KR102562459B1 - System for classification of driver's longitudinal driving behavior - Google Patents

Abstract

A longitudinal driving habit classification system of a driver according to an embodiment of the present invention includes a vehicle data analyzer for collecting and analyzing motion data of a vehicle in which the driver is riding; a machine learning unit that performs unsupervised learning using trend information of the speed data calculated by the vehicle data analysis unit; a driving habit classification unit which classifies clusters calculated as learning results from the machine learning unit according to driving habits; and a driving habit determining unit that compares the classified clusters with the driver's driving habit.

Description

Driver's Longitudinal Driving Habit Classification System {SYSTEM FOR CLASSIFICATION OF DRIVER'S LONGITUDINAL DRIVING BEHAVIOR}

The present invention relates to a longitudinal driving habit classification system of a driver, and more particularly, provides a driver's longitudinal driving habit classification system capable of classifying and determining the driver's habit using only vehicle speed change information.

In recent years, as the number of vehicles supplied has rapidly increased, safety accidents and traffic accidents occurring while driving vehicles are increasing. Among the safety accidents and traffic accidents that occur while driving a vehicle, safety accidents and traffic accidents that occur due to incorrect driving habits occupy a large portion.

Even if a driver suddenly starts, accelerates, or stops while driving a vehicle, the driver himself does not easily feel it, and violent driving becomes a habit due to the driver's longitudinal driving habit, resulting in a major safety accident. Here, the longitudinal direction means the driving direction of the vehicle.

In addition, in the past, acceleration (ax, ay, az), speed (vx, vy, vz), roll, Efforts have been made to classify drivers' longitudinal driving habits by measuring pitch, yaw, roll rate, pitch rate, and yaw rate. However, this has limitations in that it requires a lot of data and a high-spec microcontroller to process it.

Furthermore, autonomous driving control systems of vehicles represented by ACC/CDM have recently been applied to mass production. However, when the vehicle performs longitudinal autonomous driving in a state that does not conform to the driver's longitudinal driving habit by such an autonomous driving control system, the driver feels uncomfortable with the autonomous driving that does not conform to his/her longitudinal driving habit, and eventually autonomous driving Do not use driving.

Therefore, it is necessary to frequently use the autonomous driving function by identifying and classifying the driver's longitudinal driving habits and then providing autonomous driving that meets the longitudinal driving habits.

The present applicant has proposed the present invention in order to solve the above problems.

Korean Registered Patent No. 10-0857820 (registered on September 3, 2008)

The present invention has been made to solve the above problems, and a longitudinal driving habit classification system of a driver capable of classifying a driver's longitudinal driving habit using only a vehicle speed change curve measured by a vehicle speed sensor installed in all vehicles. provides

The present invention provides a longitudinal driving habit classification system of a driver capable of increasing driving habit classification accuracy as vehicle speed data continues to accumulate because machine learning is used.

A longitudinal driving habit classification system of a driver according to an embodiment of the present invention for achieving the above objects includes a vehicle data analyzer for collecting and analyzing motion data of a vehicle in which the driver rides; a machine learning unit that performs unsupervised learning using trend information of the speed data calculated by the vehicle data analysis unit; a driving habit classification unit which classifies clusters calculated as learning results from the machine learning unit according to driving habits; and a driving habit determining unit that compares the classified clusters with the driver's driving habit.

The vehicle data analysis unit may include: a vehicle speed data acquisition unit collecting vehicle speed data and generating a vehicle speed signal change curve; a vehicle speed change trend line derivation unit for deriving a vehicle speed change trend line equation from the vehicle speed signal change curve generated by the vehicle speed data acquisition unit; and a trend line equation coefficient extractor extracting coefficients of each term from the vehicle speed change trend line equation derived by the vehicle speed change trend line derivation unit.

It includes a big data unit that converts the coefficients of each term extracted from the trend line equation coefficient extractor into a database, and the machine learning unit receives the coefficients from the big data unit and uses them as inputs for unsupervised learning.

