KR20200004778A - Smart helmet and system - Google Patents

Abstract

The present invention provides a smart safety helmet and a safety helmet management system, wherein the smart safety helmet comprises: a body that can be worn on the head of a wearer; a brainwave detection sensor that detects brainwave of the wearer by being mounted to contact at least one of the frontal, occipital and parietal lobes of the wearer′s head; a wireless communication unit that transmits and receives wirelessly to the outside; and a power supply unit The smart safety helmet of the present invention can monitor the wearer′s condition in real time.

Description

EEG-based smart hard hat and management system {Smart helmet and system}

The present invention relates to an EEG-based smart helmet and management system, and more particularly, to a smart helmet and smart helmet management system for detecting and analyzing the brain waves of the wearer by the sensor mounted on the helmet.

In general, a hard hat is a protective tool used in an industrial site, such as a construction site, to protect the head of the worker from external impact.

Workers at industrial sites are often far from the outside and only a small number of people work together, and even in danger, they may be in a difficult situation because others are not aware of it. In addition, even though people who enter and leave the industrial site is legally required to wear a hard hat, industrial accidents due to the wearing of a hard hat is increasing.

Recently, hard hats with various functions have been introduced to induce the wearing of hard hats by workers. For example, safety helmets with forced ventilation, weighing, LED lighting, and stress relief have been developed.

Brain waves, on the other hand, appear differently according to physical and mental states, which is a very important indicator of brain activity. EEG analysis is used in various fields because it is possible to determine whether drowsiness and danger according to EEG waveforms.

Smart helmet according to the present invention is to solve the problems of the prior art, to provide a smart hard hat and smart hard hat management system that protects the wearer from external shocks, and analyzes the brain waves of the wearer in real time when an emergency situation occurs The purpose.

It is not limited to the said objective, A further description is given below.

To achieve the above object,

The present invention can be worn on the wearer's head; An EEG sensor which is mounted to contact at least one of the frontal lobe, the occipital lobe, and the parietal lobe of the wearer's head and detects the EEG of the wearer; Wireless communication unit for transmitting and receiving wirelessly with the outside; It provides a smart helmet comprising a; and a power supply.

In addition, the EEG obtained from the EEG sensor is filtered in the effective range using a bandpass filter, the EMG noise generated by the severe movement of the wearer's neck and face is generated by a predetermined value or more diluted in EEG If the EEG signal information of the relevant time is treated as noise and deleted, and if the electrode of the EEG sensor is temporarily detached or poor contact due to the severe movement of the wearer, the EEG signal information of the relevant time is interpreted as temporary electrode detachment. It provides a smart hard hat further comprising a noise removing unit for treating the noise as noise.

In addition, the wearer further comprises an operation switch for turning on and off the operation, when the operation switch is in the on state, when the electrode of the EEG sensor detached for more than a predetermined time, the wireless communication unit is recognized as an emergency situation Sending an emergency notification to the outside through the smart switch, when the operation switch is in the off state for more than a predetermined time when the electrode of the EEG sensor is detached smart emergency further comprises an emergency notification unit to handle as taken off the smart hard hat Provide a hard hat.

In addition, the wireless communication unit recognizes the wearer's RFID tag to determine the wearer's identity, and provides a smart helmet to transmit the identity of the wearer to the management server.

In addition, a smart safety helmet further comprising an EEG signal processing unit for obtaining at least one EEG value of the delta wave, theta wave, alpha wave, beta wave, gamma wave in real time based on the spectrum analysis of the signal from which the noise is removed to provide.

In addition, if the following A value calculated by the EEG value obtained by the EEG signal processing unit is higher than a predetermined value compared to the following B value and the state continues for a predetermined time, the smart helmet for transmitting the EEG value to the management server through the wireless communication unit To provide.

A = alpha wave / (alpha + beta wave)

B = beta wave / (alpha + beta wave)

The invention also, the above-mentioned smart helmet; And a management server that receives and manages brain wave information from the smart hard hat, wherein the management server is configured to receive the brain waves transmitted using an Improved Complete EEMD (Ensemble Empirical Mode Decomposition) with Adaptive Noise (ICEEMDAN). It provides a smart helmet management system including an electroencephalogram analysis unit that decomposes a plurality of intrinsic mode functions (IMFs), obtains IMFs above a threshold value from harmonic characteristics and power ratios by obtaining spectral values for each IMF.

