KR20210028471A - System for accident prediction airbag using artificial intelligence and method for activate airbag - Google Patents

Abstract

The present invention relates to an accident prediction airbag system using an artificial intelligence and an airbag operation control method. According to the present invention, the airbag operation control method using the accident prediction airbag system using the artificial intelligence comprises: a step of using one or more sensors installed on a front surface unit of a driving vehicle, and measuring an approaching speed of a moving object approaching the driving vehicle; a step of determining whether the approaching speed exceeds a reference value or not, and if the approaching speed is determined to exceed the reference value, using an image photographed by a camera installed on the driving vehicle to classify whether the moving object is a person or a vehicle; and a step of, if the moving object is classified as a vehicle, immediately activating an airbag installed on a bumper, and if the moving object is classified as a person, immediately activating an airbag installed on a side surface. As such, according to the present invention, the accident prediction airbag system using the artificial intelligence is able to classify the types of objects approaching from a front side of the vehicle, to predict a collision by using the artificial intelligence, to activate the airbag thereby, and to reduce driver or pedestrian injury in a collision. In addition, the accident prediction airbag system uses artificial intelligence, so it is able to more efficiently identify an object, and to reduce errors.

Description

Accident prediction airbag system and airbag operation control method using artificial intelligence {SYSTEM FOR ACCIDENT PREDICTION AIRBAG USING ARTIFICIAL INTELLIGENCE AND METHOD FOR ACTIVATE AIRBAG}

The present invention relates to an accident prediction airbag system and airbag operation control method using artificial intelligence, and to an accident prediction airbag system and airbag operation control method using artificial intelligence that predicts an accident in advance and operates an airbag installed outside a vehicle.

Conventional airbags consist of a detection system and an airbag module, and the detection system consists of various components including a sensor, a battery, and a diagnosis device.The airbag module consists of an airbag and an operating gas inflator.When a collision is detected by the sensor, the operating gas device Is exploded and the airbag inflates momentarily by the explosive gas.

In general, automobiles are equipped with various safety devices to protect the driver and occupants in case of an instantaneous accident, and a representative example thereof is an airbag that directly protects the driver and passengers in the front seat of the vehicle.

On the other hand, in the method of inflating the airbag cushion, there are two types: an electric type in which a sensor electrically detects a collision and energizes and ignites the inflator, and a mechanical type in which a sensing device detects a collision and the sinking ignites an igniter.

When a vehicle crashes, the compressed gas is momentarily injected into the airbag by the impact force, and the airbag inflates at a high speed and wraps around the front of the driver and passengers in the front seat. It serves to protect from objects.

However, in the case of such an airbag system, there is a disadvantage that it can only protect people inside and cannot protect people outside.

Accordingly, there is a need for an external airbag system capable of predicting an accident in order to protect an external person or object.

The technology behind the present invention is disclosed in Korean Patent Application Publication No. 10-2011-0000815 (2011.01.05).

An object of the present invention is to provide an accident prediction airbag system and an airbag operation control method using artificial intelligence that predicts an accident in advance and operates an airbag installed outside a vehicle.

According to an embodiment of the present invention for achieving such a technical problem, in the airbag operation control method using an accident prediction airbag system using artificial intelligence, the vehicle is approached by using one or more sensors installed on the front of the vehicle. Measuring the approach speed of the moving object, determining whether the approach speed exceeds a reference value, and if it is determined that the approach speed exceeds the reference value, the moving object using an image photographed by a camera installed in the vehicle Classifying whether is a person or a vehicle, and if the moving object is classified as a vehicle, immediately activating an airbag installed on the bumper, and if the moving object is classified as a human, immediately activating an airbag installed on the side.

The sensor may include any one of infrared, lidar, radar, and ultrasonic waves.

In the step of classifying whether the moving object is a person or a vehicle, a learning model is generated by learning the type of a moving object corresponding to an image captured by the camera using artificial intelligence, and the captured image is applied to the learning model. Thus, it is possible to classify whether the moving object is a person or a vehicle.

In the step of activating the airbag, the airbag may be activated by calculating an impact amount according to the type of the moving object, an approach speed, an approach direction, and a speed of the traveling vehicle, and adjusting the strength of the external airbag.

According to another sealing example of the present invention, in an accident prediction airbag system using artificial intelligence, a sensor unit for measuring an approach speed of a moving object approaching the moving vehicle using one or more sensors installed on the front of the moving vehicle, A classification unit that determines whether the approach speed exceeds a reference value, and when it is determined that the approach speed exceeds the reference value, classifies whether the moving object is a person or a vehicle using an image photographed by a camera installed in the vehicle, Further, when the moving object is classified as a vehicle, the airbag installed on the bumper is immediately activated, and when the moving object is classified as a person, the control unit immediately activates the airbag installed on the side.

