KR20220063079A - Device and method for determining human error using brain waves - Google Patents

Abstract

The present invention relates to a device and a method for determining a human error using brain waves. The device for determining the human error using the brain waves according to the present invention comprises: a brain wave measuring module for measuring a user's brain wave signal; a classification module for classifying the state of the measured brain waves; a calculation module for calculating an integrated index using the classified state of the brain waves; and a determination module for determining the user's human error if the integrated index is equal to or greater than a preset threshold. The present invention is performed by Electronics and Telecommunications Research Institute as a part of subsidy projects of Ulsan metropolitan city. Specifically, the project identification number is 20AS1100, the title of the research business is the smart HSE system development and digital cockpit system development for advancement of ICT convergence based core industries, and the title of the research project is the development for a safety management technology using brain wave signals. The present invention can prevent accident occurrences caused by the human error.

Description

Device and method for determining human error using brain waves

The present invention relates to an apparatus and method for determining human error using EEG, and more particularly, to measure a user's EEG signal to classify the EEG state, calculate an integrated index using the classified EEG state, and to calculate an integrated index The present invention relates to an apparatus and method for determining human error using EEG that determines that it is a human error and provides a notification when it is greater than or equal to a preset threshold.

Human error refers to a state in which the normality and appropriateness of information transmission, decision-making due to delivery, and operation according to commands through human sensory organs are insufficient. The causes of human error include cases caused by humans (individuals) themselves and those that occur under the influence of external factors. Statistics show that 75 to 80% of the causes of industrial accidents are due to human error. Such human error is also a major cause of various kinds of huge industrial accidents such as explosion accidents at various work sites, and an organizational environment vulnerable to errors and negligence in management can cause more human errors and related accidents.

For this reason, studies have been conducted to analyze the causes and types of human errors and to prevent them in advance, and efforts to prevent human errors and the establishment of a safe environment are required at the individual and organizational level.

The present invention is to solve the problems of the prior art described above, the present invention accurately calculates the user's stress and fatigue index using a plurality of EEG signals, and uses EEG to determine the user's human error according to the index. An object of the present invention is to provide an error determination apparatus and method.

Another object of the present invention is to provide an apparatus and method for determining human error for classifying into a stress or fatigue state by generating an artificial intelligence-based classification algorithm for stress and fatigue through features extracted from EEG signals.

Objects of the present invention are not limited to those mentioned above, and other objects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description.

The present invention for achieving the above object is an EEG measurement module for measuring the user's EEG signal, a classification module for classifying the measured EEG state, and a calculation module for calculating an integrated index using the classified EEG condition. and a determination module for determining a user's human error when the integrated index is greater than or equal to a preset threshold value.

The EEG measurement module includes a band portion wearable on the head, a plurality of measurement electrodes mounted on the band portion, and the measurement electrode portion first to fourth electrodes arranged in a line to measure a plurality of EEG signals; It is configured to include a reference electrode to measure the reference EEG signal.

The classification module includes an extractor for extracting EEG features from the EEG signal, a generator for generating a classification algorithm using the extracted EEG features, and a stress or fatigue state from the extracted EEG features using the classification algorithm. It includes a classification unit to classify.

The calculation module includes a first calculation unit that calculates a stress or fatigue index by using the classified stress or fatigue state, and a second calculation unit that calculates an integrated index by applying a weight to the stress or fatigue index.

The first calculation unit is characterized in that it calculates the stress or fatigue index according to the following [Equation 1] and [Equation 2].

[Equation 1]

는 시간 t(초)에서 스트레스 또는 피로 상태로 분류된 개수,

는 한 시간 동안 지속적으로 스트레스 또는 피로 상태로 분류된 개수를 나타낸다.here,

is the number classified as a stress or fatigue state at time t (seconds),

indicates the number of categories classified as stressed or fatigued continuously for one hour.

[Equation 2]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표를 나타낸다.here,

denotes an indicator for the stress or fatigue state at time x (seconds).

The second calculation unit is characterized in that it calculates the integrated index according to the following [Equation 3].

