KR20210123704A - Apparatus and method for workload assessment - Google Patents

Abstract

The present invention relates to a workload assessment apparatus and a workload assessment method using the apparatus including: inertia measurement units attached to a worker's first body part and measuring inertia data on the first body part; at least one pressure sensor attached to the worker's second body part and measuring the pressure on the second body part applied during work; a data processing unit implementing human body modeling on the worker using the measured inertia data and preprocessing the measured pressure value; and a workload assessment unit assessing the workload attributable to a work posture using the human body modeling and assessing the workload attributable to a working load using the preprocessed measured pressure value. According to the present invention, inertia data and pressure measurement is performed with the inertia measurement unit and the pressure sensor attached to a worker's body, and then preprocessing is performed and workload assessment methodology is applied. As a result, the worker's workload can be assessed in real time, accurately, and conveniently.

Description

Apparatus and method for work load assessment {APPARATUS AND METHOD FOR WORKLOAD ASSESSMENT}

The present invention relates to an apparatus and method for evaluating an operator's workload for the prevention of musculoskeletal disorders.

The burden imposed on the worker during work is a cause of musculoskeletal disorders, and these musculoskeletal disorders cause accidents and injuries. Therefore, it is important to find and prevent major causes of musculoskeletal disorders during work. Work load assessment is one of the methods to find the main causes of musculoskeletal disorders.

The conventional workload evaluation is an observational evaluation method, and a trained evaluator actually sees the work in the workplace or watches an image recorded for a long time, and manually writes a preset sheet according to the evaluation methodology. Here, as the evaluation methodology, OWAS (Ovako Work Posture Assessment System), REBA (Rapid Entire Body Assessment), RULA (Rapid Upper Limb Assessment), AWBA (Agricultural Whole Body Assessment), etc. are used.

However, the conventional observational evaluation method has a problem in that it is difficult to evaluate for a long time because the evaluator evaluates it manually, and the evaluation result is different depending on the viewing angle of the operator, so that an accurate result is not derived. In other words, since a large number of evaluators are required for long-term evaluation, the operator's posture is actually evaluated once every 10 minutes, and a qualitative evaluation is made according to the evaluator's judgment, so it is difficult to accurately evaluate.

In addition, although the above-described evaluation methodologies have the advantage of being able to evaluate the work load of the operator performed at the job site, they do not consider time or load as a posture-oriented evaluation methodology.

Korean Patent Application Laid-Open No. 10-2016-0076488 (Apparatus and method for determining the possibility of musculoskeletal disorders, Digigate Co., Ltd., 2016.06.30.)

An object of the present invention for solving the above-mentioned problems is to provide a work load evaluation apparatus and method with increased accuracy and convenience by enabling automatic quantitative evaluation for a long time rather than a manual qualitative evaluation by an evaluator. In addition, it is to provide a work load evaluation device and method that can accurately evaluate not only the work posture, but also the work time and work load.

However, the problem to be solved by the present invention is not limited thereto, and may be variously expanded without departing from the spirit and scope of the present invention.

A work load evaluation apparatus according to an embodiment of the present invention for solving the above-described problems is attached to a first body part of an operator, and a plurality of inertia measurement devices (IMU, Inertia) for measuring inertia data of the first body part Measurement Unit), at least one pressure sensor attached to the second body part of the worker and measuring the pressure of the second body part applied during work, and modeling the worker's body using the measured inertia data and a data processing unit for pre-processing the pressure measurement value, evaluating the work load due to the working posture using the human body modeling, and evaluating the work load due to the work load using the pre-processed pressure measurement value Includes burden assessment department.

According to an aspect, the inertial data may include at least one of 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data. .

According to one aspect, the first body part may include at least one of a head, chest, waist, arm, and leg of the worker.

According to one aspect, the at least one pressure sensor may be attached to an insole and installed inside the worker's shoe.

According to one aspect, the at least one pressure sensor may be coupled to a patch and attached to the hip portion of the operator.