As described above, according to the present invention, since the airbag is activated by distinguishing the type of object approaching by the vehicle and predicting a collision situation using artificial intelligence, it is possible to reduce injuries to a driver or a pedestrian during a collision.

In addition, because artificial intelligence is used, objects can be identified more efficiently, and errors can be reduced compared to a system that activates an airbag according to an existing amount of impact.

1 is a block diagram illustrating an accident prediction airbag system according to an embodiment of the present invention.
2 is a flowchart illustrating a method of controlling an airbag operation according to an exemplary embodiment of the present invention.
3 is a diagram for explaining step S210 of FIG. 2.
4A and 4B are views for explaining step S220 of FIG. 2.
5 is a view for explaining step S230 of FIG. 2.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part "includes" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary.

Then, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those of ordinary skill in the art can easily implement the present invention.

1 is a block diagram illustrating an accident prediction airbag system according to an embodiment of the present invention.

As shown in FIG. 1, the accident prediction airbag system 100 includes a sensor unit 110, a classification unit 120, and a control unit 130.

First, the sensor unit 110 measures an approach speed of a moving object approaching the moving vehicle using one or more sensors installed on the front side of the moving vehicle.

In this case, the sensor unit 110 may include any one of an infrared sensor, a lidar sensor, a radar sensor, and an ultrasonic sensor, and measures the approach speed of the moving object.

Here, the infrared sensor detects a physical quantity or a stoichiometric quantity such as temperature, pressure, and intensity of radiation by using infrared rays, and can measure an approach speed of a moving object.

In addition, the LiDAR sensor detects an object using optical pulses and measures the characteristics of a signal reflected from the detected object, and measures the approach speed of a moving object using the speed and strength of the reflected signal. can do.

In addition, a radar sensor generates electromagnetic waves and measures the distance and direction to the object by receiving the electromagnetic waves reflected from the object.It can measure the approach speed of a moving object using the received distance and direction data. I can.

Next, the classification unit 120 determines whether the approach speed exceeds the reference value, and when it is determined that the approach speed exceeds the reference value, classifies whether the moving object is a person or a vehicle using a camera installed in the vehicle in progress.

At this time, the classification unit 120 generates a learning model by learning the type of a moving object corresponding to an image captured from the camera using artificial intelligence, and applies the captured image to the learning model to determine whether the moving object is a person or a vehicle. Classify.

Here, artificial intelligence is to learn human learning ability, reasoning ability, perceptual ability, and language comprehension ability to generate results for input values, and whether the moving object in the captured image is a person or a vehicle. Can be classified.

Next, when the moving object is classified as a vehicle, the controller 130 immediately activates the airbag installed on the bumper, and if the moving object is classified as a person, the control unit 130 immediately activates the airbag installed on the side.

In addition, the control unit 130 activates the airbag by calculating the amount of impact according to the type of the moving object, the approaching speed, the approaching direction, and the speed of the traveling vehicle, and adjusting the strength of the external airbag.

Hereinafter, an airbag operation control method using an accident prediction airbag system according to an embodiment of the present invention will be described with reference to FIGS. 2 to 5.

2 is a flowchart illustrating a method of controlling an airbag operation according to an exemplary embodiment of the present invention.

First, as shown in FIG. 2, the sensor unit 110 measures the approach speed of a moving object approaching the moving vehicle using one or more sensors installed on the front side of the moving vehicle (S210).

3 is a diagram for explaining step S210 of FIG. 2.

As shown in FIG. 3, the sensor unit 110 measures the approach speed of a moving object approaching a moving vehicle using a plurality of sensors 111a to 111d installed on the front side of the vehicle.

In this case, the sensor unit 110 measures the approach speed of the moving object using the speed of the moving object approaching the moving vehicle and the speed of the moving vehicle.

Here, the plurality of sensors 111a to 111d installed on the front of the vehicle may be any one of an infrared sensor, a lidar sensor, a radar sensor, and an ultrasonic sensor.

In FIG. 3, only a plurality of sensors installed on the front side are shown, but they are installed on the rear and side of the vehicle to measure the approach speed of the moving object.

Next, the classification unit 120 determines whether the approach speed of the moving object exceeds the reference value, and if it is determined that the approach speed exceeds the reference value, classifies whether the moving object is a person or a vehicle using a camera installed in the vehicle in progress. (S220).