[Equation 3]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표, w는 각 사용자별 피로 대비 스트레스 가중치(단,

),

는 시간 x(초)에서의 스트레스 및 피로상태를 고려한 통합 지표를 나타낸다.here,

is an index for stress or fatigue state at time x (seconds), and w is the stress weight for each user versus fatigue (however,

),

represents the integrated index considering the stress and fatigue states at time x (seconds).

and a monitoring module for monitoring the EEG signal and the integrated indicator.

It further includes a communication module for transmitting and receiving the EEG signal to the control center, a battery module for supplying power required for driving, and a notification module for providing a notification about the human error.

According to another feature of the present invention, the measuring module is a measuring step of measuring the user's EEG, the classification module is a classification step of classifying the measured EEG state, and the calculation module is an integrated index using the classified EEG state. The calculation step of calculating, the determination module determines the user's human error when the integrated indicator is greater than or equal to a preset threshold, and the notification module provides notification of information about the determined human error to the user or the control center It provides a human error determination method using brain waves comprising the steps.

Before measuring the user's EEG, it further includes a setting step of setting a weight according to the user.

In the classification step, the extraction unit extracts EEG features from the signal from the measured EEG signal, the generation unit generates a classification algorithm using the extracted EEG features, and the classification unit extracts the EEG features using the classification algorithm. and classifying the stress or fatigue state from the EEG characteristics.

In the calculation step, a first calculation step in which a first calculation unit calculates a stress or fatigue index by using the classified stress or fatigue state, a second calculation unit calculates an integrated index by applying a weight to the stress or fatigue index and a second calculation step.

The first calculation step is characterized in that the stress or fatigue index is calculated according to the following [Equation 1] and [Equation 2].

[Equation 1]

는 시간 t(초)에서 스트레스 또는 피로 상태로 분류된 개수,

는 한 시간 동안 지속적으로 스트레스 또는 피로 상태로 분류된 개수를 나타낸다.here,

is the number classified as a stress or fatigue state at time t (seconds),

indicates the number of categories classified as stressed or fatigued continuously for one hour.

[Equation 2]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표를 나타낸다.here,

denotes an indicator for the stress or fatigue state at time x (seconds).

The second calculation step is characterized in that the integrated index is calculated according to the following [Equation 3].

[Equation 3]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표, w는 각 사용자별 피로 대비 스트레스 가중치(단,

),

는 시간 x(초)에서의 스트레스 및 피로상태를 고려한 통합 지표를 나타낸다.here,

is an index for stress or fatigue state at time x (seconds), and w is the stress weight for each user versus fatigue (however,

),

represents the integrated index considering the stress and fatigue states at time x (seconds).

According to the apparatus and method for determining human error using EEG of the present invention as described above, the user's stress and fatigue index is accurately calculated using the user's EEG signal, the user's human error is determined according to the index, and a notification is provided. , it can prevent accidents caused by human error.

The present invention can classify the EEG signal of each user more accurately by classifying the stress or fatigue state using the EEG classification algorithm.

The present invention can monitor the user's condition in real time in the control center by transmitting and receiving EEG signals to the control center using wireless communication.

The effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned may be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be.

1 is a configuration diagram for explaining an apparatus for determining human error using brain waves according to the present invention;
Figure 2 is a configuration diagram for explaining the EEG measurement module of Figure 1;
3 is a configuration diagram for explaining the classification module of FIG. 1
4 is a configuration diagram for explaining the calculation module of FIG. 1
5 is an exemplary view for explaining the determination module of FIG. 1
6 is an exemplary view for explaining the monitoring module of FIG. 1
7 is an exemplary diagram of a human error determination device using the brain waves of FIG. 1
8 is a view for explaining a wearing position of the human error determination device using the brain wave of FIG. 1
9 is a flowchart for explaining a human error determination method using brain waves according to the present invention

Objects and effects of the present invention, and technical configurations for achieving them will become clear with reference to the embodiments described below in detail in conjunction with the accompanying drawings. In the description of the present invention, if it is determined that a detailed description of a well-known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

And, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to intentions or customs of users and operators.

