KR102276393B1 - Vehicle Control System For Preventing The Vehicle Crashing On front Side, And Control Method Thereof - Google Patents

Abstract

Disclosed are a vehicle control system and a control method for preventing forward collision. A vehicle control system for preventing forward collision according to an embodiment of the present invention is a vehicle control system for preventing forward collision that detects a crosswalk situation in front of a vehicle and controls the vehicle based on the detected data, and the road situation in front of the vehicle an image information collecting unit that detects and transmits the detected data to the recognition processing unit; a recognition processing unit that selects a crosswalk sign and the rear side of a stopped vehicle from the data detected through the image information collection unit, detects the crosswalk sign data and data related to the identified vehicle location information, and transmits it to the determination transmission unit; a determination transmission unit for calculating an appearance occurrence time of a pedestrian moving through a crosswalk based on the data received through the recognition processing unit, and changing the gear state of the vehicle using the calculated time value; and an output unit for outputting an audible or visual warning to the vehicle driver by a timer set according to the time value calculated through the determination transmission unit.

Description

Vehicle Control System For Preventing The Vehicle Crashing On front Side, And Control Method Thereof

The present invention relates to a vehicle control system for preventing forward collision, and more particularly, to a vehicle control system capable of performing preventive measures regarding a collision between a pedestrian and a vehicle.

In general, cameras are introduced and utilized for various purposes in the field of active safety vehicle (ASV).

For example, a camera may be installed in a vehicle to secure a driver's field of view. In this case, the camera is installed so that the driver can provide the driver with a photograph of a blind spot that cannot be seen at all through a rear mirror or a room mirror. As a result, the driver can control the vehicle by easily checking the blind spot by viewing the image captured by the camera.

Recently, cameras have been widely used not only to broaden the driver's field of vision, but also to prevent traffic accidents while driving. The reason is that a camera is much cheaper than a sensor such as a radar or a lidar, and has the advantage that various functions can be implemented using a single camera.

Meanwhile, accidents in which the driver's vehicle collides with the front vehicle frequently occur due to the driver's negligence or sudden braking of the vehicle in front.

Some drivers do not recognize the braking distance even when a yellow or red traffic light indicating a stop signal is lit, so they stop over the vehicle stop line or continue driving by ignoring the stop signal.

In addition, if the driver cannot identify the pedestrian due to the vehicle stopped in the front when the vehicle driving ahead enters the crosswalk, the safety of the pedestrian cannot be guaranteed.

Therefore, there is a demand for measures to reduce traffic accidents and protect both drivers and pedestrians.

Korean Patent Publication No. 10-2019-0074339 (published on June 28, 2019)

An object of the present invention is to prevent a collision with a pedestrian by notifying the driver of a pre-collision warning and limiting the vehicle speed when the driver cannot identify the pedestrian due to the vehicle stopped in the side or front when the vehicle driving ahead enters the crosswalk area It is to provide a vehicle control system and vehicle control method for preventing forward collision including a preventable configuration.

According to one aspect of the present invention as a means of solving the problem,

A vehicle control system for preventing forward collision that detects a crosswalk situation in front of a vehicle and controls the vehicle based on the detected data,

an image information collecting unit that detects the road condition in front of the vehicle and transmits the detected data to the recognition processing unit;

a recognition processing unit that selects a crosswalk sign and the rear side of a stopped vehicle from the data detected through the image information collection unit, detects the crosswalk sign data and data related to the identified vehicle location information, and transmits it to the determination transmission unit;

a determination transmission unit for calculating an appearance occurrence time of a pedestrian moving through a crosswalk based on the data transmitted through the recognition processing unit, and changing the gear state of the vehicle using the calculated time value; and

An output unit for outputting an audible or visual warning to the vehicle driver by a timer set according to the time value calculated through the determination transmission unit; provides a vehicle control system for front collision prevention comprising a.

In some cases, the recognition processing unit may include an external image recognition unit that receives and stores the image information detected through the image information collection unit.

In addition, the recognition processing unit, from the image information detected through the image information collecting unit, a crosswalk sign located on the road ahead, a crosswalk, an object sorting unit for recognizing a vehicle and an obstacle; may include.

