JP7338282B2 - Skill comparison device and method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a skill comparison device and a skill comparison method for comparing each skill of a reference person and a target person.

Skills are acquired by repeatedly performing actual work and through experience, and are therefore difficult to verbalize, document, and illustrate. Therefore, skills are difficult to convey to others and difficult to quantify. Techniques related to such skills include, for example, the techniques disclosed in Patent Literature 1 and Patent Literature 2.

The skill acquisition support system disclosed in Patent Document 1 is a skill acquisition support system that detects the difference in the user's motion with respect to the modeler's motion. a sensor attached to detect data relating to the motion of each of the model person and the user; motion data acquisition means for acquiring data relating to the motion detected by the sensor; storage means for recording such data; and data relating to the behavior of the model person recorded in the storage means are graphed in chronological order, a first graph is formed and displayed, and data relating to the behavior of the user is displayed. and graphing means for graphing in chronological order and forming and displaying a second graph. The motion is a motion of a coating work, and the monitored parts are the spray gun, the wrist, the arm, the waist, etc. of the worker. The angle and angular velocity are detected, and the azimuth angle of the monitoring site is detected by the triaxial geomagnetic sensor.

The skill evaluation system disclosed in Patent Document 2 is a device that evaluates the skill level of a worker who needs to acquire work skills. Work measurement detection means for measuring a state and detecting it as quantitative behavior data and work state data; signal processing means for extracting and processing necessary time-series signals from the behavior data and the work state data; and said signal processing means. pattern feature extracting means for extracting a specific significant pattern from the results of signal processing; pattern storing means for storing pre-constructed exemplary data and information on its data range/pattern; and said pattern storing means. pattern comparison means for comparing the pattern of the exemplary data stored by and the pattern of the worker extracted by the feature extraction means; and skill evaluation result presenting means for presenting, as an evaluation result, an index obtained by one-dimensionalizing the multidimensional feature quantity that can be expressed by the feature expressing means. The work is a manual welding work, and the behavior data and the work state data are video information of the worker's behavior taken by a CCD camera, a worker's work detected by a gyro sensor, an acceleration sensor, and a motion sensor. and the temperature of the object to be welded detected by a temperature sensor. value, minimum value and variation range (maximum value - minimum value) (see, eg, paragraph [0077]).

JP 2013-088730 A JP 2006-171184 A

The systems disclosed in these Patent Documents 1 and 2 collect, analyze, and compare data of many items. Therefore, it takes time to analyze the data of many items. It is presumed that each importance in the data of a large number of items is not necessarily equivalent to the operation of the work, but in the systems disclosed in Patent Document 1 and Patent Document 2, each importance is unknown. Therefore, it is difficult to make a selection according to the degree of importance.

The present invention is an invention made in view of the above circumstances, and its purpose is to compare each skill of the reference person and the target person with a smaller number of items with high importance based on new knowledge. It is to provide a skill comparison device and a skill comparison method.

As a result of various studies, the inventors of the present invention have found that the above object can be achieved by the present invention described below. That is, a skill comparison device according to one aspect of the present invention is a device that compares the skills of a reference person and a subject person, which are performed by body movements and are related to a predetermined work of handling a predetermined object, wherein the reference person A reference information storage unit that stores a reference value related to a predetermined item in the above; a motion measurement unit that measures the body motion of the subject; A target value processing unit that obtains a target value related to the item for the target person, the reference value of the referrer stored in the reference information storage unit, and the target value of the target person obtained by the target value processing unit wherein the items are the displacement of the body part handling the predetermined object, among the plurality of body parts obtained by dividing the body, and the rotation of the trunk of the plurality of body parts The velocity and the acceleration of the waist among the plurality of body parts, and the items further include the left foot load applied to the left foot, the stepping motion to shift the weight from the left foot to the right foot, and the weight applied to the right foot. including a right foot load, further comprising: a left foot load acquisition unit that acquires the left foot load; a stepping motion acquisition unit that acquires the stepping motion; and a right foot load acquisition unit that acquires the right foot load, and handles the predetermined object. The displacement of the body part, the turning speed of the trunk, and the acceleration of the waist are each displacement, each velocity, and each acceleration of the center of gravity of the plurality of body parts, the displacement, velocity, and acceleration of the center of gravity of the object, the Main components in expressing the motion of the body using the XYZ components of the external force acting on the right leg, the XYZ components of the external force acting on the left leg, and the XYZ components of the external force acting on the hand holding the object as explanatory variables obtained based on the first to third principal components of component analysis, the reference value of the referrer is the principal component score of the referrer, and the target value of the subject is the principal component of the subject is the score . A skill comparison method according to another aspect of the present invention is a method of comparing, between a reference person and a subject, the skills related to a predetermined work involving a predetermined object performed by body motion, wherein the reference person a reference information storage step of storing a reference value related to a predetermined item in the reference information storage unit; a motion measurement step of measuring the body motion of the subject; and the body motion of the subject measured in the motion measurement step a target value processing step of obtaining a target value related to the item in the subject, the reference value of the referrer stored in the reference information storage unit, and the target obtained in the target value processing step based on and an output step of outputting the target value of the person, wherein the item includes, among a plurality of body parts obtained by dividing the body into a plurality of parts, displacement of a body part handling the predetermined object, and the acceleration of the waist among the plurality of body parts, and the items further include the left foot load applied to the left foot, the stepping motion to shift the weight from the left foot to the right foot, and , a left foot load obtaining step of obtaining the left foot load, a stepping motion obtaining step of obtaining the stepping motion, and the right foot load, which are performed together with the performance of the motion measuring step, including the right foot load applied to the right foot. The displacement of the body part handling the predetermined object, the turning speed of the trunk, and the acceleration of the waist are obtained by obtaining the displacement, the velocity, and the displacement of the center of gravity of the plurality of body parts. Each acceleration, the displacement of the center of gravity of the object, velocity and acceleration, the XYZ components of the external force acting on the right foot, the XYZ components of the external force acting on the left foot, and the external force acting on the hand holding the object obtained based on the first to third principal components of principal component analysis when representing the motion of the body using the XYZ components as explanatory variables, and the reference value of the referrer is the principal component score of the referrer; The target value of the subject is a principal component score of the subject .

According to the new findings described below, the skill related to a given task of handling a given object, which is performed by body motion, is characterized by the displacement of the body part handling the given object, the turning speed of the trunk, and the acceleration of the waist. Each of these items is important in that order. The skill comparison device and skill comparison method output reference values and target values for the displacement of the body part handling a given object, the turning speed of the trunk, and the acceleration of the waist. It is possible to compare each skill in the referrer and the subject with a smaller number of items with higher degrees.

According to the novel findings, the skills involved in a given task of handling a given object performed by body movements are further characterized by left foot load, stepping motion and right foot load following each of the above three items. Each of these items is important in that order. Since the skill comparison device and skill comparison method further output each reference value and each target value for each of the left foot load, the stepping motion, and the right foot load, the items with high importance based on the new knowledge and fewer Items can be used to compare each skill of the referrer and the target in a more multifaceted manner.