However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. Only the present embodiments are provided so that the disclosure of the present invention is complete, and to completely inform those of ordinary skill in the art to which the present invention belongs, the scope of the invention, the present invention is defined by the scope of the claims will only be Therefore, the definition should be made based on the content throughout this specification.

Although the present invention is described with reference to limited embodiments and drawings, the present invention is not limited thereto, and it is described below with the technical spirit of the present invention by those of ordinary skill in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of equivalents of the claims to be made.

Hereinafter, the present invention will be described in more detail based on the embodiments shown in the drawings. However, the present application is not limited to these examples and drawings.

Hereinafter, the configuration of the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram for explaining an apparatus 100 for determining human error using brain waves according to the present invention.

As shown in FIG. 1, the apparatus 100 for determining human error using brain waves according to the present invention classifies the state of the brain waves using the brain wave measurement module 110 for measuring the user's brain wave signals, and the measured brain wave signals. Classification module 120, calculation module 130 for calculating an integrated index using the classified EEG state, a determination module 140 for determining a user's human error when the integrated index is greater than or equal to a preset threshold value, EEG signal A communication module 150 for transmitting and receiving to the control center, a monitoring module 160 for monitoring EEG signals and the integrated indicator, a battery module 170 for supplying power required for driving, a notification module for providing a notification about human error (180).

The EEG measurement module 110 measures the user's EEG signal. According to an embodiment, the EEG measurement module 110 may be mounted on the head for EEG measurement. For example, the EEG measurement module 110 may be mounted toward the forehead in order to measure the EEG of the frontal lobe. In addition, the EEG measurement module 110 may be mounted to be in contact with at least a portion of the head to measure EEG of the frontal lobe, parietal lobe, occipital lobe, or temporal lobe. .

The EEG measurement module 110 may use an electroencephalogram (EEG) sensor.

As shown in FIGS. 2 and 7 , the brain wave measurement module 110 includes a band part 111 that can be worn on the head, and a plurality of measurement electrode parts 113 mounted on the band part.

The band part 111 may extend in one direction so as to be adjacent to at least a part of the head in a form that can be worn on the head. For example, the band portion 111 may be a curved shape extending in the circumferential direction of the head or a flat shape extending in one direction.

The measuring electrode unit 113 includes first to fourth electrodes 113a to 113d arranged in a line to measure a plurality of EEG signals, and a reference electrode 113e to measure a reference EEG signal.

The first to fourth electrodes 113a to 113d according to an embodiment of the present invention may simultaneously acquire EEG signals through four channels. Of course, it is also possible to acquire an EEG signal using fewer or more channels than four.

The reference electrode 113e measures EEG as a reference of EEG signals measured by the first to fourth electrodes 113a to 113d in order to measure more reliable EEG signals. This is to obtain reliable data.

The classification module 120 classifies the measured EEG signals into stress and fatigue states.

As shown in FIG. 3 , the classification module 120 uses an extraction unit 121 for extracting features from an EEG signal, a generation unit 123 for generating a classification algorithm using the extracted EEG features, and a classification algorithm. It is configured to include a classification unit 125 for classifying the stress or fatigue state from the extracted EEG features.

The extraction unit 121 may amplify the measured EEG signal to extract EEG features. Here, the characteristic of the EEG may be an EEG average, standard deviation, first difference, second difference, power spectrum density (PSD) for each EEG band, entropy, and the like.

The generator 123 may generate a stress and fatigue learning model by using the EEG features extracted by the extractor 121 , and may generate a classification algorithm by learning the stress and fatigue learning model. In this case, the generator 123 may apply the transfer learning algorithm to the stress and fatigue learning model according to the data of the stress and fatigue state in order to learn the classification algorithm.

The classification unit 125 classifies the EEG signal into a stress state or a fatigue state per unit time.

The calculation module 130 calculates an index of stress or fatigue and an integrated index by using the classified stress or fatigue state.

