JP7122090B2 - Skilled motion teaching system - Google Patents

Description

TECHNICAL FIELD The present invention relates to a skilled motion teaching system, and for example, to a system for teaching unskilled workers the motions of a skilled worker in civil engineering work or construction work.

Among civil engineering work and construction work, there is work in which the quality of completion is completely different between a skilled worker and an unskilled worker. For example, plastering requires a high degree of finish that only skilled workers can perform, and it took a long time for unskilled workers to be able to perform the same work as skilled workers. There are various types of work that can only be performed by skilled workers in other fields as well. It is not easy to teach an inexperienced person the motions of an expert, and it is also true that there are so-called tricks that only an expert can understand. Therefore, various techniques have been proposed for teaching an inexperienced person the motions of an expert (see, for example, Patent Documents 1 to 4).

In Patent Document 1, users are allowed to experience handshakes with celebrities, athletes, etc., or experience the actions of experts who have special skills such as calligraphy, sports, and operation of machines and vehicles. Techniques for doing so are described. This experiential movement restoration device conforms to the structure of a hand-shaped device, and a board with a hinge, a movable hinge, and a bending mechanism, actuators, and links are attached inside the device that has a shape like a bodysuit. Each part of the body is appropriately bent or moved according to data sent from the first terminal device and the second terminal device via the network. By doing so, the experiencer can experience the motions of the experience provider by wearing the device.

Patent Literature 2 describes a sensory-type training system for experiencing the work skill know-how of an on-site skilled worker. This sensation application device is, for example, a pressure application device built into work gloves or work shoes, and based on the data acquired by the sensation acquisition device, the resistance or resistance from the skill acquisition target, the skill acquisition tool, and the ground is generated. are coming to join. Further, the computer for the training simulator reproduces the work scene image, the work feeling of the skilled worker and the physical quantity generation on the terminal display means for the trainee, the work feeling applying means and the physical quantity generating means. As a result, it is possible to learn and imitate the skilled work of a skilled worker by experience, and to acquire the work skill know-how of the on-site skilled worker.

Patent Literature 3 describes a technique for enabling a competition coach to provide real-time support information to a competitor during a competition. In this athletic performance improvement support method, the athlete is made to wear a device for stimulating the skin. Then, an arbitrary skin sensation signal is generated from the right output section as meaning "move right", and an arbitrary skin sensation signal is generated from the left output section as meaning "move left". and generate an arbitrary skin sensory signal from the output part of the front part as meaning "move forward", and generate an arbitrary skin sensory signal from the output part of the rear part as meaning "move backward". I am trying to let

Patent Document 4 describes a technique for allowing a coating robot to easily reproduce the movement of a hand-blown coating gun when a skilled technician uses the hand-blown coating gun to paint an object to be coated. there is This coating robot teaching device comprises an acceleration sensor attached to the hand-blown coating gun for detecting acceleration acting on the hand-blown coating gun in time series, and storage means for storing the acceleration detected by the acceleration sensor. and generating means for generating position control data for causing the painting robot 20 to reproduce the movement of the hand-blown painting gun 10 based on the acceleration stored in the storage means.

JP-A-2002-123165 JP 2004-240264 A JP 2008-48972 A JP 2011-45951 A

However, the conventional techniques described above do not take into consideration the difference in attributes between the expert and the learner, such as gender, age, physique, muscle strength, height, grip strength, arm length, leg length, and the like. That is, it is sufficient if the attributes of the expert and the learner are the same, but if the attributes of the two are different, it may not be possible to teach an accurate motion.

In addition, for example, when the learner is instructed to perform the movements of an expert by wearing a harness, if the attributes of the expert and the learner are different, the learner will not be able to perform unreasonable movements (movements outside the movable range of joints, muscle strength, etc.). There is a risk that the learner will be injured by being forced to lift the above heavy objects.

The present invention has been proposed in view of the above-described circumstances, and considers the difference in attributes such as gender, age, physique, muscle strength, etc. It is an object of the present invention to provide a skilled movement teaching system capable of teaching.

Moreover, it is preferable that the device for teaching the learner the actions of the expert is as small as possible in order to reduce the burden on the learner.