In another aspect, in the above-described skill comparison device, the plurality of workers are divided into a plurality of classes based on work results of having the plurality of workers perform the predetermined work, and the work results are divided into a plurality of classes. It further comprises a referrer determination processing unit that selects the referrer from the workers classified into the best class . In another aspect, in the above-described skill comparison method, the plurality of workers are divided into a plurality of classes based on the work results of having the plurality of workers perform the predetermined work, and the work result is the best class. A referrer determination processing step of selecting the referrer from the workers classified into classes is further provided .

In the above skill comparison device and skill comparison method, since the work result of the referee is superior to the work result of the subject, the subject can improve his/her skill by comparing the work result of the referee who is superior to himself/herself. A reference value can be compared with its own target value. The comparison can be performed on an item-by-item basis so that the subject can recognize items for improvement.

In another aspect, in each of the above-described skill comparison device and skill comparison method, the work result is the accuracy of the work product and the time required for the work.

According to this, it is possible to provide a skill comparison device and a skill comparison method, each of which uses the accuracy of the product of the work and the time required for the work as the work result.

In another aspect, in any of these above-described skill comparison devices, the motion measurement unit comprises optical motion capture. In another aspect, in these above-described skill comparison methods, the motion measuring step measures motion of the body using optical motion capture.

Since the skill comparison device and skill comparison method use optical motion capture, it is possible to measure the body movements of the subject in a non-contact manner without interfering with the body movements.

In another aspect, in each of the skill comparison devices and skill comparison methods described above, the work is any one of grinding or polishing work, welding work, and painting work.

When the work is grinding or polishing work, the skill comparison device and skill comparison method can compare the skills of the reference person and the subject person regarding the work of grinding or polishing. When the work is a welding work, the skill comparison device and the skill comparison method can compare the skills related to the welding work between the reference person and the target person. When the work is a painting work, the skill comparison device and skill comparison method can compare the skills related to the painting work between the reference person and the target person.

In another aspect, in each of the above-described skill comparison device and skill comparison method, the work is a die grinding or polishing work.

The skill comparison device and skill comparison method described above can compare the skills of a reference person and a subject person regarding the work of grinding or polishing a mold.

In another aspect, in each of the above-described skill comparison device and skill comparison method, the mold is a vehicle component molding mold for molding a component used in a vehicle.

The skill comparison device and skill comparison method described above can compare the skills of a reference person and a subject person regarding the work of grinding or polishing a mold for molding a vehicle member.

In another aspect, in each of the above-described skill comparison device and skill comparison method, the object is held by a hand of the plurality of body parts, and the plurality of body parts are a head, a neck, and a chest. , right shoulder, left shoulder, right upper arm, left upper arm, right forearm, left forearm, right hand, left hand, waist, buttocks, right ilium, left ilium, right knee, left knee, right foot and left foot .

In the above skill comparison device and skill comparison method, the displacement of the body part handling a predetermined object, the turning speed of the trunk, and the acceleration of the waist are obtained based on the first to third principal components of the principal component analysis. Items can be appropriately selected in order of importance.

The skill comparison device and skill comparison method according to the present invention can compare each skill of the reference person and the target person with a smaller number of items that are highly important based on new knowledge.

It is a block diagram which shows the structure of the skill comparison apparatus in embodiment. FIG. 4 is a diagram for explaining attachment positions of a plurality of markers in optical motion capture used in the skill comparison device and arrangement positions of force plates used in the skill comparison device with respect to a reference person and a target person; It is a figure for demonstrating each center-of-gravity position of each part in a body. FIG. 4 is a diagram for explaining the contents of work and test pieces used in the work; It is a flow chart which shows the operation of the skill comparison device. FIG. 4 is a diagram for explaining a skill comparison display screen displayed on the skill comparison device;

One or more embodiments of the invention are described below with reference to the drawings. However, the scope of the invention is not limited to the disclosed embodiments. It should be noted that the configurations denoted by the same reference numerals in each figure indicate the same configurations, and the description thereof will be omitted as appropriate. In the present specification, reference numerals with suffixes omitted are used when referring to generically, and reference numerals with suffixes are used when referring to individual configurations.

FIG. 1 is a block diagram showing the configuration of a skill comparison device according to an embodiment. FIG. 2 is a diagram for explaining the attachment positions of a plurality of markers in the optical motion capture used in the skill comparison device and the arrangement positions of the force plates used in the skill comparison device with respect to the reference person and the target person; is. FIG. 3 is a diagram for explaining the position of the center of gravity of each part of the body. In FIG. 3, each center of gravity position of each part is indicated by ●.

The skill comparison device according to the present embodiment is a device that compares the skills of a reference person and a subject person that are performed by bodily movements and are related to a predetermined task of handling a predetermined object. The referrer is a person whose work result of work is superior to that of the target person. The referrer is one whose skills are referred to by the subject in order for the subject to improve his or her skills. Then, in order to select the reference person from a plurality of workers, the skill comparison device in this embodiment acquires predetermined information for classifying the plurality of workers according to their work results. is configured to Furthermore, in this embodiment, the skill comparison device also has a selection function for selecting the predetermined item from among a plurality of various items. For example, as shown in FIG. 1, such a skill comparison device D includes a plurality of imaging units 11-1 to 11-k (k is an integer of 2 or more), an external force measurement unit, and an external force measurement unit, mainly for comparing skills. 12 , a position processing unit 32 , a motion information processing unit 33 and a target value processing unit 35 in the control processing unit 3 , an output unit 5 , and a reference information storage unit 71 in the storage unit 7 . Mainly in order to acquire predetermined information for classifying, this skill comparison device D includes a shape measuring unit 21, a timer unit 22, and a work result processing unit 34 in the control processing unit 3, and a referrer (to be described later). A determination processing unit 36 is provided. Mainly for the selection function, the skill comparison device D includes a principal component analysis processing section 37 in the control processing section 3, which will be described later. In this embodiment, the skill comparison device D includes a control section 31 , an input section 4 and an interface section (IF section) 6 in the control processing section 3 , which will be described later.

The plurality of imaging units 11-1 to 11-k are respectively connected to the control processing unit 3, and according to the control of the control processing unit 3, a plurality of marked workers (for example, the reference person, the subject person, and This is an apparatus for generating images (image data) by capturing images of the plurality of markers from a plurality of mutually different directions with respect to a subject or the like. For example, in a rectangular parallelepiped space (for example, 4 m×4 m×2 m) in which an operator enters, the four imaging units 11-1 to 11-4 are positioned at the upper four corners so that the optical axes of the four imaging units 11-1 to 11-4 face the central region of the space. is arranged obliquely downward, and four imaging units 11-5 to 11-8 are arranged obliquely upward so that the four imaging units 11-5 to 11-8 face the central region of the space at the lower four corners of the space. be done. As a result, the eight imaging units 11-1 to 11-8 respectively capture the markers attached to the worker from eight different directions and generate images. The plurality of imaging units 11-1 to 11-k respectively output the images generated by the imaging to the control processing unit 3, and the control processing unit 3 outputs the images from the plurality of imaging units 11-1 to 11-k. are stored in the storage unit 7 in association with each other. More specifically, the imaging units 11-1 to 11-k each include, for example, an imaging optical system that forms an infrared optical image of an object to be imaged on a predetermined imaging plane. An area image sensor arranged with a light-receiving surface aligned with an image plane to convert an infrared optical image of the object to be imaged into an electrical signal; It is a digital infrared camera including an image processing unit that generates image data representing an infrared image.