According to one aspect, the data processing unit compensates the measured value of the inertial data using a compensation algorithm, calculates a quaternion relative angle for each inertial measuring unit (IMU), and each inertial measuring unit (IMU) A data calibration operation of calculating the quaternion and Euler triaxial alignment angles of each inertial measurement unit (IMU) is performed by converting the relative angle into a quaternion rotation, and the quaternion and the Euler 3 of each inertial measurement unit (IMU) The relative angles of the plurality of inertial measurement units (IMUs) may be calculated using the axis alignment angle, and the human body modeling of the operator may be implemented using the relative angles of the plurality of inertial measurement units (IMUs).

According to one aspect, the data processing unit normalizes the received pressure data, classifies it into a training set and a test set, and extracts feature points using a window segmentation method for the training set and , and by calculating the rank of each feature point according to mutual information-forward feature selection (MI-FS), and selecting the feature point based on the rank, the preprocessing may be performed.

According to one aspect, The work burden assessment unit uses the human body modeling, at least of OWAS (Ovako Work Posture Assessment System), REBA (Rapid Entire Body Assessment), RULA (Rapid Upper Limb Assessment), AWBA (Agricultural Whole Body Assessment), and Cube method. According to one methodology, it is possible to evaluate the workload due to the above working posture.

According to one aspect, the work load evaluation unit evaluates the work load due to the work load using an artificial neural network model, and the artificial neural network model is a low risk, possible risk using the pressure measurement value. risk) and high risk (high risk) can output the work load evaluation results in any one of the three types.

In a work load evaluation method according to another embodiment of the present invention for solving the above-described problems, a plurality of inertia measurement devices (IMUs) are attached to a first body part of an operator, and a second body part of the worker is attached. attaching at least one pressure sensor to the second body part, wherein the plurality of inertial measurement units (IMUs) measure inertia data of the first body part, and the at least one pressure sensor is applied during operation measuring the pressure of the, implementing the human body modeling of the worker using the measured inertia data, preprocessing the pressure measurement value, and evaluating the workload due to the working posture using the human body modeling, , using the pre-processed pressure measurement value to evaluate the work load due to the work load.

According to an aspect, the inertial data may include at least one of 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data. .

According to one aspect, the first body part may include at least one of a head, chest, waist, arm, and leg of the worker.

According to one aspect, the at least one pressure sensor may be attached to an insole and installed inside the worker's shoe.

According to one aspect, the at least one pressure sensor may be coupled to a patch and attached to the hip portion of the operator.

According to one aspect, the step of implementing the human body modeling comprises compensating for the inertial data measurement value using a compensation algorithm and calculating a quaternion relative angle for each inertial measurement unit (IMU); Performing a data calibration operation of calculating the quaternion and Euler triaxial alignment angles of each inertial measurement unit (IMU) by converting the relative angle of each measuring unit (IMU) into a quaternion rotation, and each inertial measuring unit ( Calculating the relative angles of the plurality of inertial measurement units (IMUs) using the quaternion and Euler triaxial alignment angles of the IMU), and the relative angles of the plurality of inertial measurement units (IMUs) of the operator It may include the step of implementing the human body modeling.

According to one aspect, the pre-processing of the pressure measurement value includes normalizing the received pressure data, classifying the received pressure data into a training set and a test set, and using a window segmentation method for the training set. performing the preprocessing by extracting a feature, calculating the rank of each feature according to MI-FS (mutual information-forward feature selection), and selecting the feature based on the rank may include

According to one aspect, the step of evaluating the work load by the working posture is using the human body modeling, OWAS (Ovako Work Posture Assessment System), REBA (Rapid Entire Body Assessment), RULA (Rapid Upper Limb Assessment), AWBA (Agricultural Whole Body Assessment), and according to at least one methodology of the Cube method, it is possible to evaluate the work load by the working posture.

According to one aspect, the step of evaluating the work load due to the work load evaluates the work load due to the work load using an artificial neural network model, and the neural network model uses the pressure measurement value to evaluate the work load. risk), possible risk, and high risk can output the work load evaluation result in any one of three types.