That is, when the approaching speed of the moving object exceeds the reference value, as shown in FIG. 3, the camera installed on the front of the vehicle photographs the moving object, and the classification unit 120 determines whether it is a person or a vehicle from the captured moving object image. Judge

4A and 4B are views for explaining step S220 of FIG. 2.

As shown in Figs. 4A and 4B, the classification unit 120 classifies whether a moving object in a captured image is a person or a vehicle using artificial intelligence.

In addition, as shown in Figs. 4A and 4B, the number shown in each moving object represents the matching rate for each moving object using artificial intelligence.

That is, if the matching rate is close to 100%, it means that the moving object is accurately identified. Here, the reference value may be from 35 km/h to 40 km/h, may be changed according to the traffic volume and moving speed of the vehicle in progress, and may be set by the user.

In addition, when the moving object is classified as a vehicle, the control unit 130 immediately activates the airbag 140 installed on the bumper, and if the moving object is classified as a person, the control unit 130 immediately activates the airbag 150 installed on the side (S230).

5 is a view for explaining step S230 of FIG. 2.

As shown in FIG. 5, the airbag includes an airbag 140 installed on the bumper outside of the vehicle and an airbag 150 installed on the side.

In FIG. 5, only the air installed on the front and side portions is shown, but airbags may be installed on the rear and side surfaces of the vehicle.

In addition, the control unit 130 learns the type of moving object, the approach speed, the approach direction, and the amount of impact according to the speed of the moving vehicle and the activation strength of the external airbag according to the amount of impact using artificial intelligence.

Then, the controller 130 applies the type of the moving object, the approach speed, the approach direction, and the speed of the traveling vehicle to the artificial intelligence to generate an activation strength of the external airbag, thereby activating the external airbag.

In addition, as the amount of impact increases, the controller 130 increases and activates the strength of the airbag.

As described above, according to an embodiment of the present invention, since the type of object approaching from the front of the vehicle is distinguished and the airbag is activated by predicting a collision situation using artificial intelligence, injuries to a driver or a pedestrian during a collision can be reduced.

In addition, because artificial intelligence is used, objects can be identified more efficiently, and errors can be reduced compared to a system that activates an airbag according to an existing amount of impact.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is only exemplary, and those of ordinary skill in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the appended claims.

100: accident prediction airbag system, 110: sensor unit,
111a, ??, 111d: sensor, 120: classification
130: control unit 140: bumper airbag,
150: side airbag

Claims (8)

In the airbag operation control method using an accident prediction airbag system using artificial intelligence,
Measuring an approach speed of a moving object approaching the moving vehicle using one or more sensors installed on the front of the moving vehicle,
Determining whether the approach speed exceeds a reference value, and if it is determined that the approach speed exceeds the reference value, classifying whether the moving object is a person or a vehicle using an image photographed by a camera installed in the vehicle, and
And immediately activating an airbag installed on a bumper when the moving object is classified as a vehicle, and immediately activating an airbag installed on a side surface when the moving object is classified as a person. The method of claim 1,
The sensor,
Airbag operation control method including any one of infrared, lidar, radar, and ultrasonic waves. The method of claim 1,
Classifying whether the moving object is a person or a vehicle,
Airbag operation for generating a learning model by learning the type of moving object corresponding to the image captured by the camera using artificial intelligence, and classifying whether the moving object is a person or a vehicle by applying the captured image to the learning model Control method. The method of claim 1,
The step of activating the airbag,
An airbag operation control method for activating the airbag by adjusting the strength of the external airbag by calculating an impact amount according to the type of the moving object, the approach speed, the approach direction, and the speed of the traveling vehicle. In the accident prediction airbag system using artificial intelligence,
A sensor unit that measures an approach speed of a moving object approaching the moving vehicle using one or more sensors installed on the front of the moving vehicle,
A classification unit that determines whether the approach speed exceeds a reference value, and when it is determined that the approach speed exceeds the reference value, classifies whether the moving object is a person or a vehicle using an image photographed by a camera installed in the vehicle, And
When the moving object is classified as a vehicle, the airbag installed on the bumper is immediately activated, and when the moving object is classified as a person, the airbag system comprising a control unit that immediately activates the airbag installed on the side. The method of claim 5,
The sensor,
An airbag system that includes any one of infrared, lidar, radar, and ultrasonic waves. The method of claim 5,
The classification unit,
Airbag system that generates a learning model by learning the type of moving object corresponding to the image captured by the camera using artificial intelligence, and classifies whether the moving object is a person or a vehicle by applying the captured image to the learning model . The method of claim 5,
The control unit,
An airbag system for activating the airbag by adjusting the strength of the external airbag by calculating an impact amount according to the type of the moving object, the approach speed, the approach direction, and the speed of the traveling vehicle.

Images