Each of the plurality of markers is a member that reflects infrared light (infrared rays), and is attached to a plurality of locations on the body of a worker (referrer, subject, subject) who handles a predetermined object and on each of the objects. be. For example, as shown in FIG. 2, a plurality of markers MK are attached to each of 41 locations on the worker's body LB in order to measure the center-of-gravity position of each of 19 body regions. Attached to each of the three points on the object to measure the position of the center of gravity of the object. In this embodiment, the 19 body parts are, for example, the head, neck, chest, right shoulder, left shoulder, right upper arm, left left arm, right forearm, left forearm, right hand, left hand, These are the waist, buttocks, right ilium, left ilium, right knee, left knee, right foot and left foot. In order to suitably analyze the movement of the body, the sampling frequency in the multiple imaging units 11-1 to 11-k is, for example, 100 [Hz]. The body motion of the worker is represented as body motion information in this embodiment by each center-of-gravity position (time displacement of each center-of-gravity position) of each of 19 body parts, for example, as shown in FIG. The movement of the object caused by the movement of the worker's body is represented by the position of the center of gravity of the object (time displacement of the position of the center of gravity) as information on the movement of the object.

The body motion of the worker may be any motion as long as it is a motion to handle an object. Preferably, in one example, the movement of the body is a movement associated with the execution of any predetermined work, and the work is, for example, not only work for manufacturing goods, work for transporting goods, etc., but also performing arts, sports, etc. may include Said object is any given object used in said work, and preferably, in one example, said object is a tool used in said work. More specifically, in this embodiment, the work is a work of grinding or polishing a mold for molding a vehicle member for molding a member used in a vehicle. In this case, the tool is, in one example, a Hand Grinder. Of course, it is not limited to this, for example, the work is welding work, the tool is a welding machine such as a torch, and the work is painting work, for example A tool is, for example, a painting tool such as a spray gun. Preferably, the work is work related to vehicle manufacturing, and the tool is a tool related to the work.

In the above description, the marker is a member that reflects infrared light, and the imaging unit 11 is a digital infrared camera accordingly. The imaging unit 11 may be a member of a different color than the imaging unit 11. In this case, the imaging unit 11 may be a visible light camera.

Returning to FIG. 1, the external force measurement unit 12 is a device that is connected to the control processing unit 3 and measures the external force acting on the body of the worker according to the control of the control processing unit 3. A measurement result (external force acting on the body) is output to the control processing unit 3 . The control processing unit 3 stores this measurement result in the storage unit 7 in association with the body movement information and object movement information obtained by the position processing unit 32 as described later. The external force measurement unit 12 includes, for example, a force plate manufactured by AMTI, which measures displacement, velocity, and acceleration in each component of XYZ. In this embodiment, three first to third force plates 12-1 to 12-3 are used, and as shown in FIG. 2, the first force plate 12-1 acts on the right leg of the operator. The second force plate 12-2 is placed on the floor so as to measure the external force acting on the worker's left leg, and the first force plate 12-1 and along the left and right direction so as to measure the external force acting on the left leg of the worker. The third force plate 12-3 is placed on the floor side by side, and the third force plate 12-3 measures the external force (reaction force from the tool) acting on the hand that grips the tool. In the experimental example, a test piece TP) is placed on the workbench on which the worker works. The sampling frequency of these first through third force plates 12-1 through 12-3 is 100 [Hz] so as to synchronize with the sampling of the imaging unit 11. FIG.

The shape measuring unit 21 is a device that is connected to the control processing unit 3 and measures the shape of an object to be measured (a test piece TP after work in an example described later) under the control of the control processing unit 3 . The shape measurement unit 21 outputs the measurement result (the shape of the object to be measured) to the control processing unit 3 . The control processing unit 3 stores the measurement result in the storage unit 7 in association with the body motion information and object movement information obtained by the position processing unit 32 . The shape measuring unit 21 includes, for example, a 3D shape measuring machine manufactured by KEYENCE, which measures the shape of an object to be measured using three-dimensional coordinate values.

The clock unit 22 is a device that is connected to the control processing unit 3 and measures the time required for the work (working time) under the control of the control processing unit 3, and is, for example, a stopwatch. The timer 22 outputs the measurement result (working time) to the control processor 3 .

The input unit 4 is connected to the control processing unit 3, for example, various commands such as a command to start comparison, and various data necessary for skill comparison, such as worker names, are input to the skill comparison device D. For example, a plurality of input switches, a keyboard, a mouse, etc., to which predetermined functions are assigned. The output unit 5 is connected to the control processing unit 3, and according to the control of the control processing unit 3, the commands and data input from the input unit 4, the image of the imaging unit 11, the measurement results (analysis information), body movement information and physical movement information acquired by the skill comparison device D, a reference value for a predetermined item for the referrer, and a target value for the item for the subject. They include display devices such as CRT displays, liquid crystal displays and organic EL displays, and printing devices such as printers.

A so-called touch panel may be configured from the input unit 4 and the output unit 5 . In the case of constructing this touch panel, the input unit 4 is a position input device for detecting and inputting an operation position, such as a resistive film method or a capacitive method, and the output unit 5 is a display device. In this touch panel, the position input device is provided on the display surface of the display device, one or a plurality of input content candidates that can be input are displayed on the display device, and the input content that the user wants to input is displayed. When a position is touched, the position is detected by the position input device, and the display content displayed at the detected position is input to the skill comparison device D as the user's operation input content. With such a touch panel, it is easy for the user to intuitively understand the input operation, so that the skill comparison device D that is easy for the user to handle is provided.

The IF unit 6 is a circuit that is connected to the control processing unit 3 and performs data input/output with an external device under the control of the control processing unit 3. For example, an interface circuit of RS-232C, which is a serial communication method. , an interface circuit using the Bluetooth (registered trademark) standard, an interface circuit for infrared communication such as the IrDA (Infrared Data Association) standard, and an interface circuit using the USB (Universal Serial Bus) standard. The IF section 6 is a circuit for communicating with an external device, and may be, for example, a data communication card or a communication interface circuit conforming to the IEEE802.11 standard.

The storage unit 7 is a circuit that is connected to the control processing unit 3 and stores various predetermined programs and various predetermined data under the control of the control processing unit 3 . The various predetermined programs include, for example, a control program for controlling each section 11, 12, 21, 22, 4 to 7 of the skill comparison device D according to the function of each section, a plurality of imaging sections 11-1 A position processing program for determining the positions of a plurality of markers based on each image captured in 11-k, and the worker's body is divided into a plurality based on each position of the plurality of markers obtained by the position processing program. a motion information processing program for obtaining the positions of the centers of gravity of a plurality of body parts and the positions of the centers of gravity of the object as body motion information representing the body motion of the worker and object motion information representing the motion of the object; Based on the work result processing program that obtains the dimensional accuracy (processing accuracy) based on the shape measured in 21, and the body movement information and object movement information obtained by the movement information processing program, the object related to the predetermined item of the subject A control processing program such as a target value processing program for obtaining a value, a referrer determination processing program for determining the referrer, and a principal component analysis processing program for executing principal component analysis and selecting the predetermined item as will be described later. is included. The various predetermined data include the image of the imaging unit 11, the measurement results of the units 12, 21, and 22, and the processing results (for example, the above-mentioned center of gravity position and main Each principal component of component analysis, etc.) and data necessary for executing each of these programs, such as the name of the operator, are included. Such a storage unit 7 includes, for example, a ROM (Read Only Memory) that is a non-volatile storage element, an EEPROM (Electrically Erasable Programmable Read Only Memory) that is a rewritable non-volatile storage element, and the like. The storage unit 7 includes a RAM (Random Access Memory) or the like that serves as a so-called working memory of the control processing unit 3 that stores data generated during execution of the predetermined program.