In a computer readable storage medium storing a computer program for performing a work load evaluation method according to another embodiment of the present invention for solving the above-mentioned problems, the computer program causes the computer to: A plurality of inertia measurement devices (IMU) attached to the part measure the inertia data of the first body part, and at least one pressure sensor attached to the second body part of the operator is applied during work. A command to measure the pressure of the second body part, a command to implement human body modeling of the operator using the measured inertia data, and a command to preprocess the pressure measurement value, and a working posture using the human body modeling and a command to evaluate the work load by the work load and evaluate the work load by the work load using the preprocessed pressure measurement value.

The disclosed technology may have the following effects. However, this does not mean that a specific embodiment should include all of the following effects or only the following effects, so the scope of the disclosed technology should not be construed as being limited thereby.

According to the work load evaluation apparatus and method according to the above-described embodiments of the present invention, the inertial measurement device and the pressure sensor are attached to the body of the worker to measure the inertia data and pressure, and by pre-processing it, the work load evaluation methodology is applied. and can conveniently and in real time evaluate the workload of the operator.

1 is a block diagram of a work load evaluation apparatus according to an embodiment of the present invention.
2 is a diagram illustrating an exemplary position where a sensor is attached to an operator's body.
3 is a flowchart illustrating a method of implementing human body modeling from measurement values of the inertial measurement device.
4 is a flowchart illustrating a method of determining the degree of risk of a working load from a measurement value of a pressure sensor.
5 is a diagram for explaining a work load evaluation method of the OWAS method.
6 is a view for explaining a work load evaluation method of the AWBA method.
7 is a diagram for explaining a work load evaluation method of the Cube method.
8 is a flowchart of a work load evaluation method according to another embodiment of the present invention.
9 is an experimental result performed by applying the work load evaluation apparatus and method according to the embodiments of the present invention.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to a specific embodiment, it should be understood to include all modifications, equivalents or substitutes included in the spirit and scope of the present invention.

Terms such as first, second, etc. 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. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

When a component is referred to as being “connected” or “connected” to another component, it may be directly connected or connected to the other component, but it should be understood that other components may exist in between. something to do. On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.

The terms used in the present application 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 the present application, terms such as “comprise” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and are not interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. .

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described clearly and in detail so that those of ordinary skill in the art to which the present invention pertains can easily practice the present invention.

1 is a block diagram of a work load evaluation apparatus according to an embodiment of the present invention, and FIG. 2 is a view showing an exemplary position at which a sensor is attached to a body of an operator.

Referring to FIG. 1 , the work load evaluation apparatus 100 according to an embodiment of the present invention includes a sensor unit 110 , a data processing unit 130 , and a work load evaluation unit 150 , and the sensor unit 110 . ) includes an inertia measurement unit (IMU) 112 and a pressure sensor 114 . According to an embodiment, the sensor unit 110 may be implemented separately from the data processing unit 130 and the workload evaluation unit 150 . In this case, the sensor unit 110 is attached to the body of the worker to measure data, and the data processing unit 130 and the work load evaluation unit 150 are implemented in a terminal including a communication module and a processor to evaluate the work load. .

The inertial measurement unit (IMU) 112 is attached to the body of the operator, and measures inertia data of the corresponding body part. Here, the inertial data includes at least one of 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data. That is, the inertial measurement unit (IMU) 112 includes at least one of an acceleration sensor, an angular velocity sensor, and a geomagnetic sensor. According to an embodiment, a plurality of inertial measurement units (IMUs) 112 may be attached to a plurality of body parts of an operator. For example, as shown in FIG. 2 , a plurality of inertial measurement units (IMUs) 112 are the operator's head 112a, chest 112b, waist 112c, arms 112d to 112g, or legs ( 112h to 112k). In this case, the inertial measurement unit (IMU) 112 may be provided in a form coupled to a hat, a band, or a patch so as to be attached to the operator's head, chest, waist, arms, legs, and the like.