As shown in FIG. 4 , the calculation module 130 includes a first calculation unit 131 that calculates a stress or fatigue index by using the stress or fatigue state classified by the classification unit 125 , the stress or fatigue index. and a second calculation unit 133 for calculating an integrated index by applying a weight.

The first calculation unit 131 calculates the stress or fatigue index according to the following [Equation 1] and [Equation 2].

[Equation 1]

는 시간 t(초)에서 스트레스 또는 피로 상태로 분류된 개수,

는 한 시간 동안 지속적으로 스트레스 또는 피로 상태로 분류된 개수를 나타낸다.here,

is the number classified as a stress or fatigue state at time t (seconds),

indicates the number of categories classified as stressed or fatigued continuously for one hour.

[Equation 2]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표를 나타낸다.here,

denotes an indicator for the stress or fatigue state at time x (seconds).

The second calculation unit 133 calculates the integrated index according to Equation 3 below.

[Equation 3]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표, w는 각 사용자별 피로 대비 스트레스 가중치(단,

),

시간 x(초)에서의 스트레스 및 피로상태를 고려한 통합 지표를 나타낸다.here,

is an index for stress or fatigue state at time x (seconds), and w is the stress weight for each user versus fatigue (however,

),

It represents the integrated index considering the stress and fatigue state at time x (seconds).

The determination module 140 determines whether the integrated indicator is greater than or less than a preset threshold value, and when it is greater than or equal to a preset threshold value, determines the user's human error.

5 is an exemplary diagram for explaining the determination module 140 . The determination module 140 according to an embodiment of the present invention is A class when the integrated index is greater than 0 to 0.60 or less, B class when 0.60 to 1.53 or less, C class when 1.53 to 3.91 or less, 3.91 to 10 or less Class D can be considered. The A to D classes mean 'normal', 'warning', 'risk', and 'not allowed' for the user's personal situation.

At this time, the determination module 140 may determine the class B (warning), class C (danger), and class D (disallowed) as human error.

The communication module 150 serves to transmit and receive EEG signals to and from the control center. Here, the user's brainwave signals can be stored in the control center to build a user state DB, which is to manage the states of multiple users in an integrated way.

The communication module 150 may transmit data using a Bluetooth method or a Wi-Fi method according to the number of channels. For example, when transmitting the EEG signal, the communication module 150 transmits EEG data of 2 channels or less using the Bluetooth method, and multi-channel EEG data of 4 or more channels uses a high-speed Wi-Fi method. do. Of course, this is only an embodiment, and various wireless communication methods may be variously applied according to environments and conditions.

The monitoring module 160 monitors the EEG signal measured by the EEG measurement module 110 and the integrated index calculated by the calculation module 130 . In addition, the monitoring module 160 may monitor the user's status according to the DB built in the control center through the communication module 150 .

6 is an exemplary diagram for explaining the monitoring module 160 . Referring to FIG. 6 , the monitoring module 160 may monitor the EEG signals of 4 channels, the stress index, the fatigue index, and the integrated index in real time according to the user.

The battery module 170 serves to supply power required for driving.

The notification module 180 serves to provide a notification about the human error determined by the determination module 140 . For example, the notification module 180 may provide a vibration notification corresponding to each human error.

7 is an exemplary diagram of the apparatus 100 for determining human error using brain waves.

7, the band portion 111 of the brain wave measurement module 110 extends in the circumferential direction of the head to be worn on the head, and the first to fourth electrodes 113a to 113d are the band portion to contact the forehead. (111) is formed above. In addition, the reference electrode 113e is formed outside the band portion 111 . At this time, the first to fourth electrodes 113a to 113d and the reference electrode 113e are electrically connected.

According to an embodiment of the present invention, the communication module 150 , the battery module 160 , and the notification module 180 may be embedded in the band unit 111 or located on one side.

For example, the communication module 150 is located at the occipital point (INSION) opposite to the forehead where the first to fourth electrodes 113a to 113d are located in order to minimize the effect generated by the electrodes of the EEG measurement module 110 . to do it

FIG. 8 is a view for explaining a wearing position of the apparatus 100 for determining human error using brain waves of FIG. 1 .