The machine learning unit may calculate an optimal cluster as coefficients received from the trend line formula coefficient extraction unit or the big data unit are accumulated.

The machine learning unit selects the largest coefficient among the coefficients of each term extracted from the vehicle speed change trend line expression expressed as a polynomial, and the vehicle speed change trend line expression expressed as an n-order function expression becomes an n-order function expression close to the order expression of the term to which the selected coefficient belongs can learn

The method may further include a self-driving mode determiner configured to compare clusters classified according to the learning result of the machine learning unit with the driver's driving habit and determine an autonomous driving mode of the vehicle to match the determined driver's driving habit.

Since the longitudinal driving habit classification system of the driver according to an embodiment of the present invention classifies the driver's longitudinal driving habit using only the vehicle speed change curve measured by the vehicle speed sensor installed in all vehicles, in order to analyze the driving habit Because expensive equipment is not required and the driving habit can be determined only with the speed of the vehicle, the specifications of the controller and computer required to process it can be lowered.

Since the longitudinal driving habit classification system of a driver according to an embodiment of the present invention uses machine learning, as vehicle speed data continues to accumulate, accuracy of driving habit classification can be increased.

The longitudinal driving habit classification system of the driver according to an embodiment of the present invention may promote the use of the autonomous driving function by activating or suggesting an autonomous driving mode suitable for the derived driving habit to the driver.

1 is a block diagram showing the configuration of a longitudinal driving habit classification system of a driver according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating a vehicle and a speed bump to describe a longitudinal driving habit classification system of a driver according to FIG. 1 .
3 is a block diagram showing the configuration of a modified example of the classification system according to FIG. 1;
4 and 5 are diagrams for explaining the relationship between vehicle speed trend line equations and coefficients used in the classification system according to FIG. 3 .
6 is a flowchart illustrating a longitudinal driving habit classification method of a driver according to an embodiment of the present invention.

Hereinafter, the embodiments disclosed in this specification will be described in detail with reference to the accompanying drawings, but the same or similar reference numerals are given to the same or similar components, and overlapping descriptions thereof will be omitted. The suffix "part" for components used in the following description is given or used interchangeably in consideration of ease of writing the specification, and does not itself have a meaning or role distinct from each other. In addition, in describing the embodiments disclosed in this specification, if it is determined that a detailed description of a related known technology may obscure the gist of the embodiment disclosed in this specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in this specification, the technical idea disclosed in this specification is not limited by the accompanying drawings, and all changes included in the spirit and technical scope of the present invention , it should be understood to include equivalents or substitutes.

Terms including ordinal numbers, such as first and second, may be used to describe various components, but the components are not limited by the terms. These terms are only used for the purpose of distinguishing one component from another.

It should be understood that when an element is referred to as being “connected” to another element, it may be directly connected to the other element, but other elements may exist in the middle.

Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, terms such as "comprise" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

It is advised that the drawings are schematic and not drawn to scale. Relative dimensions and proportions of parts in the drawings are shown exaggerated or reduced in size for clarity and convenience in the drawings, and any dimensions are illustrative only and not limiting. And like structures, elements or parts appearing in two or more drawings, like reference numerals are used to indicate like features.

The embodiments of the present invention specifically represent ideal embodiments of the present invention. As a result, various modifications of the drawings are expected. Therefore, the embodiment is not limited to the specific shape of the illustrated area, and includes, for example, modification of the shape by manufacturing.

Hereinafter, a longitudinal driving habit classification system 100 and a classification method of a driver according to the present invention will be described with reference to drawings.

1 is a block diagram showing the configuration of a driver's longitudinal driving habit classification system according to an embodiment of the present invention, and FIG. 2 is a vehicle and a speed bump to explain the driver's longitudinal driving habit classification system according to FIG. 1 3 is a block diagram showing the configuration of a modified example of the classification system according to FIG. 1, and FIGS. 4 and 5 are diagrams for explaining the relationship between the vehicle speed trend line equation and coefficients used in the classification system according to FIG. 3 6 is a flowchart illustrating a longitudinal driving habit classification method of a driver according to an embodiment of the present invention.

