KR102095673B1 - Apparatus and method for real time detection of driving style - Google Patents

Abstract

A real-time driving tendency detection device and a method thereof are disclosed. The real-time driving tendency detection device includes: a sensor unit for detecting at least one driving information; and a driving tendency calculation unit which detects driving tendency by assigning weights to the driving information detected by the sensor unit, and detects the driving tendency through an exponentially weighted moving average (EWMA) technique.

Description

Real-time driving propensity detection device and method {APPARATUS AND METHOD FOR REAL TIME DETECTION OF DRIVING STYLE}

The present invention relates to a real-time driving propensity detection apparatus and method, and more particularly, to a real-time driving propensity detection apparatus and method for calculating and detecting a driver's driving propensity in real time.

In general, drivers driving a vehicle have different driving characteristics. For example, a driver having a high average driving speed, a driver having a low acceleration / deceleration characteristic compared to other drivers, a driver driving fuel economy, or a driver driving a long distance using a highway mainly on a weekend may have various driving characteristics.

The satisfaction of the driver related to the driving performance of the vehicle depends on how much the vehicle is driven according to the driver's propensity.

However, since the propensity of the driver is various as described above, since the performance characteristic of the vehicle is determined as one performance characteristic for the same vehicle type, a difference may occur between the driver's propensity to drive and the reaction of the vehicle.

Therefore, the driver often complains about the driving performance of the vehicle. That is, by grasping the driving propensity of the driver and controlling the shift so that the vehicle responds to the driving propensity of the driver, the satisfaction of the driver with respect to driving performance can be maximized.

In order to meet the needs for the driving performance of the driver as described above, there is a need for a vehicle control system for grasping the driver's driving propensity and controlling the vehicle to suit the identified driver's driving propensity.

Korean Registered Patent Publication No. 10-1484249 (Announcement date 2015.01.16.)

The present invention has been devised to solve this problem, and the driving propensity index is calculated by quantifying the driving propensity, but using the EWMA (Exponentially Weighted Moving Average), the driving propensity index is calculated in real time without using much memory. It is an object of the present invention to provide a real-time driving propensity detection device and method that enable detection.

Real-time driving propensity detection device according to an embodiment of the present invention for achieving the above object, the sensor unit for detecting at least one driving information; It is preferable to include a driving propensity calculating unit that detects driving propensity by assigning weights to the driving information detected by the sensor part, and detects driving propensities through an exponentially weighted moving average (EWMA) technique.

In one embodiment of the present invention, the driving information includes a yaw rate sensor value, an acceleration position sensor (APS) value, a longitudinal / lateral acceleration sensor value, a steering angle sensor value, a wheel speed sensor value, and an IMU (Inertial Measurement). Unit) It is preferable to include at least one of a sensor value and a curvature value.

A real-time driving propensity detection apparatus according to another embodiment of the present invention includes: a sensor unit for detecting at least one driving information; A plurality of driving propensity index calculation units for calculating a driving propensity index by assigning a weight to each driving information detected by the sensor unit, and calculating a driving propensity index through an EWMA technique; It is preferable to include; and a final driving propensity index calculating unit for calculating a final driving propensity index using the driving propensity indexes respectively calculated by the plurality of driving propensity index calculating units.

In another embodiment of the present invention, the driving propensity index calculating unit includes: an index calculating unit that converts an input value input from the sensor unit into an index signal indicating a driving tendency by applying a weight according to a vehicle speed; It is set as time windows of different sizes, and calculates an average value of index signals applied by the index calculator during the set time window period, but calculates a tendency index by calculating an average value by giving higher weight to recent data EWMA Division; It is preferable to include; and a weight-based driving propensity index calculating unit for calculating a driving propensity index corresponding to the input value by calculating an average value of the propensity index output from the plurality of EWMA units.