The present invention has been proposed in view of the above-described circumstances, and the attributes (gender, age, physique, muscle strength, height, grip strength, arm length, leg length, etc.) between experts and learners To provide a skilled motion teaching system capable of appropriately teaching a skilled motion without imposing an excessive burden on a learner in consideration of differences.

A skilled motion teaching system according to the present invention has the following features in order to achieve the above-described object. That is, the skilled movement teaching system according to the present invention includes expert attribute data acquisition means, motion data acquisition means, motion data storage means, learner attribute data acquisition means, motion data correction means, and motion reproduction means. characterized by comprising

The expert attribute data acquisition means is means for acquiring data relating to expert attributes. The motion data acquisition means is means for acquiring the motion data of the expert. The action data storage means is a means for storing the acquired action data in association with data relating to the attributes of the expert. The learner attribute data acquisition means is means for acquiring data relating to learner attributes. The action data correcting means is means for correcting the stored action data in accordance with the attribute of the learner based on the difference between the attribute of the expert and the attribute of the learner. The action reproduction means is means for causing the learner to reproduce the action of the expert based on the corrected action data. Based on the corrected motion data, the motion reproducing means permits or restricts the learner's motion so as not to exceed the range of the expert's motion data . In addition, it is possible to make a learner into a robot.

Further, in the skilled movement teaching system configured as described above, the movement data acquisition means can acquire movement data from a plurality of experts each having different attribute data. In this case, the motion data correcting means performs correction by selecting motion data matching or approximating the attribute data of the learner from among the plurality of motion data.

Further, in the skilled motion teaching system having the configuration described above, the motion data acquired by the motion data acquiring means can be created by combining three-dimensional data and time data relating to the motion of the expert.

Further, in the skilled motion teaching system having the above configuration, the motion data acquired by the motion data acquiring means can be myoelectric information when the expert moves. In this case, the motion reproducing means directly inputs motion data to the muscles of the learner to reproduce the motion of the expert.

According to the skilled movement teaching system of the present invention, the difference in attributes (sex, age, physique, muscle strength, height, grip strength, arm length, leg length, etc.) between the expert and the learner is considered. Then, by correcting the motion data of the expert so as to match the attribute of the learner, it is possible to teach the learner an appropriate skilled motion.

1 is an explanatory diagram showing a schematic configuration of a skilled motion teaching system according to an embodiment of the present invention; FIG. 1 is a block diagram showing a schematic configuration of a skilled motion teaching system according to an embodiment of the present invention; FIG. FIG. 4 is an explanatory diagram showing the configuration of skilled movement data used in the skilled movement teaching system according to the embodiment of the present invention;

Hereinafter, a skilled motion teaching system according to an embodiment of the present invention will be described with reference to the drawings. 1 to 3 illustrate a skilled motion teaching system according to an embodiment of the present invention. FIG. 1 is an explanatory diagram showing the schematic configuration of the skilled motion teaching system, and FIG. 2 shows the schematic configuration of the skilled motion teaching system. FIG. 3 is a block diagram, and FIG. 3 is an explanatory diagram showing the configuration of skillful motion data.

<Outline of Skilled Motion Teaching System>
The skilled motion teaching system according to the embodiment of the present invention is suitable for, for example, civil engineering work (power shovel, bulldozer, crane operation), construction work (wall painting, planing, sawing), and other work (massage, cooking, performance, etc.), it is a system for teaching a learner the actions of an expert. As shown in FIG. 2, this skilled motion teaching system includes expert attribute data acquisition means 10, motion data acquisition means 20, motion data storage means 30, learner attribute data acquisition means 40, and motion data correction means. 50 and motion reproduction means 60 .

As shown in FIG. 1, the skilled motion teaching system according to the embodiment of the present invention includes motion data of a skilled person, sex, age, physique, muscle strength, height, grip strength, arm length, leg length, and the like. Attribute information is acquired and stored as skillful motion data. Also, attribute information such as the learner's sex, age, physique, muscle strength, height, grip strength, arm length, leg length, etc. is acquired, and a learner attribute data table is created and stored.