Referring to FIGS. 1 and 2 , a longitudinal driving habit classification system 100 of a driver according to an embodiment of the present invention collects and analyzes motion data of a vehicle 10 in which the driver rides. vehicle data analysis unit 110; a machine learning unit 130 that performs unsupervised learning using trend information of the speed data of the vehicle 10 calculated by the vehicle data analysis unit 110; a driving habit classifying unit 140 that classifies clusters calculated as learning results from the machine learning unit 130 by driving habit; and a driving habit determining unit 150 that compares the classified clusters with the driver's driving habit.

A longitudinal driving habit classification system (100, hereinafter referred to as "longitudinal driving habit classification system") of a driver according to an embodiment of the present invention is a method for rapidly accelerating, stopping or shifting the vehicle rapidly while driving a vehicle. We propose a system that analyzes and classifies the driver's driving habits in the longitudinal direction, that is, the driving direction of the vehicle related to the speed of the vehicle.

For example, as shown in FIG. 2 , when the vehicle 10 crosses a speed bump 20, each driver passes the speed bump at a different speed. The longitudinal driving habit classification system 100 according to an embodiment of the present invention analyzes and classifies the longitudinal driving habit when passing a speed bump 20, so that the driver feels comfortable when crossing the speed bump 20. It may be induced to select a driving mode.

The longitudinal driving habit classification system 100 according to an embodiment of the present invention can classify driving habits using only a car speed change curve measured by a vehicle speed sensor (not shown) mounted on the vehicle 10. can More specifically, driving habits can be classified by separating coefficients from vehicle speed change curves and applying machine learning to the accumulated coefficients.

Referring to FIG. 1 , a longitudinal driving habit classification system 100 according to an embodiment of the present invention includes a vehicle data analysis unit 110, a big data unit 120, a machine learning unit 130, and a driving habit classification unit. 140, a driving habit determination unit 150, and an autonomous driving mode determination unit 160 may be included.

The vehicle data analyzer 110 may receive and analyze vehicle speed data measured by the vehicle speed sensor, that is, vehicle speed data or information. Since the vehicle data analyzer 110 continuously receives vehicle speed information from the vehicle speed sensor in real time, it may obtain real-time vehicle speed change information, that is, a vehicle speed change curve.

The vehicle speed change curve obtained by the vehicle data analyzer 110 may reflect information for controlling the vehicle speed, such as how fast the driver accelerates or decelerates the vehicle, in real time. The longitudinal driving habit classification system 100 according to an embodiment of the present invention is characterized in that the longitudinal driving habit of the driver is classified using only the vehicle speed change curve.

The vehicle data analyzer 110 may collect vehicle-related data only while the driver directly drives the vehicle. Since vehicle-related data when the vehicle is stationary or not driving has little or no relation to longitudinal driving habits, the vehicle data analyzer 110 preferably does not collect such data.

The vehicle data analyzer 110 may collect or derive vehicle speed change curves, but cannot directly classify longitudinal driving habits from the vehicle speed change curves. In order to analyze and classify the driver's longitudinal driving habits, a lot of vehicle speed change curves or vehicle speed change information must be accumulated, and a common phenomenon or standard must be extracted by analyzing a large number of accumulated data.

To this end, the longitudinal driving habit classification system 100 according to an embodiment of the present invention may include a big data unit 120 . The big data unit 120 is a kind of database that receives and stores vehicle speed related data, vehicle speed change information, or a vehicle speed change curve from the vehicle speed data analysis unit 110 .

In addition, the big data unit 120 may analyze accumulated or stored vehicle speed-related data or vehicle speed change curves and transmit the analysis result to the vehicle data analyzer 110 or to the machine learning unit 120 .