In another embodiment of the present invention, when the index calculation unit assigns a weight according to a vehicle speed to an input value input from the sensor unit, the index calculation unit refers to an S-shaped weight table that increases in weight as the vehicle speed increases. It is desirable to give.

In another embodiment of the present invention, the weight-based driving propensity index calculating unit calculates a driving propensity index corresponding to the input value by weighting each propensity index based on the driving propensity index calculated in the previous step. When the driving propensity index calculated in the previous step belongs to a preset intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is increased, and the weight of the propensity index output from the EWMA unit having a large time window. Calculate the driving propensity index by reducing, and if the driving propensity index calculated in the previous step exceeds a preset intermediate value, reduce the weight of the propensity index output from the EWMA unit with a small time window and the EWMA with a large time window. It is preferable to increase the weight of the propensity index output from the unit to calculate the driving propensity index.

In another embodiment of the present invention, the final driving propensity index calculating unit calculates the final driving propensity index by calculating the average value after weighting each driving propensity index based on the curvature value applied from the sensor unit. However, if the curvature value applied from the sensor unit is greater than or equal to a predetermined value, the weight of the driving propensity index related to the longitudinal direction is reduced among the driving propensity index, and the weight of the driving propensity index related to the transverse direction and the driving propensity index related to the longitudinal and transverse directions. To increase the final driving propensity index, and if the curvature value is less than or equal to a preset value, increase the weight of the driving propensity index related to the longitudinal direction, and decrease the weight of the driving propensity index related to the transverse direction and the driving propensity index related to the longitudinal and transverse directions. It is desirable to calculate the final driving propensity index.

A real-time driving propensity detection method according to an embodiment of the present invention includes: detecting at least one driving information in a sensor unit; Calculating a driving propensity index by assigning a weight to each driving information detected by the sensor unit in a plurality of driving propensity index calculating units, and calculating a driving propensity index through an EWMA technique; And calculating, by the final driving propensity index calculating unit, a final driving propensity index using the calculated plurality of driving propensity indices.

In one embodiment of the present invention, the calculating of the driving propensity index is a step of converting an index calculation unit into an index signal indicating a driving propensity by assigning a weight according to a vehicle speed to an input value input from the sensor part. ; In a plurality of EWMA units set to time windows of different sizes, the average value of the index signal applied by the index calculator during the set time window period is calculated, but a higher weight is applied to recent data to calculate the average value to calculate the propensity index Calculating; And calculating, by the weighted driving propensity index calculating unit, an average value of the propensity index output from the plurality of EWMA units and calculating a driving propensity index corresponding to the input value.

In one embodiment of the present invention, in the step of converting the index signal, when a weight according to a vehicle speed is applied to an input value input from the sensor unit, an S-shaped weight table that increases in weight as the vehicle speed increases It is desirable to assign weights with reference.

In one embodiment of the present invention, the step of calculating the driving propensity index, weights each propensity index based on the driving propensity index calculated in the previous step, and calculates the average value to correspond to the input value Calculate the driving propensity index, but when the driving propensity index calculated in the previous step falls within a preset intermediate value, increase the weight of the propensity index output from the EWMA part having a small time window and the EWMA part having a large time window. The driving propensity index is calculated by reducing the weight of the propensity index output from, and when the driving propensity index calculated in the previous step exceeds a preset intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is calculated. It is desirable to reduce and increase the weight of the propensity index output from the EWMA unit with a large time window to calculate the driving propensity index. It is.

In one embodiment of the present invention, the step of calculating the final driving propensity index is based on a curvature value applied from the sensor unit, weighting each driving propensity index, and calculating the average value to calculate the final driving propensity Calculate the index, but if the curvature value applied from the sensor unit is greater than or equal to a preset value, reduce the weight of the driving propensity index related to the longitudinal direction among the driving propensity indexes, and the driving propensity index related to the transverse direction and the driving propensity related to the longitudinal and transverse directions. The final driving tendency index is calculated by increasing the weight of the index, and if the curvature value is equal to or less than a predetermined value, the weight of the driving tendency index related to the longitudinal direction is increased, and the driving tendency index related to the lateral direction and the driving tendency index related to the longitudinal direction and the lateral direction are increased. It is preferable to reduce the weight to calculate the final driving propensity index.