Then, when teaching the learner the motions of the expert, the stored skillful motion data is corrected according to the attributes of the learner based on the difference between the attributes of the expert and the attributes of the learner. By providing the skillful motion data as reproduction data to the learner, the learner is taught the motion of the expert.

<Expert attribute data acquisition means>
The expert attribute data acquisition means 10 is means for acquiring data relating to the attributes of an expert, and can be composed of a personal computer, software installed therein, and various devices attached to the personal computer.

For example, it acquires data on the attributes of an expert input from input means such as a keyboard, mouse, touch panel, etc. connected to a personal computer. Alternatively, data relating to the attributes of the expert may be acquired by photographing the movements of the expert, recognizing the photographed image, and extracting feature points. or the like, and acquire data on the attributes of the expert based on sensing data from the measuring device.

The acquired data on the expert's attribute is stored as expert's attribute data in an HDD or the like attached to the personal computer. At this time, the expert attribute data is created by associating the ID for identifying the expert with the data on the attribute of the expert. The expert attribute data may be part of the skilled action data stored in the action data storage means 30, or may be stored in association with the expert action data.

The attributes of the expert are sex, age, physique, muscle strength, height, grip strength, arm length, leg length, etc., and various physical attributes that affect the actions of the expert. Factors included. The same applies to learner attributes.

<Operation data acquisition means>
The motion data acquiring means 20 is means for acquiring the motion data of an expert, and can be composed of a personal computer, software installed therein, and various devices attached to the personal computer.

For example, an expert wears measuring instruments such as acceleration sensors, strain gauges, pressure gauges, etc. on appropriate parts of the body (near joints, head, waist, fingertips, toes, etc.), and based on the sensing data from the measuring instruments Operation data of the expert can be acquired. When various measuring instruments are used as components of the motion data acquiring means 20 in this way, the functions of the expert attribute data acquiring means 10 can also be provided. Further, the motion data may be myoelectric information when a skilled person moves.

The acquired action data of the expert is stored in an HDD or the like attached to the personal computer as skillful action data associated with information on the attributes of the expert. At this time, the motion data of the expert and the attribute data of the expert may be included in the same file, i.e., the motion data file of the expert, or the motion data of the expert and the attribute data of the expert may be stored as separate files. may be associated and stored.

The motion data acquisition means 20 is preferably configured to acquire motion data from a plurality of experts each having different attribute data. In this way, by acquiring motion data from a plurality of experts with different gender, age, physique, muscle strength, height, grip strength, arm length, leg length, etc., learners with diverse attributes can acquire motion data. On the other hand, more approximate motion data can be obtained.

The motion data acquired by the motion data acquiring means 20 is preferably created by combining three-dimensional data and time data relating to the motion of the expert. That is, by acquiring the three-dimensional motions of an expert in time series, even complicated motions can be accurately reproduced.

<Operation data storage means>
The action data storage means 30 is a means for storing the action data acquired by the action data acquisition means 20 in association with data relating to the attributes of the expert, and is constituted by an HDD or the like attached to a personal computer. Storage means such as an HDD attached to the personal computer also functions as a storage means for storing other data.

FIG. 3 shows an example of skillful motion data stored in the motion data storage means 30. As shown in FIG. As shown in FIG. 3, the skilled motion data includes a motion type indicating what kind of motion is being performed, a skilled worker ID for identifying the skilled worker, attribute data of the skilled worker, and actually acquired motion data. It consists of the file name of the actual operation data file. The data included in the skill action data is not limited to that shown in FIG. 3, and may include only necessary data or may include other data.

The attribute data may be included in the skillful motion data file, or may be stored as a separate data file in association with the skillful motion data file. Further, the actual action data may be included in the skillful action data file, or may be stored as a separate data file in association with the skillful action data file.

<Learner attribute data acquisition means>
The learner attribute data acquisition means 40 is a means for acquiring data relating to learner attributes, and may have the same configuration as the expert attribute data acquisition means 10 . That is, the learner attribute data acquisition means 40 can be configured by a personal computer, software installed therein, and various devices attached to the personal computer.