In addition, the EEG analyzer decomposes the EEG received using a DWT (discrete wavelet transform) into a plurality of subbands, obtains at least one or more of an average, variance, skewness, and kurtosis of each band to obtain each frame. It provides a smart helmet management system that selects the maximum change rate subband that has the largest change rate among the rate of change of the value obtained in the subband.

The management server may further include an artificial intelligence unit, wherein the artificial intelligence unit defines the maximum change rate subband and IMFs above a threshold value as effective feature values, and is robust to noise among AR (auto-regressive) models. Variable) model, and a smart helmet management system for determining whether the wearer's state is arousal, drowsiness, gas inhalation, fatigue, fainting state by using the AR coefficient as an input vector of the AI model.

Smart helmet of the present invention is equipped with a brain wave sensor and a wireless communication module to detect and analyze the brain wave in real time it can monitor the status of the wearer in real time.

In addition, the smart hard hat of the present invention can be analyzed in the server database as well as the self-analysis through the application embedded in the hard hat when an error occurs, even if wireless communication is not possible to prevent a safety accident.

In addition, the smart helmet of the present invention is provided with an RFID sensor, by having a unique RFID tag for each wearer can provide a user-specific brainwave discrimination function and wearer recognition function.

Not limited to the following effects, additional effects are described below.

1 is a perspective view of a smart hard hat according to an embodiment of the present invention.
2 is a cross-sectional view according to another embodiment of the present invention.

Before describing the present invention in detail, it is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the invention as defined only by the appended claims.

All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise indicated.

Unless otherwise stated throughout this specification and claims, the term “comprise”, “consisting”, “comprising” means including an article or step referred to, or a group of things or a group of steps; It is not intended to exclude other objects or steps, a group of things, or a group of steps.

On the other hand, various embodiments of the present invention can be combined with any other embodiment unless clearly indicated to the contrary. In particular, any feature indicated as being preferred or advantageous may be combined with any other feature indicated as being preferred or advantageous.

Smart helmet according to an embodiment of the present invention is a wearable body on the wearer's head; An EEG sensor which is mounted to contact at least one of the frontal lobe, the occipital lobe, and the parietal lobe of the wearer's head and detects the EEG of the wearer; Wireless communication unit for transmitting and receiving wirelessly with the outside; And a power supply unit.

Hereinafter, a smart helmet according to an embodiment of the present invention will be described with reference to the accompanying drawings. 1 is a perspective view of a smart hard hat according to an embodiment of the present invention.

Accordingly, the smart hard hat includes a body 100, an EEG sensor 200, an alarm generating device 300, a module 400, and a power supply unit 500.

Body 100 is composed of the outer shell 110 and the endothelial 120, mounted to chuck at least one or more of the frontal lobe, occipital lobe, parietal lobe region of the endothelial EEG sensor 200 for detecting the brain waves of the wearer, EEG Alarm generating device 300 connected to the detection sensor 200 and the application to generate an alarm, a module 400 for detecting near-field wireless communication and the environment around the wearer mounted on the back of the endothelial region, EEG sensor and alarm And a power supply unit 500 for supplying power to the generator and the module. The module may include a wireless communication unit.

EEG sensor 200 may be mounted to the frontal lobe, occipital lobe, parietal lobe contact between the endothelial and the wearer, and is additionally mounted to the central portion between the contact portion to detect the EEG of the wearer in real time.

The alarm generating device 300 is controlled by an application and / or a server mounted on the smart helmet itself, and if it is determined that the EEG information measured by the EEG sensor is different from the EEG of the wearer, voice, vibration, or electricity to inform the wearer. Generate a signal.

The alarm generating device 300 may be provided as a warning sound generating device, or may be provided as a vibration motor generating vibration. Or it may be provided as a low current magnetic pole plate for generating an electrical signal.