The longitudinal driving habit classification system 100 according to an embodiment of the present invention transmits the vehicle speed-related data analyzed or acquired by the vehicle data analysis unit 110 to the machine learning unit 120, and the machine learning unit 120 By performing unsupervised learning with the corresponding data as an input value, classification results for longitudinal driving habits can be learned.

Meanwhile, in order to increase the accuracy of machine learning in the machine learning unit 120 and obtain an optimal classification result, data input to the machine learning unit 120 is important.

The vehicle data analysis unit 110 may process or extract an input value input to the machine learning unit 120 .

To this end, the vehicle data analysis unit 110, as shown in FIG. 3, includes a vehicle speed data acquisition unit 111 that collects vehicle speed data and generates a vehicle speed signal change curve; a vehicle speed change trend line derivation unit 112 for deriving a vehicle speed change trend line equation from the vehicle speed signal change curve generated by the vehicle speed data acquisition unit 111; and a trend line equation coefficient extractor 113 for extracting coefficients of each term from the vehicle speed change trend line equation derived by the vehicle speed change trend line derivation unit 112.

First, the vehicle speed data acquisition unit 111 may generate a vehicle speed signal change curve using the vehicle speed change curve measured by the vehicle speed sensor. Here, the vehicle speed change curve may be interpreted as the same as the vehicle speed signal change curve.

The vehicle speed change trend line derivation unit 112 may derive a vehicle speed change trend line equation from the vehicle speed signal change curve received from the vehicle speed data acquisition unit 111 . The vehicle speed change trend line derivation unit 112 receives many vehicle speed signal change curves, derives a vehicle speed change trend line by applying curve fitting to a plurality of vehicle speed signal change curves, and formulas or functions defining or expressing the derived vehicle speed change trend line. can also be derived.

For example, if the equation of the vehicle speed change trend line derived by the vehicle speed change trend line derivation unit 112 is an n-order function, a trend line equation such as [Equation 1] can be obtained.

[Equation 1]

The vehicle speed change trend line equation defined or expressed as in [Equation 1] is an n-order function equation having n terms. Therefore, the number of coefficients of each term is also n.

The longitudinal driving habit classification system 100 according to an embodiment of the present invention does not apply machine learning by using the vehicle speed change trend line formula as it is, but extracts only coefficients of each term from the trend line formula and applies it to machine learning. That is, the trend line equation coefficient extractor 113 may extract coefficients of each term from the vehicle speed change trend line equation derived by the vehicle speed change trend line deriver 112 and transmit them to the machine learning unit 120 . In addition, the trend line equation coefficient extractor 113 may transmit coefficients of each term to the big data module 120 .

The big data unit 120 databases the coefficients of each term extracted from the trend line equation coefficient extraction unit 113, and the machine learning unit 130 receives the coefficients from the big data unit 120 and uses them as inputs for unsupervised learning. Available.

The machine learning unit 130 may calculate an optimal cluster as more coefficients received from the trend line formula coefficient extraction unit 113 or the big data unit 120 are accumulated.

The machine learning unit 130 may derive an optimal cluster by inputting coefficients of each term of the trend line equation as input values of unsupervised learning as predictors. Here, the cluster is a result of machine learning and means classification of longitudinal driving habits.

ve Bayes, Multiclass Support Vector Machine, Ensemble 등의 분류 기법을 사용하여 클러스터를 도출할 수 있다.The machine learning unit 130 may perform unsupervised learning by inputting coefficients of each term of the trend line equation, and derive a cluster for longitudinal driving habits as a result. At this time, the machine learning unit 130 is k-NN, decision tree, Na

Clusters can be derived using classification techniques such as ve Bayes, Multiclass Support Vector Machine, and Ensemble.

Meanwhile, the machine learning unit 130 of the longitudinal driving habit classification system 100 according to an embodiment of the present invention, among the coefficients of each term extracted from the vehicle speed change trend line equation expressed as a polynomial as in [Equation 1], It can be learned that by selecting a large coefficient, the vehicle speed change trend line expression expressed as an n-order function expression becomes an n-order function expression close to the order expression of the term to which the selected coefficient belongs.