The first to fourth electrodes 113a to 113d according to an embodiment of the present invention may be positioned on the user's forehead (NASION, near the nasal root point), such as F7, Fp1, Fp2, and F8. Of course, the first to fourth electrodes 113a to 113d may be disposed at appropriate positions on the scalp, such as F3, Cz, and P4.

The reference electrode 113e according to an embodiment of the present invention may be located at the left ear A1 or the right ear A2 of the user.

9 is a flowchart for explaining a human error determination method using brain waves according to the present invention.

First, the EEG measurement module 110 measures the user's EEG (S100).

According to an embodiment of the present invention, the reference electrode 113e is located on the user's left or right ear to measure the user's reference EEG signal. In addition, the first to fourth electrodes 113a to 113d are positioned on the user's forehead to measure EEG signals of 4 channels.

Then, the classification module 120 classifies the state of the measured EEG (S200).

The extraction unit 121 of the classification module 120 extracts EEG features from the measured EEG signals, and the generator 123 generates and learns a stress and fatigue learning model using the extracted EEG features. A classification algorithm is generated by learning the model, and the classification unit 125 classifies the stress or fatigue state from the extracted EEG features using the classification algorithm.

The calculation module 130 calculates an integrated index by using the classified EEG state (S300).

The first calculator 131 calculates a stress or fatigue index according to the following [Equation 1] and [Equation 2] using the classified stress or fatigue state, and the second calculator 133 calculates the An integrated index is calculated according to the following [Equation 3] by applying a weight to the stress or fatigue index.

[Equation 1]

는 시간 t(초)에서 스트레스 또는 피로 상태로 분류된 개수,

는 한 시간 동안 지속적으로 스트레스 또는 피로 상태로 분류된 개수를 나타낸다.here,

is the number classified as a stress or fatigue state at time t (seconds),

indicates the number of categories classified as stressed or fatigued continuously for one hour.

[Equation 2]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표를 나타낸다.here,

denotes an indicator for the stress or fatigue state at time x (seconds).

[Equation 3]

는 시간 x(초)에서의 스트레스 또는 피로 상태에 대한 지표, w는 각 사용자별 피로 대비 스트레스 가중치(단,

),

시간 x(초)에서의 스트레스 및 피로상태를 고려한 통합 지표를 나타낸다. here,

is an index for stress or fatigue state at time x (seconds), and w is the stress weight for each user versus fatigue (however,

),

It represents the integrated index considering the stress and fatigue state at time x (seconds).

The determination module 140 determines whether the integrated index is above or below a preset threshold (S410), and when the integrated index is greater than or equal to a preset threshold, determines the user's human error (S420).

The determination module 140 according to an embodiment of the present invention determines the human situation as 'normal', 'warning', 'danger', 'do not allow', 'warning', 'risk', The human condition of 'not allowed' is judged as human error. This is to stop the corresponding operation when the user's human error is judged.

Then, the notification module 180 provides a notification about human error to the server or control center (S500).

The notification module 180 according to an embodiment of the present invention may provide a vibration notification corresponding to each of the human errors of 'warning', 'danger', and 'not allowed' determined by the determination module 140 to the user. there is.

Meanwhile, a weight may be set according to the user's stress or fatigue. The weight is a value corresponding to fatigue versus stress for each user according to the result of the integrated index.

As described above, according to the present invention, the apparatus 100 and method for determining human error using brain waves can prevent accidents due to human errors by determining human errors using the user's brain wave signals and providing notifications.

Although described with reference to the illustrated embodiments of the present invention as described above, these are merely exemplary, and those of ordinary skill in the art to which the present invention pertains can use various functions without departing from the spirit and scope of the present invention. It will be apparent that modifications, variations and equivalent other embodiments are possible. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the appended claims.