In this case, it is preferable that the object sorting unit selects a crosswalk sign, a crosswalk, a vehicle and an obstacle located on the road ahead by executing the deep learning-based real-time object classification algorithm YOLO (You Only Look Once).

In addition, the recognition processing unit, in the image information detected through the image information collection unit, transmits messages about crosswalk signs, crosswalks, vehicles, obstacles and pedestrians to the determination transmission unit using CAN communication, which is a vehicle communication protocol. It may include a signal processing unit.

Preferably, the determination transmission unit, from the data received from the recognition processing unit, calculates the maximum time that a pedestrian may appear based on a crosswalk sign, a crosswalk, a vehicle, and an obstacle, and sets a timer based on this It may include a collision determination unit;

In this case, the output unit may output an audible or visual warning to the vehicle driver by a timer set in the collision determination unit.

Preferably, the shift determination unit may include a shift request processing unit that changes the gear state of the vehicle based on the vehicle inclination value and the accelerator position sensor value (APS value) received through a vehicle sensor mounted on the vehicle. .

According to another aspect of the present invention as a means of solving the problem,

A vehicle control method for preventing forward collision that detects a crosswalk situation in front of a vehicle and controls the vehicle based on the detected data,

a road image information acquisition step of detecting a road condition in front of the vehicle;

a road image information identification step of selecting a crosswalk sign, a crosswalk, a vehicle and an obstacle located on the road ahead from the detected road condition data in front of the vehicle;

A pedestrian appearance time calculation step of calculating the appearance time assuming that the pedestrian is not visible to the driver because the vehicle is covered by an obstacle, based on the data selected through the road image information identification step;

A timer setting step of setting a timer according to the appearance time calculated from the pedestrian appearance time calculation step;

a warning output step of outputting an audible or visual warning of the possibility of collision with a pedestrian during the timer time set from the timer setting step; and

When the speed of the vehicle is not decelerated for the timer time set from the timer setting step, a shift control step of forcibly shifting the gear state of the vehicle to a lower gear to reduce the speed of the vehicle; can provide

Preferably, the road image information identification step is executed based on YOLO (You Only Look Once), a real-time object classification algorithm based on deep learning, to select a crosswalk sign, a crosswalk, a vehicle and an obstacle located on the road ahead. have.

In addition, in the step of calculating the pedestrian appearance time, in the data received from the road image information identification step, the appearance of a pedestrian is determined based on location data, size data, and time data of crosswalk signs, crosswalks, vehicles, obstacles and pedestrians. The maximum time that can occur can be calculated.

Preferably, the shift control step may change the gear state of the vehicle based on a vehicle inclination value and an accelerator position sensor value (APS value) received through a vehicle sensor mounted on the vehicle.

Also, in the shift control step, when the accelerator position sensor value (APS value) is decreased by a predetermined amount by the driver during the timer time set from the timer setting step, the upshift of the vehicle may be blocked.

Preferably, in the shift control step, the amount of change per hour (APS Grad) of the value of the accelerator position sensor, the slope of the vehicle, the value of the accelerator position sensor (APS value), the current gear number of the vehicle, and data related to the set timer Only when the set reference value is satisfied, the gear state of the vehicle can be forcibly shifted to a lower gear.

In this case, in the shift control step, the engine torque may be reduced by shifting the gear state of the vehicle to a lower gear and simultaneously outputting an engine torque reduction request.

According to the vehicle control system for preventing forward collision of the present invention, by having an image information collecting unit, a recognition processing unit, a shift determining unit, and an output unit that perform a specific role, the vehicle is stopped in the front side when entering the crosswalk area of the vehicle driving ahead. When the driver fails to identify the pedestrian due to the vehicle, it is possible to provide a vehicle control system capable of preventing a collision with a pedestrian by notifying the driver of a prior collision warning and limiting the vehicle speed.

In addition, according to the vehicle control method for preventing forward collision of the present invention, the road image information acquisition step, the road image information identification step, the pedestrian appearance time calculation step, the timer setting step, the warning output step, and the shift control step of performing a specific process By including, if the driver fails to identify pedestrians due to vehicles stopped in the side-front when entering the crosswalk area of vehicles traveling in front, collisions with pedestrians can be prevented by notifying the driver of a pre-collision warning and limiting the vehicle speed. A vehicle control method can be provided.