The storage unit 7 functionally includes a reference information storage unit 71 that stores a reference value related to a predetermined item for the referrer. The predetermined item is an item that characterizes a skill related to a predetermined task of handling a predetermined object. , the turning speed of the trunk of the plurality of body parts, and the acceleration of the waist of the plurality of body parts. In order to compare each skill of the reference person and the subject person from various angles, the above items include the left foot load applied to the left foot, the stepping action to shift the weight from the left foot to the right foot, and the right foot load applied to the right foot. may include one or more of Such items are selected in order of importance and determined by principal component analysis as described later in order to reduce the number of items.

The control processing unit 3 controls each unit 11, 12, 21, 22, 4 to 7 of the skill comparison device D according to the function of each unit, and performs a predetermined work of handling a predetermined object, which is performed by body motion. It is a circuit for comparing the skills related to the reference person and the target person. The control processing unit 3 is configured by, for example, a CPU (Central Processing Unit) and its peripheral circuits. By executing the control processing program, the control processing unit 3 includes a control unit 31, a position processing unit 32, a motion information processing unit 33, a work result processing unit 34, a target value processing unit 35, and a referrer determination processing unit. 36 and a principal component analysis processing unit 37 are functionally configured.

The control unit 31 controls each unit 11, 12, 21, 22, 4 to 7 of the skill comparison device D according to the function of each unit, and controls the skill comparison device D as a whole. The control unit 31 associates each image captured by the plurality of image capturing units 11-1 to 11-k with a predetermined identifier (for example, a serial number, a worker name of a worker shown in each image, etc.). is stored in the storage unit 7. The control unit 31 stores the work time measured by the clock unit 22 as one of the work results in association with the identifier in the storage unit 7, and outputs it to the output unit 5 and the IF unit 6 as necessary.

The position processing section 32 obtains the positions of the plurality of markers based on the images captured by the plurality of imaging sections 11-1 to 11-k. The position processing unit 32 associates each of the obtained positions of the plurality of markers with each of the images (for example, associates them with the identifier associated with each of the images) and stores them in the storage unit 7. output to the output unit 5 or the IF unit 6.

More specifically, in an image generated by capturing an image of a marker that reflects infrared light with the image capturing unit 11 (11-1 to 11-k), the image area in which the marker is captured has its pixel value is different from the pixel values around it (for example, the pixels in the image area where the marker is projected are brighter and have higher pixel values than the pixels around it). By subjecting the image to image processing such as binarization processing, an image area in which the marker is projected is extracted from the image, and the two-dimensional coordinate values of the extracted image area in the image are obtained. Then, for the same marker, the position processing unit 32 calculates camera parameters (the different imaging units 11 Three-dimensional coordinate values of the marker are obtained as the position of the marker by so-called epipolar matching using each arrangement position and each orientation (in each optical axis direction).

In the present embodiment, the plurality of imaging units 11-1 to 11-k and the position processing unit 32 constitute an example of a position measuring unit that measures each position of the plurality of markers. configured with motion capture.

Based on the positions of the plurality of markers obtained by the position processing unit 32, the motion information processing unit 33 calculates the positions of the center of gravity of each of the plurality of body parts obtained by dividing the worker's body into a plurality of parts and the position of the center of gravity of the object. It is obtained as body motion information representing the body motion of the worker and object motion information representing the motion of the object. The motion information processing unit 33 associates the obtained body motion information and object motion information with each of the positions (for example, with the identifier associated with each of the positions via each of the images), and stores the information in the storage unit 7. , and output to the output unit 5 or the IF unit 6 as required.

More specifically, at each sampling timing (in the above example, each sampling timing at the sampling frequency of 100 [Hz]), the motion information processing unit 33 detects each position of the plurality of markers obtained by the position processing unit 32. Based on this, the position of the center of gravity of each of the 19 body parts is determined. For example, for each of the 19 body parts, a first arithmetic expression for obtaining the center-of-gravity position of the body part from the positions of a plurality of markers related to the body part is obtained in advance and stored in the storage unit 7, and the motion information processing unit. For each of the 19 body parts, 33 obtains the position of the center of gravity of the body part based on the first arithmetic expression from each position of a plurality of markers related to the body part obtained by the position processing unit 32 . Similarly, a second arithmetic expression for obtaining the center-of-gravity position of an object (a tool in this embodiment) from the positions of a plurality of markers related to the object is obtained in advance and stored in the storage unit 7. The position of the center of gravity of the object is obtained from the respective positions of the plurality of markers related to the object obtained by the processing unit 32 based on the second arithmetic expression.

The work result processing unit 34 obtains dimensional accuracy (processing accuracy) based on the shape measured by the shape measuring unit 21 . The shape of the work product measured by the shape measuring unit 21 may be used as another work result. machining accuracy) is obtained as another work result. The work result processing unit 34 associates the obtained dimensional accuracy with the identifier, stores it in the storage unit 7, and outputs it to the output unit 5 or the IF unit 6 as necessary.

The target value processing unit 35 obtains a target value related to a predetermined item of the target person based on the body motion information and the physical movement information of the target person obtained by the motion information processing unit 33 . More specifically, in the present embodiment, when the item includes one or more of the load on the left leg, the stepping motion, and the load on the right leg, the target value processing unit 35 calculates the target person's The target value is obtained based on the body motion information, the object motion information, and the measurement result measured by the external force measurement unit 12 . The target value processing unit 35 stores the obtained target value in the storage unit 7 in association with the body motion information and the object movement information (for example, in association with the identifier associated with the body motion information and the object movement information). The information is stored and output to the output unit 5 or the IF unit 6 as required. In this embodiment, the target value processing unit 35 obtains a reference value related to a predetermined item of the referrer based on the body motion information and the object movement information of the referrer obtained by the motion information processing unit 33, and The obtained reference value is stored in the reference information storage unit 71 . More specifically, in the present embodiment, when the item includes one or more of the load on the left foot, the stepping motion, and the load on the right foot, the target value processing unit 35 calculates the The reference value is obtained based on the body motion information, the object motion information, and the measurement result measured by the external force measurement unit 12 .

The referrer determination processing unit 36 determines a referrer as described later. Note that the operator (user) inputs a worker name (subject name) as a referrer from the input unit 4, and the referrer determination processing unit 36 performs a work having the worker name (subject name) input from the input unit. A person (subject) may be determined as a reference person.