100: human error determination device using brain waves
110: EEG measurement module
120: classification module
130: calculation module
140: judgment module
150: communication module
160: monitoring module
170: battery module
180: notification module

Claims (14)

EEG measurement module for measuring the user's EEG signal;
a classification module for classifying the measured EEG state;
a calculation module for calculating an integrated index by using the classified EEG state; and
Human error determination apparatus using brain waves including a determination module for determining the user's human error when the integrated indicator is greater than or equal to a preset threshold.
According to claim 1,
The brain wave measurement module,
A band part that can be worn on the head;
It includes a plurality of measuring electrode parts mounted on the band part,
The measuring electrode unit comprises first to fourth electrodes arranged in a line to measure a plurality of EEG signals, and a reference electrode to measure a reference EEG signal.
According to claim 1,
The classification module is
an extraction unit for extracting EEG features from the EEG signal;
a generator for generating a classification algorithm by using the extracted EEG features;
The apparatus for determining human error using brain waves, including a classification unit for classifying stress or fatigue states from the extracted brain wave features by using the classification algorithm.
4. The method of claim 3,
The calculation module is
a first calculator for calculating a stress or fatigue index by using the classified stress or fatigue state;
A human error determination device using brain waves, including a second calculator for calculating an integrated index by applying a weight to the stress or fatigue index.
5. The method of claim 4,
The first calculation unit,
Human error determination apparatus using EEG, characterized in that the stress or fatigue index is calculated according to the following [Equation 1] and [Equation 2].
[Equation 1]

here,

is the number classified as a stress or fatigue state at time t (seconds),

indicates the number of categories classified as stressed or fatigued continuously for one hour.
[Equation 2]

here,

denotes an indicator for the stress or fatigue state at time x (seconds).
5. The method of claim 4,
The second calculation unit,
Human error determination device using EEG, characterized in that calculating the integrated index according to the following [Equation 3].
[Equation 3]

here,

is an index for stress or fatigue state at time x (seconds), and w is the stress weight for each user versus fatigue (however,

),

represents the integrated index considering the stress and fatigue states at time x (seconds).
According to claim 1,
Human error determination device using EEG comprising a monitoring module for monitoring the EEG signal and the integrated indicator.
According to claim 1,
a communication module for transmitting and receiving the EEG signal to a control center;
a battery module for supplying power required for driving; and
Human error determination device using brain waves further comprising a notification module for providing a notification about the human error.
The measurement module includes a measurement step of measuring the user's brain waves;
The classification module includes a classification step of classifying the measured EEG state;
The calculation module includes a calculation step of calculating an integrated index using the classified EEG state;
The determination module may include: a determination step of determining a human error of a user when the integrated index is greater than or equal to a preset threshold; and
The notification module is a human error determination method using brain waves, comprising a notification step of providing a notification to a user or a control center with information on the determined human error.
10. The method of claim 9,
Before measuring the user's EEG,
Human error determination method using brain waves further comprising a setting step of setting weights according to users.
10. The method of claim 9,
The classification step is
extracting, by an extraction unit, EEG features from the measured EEG signal;
generating, by a generator, a classification algorithm using the extracted EEG features;
A method for determining human error using EEG comprising the step of a classification unit classifying a stress or fatigue state from the extracted EEG features by using the classification algorithm.
12. The method of claim 11,
The calculation step is
a first calculation step in which a first calculation unit calculates a stress or fatigue index by using the classified stress or fatigue state;
A method for determining human error using brain waves, comprising a second calculation step of calculating, by a second calculation unit, an integrated index by applying a weight to the stress or fatigue index.
13. The method of claim 12,
The first calculation step is,
Human error determination method using EEG, characterized in that the stress or fatigue index is calculated according to the following [Equation 1] and [Equation 2].
[Equation 1]

here,

is the number classified as a stress or fatigue state at time t (seconds),

indicates the number of categories classified as stressed or fatigued continuously for one hour.
[Equation 2]

here,

denotes an indicator for the stress or fatigue state at time x (seconds).
13. The method of claim 12,
The second calculation step is
Human error determination method using EEG, characterized in that calculating the integrated index according to the following [Equation 3].
[Equation 3]

here,

is an index for stress or fatigue state at time x (seconds), and w is the stress weight for each user versus fatigue (however,

),

represents the integrated index considering the stress and fatigue states at time x (seconds).

Images