1 is a configuration diagram illustrating a vehicle control system for preventing forward collision according to an embodiment of the present invention.
2 is a flowchart illustrating a vehicle control method for preventing forward collision according to an embodiment of the present invention.
3 is a schematic diagram showing the situation on the crosswalk.

Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the drawings.

In describing the present invention, the terms used in the following specification are only used to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise.

In addition, in this specification, terms such as "comprise" or "have" are intended to designate that the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification exist, but one or more other It is to be understood that this does not preclude the possibility of addition or presence of features or numbers, steps, operations, components, parts, or combinations thereof.

Also, terms such as first and second may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

In addition, terms such as “…unit”, “…unit”, “…module”, etc. described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software. can

In the description with reference to the accompanying drawings, the same reference numerals will be assigned to the same components, and repeated descriptions of the same components will be omitted. In the description of the present invention, when it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

1 is a block diagram illustrating a vehicle control system for preventing forward collision according to an embodiment of the present invention, and FIG. 2 is a flowchart illustrating a vehicle control method for preventing forward collision according to an embodiment of the present invention. is shown. In addition, Fig. 3 is a schematic diagram showing the situation on the crosswalk.

First, referring to FIG. 1 , the vehicle control system 100 for preventing forward collision according to the present embodiment is a system for preventing forward collision that detects a crosswalk situation in front of a vehicle and controls the vehicle based on the detected data. As such, it may include an image information collection unit 110 , a recognition processing unit 120 , a determination transmission unit 130 , and an output unit 140 performing a specific role.

Specifically, the image information collection unit 110 may detect a road condition in front of the vehicle, and then transmit the detected data to the recognition processing unit 120 .

Preferably, the recognition processing unit 120 according to the present embodiment may include an external image recognition unit 121 , an object selection unit 122 , and a signal processing unit 123 .

In this case, the external image recognition unit 121 receives and stores the image information detected through the image information collection unit 110 .

The object selection unit 122 may recognize a crosswalk sign, a crosswalk, a vehicle, and an obstacle located on the road ahead from the image information detected through the image information collection unit 110 . At this time, the object selection unit 122 may select a crosswalk sign, a crosswalk, a vehicle, and an obstacle located on the road ahead by executing the deep learning-based real-time object classification algorithm YOLO (You Only Look Once). .

The signal processing unit 123, from the image information detected through the image information collection unit 110, a message about a crosswalk sign, a crosswalk, a vehicle, an obstacle, and a pedestrian using CAN communication, a vehicle communication protocol, a decision transmission unit may be sent to 130 .

The recognition processing unit 120 selects the crosswalk sign and the rear side of the stopped vehicle from the data detected through the image information collection unit 110, and detects the data on the crosswalk sign data and the location information of the identified vehicle. It can be transmitted to the determination transmission unit (130).

The determination transmission unit 130 calculates the occurrence time of the appearance of a pedestrian moving through the crosswalk based on the data received through the recognition processing unit 120, and changes the gear state of the vehicle using the calculated time value. can

Specifically, as shown in FIG. 1 , the determination transmission unit 130 according to the present embodiment may include a collision determination unit 131 and a shift request processing unit 132 performing a specific role.

The collision determination unit 131, in the data received from the recognition processing unit 120, based on the location data and size data and time data of a crosswalk sign, a crosswalk, a vehicle, an obstacle, and a pedestrian, the appearance of a pedestrian may occur You can calculate the maximum time and set a timer based on it. In this case, the output unit 140 may output an audible or visual warning to the vehicle driver by a timer set by the collision determination unit 131 .

The shift request processing unit 132 may change the gear state of the vehicle based on the vehicle inclination value and the accelerator position sensor value (APS value) received through the vehicle sensor mounted on the vehicle.

Meanwhile, the output unit 140 may output an audible or visual warning to the vehicle driver by a timer set according to the time value calculated through the determination transmission unit 130 .

Hereinafter, a vehicle control method ( S100 ) for preventing forward collision will be described using each configuration constituting the vehicle system 100 for preventing forward collision described above.