The real-time driving propensity detection apparatus and method of the present invention can use EWMA to calculate and detect the driving propensity of the driver in real time without using much memory.

In addition, it is possible to improve driving habits of drivers by feeding the current driving propensity index to drivers who want to maximize fuel efficiency.

In addition, for a driver who enjoys driving, an individual vehicle chassis system according to a driving propensity index can provide an optimal / customized vehicle response to an individual driver.

In addition, by combining the driving situation prediction logic and the driving propensity detection logic, the individual chassis system of the vehicle is changed according to the driving characteristics of the usual driver, thereby maximizing customer satisfaction of autonomous driving.

1 is a view schematically showing the configuration of a real-time driving propensity detection device according to an embodiment of the present invention.
2 is a view showing an exemplary weight graph applied to the present invention.
3 is a view showing a graph of a ratio limit applied to the present invention by way of example.
4 is a view showing an example signal output from the EWMA unit applied to the present invention.
5 is a view for explaining the operation of the weight-based driving propensity index calculation unit applied to the present invention.
6 and 7 are process diagrams for describing a real-time driving propensity detection method according to an embodiment of the present invention.

For a detailed description of the present invention, which will be described later, reference is made to the accompanying drawings that illustrate, by way of example, specific embodiments in which the invention may be practiced. These examples are described in detail enough to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different, but need not be mutually exclusive. For example, certain shapes, structures, and properties described herein may be implemented in other embodiments without departing from the spirit and scope of the invention in connection with one embodiment. In addition, it should be understood that the location or placement of individual components within each disclosed embodiment can be changed without departing from the spirit and scope of the invention. Therefore, the following detailed description is not intended to be taken in a limiting sense, and the scope of the present invention, if appropriately described, is limited only by the appended claims, along with all ranges equivalent to those claimed. In the drawings, similar reference numerals refer to the same or similar functions across various aspects.

Hereinafter, a real-time driving propensity detection apparatus and method according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a view schematically showing the configuration of a real-time driving propensity detection device according to an embodiment of the present invention.

As shown in FIG. 1, the real-time driving propensity detection apparatus 100 according to an embodiment of the present invention may include a sensor unit 110, a driving propensity calculation unit 120, and a control unit 130.

In this configuration, the sensor unit 110 detects various types of driving information necessary to quantify the driving tendency of the driver, and applies each detected driving information to the driving tendency calculating unit 120.

The driving information detected by the sensor unit 110 includes a yaw rate sensor value, an Accel Position Sensor (APS) value, a longitudinal / lateral acceleration sensor value, a steering angle sensor value, a wheel speed sensor value, an IMU ( Inertial Measurement Unit) sensor values, curvature values, and the like. Here, the curvature value may be obtained by reading an applied image from a camera (not shown) photographing a vehicle traveling direction, or may be obtained through calculation using a yaw rate sensor value and a vehicle speed.

The driving tendency calculating unit 120 detects driving tendencies by assigning weights to various driving information detected by the sensor unit 110, but calculates an average value by assigning low weights to old data and higher weights to recent data. The driving tendency is detected through an exponentially weighted moving average (EWMA) technique, and the detected driving tendency is applied to the controller 130.

The driving tendency calculating unit 120 may include a plurality of driving tendency index calculating units 121 and a final driving tendency index calculating unit 129.

The driving propensity index calculating unit 121 calculates a driving propensity index by assigning a weight to each driving information detected by the sensor unit 110, and calculates the calculated driving propensity index to the final driving propensity index calculating unit 129 Apply as.

The driving propensity index calculating unit 121 may calculate the driving propensity index through an EWMA technique that assigns a low weight to old data and an exponentially higher weight to recent data.

