Vehicle with driver warning system and method of control

Patent metadata

Field	Value
Publication number	US-10723348-B2
Application number	US-201816029010-A
Country	US
Kind code	B2
Filing date	Jul 6, 2018
Priority date	Feb 20, 2018
Publication date	Jul 28, 2020
Grant date	Jul 28, 2020

Abstract

A vehicle, which adjusts a collision warning time based on driver's field of view and visibility of the object, is provided. The vehicle includes: a distance sensor that collects data of an object, a camera that collects image data, a driver recognition sensor that collects driver's facial data, and a controller that analyzes driver's gaze and field of view using the driver's facial data, analyzes visibility of the object using the image data, and calculates a collision warning time based on a result of analyzing the driver's gaze and field of view and the visibility of the object, and an alarm device that informs the driver of the risk of collision at the collision warning time under the control of the controller.

First claim

What is claimed is:

1. A vehicle comprising:
a distance sensor configured to collect data of an object;
a camera configured to collect image data;
a driver recognition sensor configured to collect facial data of a driver; and
a controller configured to analyze a gaze of the driver and a field of view using the facial data, to analyze visibility of the object using the image data, and to calculate a collision warning time based on a result of analyzing the gaze and the field of view and the visibility of the object,
wherein the controller is configured to:
control at least one of a warning light, an audio system or an audio video navigation (AVN) system to inform the driver of a risk of collision at the collision warning time,
calculate an attention distribution function of the driver by analyzing the gaze and the field of view,
calculate a visibility index by analyzing the visibility of the object,
calculate a recognition index using the attention distribution function and the visibility index, and
calculate the visibility index of the object by extracting a color coordinate value of the object and a color coordinate value of a surrounding environment of the object from the image data and subtracting a sum of the color coordinate value of the object and the color coordinate value of the surrounding environment of the object from a maximum color coordinate value.

2. The vehicle according to claim 1 , wherein the controller is configured to calculate the collision warning time using the recognition index and a predetermined reference warning time.

3. The vehicle according to claim 2 , wherein the collision warning time is calculated according to the following equation:

Th
p

=

(

2

1
+
DI

)

×

Th
basic

where, Th p is collision warning time, DI is a recognition index, and Th basic is reference warning time.

4. The vehicle according to claim 1 , wherein the controller is configured to calculate the attention distribution function on a two-dimensional coordinate system in which a center line of the field of view is set as a vertical axis and a straight line connecting coordinates of the object and the vertical axis is set as a horizontal axis.

5. The vehicle according to claim 4 , wherein the attention distribution function is defined as a normal distribution density function.

6. The vehicle according to claim 1 , wherein the recognition index is calculated according to the following equation:

DI
=

765
-
I

765

×

f
⁡

(
α
)

where, DI is a recognition index, I is a visibility index, f(a) is an attention distribution function value of a driver.

7. The vehicle according to claim 1 , further comprising a braking device configured to brake the vehicle at the collision warning time under a control of the controller.

8. A control method of a vehicle, comprising:
collecting data of an object, image data, and facial data of a driver;
analyzing, by a controller, a gaze of the driver, and a field of view using the facial data of the driver;
analyzing, by the controller, visibility of the object using the image data;
calculating, by the controller, a collision warning time based on a result of analyzing the gaze, the field of view and the visibility of the object;
informing, by an alarm device, the driver of a risk of collision at the collision warning time,
wherein:
analyzing the gaze and the field of view further comprises calculating, by the controller, an attention distribution function of the driver, and
analyzing the visibility of the object using the image data comprises calculating, by the controller, a visibility index of the object; and

calculating, by the controller, a recognition index using the attention distribution function of the driver and the visibility index of the object,
wherein:
calculating the visibility index of the object comprises extracting, by the controller, a color coordinate value of the object and a color coordinate value of a surrounding environment of the object from the image data, and
the visibility index is calculated by subtracting a sum of the color coordinate value of the object and the color coordinate value of the surrounding environment of the object from a maximum color coordinate value.

9. The method according to claim 8 , wherein the calculation of the collision warning time is performed by the controller using the recognition index and a predetermined reference warning time.

10. The method according to claim 9 , wherein the collision warning time is calculated according to the following equation:

Th
p

=

(

2

1
+
DI

)

×

Th
basic

where, Th p is collision warning time, DI is a recognition index, and Th basic is reference warning time.

11. The method according to claim 8 , wherein calculating the attention distribution function of the driver further comprises:
setting, by the controller, a two-dimensional coordinate system in which a center line of the field of view of the driver is set as a vertical axis and a straight line connecting coordinates of the object and the vertical axis is set as a horizontal axis; and
wherein the attention distribution function of the driver is calculated, by the controller, on the two-dimensional coordinate system.

12. The method according to claim 11 , wherein the attention distribution function of the driver is defined as a normal distribution density function.

13. The method according to claim 8 , wherein the recognition index is calculated according to the following equation:

DI
=

765
-
I

765

×

f
⁡

(
α
)

where, DI is a recognition index, I is a visibility

Assignees

• Hyundai Motor Co Ltd
• Kia Motors Corp

Inventors

• Lee Tae Young
• Kim Eungseo
• Sung Donghyun
• Park Jonghyeok
• Lee Sangmin
• Kim Junghyun
• Kim Donghyuk
• Kwon Yongseok

Classifications

• G06V40/193

  Preprocessing; Feature extraction · CPC title

• G06V20/597

  Recognising the driver's state or behaviour, e.g. attention or drowsiness · CPC title

• B60T2201/022

  Collision avoidance systems · CPC title

• B60W2710/182

  Brake pressure, e.g. of fluid or between pad and disc · CPC title

• B60W2540/229

  Attention level, e.g. attentive to driving, reading or sleeping · CPC title