The principal component analysis processing unit 37 executes principal component analysis and selects the predetermined items, as will be described later.

Then, the control unit 31 outputs the reference value of the referrer stored in the reference information storage unit 71 and the target value of the target person obtained by the target value processing unit 35 to the output unit 5. Output to part 6.

Here, the plurality of imaging units 11-1 to 11-k, the external force measurement unit 12, the position processing unit 32, and the motion information processing unit 33 measure the body motion of the worker (referrer, subject). It corresponds to an example of a motion measuring unit that configures the measuring unit 1 and measures the body motion of the subject. In this embodiment, as will be described later, the explanatory variables of the principal component analysis are the XYZ components of the external force acting on the right foot, the XYZ components of the external force acting on the left foot, and the Since the XYZ components of the external force are included, the external force measuring section 12 (first to third force plates 12-1 to 12-3) is included in the motion measuring section 1. FIG. The first force plate 12-1 corresponds to an example of a right foot load acquisition section that acquires the right foot load. The second force plate 12-2 corresponds to an example of a left foot load acquisition section that acquires the left foot load. The first and second force plates 12-1 and 12-2 correspond to an example of a stepping motion acquisition unit that acquires the stepping motion. The shape measuring section 21, the timer section 22, and the work result processing section 34 constitute the work result acquisition section 2 that acquires the work result of the predetermined work.

The new knowledge and operation of the skill comparison device will now be described. FIG. 4 is a diagram for explaining the contents of the work and test pieces used in the work. FIG. 4A is a perspective view schematically showing the test piece before work, and FIG. 4B is a perspective view schematically showing the test piece after work. FIG. 5 is a flow chart showing the operation of the skill comparison device. FIG. 6 is a diagram for explaining a skill comparison display screen displayed on the skill comparison device.

In order to compare the skills involved in a given task of handling a given object performed by bodily movements in referees and subjects with a smaller number of items of higher importance, as follows: Using the skill comparison device D in the embodiment, a reference person is selected from among the plurality of workers by having a plurality of workers perform the work, and the principal component using the data obtained in the work A smaller number of items with high importance were selected by the analysis, and new findings were obtained.

More specifically, the work is a work of grinding or polishing the above-described mold using a tool grinder as the object (grinding work). This grinder work consists of two processes: a rough process in which the die is roughly cut and a finishing process in which the roughly cut surface is polished to a smooth surface. In order to improve the precision), it is usually important to perform the work in the rough process with high precision and high efficiency. For this reason, here, the worker (subject, reference person, subject person) is made to perform the work of the rough process in the grinder work on the test piece TP that looks like a mold, and each measurement is performed by the skill comparison device D. .

For example, as shown in FIG. 4A, this test piece TP is a 100×100×35 mm SS material (general structure) in which a groove TPb extending in one direction with a depth of 0.1 mm is machined in the center. rolled steel). For this test piece TP, the operation of the rough process is performed by forming a groove TPb, and one of the protrusions TPa and TPc having a height of 0.1 mm formed on both sides of the groove TPb, for example, a protrusion This is a work of uniformly grinding TPc by 0.1 mm with a handy grinder. After such a rough process, the test piece TP is formed with a ground plane portion TPd that is substantially flush with the bottom surface of the groove TPb, as shown in FIG. 4B. There are 46 workers, including beginners to experts.

Each of these 46 workers sequentially performs the work of the rough process on the test piece TP. During the implementation, the above-described skill comparison device D is used, and in FIG. is stored in the storage unit 7 in association with the worker name (worker ID) (S1-1), and the external force acting on the body of the worker is measured by the external force measurement unit 12 (12-1 to 12-3). Measured, associated with the operator name of the operator and stored in the storage unit 7 (S1-2), by the position processing unit 32, by the plurality of imaging units 11-1 to 11-k in processing S1-1 Each position of the plurality of markers is obtained based on each generated image, is associated with the worker name of the worker, and stored in the storage unit 7 (S2). Based on the positions of the plurality of markers obtained by the position processing unit 32, body movement information (the position of the center of gravity of each body part) and object movement information (the position of the center of gravity of the tool) of the worker are obtained. It is stored in the storage section 7 in association with the worker name of the worker (S3). The worker name is input from the input unit 4 before starting the work, and is stored in the storage unit 7 by the control unit 31, for example.

During the execution, the working time of the worker is measured by the timer unit 22 as one of the work results, and is stored in the storage unit 7 in association with the worker name of the worker (S1-3).

After completion of the work, the shape of the test piece TP after work is measured by the shape measuring unit 21, and stored in the storage unit 7 in association with the worker name of the worker (S1-4). The dimensional accuracy is obtained by the processing unit 34 based on the measured shape, and is stored in the storage unit 7 in association with the operator name of the operator (S4). It should be noted that each measurement (duration of motion analysis) of the worker's body motion and external force was set for 2 minutes from the start of grinding in the rough process.

Subsequently, a referrer is determined by the referrer determination processing unit 36 (S11). As described above, the referrer has excellent work results, so the workers are classified into a plurality of classes according to the work results, and the referrer is determined based on the work results. In this embodiment, five A-class to E-class are set. The A class is a class for beginners with poor work results, ie, the lowest level of proficiency. The B class is a class for intermediates whose work results are normal, that is, their proficiency level is normal (intermediate). Class C is a class of semi-advanced workers whose work results are semi-excellent, that is, their proficiency is semi-advanced. The D class is a class for advanced workers whose work results are excellent, that is, their proficiency is advanced. The E class is the class for the most advanced workers with the best work results, ie, the highest level of proficiency. In the rough process of the grinder work described above, the higher the proficiency of the skill, the higher the dimensional accuracy (processing accuracy), the higher the work efficiency, that is, the shorter the work time, and the less variation. be done. Therefore, in this embodiment, the standard deviation σ of each of the dimensional accuracy and the working time is obtained, and the range within about 3σ from the one with the highest dimensional accuracy and the shortest working time is set to the E class, exceeding about 3σ, A range within approximately 2.5σ is set to class D, a range exceeding approximately 2.5σ and within approximately 2σ is set to class C, and a range exceeding approximately 2σ and within approximately 1σ is set to class D. Thresholds for dimensions and work times (dimension thresholds, work time thresholds) for classifying each class were set so that a range exceeding approximately 1σ was set for A class. Specifically, Table 1 shows each dimension threshold [mm] and each work time threshold [min] for dividing each class.

Here, the dimensional accuracy of the test piece TP is measured using, for example, a plane including the bottom surface of the groove TPb as a reference plane and each grid point of a square grid virtually set on the grinding plane portion TPd as each measurement point. At each measurement point, the difference in the height direction with respect to the reference plane (height or depth from the reference plane) is obtained, and the average value of the absolute values of the differences at each measurement point is obtained. The average dimensional accuracy Aves is 0.032 [mm], the standard deviation σs is 0.022 [mm], and the average working time Avet is 14.4 [min]. The standard deviation σt was 8.0 [min].