The pressure sensor 114 measures the pressure of the body part applied during work. The pressure sensor 114 may be attached to a body part to which pressure (ie, load) is applied during work. For example, as shown in FIG. 2 , the pressure sensor 114 is attached to the insole for evaluation of the workload of a standing job, and the pressure applied to both feet of the worker while being installed inside the worker's shoes (that is, , load) can be measured (114a, 114b). In addition, the pressure (ie, load) applied to both hips of the operator may be measured while attached to the patch and attached to the hip portion of the operator (eg, the back pocket) for evaluation of the workload of the sitting task. At this time, at least one pressure sensor 114 may be used to measure the pressure applied to one body part. That is, in the above example, at least one pressure sensor 114 may be attached to each insole in the case of performing a work load evaluation of a standing job.

The data processing unit 130 implements the human body modeling of the worker using the measured values of the inertial measurement unit (IMU) 112, and the work load evaluation unit 150 evaluates the risk level of the work load using an artificial neural network. The measured value of the pressure sensor 114 is pre-processed so that

The work load evaluation unit 150 evaluates the work load due to the working posture using the human body modeling implemented by the data processing unit 130, and the work load due to the work load using the measured value of the pre-processed pressure sensor 114 . evaluate As a methodology for evaluating work load by working posture, for example, Ovako Work Posture Assessment System (OWAS), Rapid Entire Body Assessment (REBA), Rapid Upper Limb Assessment (RULA), Agricultural Whole Body Assessment (AWBA), and At least any one of the Cube methods may be used, and a plurality of evaluation methodologies may be applied to calculate the work load evaluation result according to each methodology.

Hereinafter, the human body modeling implementation will be described with reference to FIG. 3 , the work load evaluation by the work load will be described with reference to FIG. 4 , and the work load evaluation unit 150 evaluated with reference to FIGS. 5 to 7 . We will explain how to apply the methodology to evaluate the workload caused by the working posture.

3 is a flowchart illustrating a method of implementing human body modeling from measurement values of the inertial measurement device.

Referring to FIG. 3 , first, the data processing unit 130 receives a measurement value from the inertial measurement unit (IMU) 112 ( S310 ). As described above, the measured value of the inertial measurement unit (IMU) 112 is at least one of 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data of the body part to which the inertial measurement device (IMU) 112 is attached. include

Next, the data processing unit 130 compensates the measured value of the inertial measurement unit (IMU) 112 using a compensation algorithm, and calculates a quaternion relative angle for each sensor ( S330 ). For example, a quaternion relative angle is calculated for each sensor by compensating for a drift phenomenon of the angular velocity sensor included in the inertial measurement unit (IMU) 112 using the Madgwick filter algorithm.

Next, the data processing unit 130 converts the compensated relative angle for each sensor into a quaternion rotation to calculate the quaternion and Euler triaxial alignment angles of the inertial measurement unit (IMU) 112. Perform a data calibration (calibration) operation. (S350).

Next, the data processing unit 130 calculates the relative angles of the plurality of inertial measurement units (IMUs) 112 using the quaternion and Euler triaxial alignment angles of the inertial measurement unit (IMU) 112 . In this case, the relative angles of the plurality of inertial measurement units (IMUs) 112 may be calculated using two quaternions (S370).

Finally, the human body modeling of the operator is implemented using the relative angles of the plurality of inertial measurement units (IMUs) 112 (S390). For example, when a plurality of inertial measurement units (IMUs) 112 are attached to the body of the operator as shown in FIG. 2 , the angle of the neck of the operator is calculated using the relative angles of 112a and 112b, and the relative angles of 112b and 112c are used. The angle can be used to calculate the angle of the operator's torso. In addition, the angle of the upper arm of the operator's right arm can be calculated using the relative angles of 112b and 112d, and the angle of the lower arm of the operator's right arm can be calculated using the relative angles of 112d and 112e. In addition, the angle of the upper arm of the operator's left arm can be calculated using the relative angles of 112b and 112f, and the angle of the lower arm of the operator's left arm can be calculated using the relative angles of 112f and 112g. Meanwhile, the right pelvis angle of the operator may be calculated using the relative angles of 112c and 112h, and the left pelvis angle of the operator may be calculated using the relative angles of 112c and 112j. In addition, the operator's right knee angle may be calculated using the relative angles of 112h and 112i, and the operator's left knee angle may be calculated using the relative angles of 112j and 112k.