The vehicle control method (S100) for preventing forward collision according to the present embodiment includes a road image information acquisition step (S110) for performing a specific process, a road image information identification step (S120), a pedestrian appearance time calculation step (S130), It includes a timer setting step (S140), a warning output step (S150), and a shift control step (S160).

Specifically, in the road image information acquisition step (S110), a road condition in front of the vehicle is detected. In this case, the detected road condition in front of the vehicle means information about a crosswalk sign, a crosswalk, a vehicle, an obstacle, and a pedestrian.

In the road image information acquisition step (S110), the corresponding information is detected using the image information collection unit 110 mounted on the front of the vehicle. In this case, as a configuration that can be employed in the image information collection unit 110, an image photographing device, an infrared camera, an ultrasonic measuring device, etc. may be exemplified, but the present invention is not limited thereto. In some cases, the image information collection unit 110 may obtain current state information of the traffic light by performing wireless communication with the corresponding traffic light. In this case, it goes without saying that a wireless communication device capable of providing traffic light information through wireless communication with the vehicle should be mounted on the corresponding traffic light.

In the road image information identification step (S120), a process of selecting a crosswalk sign, a crosswalk, a vehicle, and an obstacle located on the road ahead from the detected road condition data in front of the vehicle is performed.

Referring to section P1 of FIG. 3 , section P1 represents a section in which the road image information acquisition step ( S110 ) is performed using the image information collecting unit 110 .

At this time, the road image information identification step (S120) is performed by the recognition processing unit 120 according to the present embodiment, and is executed based on YOLO (You Only Look Once), a real-time object classification algorithm based on deep learning, You can select crosswalk signs, crosswalks, vehicles and obstacles located in Through a deep learning-based real-time object classification algorithm (YOLO), pre-trained signs, vehicles, and crosswalks can be identified. Objects identified by the Real-Time Object Classification Algorithm (YOLO) are always created around each frame.

In the pedestrian appearance time calculation step ( S130 ), based on the data selected through the road image information identification step ( S120 ), the appearance time is calculated by assuming that the pedestrian is not visible to the driver because it is hidden by a vehicle or obstructed by an obstacle. As illustrated in FIG. 3 , there may be a case in which the pedestrian is not recognized by the vehicle V2 stopped in front of the vehicle V1 on which the driver is riding. At this time, in the pedestrian appearance time calculation step ( S130 ), an expected time for the pedestrian to appear is calculated.

Specifically, in the data received from the road image information identification step (S120), the maximum time that a pedestrian can appear based on location data, size data, and time data of crosswalk signs, crosswalks, vehicles, obstacles and pedestrians can be calculated.

As shown in FIG. 3, the position and width of the vehicle are detected based on the image information about the vehicle V2 stopped in front, and the crosswalk covered by the position and width of the vehicle V2 stopped in front ( 20) can be calculated. By dividing the width of the obscured crosswalk 20 by the pedestrian's general average walking speed (eg, 3 to 5 km/h), it is assumed that the pedestrian is present, and the appearance time can be calculated. Specifically, when the width obscured by the vehicle V2 stopped in front is 2 m, the pedestrian's average walking speed value is divided by 3 km/h (0.83 m/sec) to calculate the pedestrian appearance time value of 2.41 sec. In this case, the maximum value of the pedestrian appearance time value may be calculated by detecting the maximum value of the width obscured by the vehicle V2 stopped in front and setting the average walking speed value of the pedestrian as the minimum value.

In the timer setting step (S140), a timer is set according to the appearance time calculated from the pedestrian appearance time calculation step (S130). The timer is set using the maximum value of the pedestrian appearance time mentioned above. In some cases, the time for the vehicle to reach the crosswalk may be calculated using the current driving speed value of the vehicle and the distance value between the crosswalk and the crosswalk. At this time, when the time value at which the vehicle arrives at the crosswalk is shorter than the maximum value of the pedestrian appearance time, it is preferable to set the timer using the time value at which the vehicle arrives at the crosswalk.

In the warning output step (S150), an audible or visual warning of the possibility of collision with a pedestrian during the timer time set from the timer setting step (S140) is outputted.

In the shift control step S160 , when the speed of the vehicle is not decelerated for the timer time set from the timer setting step S140 , the speed of the vehicle may be reduced by forcibly shifting the gear state of the vehicle to a lower gear.