In such a classification, 46 workers, 3 belong to A class, 11 belong to B class, 13 belong to C class, 14 belong to D class, 5 belong to E Divided as belonging to a class. Five people who belong to E class are sensory evaluations, but they are considered by other workers to have the best work results and the highest level of skill, and three people who belong to A class are newcomers. Therefore, the classification according to Table 1 is appropriate. In the above description, three classes, B class, C class and D class, are provided between A class and E class, but the number of classes between A class and E class may be arbitrary. , may be one, two, or four or more.

In this embodiment, the operator (user) presets the dimension threshold value of 0.03 [mm] and the work time threshold value of 10 [min] for discriminating class E, so that a worker of class E can be a reference person in advance. is set, and the referrer determination processing unit 36 selects a work whose dimensional accuracy is 0.03 [mm] or less from a plurality of subjects and whose work time is 10 [min] or less. determine (select, extract) a person as a referrer.

Subsequently, the principal component analysis is performed by the principal component analysis processing unit 37, and the predetermined items are determined (selected) (S12). More specifically, the principal component analysis processing unit 37 analyzes each displacement, each velocity, and each acceleration of the center of gravity of the plurality of body parts, the displacement, velocity, and acceleration of the center of gravity of the object, and the external force acting on the right leg. , the XYZ components of the external force acting on the left leg, and the XYZ components of the external force acting on the hand that grips the object as explanatory variables to express the motion of the body. Based on the third principal component, the displacement of the body part handling the predetermined object, the turning speed of the trunk, and the acceleration of the waist are obtained as the predetermined items, and the fourth to third principal component analysis are performed. Based on the six principal components, the left foot load, the stepping motion and the right foot load are obtained as the predetermined items. For each sampling timing, the displacement, velocity, and acceleration of the center of gravity at the sampling timing are obtained from the position of the center of gravity at the sampling timing, the position of the center of gravity at the sampling timing temporally adjacent to the sampling timing, and the sampling frequency. be done.

More specifically, the principal component analysis processing unit 37 first calculates the displacements, velocities, and accelerations of the centers of gravity of the plurality of body parts, the displacements, velocities, and accelerations of the centers of gravity of the objects, and the external force acting on the right leg. , the XYZ components of the external force acting on the left leg, and the XYZ components of the external force acting on the hand that grips the object are used as explanatory variables to represent the motion of the body. . That is, in this embodiment, there are 57 explanatory variables consisting of displacements, velocities and accelerations of the center of gravity of 19 body parts, 3 explanatory variables consisting of the displacement, velocity and acceleration of the center of gravity of the object, Three explanatory variables consisting of the XYZ components of the external force acting on the right foot (measurement results of the first force plate 12-1), the XYZ components of the external force acting on the left foot (measurement results of the second force plate 12-2 result) and three explanatory variables consisting of the XYZ components of the external force acting on the hand that grips the object (measurement result of the third force plate 12-3), a total of 69 In a 69-dimensional space consisting of explanatory variables, each displacement, each velocity and each acceleration of the center of gravity in 19 body parts obtained from the 46 workers, the displacement, velocity and acceleration of the center of gravity in the object, Principal component analysis of the correlation matrix using each data of the XYZ components of the external force acting on the right foot, the XYZ components of the external force acting on the left foot, and the XYZ components of the external force acting on the hand holding the object is executed. Some of its eigenvectors are shown in Table 2. Table 2 shows the six first to sixth principal components PC1 to PC6 for three persons, Mr. α, Mr. β, and Mr. γ, who belong to the E class. The cumulative contribution rate of the first to third principal components is fifty-odd percent, and the three first to third principal components can cover 50% or more of the data in the 69-dimensional space (description can). The cumulative contribution rate of the first to sixth principal components is more than 70%, and the six first to sixth principal components can cover 70% or more of the data in the 69-dimensional space (description can).

Then, not individual referees, but five referees belonging to class E (Mr. α, Mr. β, Mr. γ, Mr. δ, Mr. ε) characterize the skills related to the predetermined work that handles the predetermined object. For each 1st to 6th principal component, each eigenvector in the five referees belonging to the E class was synthesized. More specifically, the principal component analysis processing unit 37 analyzes the first principal component PC1 of Mr. α, the first principal component PC1 of Mr. β, the first principal component PC1 of Mr. γ, the first principal component PC1 of Mr. γ, and Mr. δ and the first principal component PC1 of Mr. .epsilon., five explanatory variables are selected according to the five persons in descending order. The principal component analysis processing unit 37 calculates the second principal component PC2 of Mr. α, the second principal component PC2 of Mr. β, the second principal component PC2 of Mr. γ, and the second principal component PC2 of Mr. δ in the second principal component PC2. , and the second principal component PC2 of Mr. .epsilon. The principal component analysis processing unit 37 similarly selects five explanatory variables for each of the third to sixth principal components PC3 to PC6. Table 3 shows the result of selection made in this manner. In this Table 3, each column shows the selection results (explanatory variables) selected from the first to sixth principal components PC1 to PC6 in order from left to right on the page. For example, for the first principal component PC1, "displacement_neck_XYZ, displacement_neck_XYZ, displacement_upper right arm_XYZ, displacement_upper right arm_XYZ, displacement_right hand_XYZ" is the selection result. Here, "displacement_neck_XYZ" (displacement_Neck_XYZ) means the displacement of the neck in the XYZ orthogonal coordinate system, and "displacement_right upper arm_XYZ" (displacement_RUAm_XYZ) means the displacement of the right upper arm in the XYZ orthogonal coordinate system. and "displacement_right hand_XYZ" (displacement_RHand_XYZ) means the displacement of the right hand in the XYZ orthogonal coordinate system. “Acceleration_grinder_XYZ” means the acceleration of the tool in the XYZ orthogonal coordinate system, and “floor reaction force_right foot_x” means the x component of the external force acting on the right foot in the XYZ orthogonal coordinate system. Others mean the same. In this way, the principal component analysis processing unit 37 selects the constituent elements that form each of the six first to sixth items.

Here, in the present embodiment, "displacement_neck_XYZ, displacement_neck_XYZ, displacement_right upper arm_XYZ, displacement_right upper arm_XYZ, displacement_right hand_XYZ" selected from the first principal component PC1 is "predetermined Displacement of the body part that handles things” was given as the item name. "Speed_Right shoulder_XYZ, Speed_Neck_XYZ, Speed_Right arm_XYZ, Speed_Left shoulder_XYZ, Acceleration_Grinder_XYZ" selected from the second principal component PC2 is defined as "rotation speed of trunk". , was used as the item name. Similarly, for each of the third to sixth principal components PC3 to PC6, each selection result is defined as "lumbar acceleration", "left leg load", "stepping motion", and "right leg load". Item name.