4 is a flowchart illustrating a method of determining the degree of risk of a working load from a measurement value of a pressure sensor.

Referring to FIG. 4 , first, the data processing unit 130 receives a measurement value from the pressure sensor 114 . As described above, the measured value of the pressure sensor 114 is pressure (ie, load) data applied to the body part to which the pressure sensor 114 is attached.

Next, the data processing unit 130 pre-processes the measured value of the pressure sensor 114 so that the work load evaluation unit 150 can evaluate the degree of risk of the work load using the artificial neural network (S430). Specifically, the received pressure data is normalized and classified into a training set and a test set. At this time, not only the measured value of the pressure sensor 114 but also the evaluation result data for each operation by experts can be additionally collected and used as a test set. For the training set, feature points are extracted using a window segmentation method. Thereafter, the rank of each feature point is calculated according to MI-FS (mutual information-forward feature selection), and a feature point to be used for training the artificial neural network is selected based on this. MI-FS is a two-step feature point selection method and can derive prediction results with higher accuracy than when only MI or FS is used.

Next, the work load evaluation unit 150 evaluates the work load due to the work load using the artificial neural network model (S450). The artificial neural network uses the measurement value of the pressure sensor to work with three types of work load: low risk (S471), possible risk (S473), and high risk (S475). print the burden

On the other hand, as described above, the work load evaluation unit 150 evaluates the work load due to the working posture using the human body modeling implemented by the data processing unit 130 . At this time, the work load evaluation unit 150 can evaluate the work load by the working posture by applying the evaluation methodologies listed above, and the OWAS method, the AWBA method, and the Cube method will be briefly described below as examples. The work load evaluation apparatus 100 according to an embodiment of the present invention automatically and real-time application of this evaluation methodology enables accurate and convenient evaluation compared to a conventional evaluation method by hand of an evaluator.

5 is a diagram for explaining a work load evaluation method of the OWAS method.

The OWAS method is, in general, the most widely used work load assessment method internationally. As described above, the OWAS method is an observational evaluation method, and a trained evaluator actually sees the work in the workplace or watches an image recorded for a long time, and writes a preset sheet by hand. For accurate evaluation, evaluation should be performed once per second, but in reality, evaluation is performed once every 10 minutes due to the lack of manpower.

Referring to FIG. 5 , the OWAS method first calculates scores according to the working posture for the waist, arms, and legs, and applies the evaluated scores to the OWAS category to finally derive the risk assessment results of steps 1 to 4.

In the work load evaluation apparatus 100 according to an embodiment of the present invention, the work load evaluation unit 150 calculates a score according to the working posture according to the OWAS method using the human body modeling implemented by the data processing unit 130, and , it is possible to apply the OWAS category by considering the working load together and derive the risk assessment result.

6 is a view for explaining a work load evaluation method of the AWBA method.

The AWBA method is a work load evaluation method developed for the purpose of evaluating the work load of Korean farmers in consideration of the agricultural characteristics of Korea. Like the OWAS method, the AWBA method evaluates the work load based on one point during the work. The AWBA method is an observational evaluation method, and a trained evaluator actually sees the work in the workplace or watches a video recorded for a long time, and writes a preset sheet by hand. For accurate evaluation, evaluation should be performed once per second, but in reality, evaluation is performed once every 10 minutes due to the lack of manpower.

Referring to FIG. 6 , the AWBA method first evaluates the risk level according to the AULA method and the ALLA method based on the working posture of the worker, and applies this to the AWBA table to finally obtain the risk evaluation results of steps 1 to 4 derive

In the work load evaluation apparatus 100 according to an embodiment of the present invention, the work load evaluation unit 150 evaluates the risk according to AULA and ALLA according to the AWBA method using the human body modeling implemented by the data processing unit 130 . and the risk assessment result can be derived according to the AWBA table.