The aforementioned driving propensity index calculating unit 121 may be made to correspond to each of the remaining driving information except for information applied to the final driving propensity index calculating unit 129 among the driving information detected by the sensor unit 110. For example, as shown in FIG. 1, the yaw rate sensor value, acceleration position sensor value, longitudinal / lateral acceleration sensor value, steering angle sensor value, wheel speed sensor value, IMU sensor value, curvature value through the sensor unit 110 When detecting, the yaw rate sensor value, acceleration position sensor value, longitudinal / lateral acceleration sensor value, steering angle sensor value, wheel speed sensor value, IMU sensor value excluding the curvature value applied to the final driving propensity index calculator 129 The driving propensity index calculation unit 121 may be implemented in five corresponding to each.

Each driving tendency index calculating unit 121 may include an index calculating unit 123, a plurality of EWMA units 125, and a weight-based driving tendency index calculating unit 127.

The index calculator 123 processes the input values input from the sensor unit 110 and converts them into index signals representing driving tendencies. That is, the input value input from the sensor unit 110 may be weighted according to the vehicle speed and converted into an index signal indicating a driving tendency.

For example, after the yaw rate sensor value is applied, the index calculation unit 123 differentiates the yaw rate sensor value received from the sensor unit 110, and then assigns a weight according to the vehicle speed, but the yaw rate is below a predetermined vehicle speed. The sensor value is set to 0, and above the preset vehicle speed, the yaw rate sensor value can be used as it is to perform the calculation. Here, when assigning a weight according to the vehicle speed, there is generally statistical data in which the size of the vehicle signals (driving information) decreases when the vehicle speed increases. As shown in FIG. 2, the S-shape in which the weight increases as the vehicle speed increases. The weight can be assigned by referring to the weight table of.

In addition, when the index calculator 123 converts an input signal input from the sensor unit 110 into an index signal indicating a driving tendency by assigning a weight according to the vehicle speed, the index ratio may be limited according to the vehicle speed. You can. That is, as illustrated in FIG. 3, when the vehicle speed is low, the conversion ratio may be limited by referring to an S-shaped look-up table in which the conversion ratio is close to 0 and the conversion ratio increases as the vehicle speed increases. .

This is to minimize fluctuations in the driving propensity index when the vehicle is stopped by a traffic light or unintentionally driving at low speed due to traffic jams.

Each EWMA unit 125 calculates the average value of the index signal applied by the index calculation unit 123, but calculates the average value of the index signal during the set time window period. Here, when calculating the average value, the old data is given a low weight, and the recent data is given an exponentially higher weight to calculate the average value to calculate a propensity index.

The EWMA unit 125 may be implemented with time windows each having a different size. For example, assuming that the EWMA unit 125 is made up of 10, the time window of the first EWMA unit may be set to 1 minute, and the time window of the second EWMA unit may be set to 2 minutes to be implemented. And?, The tenth EWMA part can be implemented by setting to 10 minutes.

When the EWMA unit 125 is expressed mathematically, it can be expressed as Equation 1, which is the same as the equation representing the low-pass filter.

), k는 시간 인덱스, T는 시정수(Time Window),

는 샘플링 타임(sampling time:예를 들어, 0.01s)이다.In Equation 1, α is a weight (

), k is the time index, T is the time constant (Time Window),

Is a sampling time (for example, 0.01s).

The EWMA unit 125, which calculates the propensity index by calculating the average value during the time window period by applying the low weight to the old data and the high weight to the latest data, averages the data during the time window period based on the k + 1 time point. Unlike SMA (Simple Moving Average) that calculates, real-time calculation is possible, and as shown in FIG. 4, the time constant T (Time Window) of Equation 1 is adjusted to derive results similar to SMA.