Subsequently, a reference value is obtained by the target value processing unit 35 and stored in the reference information storage unit 71 (S13). More specifically, for each of the first to sixth principal components (first to sixth items), the target value processing unit 35 calculates the eigenvector , by multiplying the average value of each measured value for five referees belonging to class E (hereinafter abbreviated as "E class measured mean value" in this paragraph) corresponding to the component, the main Component scores are obtained as reference values. For example, the first principal component score, that is, in this embodiment, the reference value for the displacement of the body part that handles a given object (the first reference value of the first item) is (displacement_neck_eigenvector of XYZ)×(said E class measured average value at displacement_neck_XYZ), (displacement_neck_XYZ eigenvector) × (E class measured average value at the displacement_neck_XYZ), (displacement_right upper arm_XYZ eigenvector) × (Eigenvector of displacement_upper right arm_XYZ) × (measured average of E class at displacement_upper right arm_XYZ), and ( Eigenvector of Displacement_Right_XYZ)×(Measured mean value of E class at Displacement_Right_XYZ). Further, for example, the second principal component score, that is, in this embodiment, the reference value for the turning speed of the trunk (the second reference value of the second item) is (speed_right shoulder_eigenvector of XYZ)×(said speed_ E class measured average value at right shoulder_XYZ), (Eigenvector of speed_neck_XYZ) × (E class measured average value at the speed_neck_XYZ), (speed_right arm_XYZ eigenvector) × (E class measured mean value at Velocity_Right Arm_XYZ), (Eigenvector of Velocity_Left Shoulder_XYZ) x (E class measured mean value at Velocity_Left Shoulder_XYZ), and (Acceleration_Grinder eigenvector of _XYZ)×(measured average value of E class at the acceleration_grinder_XYZ). The third to sixth reference values of the third to sixth principal component scores are similarly obtained.

Subsequently, the target value is obtained by the target value processing unit 35 and stored in the storage unit 71 (S5). More specifically, for each of the first to sixth principal components (first to sixth items), the target value processing unit 35 calculates the eigenvector is multiplied by the actual measured value of the subject corresponding to the component to obtain the principal component score as a reference value. For example, the first principal component score, that is, in this embodiment, the target value of the displacement of the body part that handles a predetermined object (the first target value of the first item) is (displacement_neck_eigenvector of XYZ)×(said Subject's measured value at displacement_neck_XYZ), (displacement_neck_XYZ eigenvector) × (subject's measured value at said displacement_neck_XYZ), (displacement_right upper arm_XYZ eigenvector) × ( Subject's measured value at said Displacement_Right Arm_XYZ), (Eigenvector of Displacement_Right Arm_XYZ)×(Subject's measured value at said Displacement_Right Arm_XYZ), and (Displacement_Right Hand_XYZ) eigenvector)×(displacement_right hand_measured value of the subject at XYZ). Further, for example, the second principal component score, that is, in the present embodiment, the target value of the turning speed of the trunk (second target value of the second item) is (speed_right shoulder_eigenvector of XYZ)×(said speed_ Subject's measured value at right shoulder_XYZ), (velocity_neck_XYZ eigenvector) x (subject's actual measured value at said velocity_neck_XYZ), (velocity_right arm_XYZ eigenvector) x (said Subject's measured value at Velocity_Right Arm_XYZ), (Eigenvector of Velocity_Left Shoulder_XYZ) x (Subject's measured value at Velocity_Left Shoulder_XYZ), and (Eigenvector of Acceleration_Grinder_XYZ) It is the sum of x (measured value of the subject at the acceleration_grinder_XYZ). The third to sixth target values of the third to sixth principal component scores are similarly obtained.

Then, the reference value of the referrer stored in the reference information storage unit 71 and the target value of the target person obtained by the target value processing unit 35 are output to the output unit 5 by the control unit 31, and if necessary, the IF The data is output to the unit 6 (S6), and the process ends. When the output unit 5 is a display device, a skill comparison display screen is displayed on the output unit 5 for comparing the skills related to the work between the reference person and the target person by executing this process S6.

For example, as shown in FIG. 6, this skill comparison display screen includes a scale SC and bar graphs a-1 to f-1; a- 2 to f-2. More specifically, below the scale SC, a bar graph a-1 of the first reference value and a bar graph a-2 of the first target value in the displacement of the body part handling the predetermined object are juxtaposed with each other. The bar graph b-1 of the second reference value and the bar graph b-2 of the second object value for the turning speed of the trunk, and the bar graph c-1 and the bar graph c-1 of the third reference value of the acceleration of the waist are juxtaposed with each other. A bar graph c-2 of three target values, a bar graph d-1 and a bar graph d-2 of the fourth reference value in the left foot load juxtaposed with each other, and a fifth reference in the stepping motion juxtaposed with each other. A value bar graph e-1 and a fifth target value bar graph e-2, and a sixth reference value bar graph f-1 and a sixth target value bar graph f-2 in the right leg load, juxtaposed with each other, are further juxtaposed. It is In the example shown in FIG. 6, by referring to this skill comparison display screen, the subject can see that the first target value is close to the first reference value in the skill of the work, but the other second to sixth target values If there is a large difference between each of the second to sixth reference values, it can be recognized, and the second to sixth items can be recognized as improvement items to be improved.

As described above, the skill comparison device D and the skill comparison method implemented therein according to the present embodiment provide reference values and Since each target value is output, each skill of the reference person and the target person can be compared with a smaller number of items having high importance based on the new knowledge described above.

The skill comparison device D and the skill comparison method further output each reference value and each target value for each of the left foot load, the stepping motion, and the right foot load. A small number of items allows a more multifaceted comparison of each skill in the referrer and the target.

In the skill comparison device D and the skill comparison method, the work result of the referee is better than the work result of the target person, so the target person can improve his/her skill by being a better referee than himself/herself. can compare the reference value and the self target value and recognize the difference. The comparison can be performed on an item-by-item basis so that the subject can recognize items for improvement.

Since the skill comparison device D and the skill comparison method use optical motion capture, it is possible to measure the body movements of the subject in a non-contact manner without interfering with the body movements.

Since the work is grinding or polishing work, the skill comparison device D and the skill comparison method described above can compare the skills related to the work of grinding or polishing between the reference person and the target person.

In the skill comparison device D and the skill comparison method, the displacement of the body part handling a predetermined object, the turning speed of the trunk, the acceleration of the waist, the load on the left leg, the stepping motion, and the load on the right leg are the first to sixth principal components of the principal component analysis. Items can be appropriately selected in order of importance since they are determined based on the ingredients.

According to this embodiment, it is possible to provide a skill comparison device D and a skill comparison method, each of which uses the accuracy of the work product and the time required for the work as the work result.

In the above-described embodiment, determination (selection) of the referrer, determination of items by principal component analysis, and calculation of the reference value of the referrer were performed in accordance with the calculation of the target value of the subject. In addition to the calculation of the target value, the determination of the referrer, the determination of the item by principal component analysis, and the calculation of the reference value of the referrer are performed in advance by this device D, another device D, or manual calculation, and the reference value is It may be stored in advance in the reference information storage unit 71 , and the calculation algorithm for the target value (for example, the calculation method described in the above process S<b>5 ) may be stored in advance in the storage unit 7 . In this case, processing S1-3, processing S-4, processing S4, processing S11, processing S12 and processing S13 shown in FIG. 5 are omitted, and each of processing S1-1, processing S2, processing S3 and processing S1-2 Following execution of the process, each process of process S5 and process S6 is executed. Accordingly, the work result acquisition unit 2, the work result processing unit 34, the referrer determination processing unit 36, and the principal component analysis processing unit 37 can also be omitted. That is, the skill comparison device D includes a control processing unit 3 including a motion measurement unit 1, a control unit 31, a position processing unit 32, a motion information processing unit 33, and a target value processing unit 35, an output unit 5, and a storage unit 7. May be configured.