7 is a diagram for explaining a work load evaluation method of the Cube method.

The Cube method is a work load evaluation method that considers all three major factors that cause work load (ie, working posture, work time, and work load). It is a dynamic workload evaluation methodology that evaluates

Referring to FIG. 7 , the Cube method classifies a previously performed task into detailed tasks according to task contents or task actions. And the detailed work is evaluated according to the table of FIG. 7, where the working time is evaluated by converting the time for performing the corresponding detailed work among the total working hours into a percentage, and the working load is evaluated according to a qualitative standard.

In the work load evaluation apparatus 100 according to an embodiment of the present invention, the work load evaluation unit 150 uses the human body modeling and work load implemented by the data processing unit 130 to detail the work performed according to the Cube method. It can be classified as a task and the workload can be evaluated. In this case, in the case of work load, quantitative evaluation can be performed using an artificial neural network rather than a qualitative criterion, unlike the conventional manual evaluation of workers.

8 is a flowchart of a work load evaluation method according to another embodiment of the present invention.

Referring to FIG. 8 , first, a plurality of inertial measurement units (IMUs) 112 are attached to a first body part of an operator, and at least one pressure sensor 114 is attached to a second body part (S810). The plurality of inertial measurement units (IMUs) 112 may be attached to, for example, the operator's head, chest, waist, arm, or leg, and at least one pressure sensor 114 may be, for example, an insole. It may be attached to the (insole) and installed on the inside of the worker's shoes, or may be coupled to the patch and attached to the worker's buttocks (eg, back pocket).

Next, a plurality of inertial measurement units (IMUs) measure the inertia data of the first body part, and at least one pressure sensor measures the pressure of the second body part applied during operation (S830). For example, the plurality of inertial measurement units (IMUs) 112 may include 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data 3 - of the first body part. axis geomagnetic data) can be measured.

Next, the human body modeling is implemented from the measured value of the inertial measurement unit (IMU) 112 and the measured value of the pressure sensor 114 is pre-processed (S850).

Specifically, the measured value of the inertial measurement unit (IMU) 112 may be compensated using a compensation algorithm, and a quaternion relative angle for each sensor may be calculated. A calibration operation of calculating the quaternion and Euler triaxial alignment angles of the inertial measurement unit (IMU) 112 may be performed by converting the compensated relative angles for each sensor into a quaternion rotation. In addition, the relative angles of the plurality of inertial measurement units (IMUs) 112 are calculated using the quaternion and Euler triaxial alignment angles of the inertial measurement unit (IMU) 112, and the The human body modeling of the operator may be implemented using the relative angle.

Meanwhile, the pressure data received from the pressure sensor 114 is normalized and classified into a training set and a test set. At this time, not only the measured value of the pressure sensor 114 but also the evaluation result data for each operation by experts may be additionally collected and used as a test set. For the training set, feature points can be extracted using a window segmentation method. Thereafter, the rank of each feature point is calculated according to mutual information-forward feature selection (MI-FS), and a feature point to be used for training the artificial neural network can be selected based on this.

Next, the work load evaluation by the working posture and the work load evaluation by the work load are performed (S870).

Work burden assessment by working posture is, for example, OWAS (Ovako Work Posture Assessment System), REBA (Rapid Entire Body Assessment), RULA (Rapid Upper Limb Assessment), AWBA (Agricultural Whole Body Assessment), and Cube methods. It may be performed using at least one, and a plurality of evaluation methodologies may be applied to calculate a work load evaluation result according to each method.

On the other hand, the work load caused by the work load can be evaluated using an artificial neural network model. The artificial neural network can output the work load due to the work load in three types of low risk, possible risk, and high risk using the measured value of the pressure sensor 114 . .

9 is an experimental result performed by applying the work load evaluation apparatus and method according to the embodiments of the present invention.