As described above, each EWMA unit 125 set to a time window having a different size calculates a propensity index by calculating an average value of an index signal applied by the index calculation unit 123, and the EWMA unit having a large time window The calculated propensity index has a long-term characteristic that does not change well, that is, the characteristics of the driver's driving habits, and the propensity index calculated in the EWMA unit having a small time window has a short-term characteristic that is frequently changed, that is, a characteristic of the driver's driving intention. Therefore, it is desirable to set the size of the time window differently so that various propensity indexes can be calculated.

The weight-based driving propensity index calculation unit 127 calculates an average value of the propensity index output from each of the plurality of EWMA units 125 to calculate a driving propensity index corresponding to the input value.

When calculating the driving propensity index, the weight-based driving propensity index calculating unit 127 assigns a weight to each propensity index based on the driving propensity index calculated in the previous step, and calculates the average value to correspond to the input value The driving propensity index can be calculated.

Specifically, when the driving tendency index calculated in the previous step belongs to an intermediate value (for example, it may be set to 35 to 65, most preferably to 50), the EWMA having a small time window is small. The driving propensity index may be calculated by increasing the weight of the propensity index output from the unit and reducing the weight of the propensity index output from the EWMA unit having a large time window.

In addition, when the driving propensity index calculated in the previous step exceeds the intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is reduced, and the driving propensity is increased by increasing the weight of the propensity index output from the EWMA unit having a large time window. The index can be calculated.

Here, as the driving propensity index calculated in the previous step deviates significantly from the intermediate value, the weight assigned to the propensity index output from the EWMA unit having a small time window becomes smaller, and the propensity index output from the EWMA unit having a large time window is smaller. The weight can be higher.

For example, if the driver accelerates with a sporting driving intention that dynamically accelerates or decelerates or turns the vehicle, the driving propensity index increases, and the more severe it is, the longer the driving propensity index increases. Is maintained for hours.

That is, as the driving propensity index goes to both extremes (0 or 100), the index is maintained for a long time.

The final driving propensity index calculating unit 129 calculates the final driving propensity index using the driving propensity indexes respectively calculated by the plurality of driving propensity index calculating units 121, and drives the driver based on the calculated final driving propensity index The propensity is detected, and the detected driving propensity is applied to the controller 130.

The driving propensity index applied from the plurality of driving propensity index calculation units 121 to the final driving propensity index calculation unit 129 includes a longitudinal direction-related driving tendency index, a lateral direction-related driving tendency index, and a type in which both the longitudinal and lateral directions are related. It can be divided into driving propensity index related to direction and transverse direction.

Therefore, the final driving propensity index calculating unit 129 calculates the average of the driving propensity indexes calculated by the plurality of driving propensity index calculating units 121, respectively, and calculates the final driving propensity index, and the curvature applied from the sensor unit 110 Each driving propensity index is weighted based on the value, and the average of the driving propensity index is averaged to calculate the final driving propensity index.

Specifically, if the curvature value applied from the sensor unit 110 is greater than or equal to a predetermined value, the weight of the driving propensity index related to the longitudinal direction is reduced, and the weight of the driving propensity index related to the transverse direction and the longitudinal and transverse direction is increased. The final driving propensity index can be calculated.

In addition, if the curvature value applied from the sensor unit 110 is equal to or less than a preset value, the weight of the driving propensity index related to the longitudinal direction is increased, and the weight of the driving propensity index related to the transverse direction and the driving propensity index related to the longitudinal direction and the transverse direction is reduced to final driving. The propensity index can be calculated.

The reason why the final driving propensity index calculating unit 129 calculates the final driving propensity index based on the curvature value applied from the sensor unit 110 is to calculate the driving propensity index suitable for the driving situation.

Meanwhile, the controller 130 may automatically change the driving mode of the vehicle based on the driving propensity detected by the driving propensity calculator 120.

6 and 7 are process diagrams for describing a real-time driving propensity detection method according to an embodiment of the present invention.