Module 400 is provided adjacent to the helmet endothelial contact and the back of the head of the wearer, there is a wireless communication unit. It may be short-range wireless communication, for example, a wireless communication sensor such as Bluetooth, Wi-Fi, RFID, etc. and a sensor for detecting the wearer's surroundings, for example, a harmful gas detection, obstacle detection, such as environmental detection sensor is provided.

In one embodiment, the wireless communication unit is provided with an RFID sensor, recognizes the wearer's RFID tag to identify the wearer's identity, and transmits the wearer's identity to the management server, and transmits the brainwave information of the wearer to the server database and It is possible to store and customize the wearer's EEG discrimination, and to recognize the wearer in an emergency situation, it is possible to respond quickly.

The power supply unit 500 supplies the driving power of the smart hard hat. In this embodiment, the power supply unit is provided with a rechargeable secondary battery, and supplies power to the EEG sensor, the alarm generating device and the module.

The application is embedded in a smart helmet to receive EEG information collected from the EEG sensor. When an EEG abnormality occurs, the application may transmit the EEG information to the server, or determine by itself in the application to instruct the alarm generating device to generate an alarm.

If the EEG information transmitted from the application to the server is analyzed through the algorithm of the server and determined as a new pattern of EEG, it may be recorded in the database or update the information, algorithm or analysis tool (TOOL) of the application.

In addition, if the EEG information is generated due to the wearer's movement and the EEG cannot be determined, feedback is provided according to whether the normal EEG returns. When returning to the normal EEG, the EEG panda or information of the impossible section can be deleted, and when not returning to the normal EEG for a predetermined time, the corresponding EEG information can be transmitted to the server to respond directly from the server.

As a specific embodiment, the smart hard hat may further include a noise removing unit.

* The noise removing unit filters the EEG acquired from the EEG sensor into a valid range using a bandpass filter, and the EMG noise generated by severe movement of the wearer's neck and face is generated by a predetermined value or more. When diluted to, the entire EEG signal information of the time can be treated as noise and deleted. If the EMG noise occurs more than a predetermined value by diluting the wearer in various ways in advance, it can be stored in the database, and whether the EMG noise is in the case of diluting the EEG in real time can be stored in the database. Can be judged by comparison.

In addition, when the electrode of the EEG sensor is temporarily detached or has a poor contact due to a heavy movement of the wearer, the electrode may be interpreted as temporary electrode detachment and treated as noise to delete the entire EEG signal information of the corresponding time.

On the other hand, the wearer may further include an operation switch to turn on or off the operation, may further include an emergency notification unit.

The emergency situation notification unit, when the operation switch is turned on, when the electrode of the EEG sensor is detached for more than a predetermined time, it is recognized as an emergency situation to notify the emergency situation to the outside of the server or workers around the wireless communication through the wireless communication unit Can transmit On the other hand, when the electrode of the EEG sensor is detached for more than a predetermined time while the operation switch is off, the wearer can be treated as if the smart hat has been taken off arbitrarily.

The smart hard hat may further include an EEG signal processor. The EEG signal processor may obtain at least one EEG value of the delta wave, theta wave, alpha wave, beta wave, and gamma wave in real time based on the spectrum analysis of the signal from which the noise is removed through the noise removing unit. EEG value can be determined or transmitted to the management server.

In the management server transmission, all the EEG may be transmitted to the management server unconditionally, but in order to reduce the overload of the wireless communication unit and the management server, the following A value calculated from the EEG value obtained by the EEG signal processor is compared with the following B value. The EEG value may be transmitted to the management server through the wireless communication unit only when the predetermined value is higher than the predetermined value. The predetermined value may be selected within a range of 1.1 to 3 times, and the predetermined time may be set within a range of 2 to 60 seconds.

A = alpha wave / (alpha + beta wave)

B = beta wave / (alpha + beta wave)

On the other hand, if the following A value is higher than a predetermined value than the following B value and the state continues for a predetermined time, it is recognized as an emergency situation and the emergency situation notification unit may transmit an emergency situation notification to the outside through the wireless communication unit, an alarm is generated. Alarms can be triggered through the device to induce awakening of the wearer.