A comparative experiment with the prior art was performed by applying the work load evaluation apparatus and method according to the embodiments of the present invention. In the experiment, when a worker attaches a sensor to the body and performs 22 operations in 11 representative agricultural operations, the results of the workload assessment by the apparatus and method for assessing workload according to the embodiments of the present invention and two observers A, the results of the workload evaluation performed according to the OWAS method, which is the most used among the prior art, were compared.

In the table of FIG. 9 , shaded cells indicate that the evaluation result of the observer matches the evaluation result applied with the work load evaluation apparatus and method according to the embodiments of the present invention. Referring to FIG. 9 , the evaluation results of the work load evaluation apparatus and method according to embodiments of the present invention are the same as the evaluation results of two evaluators or the same as the evaluation results of one person. On the other hand, evaluation results of two observers were found to be different from each other in 7 movements out of 22, which is due to a limitation of the conventional observer evaluation method. That is, a difference occurs in the evaluation result of the evaluator depending on the angle from which the operator is viewed, and this is due to the qualitative evaluation in which the operator's subjectivity is reflected. The work load evaluation apparatus and method according to the embodiments of the present invention quantitatively calculates the joint angle of the human body and evaluates the work load, so the reproducibility is high, the reliability is high, and the convenience is high.

The above-described work load assessment method according to the present invention may be implemented as a computer-readable code on a computer-readable recording medium. The computer-readable recording medium includes any type of recording medium in which data that can be read by a computer system is stored. For example, there may be a read only memory (ROM), a random access memory (RAM), a magnetic tape, a magnetic disk, a flash memory, an optical data storage device, and the like. In addition, the computer-readable recording medium may be distributed in computer systems connected through a computer communication network, and stored and executed as readable codes in a distributed manner.

Although described above with reference to the drawings and embodiments, it does not mean that the protection scope of the present invention is limited by the drawings or embodiments, and those skilled in the art will appreciate the spirit and scope of the present invention described in the claims below. It will be understood that various modifications and variations of the present invention can be made without departing from the scope thereof.

100: work load evaluation device
110: sensor unit
112: inertial measurement unit (IMU)
114: pressure sensor
130: data processing unit
150: work load evaluation department

Claims (19)