First, the driving propensity detection device 100 detects various driving information through the sensor unit 110 to quantify the driving propensity of the driver (S100).

The driving information detected in step S100 may include yaw rate sensor values, acceleration position sensor values, longitudinal / lateral acceleration sensor values, steering angle sensor values, wheel speed sensor values, IMU sensor values, curvature values, and the like.

Thereafter, a driving tendency is detected by assigning a weight to each driving information detected in step S100, but the driving tendency is detected through an EWMA technique (S200).

Hereinafter, the above-described step S200 will be described in more detail with reference to FIG. 7.

Each of the driving information detected in the above step S100 is applied to each driving propensity index calculating unit 121 provided in the driving propensity calculating unit 120, and the driving information detected from the sensor unit 110 is received. The index calculating unit 123 of the driving tendency index calculating unit 121 processes each input value input from the sensor unit 110 and converts it into an index signal indicating the driving tendency (S210).

In the above-described step S210, the index calculator 123 converts the input value input from the sensor unit 110 into an index signal indicating a driving tendency by assigning a weight according to the vehicle speed, as shown in FIG. 2. As the weight increases, the weight can be converted into an index signal by assigning weights according to an S-shaped weight table.

For example, after the yaw rate sensor value is applied, the index calculation unit 123 differentiates the yaw rate sensor value received from the sensor unit 110, and then assigns a weight according to the vehicle speed, but the yaw rate is below a predetermined vehicle speed. The sensor value is set to 0, and above the preset vehicle speed, the yaw rate sensor value can be used as it is to perform the calculation. Here, the index calculating unit 123 may convert the yaw rate sensor value into an index signal by giving an S-shaped weight that increases in weight as the vehicle speed increases.

In addition, in the above step S210, the index calculator 123 stops by a traffic light, or when driving at an unintended low speed due to traffic jams, etc., in order to minimize fluctuations in the driving propensity index, convert when the vehicle speed is low. The conversion ratio may be limited by referring to an S-shaped look-up table in which the ratio is close to 0 and the conversion ratio increases as the vehicle speed increases.

The index signals converted in step S210 are applied to a plurality of EWMA units 125 each having a time window of a different size, and each EWMA unit 125 is applied by the index calculation unit 123 during a set time window period. The inclination index is calculated by calculating the average value of the index signals (S220).

In the above-described step S220, each EWMA unit 125 may calculate a propensity index by calculating an average value during a time window period by giving a low weight to old data and a higher weight to recent data.

By applying the index signal to the plurality of EWMA units 125 set in time windows of different sizes in step S220, the propensity index with long-term characteristics (driver's driving habits) and short-term characteristics (driver's driving will) Various propensity indexes such as propensity indexes can be calculated.

The propensity index calculated through the above step S220 is applied to the weight-based driving propensity index calculating unit 127, and the weight-based driving propensity index calculating unit 127 is the propensity index applied from the plurality of EWMA units 125, respectively. The driving tendency index corresponding to the input value is calculated by calculating the average value (S230).

In the above-described step S230, the weight-based driving tendency index calculating unit 127 calculates the driving tendency index, assigns a weight to each tendency index based on the driving tendency index calculated in the previous step, and calculates the average value thereof. When the driving propensity index corresponding to the input value is calculated, when the driving propensity index falls within an intermediate value (for example, it may be set to 35 to 65, most preferably 50), time The driving propensity index may be calculated by increasing the weight of the propensity index output from the EWMA unit having a small window and reducing the weight of the propensity index output from the EWMA unit having a large time window. In addition, when the driving propensity index calculated in the previous step exceeds the intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is reduced, and the driving propensity is increased by increasing the weight of the propensity index output from the EWMA unit having a large time window. The index can be calculated.

As described above, when the driving inclination index is calculated by the plurality of driving inclination index calculating units 121 through the above-described step S200, the driving inclination index calculated by the plurality of driving inclination index calculating units 121 is the final driving tendency. Applied to the index calculation unit 129, the final driving propensity index calculating unit 129 calculates the final driving propensity index by obtaining the average of the driving propensity indexes received from the plurality of driving propensity index calculating units 121 (S300). .