As another embodiment of the present invention, it provides a smart helmet management system including a smart server and a management server for receiving and managing the brain wave information from the smart helmet.

The management server may accurately analyze the brain waves, including the brain wave analysis unit. The harmonic analyzer decomposes the received EEG into a plurality of intrinsic mode functions (IMFs) using an Improved Complete EEMD (Ensemble Empirical Mode Decomposition) with Adaptive Noise (ICEEMDAN), and obtains spectral values for each IMF. IMFs above the threshold can be obtained from.

In addition, the management server decomposes the EEG received using a DWT (discrete wavelet transform) into a plurality of subbands, obtains at least one or more of the average, variance, skewness, and kurtosis of each band, and then each frame. You can select the maximum rate of change subband that has the largest rate of change among the values of the subbands.

The management server may further include an artificial intelligence unit, wherein the artificial intelligence unit defines IMFs above the maximum change rate subband and a threshold value as effective feature values, and is robust to noise among AR (auto-regressive) models. The AR coefficient is obtained from the model, and the AR coefficient is used as the input vector of the artificial intelligence model to determine whether the wearer is in an awake state, drowsiness, gas inhalation, fatigue, or fainting state (see FIG. 3). Meanwhile, the brain wave analysis unit and / or the artificial intelligence unit may be placed in the management server, or may be placed in the smart helmet itself.

The management server can perform emergency measures such as telephone connection to the wearer, emergency broadcast, alarm to other workers, prompt on-site safety manager when the wearer has an emergency such as drowsiness, inhalation of gas, fainting.

In order to improve the accuracy of the AI department, EEG information of several workers is collected and stored in a database, EEG analysis information for each situation is stored in a database, and data statistics are analyzed according to worker's time, age, gender, and region. Can be stored in the database. In addition, it is possible to systematically schedule workers by analyzing workers' stresses by time or task from such information.

The following method is mentioned as an example of the method of calculating | requiring AR coefficient using EIV.

차 AR 모델로 표현되는 신호를 나타낸 것이다.In the system identification problem, the error and noise at the input and the output are called EIV models. Figure 2 shows the EIV method for estimating the AR coefficient.

The signal represented by the difference AR model is shown.

(Equation 1)

는 AR 모델의 계수를 나타내고,

는 평균이

이고 분산이

인 가산성 백색 잡음(additive white Gaussian nosie; AWGN)이다. 신호

는 평균이

이고 분산이

인 AWGN

가 더해져서 관측이 된다.here

Represents the coefficients of the AR model,

Has an average

And dispersion

Phosphoric Additive White Gaussian nosie (AWGN). signal

Has an average

And dispersion

AWGN

Is added to the observation.

(Equation 2)

In the present invention, using the EIV having the following characteristics to measure the dispersion of the AR coefficient, input and measurement noise.

Iii) using the properties of the solution set of the Frisch technique

Ii) using the shift nature of a time-invarint dynamic system

I) Simultaneously estimating AR coefficients, input and measured noise variances that satisfy the positive-definite matrix of the covariance matrix

Frisch technique and shift properties can be used to find the point where noise variance is minimized in the noise plane. That is, since the error of the AR coefficients predicted in the noise environment can be minimized, a feature vector more robust to noise can be extracted.

개의 유한 데이터에서 EIV에 기반한 AR 계수 예측 알고리즘에 대해 간략히 기술한 것이다.next

It is a brief description of the AR coefficient prediction algorithm based on EIV in two finite data.

에 대한 공분산 행렬

은 다음과 같이 정의된다. Step 1:

Covariance matrix for

Is defined as

(Equation 3)

는

개의 관측된 신호

로 구성되는

Hankel 행렬이고 다음과 같다.here

Is

Observed signals

Composed of

It is a Hankel matrix and is

(Equation 4)

은 유한하므로, AR 계수를 추정함에 있어 공분산 행렬에 대한 추정치를 계산하여 이용해야 한다. AR 차수

,

에 대해 다음과 같이 공분산의 추정을 계산한다.The number of actual observations

Since is finite, an estimate of the covariance matrix should be used to estimate the AR coefficient. AR order

,

Calculate the covariance estimate for

(Eq. 5)

(Equation 6)

에서 식 (7)을 만족하는 점

에 대해 연속된 convex 곡선의 집합을

라 하자.Step 2: noise plane

Satisfying equation (7)

A set of consecutive convex curves for

Let's do it.

