: learning logic
214: classification logic

Claims (30)

Excluding unique characteristics of the vehicle from the vehicle driving information by converting vehicle driving information acquired through driving of the vehicle into a form of predetermined standard vehicle driving information;
distinguishing between driving information generated by a driver's manipulation and driving information generated by a driving environment in order to exclude the unique characteristics of the vehicle;
extracting statistical characteristics during a preset section at preset time intervals from vehicle travel information acquired through vehicle travel;
extracting statistical characteristics from the converted vehicle driving information and labeling the vehicle driving information to extract a driver's driving tendency from the extracted statistical characteristics;
and generating a classification logic for classifying a driver's driving tendency based on a result of the labeling. delete delete According to claim 1,
The driving information generated by the driver's manipulation includes at least one of damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, and yaw rate. According to claim 4,
The driving information generated by the driver's manipulation includes at least one of a steering angular velocity, an accelerator pedal manipulation amount, and a brake pedal manipulation amount. According to claim 5,
A vehicle control method for excluding inherent characteristics of the vehicle by selectively taking some of the damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, yaw rate, steering angular velocity, accelerator pedal operation amount, and brake pedal operation amount. According to claim 1,
In the step of extracting the statistical characteristics, each of the preset sections has sections overlapping with each other. According to claim 7,
The statistical characteristic includes a feature vector, a maximum deviation, a minimum deviation, and a standard deviation. According to claim 1,
A vehicle control method applying a machine learning-based neural network learning theory for the labeling. According to claim 1,
A vehicle control method in which an error matrix is used for the labeling. 11. The method of claim 10, wherein the labeling,
generating a classification result from the data set of the vehicle driving information through driver classification logic;
and generating the rearranged error matrix by rearranging the generated classification result through error matrix rearrangement logic. According to claim 11,
The error matrix is represented by an image in which a cluster is expressed through reordering. According to claim 1,
The vehicle control method further comprising establishing a database by acquiring the vehicle driving information through a test drive of the vehicle. According to claim 13,
The vehicle control method further comprising receiving the vehicle driving information stored in the database and removing noise included in the vehicle driving information. a section-by-section feature extraction logic for extracting statistical characteristics during a preset section at a preset time interval from vehicle driving information acquired through vehicle driving;
a vehicle feature standardization logic configured to exclude unique characteristics of the vehicle from the vehicle driving information by converting the vehicle driving information obtained through driving of the vehicle into a form of predetermined standard vehicle driving information;
logic for distinguishing between driving information generated by a driver's manipulation and driving information generated by a driving environment in order to exclude the unique characteristics of the vehicle from the vehicle driving information;
a driving tendency labeling logic for extracting statistical characteristics from the converted vehicle driving information and labeling the vehicle driving information to extract a driver's driving tendency from the extracted statistical characteristics;
and learning logic for generating classification logic for classifying a driver's driving tendency based on a result of the labeling. delete delete According to claim 15,
The driving information generated by the driver's manipulation includes at least one of damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, and yaw rate. According to claim 18,
The driving information generated by the driver's manipulation includes at least one of a steering angular velocity, an accelerator pedal manipulation amount, and a brake pedal manipulation amount. According to claim 19,
A vehicle that excludes the unique characteristics of the vehicle by selectively taking some of the damper speed, vehicle speed, longitudinal acceleration, lateral acceleration, yaw rate, steering angular velocity, accelerator pedal operation amount, and brake pedal operation amount. The method of claim 20, wherein the feature extraction logic for each section,
In the step of extracting the statistical characteristics, each of the preset sections has sections overlapping each other. According to claim 21,
The statistical characteristics include a feature vector, a maximum deviation, a minimum deviation, and a standard deviation. According to claim 15,
A vehicle that applies machine learning-based neural network learning theory for the labeling. According to claim 15,
A vehicle whose error matrix is used for the labeling. The method of claim 15, wherein the driving tendency labeling logic,
generating a classification result from the data set of the vehicle driving information through driver classification logic;
A vehicle that generates a rearranged error matrix by rearranging the generated classification result through an error matrix rearrangement logic. 26. The method of claim 25,
The error matrix is a vehicle represented by an image in which a cluster is expressed through reordering. According to claim 15,
The vehicle further comprising a database for storing the vehicle driving information obtained through test driving of the vehicle. 28. The method of claim 27,
and a noise canceling logic configured to receive the vehicle driving information stored in the database and remove noise included in the vehicle driving information. Excluding unique characteristics of the test vehicle from the converted vehicle driving information of the test vehicle by converting vehicle driving information obtained through driving of the test vehicle into predetermined standard vehicle driving information;
classifying driving information generated by a driver's manipulation and driving information generated by a driving environment in order to exclude the unique characteristics of the vehicle;
extracting statistical characteristics during a preset section at preset time intervals from the converted vehicle driving information;
performing labeling of the vehicle driving information to extract a driver's driving tendency by extracting statistical characteristics from the converted vehicle driving information and applying machine learning-based neural network learning theory to the extracted statistical characteristics;
and generating a classification logic for classifying a driver's driving tendency based on a result of the labeling. converting vehicle driving information acquired through driving of the mass-produced vehicle into a predetermined standard vehicle driving information form so that unique characteristics of the mass-produced vehicle are excluded from the converted vehicle driving information of the mass-produced vehicle;
classifying driving information generated by a driver's manipulation and driving information generated by a driving environment in order to exclude the unique characteristics of the vehicle;
extracting statistical characteristics during a preset section at preset time intervals from the converted vehicle driving information;
classifying a driving tendency of a driver driving the mass-produced vehicle by using a classification logic provided through labeling of the vehicle driving information to extract the driver's driving tendency;
and determining an automatic driving mode of the vehicle based on the classified driving tendency.

Images