Referring to section P2 of FIG. 3 , section P2 is a warning output step (S150) and shift control step (S160) of the vehicle (V1) at the point in time when the crosswalk guide sign in front and the vehicle (V2) stopping are identified. ) indicates the section in which it is executed.

The vehicle V1 in which the driver is boarded receives a prior collision warning notice according to the identification of the crosswalk sign S and the vehicle V2 stopped in front. At this time, the maximum time for the pedestrian to appear is calculated according to the width of the vehicle frame detected at the same time, and the collision warning guidance continues while the time is consumed. The shift control method of the vehicle is applied differently depending on how much the driver recognizes this and presses the accelerator pedal.

Specifically, the gear state of the vehicle may be changed based on the vehicle inclination value and the accelerator position sensor value (APS value) received through the vehicle sensor mounted on the vehicle. If the vehicle's inclination value is smaller than the preset reference value (in the case of a downhill slope) and the vehicle's gear stage is greater than the preset reference value (in high-speed driving), the vehicle's speed is reduced by forcibly shifting the vehicle's gear to a lower gear. can be slowed down.

On the other hand, when the slope value of the vehicle is greater than the preset reference value (in case of flat ground or uphill slope), and the gear stage of the vehicle is greater than the preset reference value (in the case of low-speed driving), during the timer time set from the timer setting step S140 Upshifting of the vehicle can be blocked.

That is, as described above, in the shift control step S160 according to the present embodiment, the amount of change per time (APS Grad) of the accelerator position sensor value, the vehicle slope (Slope), the accelerator position sensor value (APS value), the vehicle's It is preferable to forcibly shift the gear state of the vehicle to a lower gear only when data regarding the current gear level and the set timer satisfy a preset reference value.

In some cases, in the shift control step S160 according to the present embodiment, the gear state of the vehicle may be prevented in order to prevent a kick-down situation that may occur as the gear number of the vehicle is forcibly changed downward. It is preferable to reduce the engine torque by outputting a request for reducing engine torque while shifting to a lower gear.

Meanwhile, section P3 of FIG. 3 is a section in which re-acceleration and upshift of the vehicle V1 in which the driver is riding are allowed after the timer set in section P2, ie, deceleration section, has expired. It is a section in which it is determined that there are no pedestrians that are no longer identified by the vehicle V2 stopped in front, and the driver's field of vision is sufficient to identify the situation ahead.

As mentioned above, according to the prior art, accidents in which the driver's vehicle collides with the front vehicle frequently occur due to the driver's carelessness or sudden braking of the vehicle in front. Some drivers do not recognize the braking distance even when the yellow or red traffic light indicating a stop signal is lit, so they stop over the vehicle stop line or continue driving by ignoring the stop signal. If the driver could not identify the pedestrian due to the vehicle stopped in front of the city, the safety of the pedestrian could not be guaranteed.

On the other hand, according to the vehicle control system for preventing forward collision of the present invention, by having an image information collecting unit, a recognition processing unit, a judgment transmission unit, and an output unit that perform a specific role, the vehicle in front of the vehicle enters the crosswalk. When a driver fails to identify a pedestrian due to a stopped vehicle, it is possible to provide a vehicle control system capable of preventing a collision with a pedestrian by notifying the driver of a prior collision warning and limiting the vehicle speed.

In addition, according to the vehicle control method for preventing forward collision of the present invention, the road image information acquisition step, the road image information identification step, the pedestrian appearance time calculation step, the timer setting step, the warning output step, and the shift control step of performing a specific process By including, if the driver fails to identify pedestrians due to vehicles stopped in the side-front when entering the crosswalk area of vehicles traveling in front, collisions with pedestrians can be prevented by notifying the driver of a pre-collision warning and limiting the vehicle speed. A vehicle control method can be provided.