In the above-described step S300, the final driving propensity index calculating unit 129 assigns a weight to each driving propensity index based on the curvature value applied from the sensor unit 110, averages the driving propensity index, and obtains the final driving propensity When calculating the index, if the curvature value is greater than or equal to a preset value, reduce the weight of the driving propensity index related to the longitudinal direction, and increase the weight of the driving propensity index related to the transverse direction and the driving propensity index related to the transverse direction to calculate the final driving propensity index. can do. In addition, if the curvature value applied from the sensor unit 110 is equal to or less than a preset value, the weight of the driving propensity index related to the longitudinal direction is increased, and the weight of the driving propensity index related to the transverse direction and the driving propensity index related to the longitudinal direction and the transverse direction is reduced to final driving. The propensity index can be calculated.

The real-time driving propensity detection method according to an embodiment of the present invention may be implemented as an application or implemented in the form of program instructions that can be executed through various computer components to be recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, or the like alone or in combination.

The program instructions recorded on the computer-readable recording medium are specially designed and configured for the present invention, and may be known and usable by those skilled in the computer software field.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs, DVDs, and magneto-optical media such as floptical disks. media), and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like.

Examples of program instructions include not only machine language codes produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

Although described above with reference to embodiments, those skilled in the art understand that various modifications and changes can be made to the present invention without departing from the spirit and scope of the present invention as set forth in the claims below. Will be able to.

110. Sensor unit, 120. Driving propensity calculation unit,
121. Driving propensity index calculation unit, 123. Index calculation unit,
125. EWMA unit, 127. Weight-based driving propensity index calculation unit,
129. Final driving propensity index calculation unit, 130. Control unit

Claims (13)