a plurality of inertia measurement units (IMUs) attached to a first body part of an operator and measuring inertia data of the first body part;
at least one pressure sensor attached to the second body part of the worker and measuring the pressure of the second body part applied during work;
a data processing unit that implements human body modeling of the operator using the measured inertia data and pre-processes the pressure measurement value; and
and a work load evaluation unit for evaluating the work load due to the work posture using the human body modeling and evaluating the work load due to the work load using the pre-processed pressure measurement value. According to claim 1,
The inertial data is
A workload assessment device comprising at least one of 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data. According to claim 1,
The first body part is
A work load evaluation device comprising at least one of the worker's head, chest, waist, arm, and leg. According to claim 1,
the at least one pressure sensor
A work load evaluation device that is attached to the insole and installed inside the worker's shoes. According to claim 1,
the at least one pressure sensor
Combined with the patch and attached to the buttocks of the worker, the work load evaluation device. According to claim 1,
The data processing unit
Compensating the measured value of the inertial data using a compensation algorithm, calculating a relative angle of a quaternion for each inertial measurement unit (IMU),
Quaternion rotation conversion of the relative angle for each inertial measurement unit (IMU) to calculate the quaternion and Euler triaxial alignment angle of each inertial measurement unit (IMU) performs a data calibration (calibration) work,
Calculating the relative angles of the plurality of inertial measurement units (IMUs) using the quaternion and the Euler triaxial alignment angle of each inertial measurement unit (IMU), and
A work load evaluation device for implementing the human body modeling of the operator by using the relative angles of the plurality of inertial measurement units (IMUs). According to claim 1,
The data processing unit
Normalize the received pressure data and classify it into a training set and a test set,
extracting features using a window segmentation method for the training set, and
A work load evaluation apparatus for calculating a rank of each feature point according to mutual information-forward feature selection (MI-FS), and performing the preprocessing by selecting a feature point based on the rank. According to claim 1,
The work load evaluation unit
Using the human body modeling, according to at least one methodology of OWAS (Ovako Work Posture Assessment System), REBA (Rapid Entire Body Assessment), RULA (Rapid Upper Limb Assessment), AWBA (Agricultural Whole Body Assessment), and Cube method A work load evaluation device that evaluates the work load by the work posture. According to claim 1,
The work load evaluation unit evaluates the work load due to the work load using an artificial neural network model,
The artificial neural network model outputs the work load evaluation result by the work load in any one of three types of low risk, possible risk, and high risk using the pressure measurement value. , a work load assessment device. attaching a plurality of inertia measurement devices (IMUs) to a first body part of an operator, and attaching at least one pressure sensor to a second body part of the operator;
measuring, by the plurality of inertial measurement units (IMUs), inertial data of the first body part, and measuring, by the at least one pressure sensor, the pressure of the second body part applied during operation;
implementing human body modeling of the operator using the measured inertia data, and pre-processing the pressure measurement value; and
Evaluating the workload due to the working posture using the human body modeling, and evaluating the workload due to the working load using the pre-processed pressure measurement value. 11. The method of claim 10,
The inertial data is
A workload assessment method comprising at least one of 3-axis acceleration, 3-axis angular velocity, and 3-axis geomagnetic data. 11. The method of claim 10,
The first body part is
Including at least one of the worker's head, chest, waist, arm, and leg, the work load assessment method. 11. The method of claim 10,
the at least one pressure sensor
A work load assessment method that is attached to an insole and installed inside the worker's shoes. 11. The method of claim 10,
the at least one pressure sensor
It is coupled to the patch and attached to the buttocks of the worker, the work load evaluation method. 11. The method of claim 10,
The step of implementing the human body modeling is
compensating for the measured value of the inertial data using a compensation algorithm, and calculating a quaternion relative angle for each inertial measurement unit (IMU);
performing a data calibration operation of calculating a quaternion and Euler triaxial alignment angle of each inertial measurement unit (IMU) by converting the relative angle of each inertial measurement unit (IMU) into a quaternion rotation;
calculating the relative angles of the plurality of inertial measurement units (IMUs) using the quaternion and the Euler triaxial alignment angles of the respective inertial measurement units (IMUs); and
and implementing the human body modeling of the operator by using the relative angles of the plurality of inertial measurement units (IMUs). 11. The method of claim 10,
The step of pre-processing the pressure measurement value is
normalizing the received pressure data and classifying it into a training set and a test set;
extracting feature points using a window segmentation method for the training set; and
Calculating a rank of each feature point according to mutual information-forward feature selection (MI-FS), and performing the pre-processing by selecting a feature point based on the rank. 11. The method of claim 10,
The step of evaluating the work load by the working posture is
Using the human body modeling, according to at least one methodology of OWAS (Ovako Work Posture Assessment System), REBA (Rapid Entire Body Assessment), RULA (Rapid Upper Limb Assessment), AWBA (Agricultural Whole Body Assessment), and Cube method A work load evaluation method for evaluating the work load by the work posture. 11. The method of claim 10,
The step of evaluating the work load due to the work load evaluates the work load due to the work load using an artificial neural network model,
The human neural network model outputs the work load evaluation result by the work load in one of three types of low risk, possible risk, and high risk using the pressure measurement value. , the work load assessment method. A computer readable storage medium storing a computer program for performing a method of evaluating a workload, wherein the computer program causes the computer to:
A plurality of inertia measurement devices (IMU) attached to the first body part of the operator measure inertia data of the first body part, and at least one pressure sensor attached to the second body part of the operator a command to measure the pressure of the second body part applied during the operation;
a command to implement human body modeling of the operator using the measured inertia data and pre-process the pressure measurement value; and
A computer-readable storage medium storing a computer program, comprising instructions to evaluate the work load due to the working posture using the human body modeling and to evaluate the work load due to the work load using the pre-processed pressure measurement value .

Images