In the above detailed description of the present invention, only specific embodiments thereof have been described. However, it is to be understood that the present invention is not limited to the particular form recited in the detailed description, but rather, it is to be understood to cover all modifications and equivalents and substitutions falling within the spirit and scope of the invention as defined by the appended claims. should be

A: accelerator pedal
S: crosswalk sign
V1: The vehicle the driver is in
V2: Vehicle parked in front
10: pedestrian
20: crosswalk
100: vehicle control system for front collision prevention
110: image information collection unit
120: recognition processing unit
121: external image recognition unit
122: object selection unit
123: signal processing unit
130: judgment transmission unit
131: collision determination unit
132: shift request processing unit
140: output unit
S100: Vehicle control method for preventing forward collision
S110: Road image information acquisition step
S120: road image information identification step
S130: Pedestrian appearance time calculation step
S140: Timer setting step
S150: warning output stage
S160: shift control stage

Claims (8)

A vehicle control system 100 for preventing forward collision that detects a crosswalk situation in front of a vehicle and controls the vehicle based on the detected data,
After detecting the road condition in front of the vehicle, the image information collecting unit 110 for transmitting the detected data to the recognition processing unit 120;
Determination transmission unit 130 by selecting a crosswalk sign and the rear side of a stopped vehicle from the data detected through the image information collection unit 110, and detecting data regarding the crosswalk sign data and the identified vehicle location information Recognition processing unit 120 to transmit to;
Based on the data received through the recognition processing unit 120, the determination transmission unit 130 calculates the occurrence time of the appearance of a pedestrian moving through the crosswalk, and changes the gear state of the vehicle using the calculated time value. ; and
an output unit 140 for outputting an audible or visual warning to the vehicle driver by a timer set according to the time value calculated through the determination transmission unit 130;
The determination transmission unit 130,
From the data received from the recognition processing unit 120, the maximum time that a pedestrian can appear is calculated based on location data, size data, and time data of crosswalk signs, crosswalks, vehicles, obstacles and pedestrians, and A collision determination unit 131 for setting a timer based on;
The calculation of the maximum time during which the appearance of the pedestrian can occur is,
Detects the position and width of the vehicle based on the image information about the vehicle stopped in front, calculates the width of the crosswalk 20 obscured by the stopped vehicle, and the crosswalk 20 obscured by the vehicle By dividing the width by the average walking speed value of the pedestrian, the appearance time value of the pedestrian is calculated, the maximum value of the width of the crosswalk 20 covered by the stopped vehicle is detected, and the average walking speed value of the pedestrian is set as the minimum value. , a vehicle control system for front collision prevention, characterized in that calculating the maximum value of the time value during which the appearance of a pedestrian can occur.
According to claim 1,
The recognition processing unit 120,
an external image recognition unit 121 for receiving and storing image information detected through the image information collection unit 110;
A vehicle control system for front collision prevention, characterized in that it comprises a.
According to claim 1,
The recognition processing unit 120,
an object sorting unit 122 for recognizing a crosswalk sign, a crosswalk, a vehicle and an obstacle located on the road ahead in the image information detected through the image information collecting unit 110;
A vehicle control system for front collision prevention, characterized in that it comprises a.
4. The method of claim 3,
The object selection unit 122 executes on the basis of YOLO (You Only Look Once), a real-time object classification algorithm based on deep learning, to select a crosswalk sign, a crosswalk, a vehicle and an obstacle located on the road ahead. A vehicle control system to prevent forward collisions.
According to claim 1,
The recognition processing unit 120,
In the image information detected through the image information collection unit 110, messages about crosswalk signs, crosswalks, vehicles, obstacles and pedestrians are transmitted to the determination transmission unit 130 using CAN communication, which is a vehicle communication protocol. signal processing unit 123;
A vehicle control system for front collision prevention, characterized in that it comprises a.
delete According to claim 1,
The determination transmission unit 130,
a shift request processing unit 132 for changing the gear state of the vehicle based on the vehicle inclination value and the accelerator position sensor value (APS value) received through the vehicle sensor mounted on the vehicle;
A vehicle control system for front collision prevention, characterized in that it comprises a.
According to claim 1,
The determination transmission unit 130,
Forcibly vehicle only when data on the amount of change per hour of the accelerator position sensor value (APS Grad), the vehicle’s slope (Slope), the accelerator position sensor value (APS value), the current gear level of the vehicle, and the set timer satisfy the preset standard value. A vehicle control system for preventing forward collision, characterized in that it shifts the gear state of the

Images