For example, when the vehicle speed change trend line expression is expressed as a cubic function expression such as [Equation 2], as shown in FIGS. It becomes a function, and as the coefficient (a2) of the quadratic term increases, it can become a cubic function close to the quadratic expression.

[Equation 2]

4 (a) and (b) and FIG. 5 (a) and (b), when the coefficient of each term increases, a change in the trend line equation can be seen. 4 and 5 are graphs showing changes in the function formula of [Equation 2] when the coefficients a1, a2, and a3 of each term of [Equation 2] increase from 1 to 100.

The machine learning unit 130 may derive a cluster by matching the order of the term to which the largest coefficient belongs in the vehicle speed change trend line equation through the learning described above and the driving habit.

The driving habit classification unit 140 of the longitudinal driving habit classification system 100 according to an embodiment of the present invention may define or classify the driving habit by using the learning result of the machine learning unit 130 . For example, the driving habit classification unit 140 may classify longitudinal driving habits into clusters of aggressive driving behavior, moderate driving behavior, and defensive driving behavior. there is.

The driving habit determination unit 150 of the longitudinal driving habit classification system 100 according to an embodiment of the present invention determines vehicle data obtained in real time from a vehicle in which a driver is riding, that is, a driving habit derived from a vehicle speed change curve. Driving habits can be classified by comparing with cluster values. The driving habit determiner 150 may notify the driver of the analyzed driving habit.

In addition, the longitudinal driving habit classification system 100 according to an embodiment of the present invention compares the driver's driving habit with a cluster classified according to the learning result of the machine learning unit 130 to match the determined driver's driving habit. An autonomous driving mode determination unit 160 for determining an autonomous driving mode of the vehicle may be further included.

As the driver's driving habit information determined by the driving habit determiner 150 is transmitted to the autonomous driving mode determiner 160, the autonomous driving mode determiner 160 proposes or determines an autonomous driving mode that matches the driver's driving habit. can

Since the autonomous driving mode determined by the autonomous driving mode determining unit 160 is a driving mode that meets the driver's driving habit, the driver can feel comfortable as if he/she is driving himself and can reduce the feeling of resistance to autonomous driving.

The device (system) described above may be implemented as a hardware component, a software component, and/or a combination of hardware components and software components. For example, devices and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), It may be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. The device can be commanded. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

Hereinafter, a longitudinal driving habit classification method of a driver according to an embodiment of the present invention will be described with reference to drawings. The driver's longitudinal driving habit classification method according to an embodiment of the present invention is a longitudinal driving habit classification method using the driver's longitudinal driving habit classification system 100 according to an embodiment of the present invention described above.

Referring to FIG. 6 , the longitudinal driving habit classification method of a driver includes analyzing vehicle speed data (1100), obtaining a vehicle speed change curve (1200), deriving a vehicle speed trend line equation (1300), and extracting trend line equation coefficients ( 1400), performing machine learning (1500), classifying driving habits (1600), determining driving habits (1700), and determining an autonomous driving mode (1800).

In the vehicle speed data analysis step 1100 , vehicle speed-related data may be collected and analyzed in real time by the vehicle data analyzer 110 .

In step 1200 of obtaining the vehicle speed change curve, the vehicle speed curve or the vehicle speed change curve may be obtained in real time from the vehicle speed sensor by the vehicle speed data acquisition unit 111 .

In the step of deriving the vehicle speed trend line equation (1300), the vehicle speed change trend line equation may be derived by applying curve fitting to the vehicle speed change curve by the vehicle speed change trend line derivation unit 112. That is, a function expression that mathematically expresses the vehicle speed change trend line may be derived.

In the trend line equation coefficient extraction step 1400 , coefficients of each term may be extracted from the function equation expressing the vehicle speed change trend line by the trend line equation coefficient extractor 113 .