(Eq. 7)

를 지나는 직선이

와 만나는 점

는 다음과 같이 구할 수 있다.Origin and Point Using Equation (8)

Straight through

Meet with

Can be obtained as

(Eq. 8)

이다. 이와 같은 방식으로 원점과 점

를 지나는 직선이

와 만나는 점

또한 구할 수 있다. here

to be. Origin and point in this way

Straight through

Meet with

Also available.

,

,

를 계산한다.Step 3: Using the relationship

,

,

Calculate

(Eq. 9)

(Eq. 10)

(Eq. 11)

는 점

를 이용하여 구한 AR 계수

를 의미한다.here

Point

AR coefficient obtained using

Means.

일 때, 식 (12)에 주어진 SR 비용함수(shifted relation cost function)를 계산한다.Step 4:

In this case, the SR shifted relation cost function given in Eq. (12) is calculated.

(Eq. 12)

를 최소화 하는 분산

를 탐색한다.Step 5: SR Cost Function

Dispersion to minimize

Navigate.

와 일치하는 AR 계수 벡터

를 구한다.Step 6:

AR coefficient vector matching

Obtain

The features, structures, and effects illustrated in the above-described embodiments may be combined or modified with respect to other embodiments by those skilled in the art to which the embodiments belong. Therefore, contents related to such combinations and modifications should be construed as being included in the scope of the present invention.

100: body
200: EEG sensor
300: alarm generating device
400: Module
500: power supply
600: ground electrode

Claims (7)

Body wearable on the wearer's head;
An EEG sensor which is mounted to contact at least one or more of the frontal lobe, occipital lobe, and parietal lobe of the wearer's head and detects the brain wave of the wearer;
Wireless communication unit for transmitting and receiving wirelessly with the outside; And
Smart helmet including a power supply.
The method of claim 1,
The EEG obtained from the EEG sensor is filtered to a valid range using a bandpass filter,
If the EMG noise generated by severe movement of the wearer's neck and face is over a predetermined value and diluted with EEG, the EEG signal information of the relevant time is treated as noise and deleted.
When the electrode of the EEG sensor is temporarily detached or poor contact due to the heavy movement of the wearer, the smart safety hat further comprises a noise removing unit for treating the entire EEG signal information of the corresponding time as noise by removing it as a temporary electrode detachment.
The method of claim 1,
Further includes an actuation switch that the wearer can turn on and off the actuation,
When the operation switch is turned on, when the electrode of the EEG sensor is detached for more than a predetermined time, it is recognized as an emergency situation and transmits an emergency notification to the outside through the wireless communication unit,
When the electrode of the EEG sensor is detached for more than a predetermined time in the operation switch is off state, the smart helmet further comprises an emergency notification unit that the wearer arbitrarily handles as taking off the smart helmet.
The method of claim 1,
The wireless communication unit recognizes the wearer's RFID tag to identify the wearer's identity, smart helmet for transmitting the identity of the wearer to the management server.
The method of claim 2,
A smart hardhat further comprising an EEG signal processor for obtaining in real time at least one or more EEG values of the delta wave, theta wave, alpha wave, beta wave, and gamma wave based on spectrum analysis of the signal from which the noise is removed.
The method of claim 5,
Smart A hard hat for transmitting the EEG value to the management server through the wireless communication unit when the following A value calculated by the EEG value obtained by the EEG signal processing unit is higher than the B value below a predetermined value and the predetermined time is continued.

A = alpha wave / (alpha + beta wave)
B = beta wave / (alpha + beta wave)
The method of claim 5,
When the following A value calculated by the EEG value obtained by the EEG signal processing unit becomes higher than the B value below the predetermined value and the state continues for a predetermined time, the emergency situation is recognized and transmitted to the outside through the wireless communication unit. Smart hard hat further comprising an emergency notification unit.

A = alpha wave / (alpha + beta wave)
B = beta wave / (alpha + beta wave)

Images