In the above-described embodiment, the referrer was a worker belonging to class E, but the present invention is not limited to this. good. For example, the referrer may belong to a higher class than the class to which the subject belongs. More specifically, for example, if the subject belongs to A class, the referrer may belong to any of B class, C class, D class and E class. If belonging to the C class, the referrer may belong to the D or E class. If the referrer belongs to a class that is one class higher than the class to which the target belongs, the difference between the target value and the reference value becomes smaller, making it easier to reach the reference value and motivating the target to improve their skills. can be granted.

Further, in the above-described embodiment, the skill comparison device D outputs the first to sixth reference values and the first to sixth target values to the output unit 5 in the process S6. and each data such as body movement information, object movement information, dimensional accuracy, work time, etc. may be output to the output unit 5 for each subject.

Although the present invention has been adequately and fully described above through embodiments with reference to the drawings in order to express the present invention, modifications and/or improvements to the above-described embodiments can be easily made by those skilled in the art. It should be recognized that it is possible. Therefore, to the extent that modifications or improvements made by those skilled in the art do not depart from the scope of the claims set forth in the claims, such modifications or improvements do not fall within the scope of the claims. is interpreted to be subsumed by

D skill comparison device 1 motion measurement unit 2 work result acquisition unit 3 control processing unit 4 input unit 5 output unit 6 interface unit (IF unit)
7 storage unit 31 control unit 32 position processing unit 33 motion information processing unit 34 work result processing unit 35 target value processing unit 36 referrer determination processing unit 37 principal component analysis processing unit

Claims (16)

A skill comparison device that compares the skills of a reference person and a target person, which are performed by body movements and are related to a predetermined task of handling a predetermined object,
a reference information storage unit that stores a reference value related to a predetermined item for the referrer;
a motion measuring unit that measures the body motion of the subject;
a target value processing unit that obtains a target value related to the item for the subject based on the body motion of the subject measured by the motion measurement unit;
An output unit that outputs the reference value of the referrer stored in the reference information storage unit and the target value of the target person obtained by the target value processing unit,
The items include, among a plurality of body parts obtained by dividing the body into a plurality of parts, the displacement of the body part that handles the predetermined object, the turning speed of the trunk of the plurality of body parts, and the plurality of body parts. is the acceleration of the waist in
The items further include the left foot load applied to the left foot, the stepping action of shifting the weight from the left foot to the right foot, and the right foot load applied to the right foot,
a left leg load acquisition unit that acquires the left leg load ;
a stepping motion acquisition unit that acquires the stepping motion ;
a right leg load acquisition unit that acquires the right leg load ,
The displacement of the body part handling the predetermined object, the turning speed of the trunk, and the acceleration of the waist are each displacement, each velocity, and each acceleration of the center of gravity of the plurality of body parts, and the displacement of the center of gravity of the object. , velocity and acceleration, the XYZ components of the external force acting on the right leg, the XYZ components of the external force acting on the left leg, and the XYZ components of the external force acting on the hand that grips the object are used as explanatory variables. obtained based on the first to third principal components of principal component analysis in the case of representing motion ,
the reference value of the referrer is a principal component score of the referrer ;
wherein the target value of the subject is a principal component score of the subject ;
Skill comparison device. The plurality of workers are divided into a plurality of classes based on the work result of having the plurality of workers perform the predetermined work, and the referrer is selected from the workers classified into the class with the best work result. Further comprising a referrer determination processing unit to
A skill comparison device according to claim 1. The work result is the accuracy of the work product and the time required for the work,
3. A skill comparison device according to claim 2 . the motion measurement unit comprises optical motion capture;
A skill comparison device according to any one of claims 1 to 3 . The work is any one of a grinding or polishing work, a welding work and a painting work.
A skill comparison device according to any one of claims 1 to 4 . The work is a mold grinding or polishing work,
A skill comparison device according to any one of claims 1 to 4 . The mold is a vehicle member molding mold for molding a member used in a vehicle,
7. A skill comparison device according to claim 6 . the object is grasped by a hand of the plurality of body parts;
The plurality of body parts include the head, neck, chest, right shoulder, left shoulder, right upper arm, left left arm, right forearm, left forearm, right hand, left hand, waist, buttocks, and right intestine. bone, left ilium, right knee, left knee , right foot and left foot,
A skill comparison device according to any one of claims 1 to 7 . A skill comparison method for comparing the skills of a referrer and a target person, which are performed by bodily movements and are related to a predetermined task of handling a predetermined object,
a reference information storage step of storing a reference value related to a predetermined item for the referrer in a reference information storage unit;
a motion measurement step of measuring a body motion of the subject;
a target value processing step of obtaining a target value related to the item of the subject based on the body motion of the subject measured in the motion measurement step;
An output step of outputting the reference value of the referrer stored in the reference information storage unit and the target value of the target person obtained in the target value processing step,
The items include, among a plurality of body parts obtained by dividing the body into a plurality of parts, the displacement of the body part that handles the predetermined object, the turning speed of the trunk of the plurality of body parts, and the plurality of body parts. is the acceleration of the waist in
The items further include the left foot load applied to the left foot, the stepping action of shifting the weight from the left foot to the right foot, and the right foot load applied to the right foot,
Further comprising a left foot load obtaining step of obtaining the left foot load, a stepping motion obtaining step of obtaining the stepping motion, and a right foot load obtaining step of obtaining the right foot load, which are performed together with the motion measuring step;
The displacement of the body part handling the predetermined object, the turning speed of the trunk, and the acceleration of the waist are each displacement, each velocity, and each acceleration of the center of gravity of the plurality of body parts, and the displacement of the center of gravity of the object. , velocity and acceleration, the XYZ components of the external force acting on the right leg, the XYZ components of the external force acting on the left leg, and the XYZ components of the external force acting on the hand that grips the object are used as explanatory variables. determined based on the first to third principal components of principal component analysis in the case of representing motion ,
the reference value of the referrer is a principal component score of the referrer ;
wherein the target value of the subject is a principal component score of the subject ;
Skill comparison method. The plurality of workers are divided into a plurality of classes based on the work result of having the plurality of workers perform the predetermined work, and the referrer is selected from the workers classified into the class with the best work result. Further comprising a referrer determination processing step to
Skill comparison method according to claim 9 . The work result is the accuracy of the work product and the time required for the work,
Skill comparison method according to claim 10 . The motion measurement step measures motion of the body using optical motion capture.
A skill comparison method according to any one of claims 9 to 11 . The work is any one of a grinding or polishing work, a welding work and a painting work.
A skill comparison method according to any one of claims 9 to 12 . The work is a mold grinding or polishing work,
A skill comparison method according to any one of claims 9 to 12 . The mold is a vehicle member molding mold for molding a member used in a vehicle,
15. A skill comparison method according to claim 14 . the object is grasped by a hand of the plurality of body parts;
The plurality of body parts include the head, neck, chest, right shoulder, left shoulder, right upper arm, left left arm, right forearm, left forearm, right hand, left hand, waist, buttocks, and right intestine. bone, left ilium, right knee, left knee , right foot and left foot,
A skill comparison method according to any one of claims 9 to 15 .