In the machine learning execution step 1500, coefficients are input to the machine learning unit 130 as input values and unsupervised learning is performed, thereby deriving clusters for each driving habit. In this step 1500, the largest coefficient among the coefficients of each term extracted from the vehicle speed change trend line expression expressed as a polynomial is selected, and the vehicle speed change trend line expression expressed as an n-order function expression is an n-order function expression close to the order expression of the term to which the selected coefficient belongs. You can learn repeatedly to become.

In the driving habit classifying step 1600 , the driving habit classifying unit 140 may classify clusters as a result of machine learning by driving habit or define the driving habit.

In the driving habit determination step 1700, the driving habit derived from the vehicle data acquired in real time from the vehicle in which the driver is riding is compared with the cluster value by the driving habit determination unit 150 to classify the driving habit. can do. In addition, the analyzed driving habit may be notified to the corresponding driver. In the self-driving mode determining step 1800 , the self-driving mode determining unit 160 may determine the self-driving mode of the vehicle that meets the driver's driving habit.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CDROMs and DVDs, and magnetic-optical media such as floptical disks. Included are hardware devices specially configured to store and execute program instructions, such as magneto-optical media and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, in one embodiment of the present invention, specific details such as specific components and limited embodiments and drawings have been described, but this is only provided to help a more general understanding of the present invention, and the present invention is based on the above embodiments. It is not limited, and those skilled in the art can make various modifications and variations from these descriptions. Therefore, the spirit of the present invention should not be limited to the described embodiments and should not be determined, and all things equivalent or equivalent to the claims as well as the following claims belong to the scope of the present invention.

100: driver's longitudinal driving habit classification system
110: vehicle data analysis unit 111: vehicle speed data acquisition unit
112: vehicle speed change trend line derivation unit 113: trend line equation coefficient extraction unit
120: big data unit 130: machine learning unit
140: driving habit classification unit 150: driving habit determination unit
160: autonomous driving mode determining unit

Claims (6)

a vehicle data analysis unit that collects and analyzes motion data of a vehicle in which a driver is riding;
a machine learning unit that performs unsupervised learning using trend information of the speed data calculated by the vehicle data analysis unit;
a driving habit classification unit which classifies clusters calculated as learning results from the machine learning unit according to driving habits; and
A driving habit determining unit comparing the classified clusters with the driver's driving habit;
The vehicle data analysis unit may include: a vehicle speed data acquisition unit collecting vehicle speed data and generating a vehicle speed signal change curve; a vehicle speed change trend line derivation unit for deriving a vehicle speed change trend line equation from the vehicle speed signal change curve generated by the vehicle speed data acquisition unit; And a trend line equation coefficient extractor for extracting coefficients of each term from the vehicle speed change trend line equation derived from the vehicle speed change trend line derivation unit;
The machine learning unit performs unsupervised learning by extracting only coefficients of each term from the vehicle speed change trend line equation expressed as a polynomial having n terms and n coefficients of each term, selecting the largest coefficient among the extracted coefficients of n and learning that the vehicle speed change trend line expression expressed as a difference function expression becomes an nth order function expression close to the order expression of the term to which the selected coefficient belongs.
delete According to claim 1,
It includes a big data unit that databases the coefficients of each term extracted from the trend line equation coefficient extraction unit,
The machine learning unit receives coefficients from the big data unit and uses them as inputs for unsupervised learning.
According to claim 3,
The machine learning unit calculates an optimal cluster as coefficients received from the trend line coefficient extraction unit or the big data unit are accumulated.
According to claim 4,
The machine learning unit derives a cluster by matching the order of the term to which the largest coefficient belongs in the vehicle speed change trend line equation through unsupervised learning and the driving habit to derive a cluster.
According to claim 5,
and an autonomous driving mode determination unit for determining an autonomous driving mode of the vehicle to match the driver's driving habit determined by comparing the cluster classified according to the learning result of the machine learning unit with the driver's driving habit. Longitudinal driving habit classification system.