A sensor unit detecting at least one driving information;
A plurality of driving propensity index calculation units for calculating a driving propensity index by assigning a weight to each driving information detected by the sensor unit, and calculating a driving propensity index through an EWMA technique; And
It includes; a final driving propensity index calculating unit for calculating a final driving propensity index by using the driving propensity indexes respectively calculated by the plurality of driving propensity index calculating parts;
The driving propensity index calculation unit,
An index calculating unit that assigns a weight according to the vehicle speed to an input signal input from the sensor unit and converts it into an index signal indicating a driving tendency, but assigns weight by referring to an S-shaped weight table whose weight increases as the vehicle speed increases;
It is set as time windows of different sizes, and calculates an average value of index signals applied by the index calculator during the set time window period, but calculates a tendency index by calculating an average value by giving higher weight to recent data EWMA Division; And
And a weight-based driving propensity index calculation unit calculating an average driving propensity index output from the plurality of EWMA units and calculating a driving propensity index corresponding to the input value. According to claim 1,
The driving information,
It includes at least one of yaw rate sensor value, acceleration position sensor (APS) value, longitudinal / lateral acceleration sensor value, steering angle sensor value, wheel speed sensor value, IMU (Inertial Measurement Unit) sensor value, and curvature value. , Real-time driving propensity detection device. According to claim 1,
The index calculation unit,
When the input value input from the sensor unit is weighted according to the vehicle speed and converted into an index signal indicating a driving tendency, the weight is referred to in an S-shaped lookup table in which the conversion ratio increases as the vehicle speed increases. Driving propensity detection device. delete delete According to claim 1,
The weight-based driving propensity index calculation unit,
Based on the driving propensity index calculated in the previous step, a weight is applied to each propensity index to calculate a driving propensity index corresponding to the input value,
When the driving propensity index calculated in the previous step belongs to a preset intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is increased, and the weight of the propensity index output from the EWMA unit having a large time window is increased. To calculate the driving propensity index,
When the driving propensity index calculated in the previous step exceeds a preset intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is reduced, and the weight of the propensity index output from the EWMA unit having a large time window is reduced. Real-time driving propensity detection device to calculate driving propensity index by increasing. According to claim 1,
The final driving propensity index calculation unit,
After weighting each driving propensity index based on a curvature value applied from the sensor unit, calculating the average value to calculate a final driving propensity index,
If the curvature value applied from the sensor unit is greater than or equal to a predetermined value, the weight of the driving tendency index related to the longitudinal direction is decreased among the driving tendency indexes, and the weights of the driving tendency index related to the lateral direction and the driving tendency index related to the longitudinal direction and the lateral direction are increased. Calculate the final driving propensity index,
When the curvature value is equal to or less than a predetermined value, real-time driving is performed by increasing the weight of the driving propensity index related to the longitudinal direction and reducing the weights of the driving propensity index related to the transverse direction and the driving propensity index related to the longitudinal and transverse directions to calculate the final driving propensity index. Propensity detection device. Detecting at least one driving information by the sensor unit;
Calculating a driving propensity index by assigning a weight to each driving information detected by the sensor unit in a plurality of driving propensity index calculating units, and calculating a driving propensity index through an EWMA technique; And
In the final driving propensity index calculation unit, using the calculated plurality of driving propensity index to calculate the final driving propensity index; includes,
The step of calculating the driving propensity index,
In the index calculation unit, an input signal input from the sensor unit is assigned a weight according to the vehicle speed to indicate a driving tendency, but an index signal is assigned to the weight by referring to an S-shaped weight table whose weight increases as the vehicle speed increases. Converting;
In a plurality of EWMA units set to time windows of different sizes, an average value of an index signal applied by the index calculator during the set time window period is calculated, but a higher weight is applied to recent data to calculate the average value to calculate the propensity index Calculating; And
And calculating, by the weight-based driving propensity index calculating unit, an average value of the propensity index output from the plurality of EWMA units and calculating a driving propensity index corresponding to the input value. The method of claim 8,
In the step of converting to the index signal,
When the input value input from the sensor unit is weighted according to the vehicle speed and converted into an index signal indicating a driving tendency,
A method for detecting a driving propensity in real-time by assigning weights by referring to an S-shaped lookup table in which the conversion ratio increases as the vehicle speed increases. delete The method of claim 8,
The step of calculating the driving propensity index,
After weighting each inclination index based on the driving inclination index calculated in the previous step, the average value is calculated to calculate a driving inclination index corresponding to the input value,
When the driving propensity index calculated in the previous step belongs to a preset intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is increased, and the weight of the propensity index output from the EWMA unit having a large time window is increased. To calculate the driving propensity index,
When the driving propensity index calculated in the previous step exceeds a preset intermediate value, the weight of the propensity index output from the EWMA unit having a small time window is reduced, and the weight of the propensity index output from the EWMA unit having a large time window is reduced. Real-time driving propensity detection method, which is the step of calculating the driving propensity index by increasing. The method of claim 8,
The step of calculating the final driving propensity index,
After weighting each driving propensity index based on a curvature value applied from the sensor unit, calculating the average value to calculate a final driving propensity index,
If the curvature value applied from the sensor unit is greater than or equal to a preset value, the weight of the driving tendency index related to the longitudinal direction is reduced among the driving tendency indexes, and the weights of the driving tendency index related to the lateral direction and the driving tendency index related to the longitudinal direction and the lateral direction are increased. Calculate the final driving propensity index,
If the curvature value is less than or equal to a predetermined value, increasing the weight of the driving propensity index related to the longitudinal direction and reducing the weights of the driving propensity index related to the transverse direction and the driving propensity index related to the transverse direction to calculate the final driving propensity index, Real-time driving propensity detection method. A computer-readable recording medium having a computer program recorded thereon for performing the real-time driving propensity detection method according to any one of claims 8, 9, 11 and 12.

Images