For example, it acquires data on learner's attributes input from input means such as a keyboard, mouse, and touch panel connected to a personal computer. In addition, it is also possible to obtain data on the learner's attributes by photographing the learner's movements, recognizing the photographed image, and extracting feature points. A measuring instrument such as a measuring instrument may be worn, and data regarding attributes of the learner may be acquired based on sensing data from the measuring instrument.

The acquired data about the attribute of the learner is stored as learner attribute data in an HDD or the like attached to the personal computer. At this time, a learner attribute data file is created by associating the ID for identifying the learner with the data regarding the attribute of the learner.

<Operation data correction means>
The motion data correction means 50 is a means for correcting the stored motion data in accordance with the attribute of the learner based on the difference between the attribute of the expert and the attribute of the learner, and is a personal computer and installed therein. configured by the software.

Differences between the attributes of experts and learners include differences between men and women, differences in age, differences in physique, and differences in muscle strength, etc., and have an effect on the movements of experts and learners. It is the difference in terms of various physical factors that give. In general, men have a stronger physique than women and thus have stronger muscles, and younger people have stronger muscles and a wider range of joint motion than older people. It is also possible to directly compare the muscle strength and obtain the difference between the two.

When motion data is obtained from a plurality of experts with different attribute data, correction is performed by selecting motion data that matches or approximates the learner's attribute data from among the plurality of motion data. can be done.

In this way, by obtaining differences in various physical factors that affect the actions of the expert and the learner, and correcting the action data of the expert to be taught to the learner so as to be suitable for the learner, It is possible to teach the learner appropriate skillful motions, and to prevent the learner from being forced to perform unreasonable motions (eg, motions outside the movable range of joints, motions to lift a heavy object exceeding muscle strength, etc.).

<Action reproduction means>
The motion reproduction means 60 is a means for causing the learner to reproduce the motion of the expert based on the corrected motion data, and reproduces appropriate parts of the learner's body (near joints, head, waist, fingertips, toes, etc.). ) and consists of an auxiliary device and a current generator. That is, by attaching an auxiliary device having an actuator to an appropriate part of the learner's body and operating the actuator, the motion of the expert can be reproduced. Also, by attaching a current generator to the learner's skin and moving the learner's muscles in accordance with the intensity of the current generated from the current generator, it is possible to reproduce the movements of an expert.

The motion reproducing means 60 preferably permits or restricts the learner's motion based on the corrected motion data. For example, when the learner learns the motion of the expert's hand, the learner can move the hand without any restrictions within the range of the expert's motion data. On the other hand, when the movement of the learner's hand is about to exceed the range of the motion data of the expert, it is sufficient to prevent the learner from moving the hand any further. In this way, on the premise that the motion data of the expert is corrected, the learner's motion is permitted or restricted within the range of the motion of the expert, so that the learner follows the model motion and appropriately behavior can be reproduced.

<Application to robots>
The skilled motion teaching system according to the present invention can be applied to robots. In this case, the robot serves as the learner, the specifications of the robot serve as data relating to the attributes of the learner, and the actuators and the like for operating the arms of the robot serve as the action reproduction means 60 . Thus, by applying the skilled motion teaching system according to the present invention to a robot, it is possible to cause the robot to reproduce the motion of an expert.

10 expert attribute data acquisition means 20 motion data acquisition means 30 motion data storage means 40 learner attribute data acquisition means 50 motion data correction means 60 motion reproduction means

Claims (5)

an expert attribute data acquisition means for acquiring data on an expert attribute;
motion data acquisition means for acquiring motion data of the expert;
motion data storage means for storing the acquired motion data in association with data relating to the attributes of the expert;
a learner attribute data acquisition means for acquiring data on learner attributes;
motion data correction means for correcting the stored motion data in accordance with the attribute of the learner based on the difference between the attribute of the expert and the attribute of the learner;
a motion reproduction means for causing the learner to reproduce the motion of the expert based on the corrected motion data;
The motion reproduction means permits or restricts the learner's motion based on the corrected motion data so as not to exceed the range of the expert's motion data .
A skilled motion teaching system characterized by: The motion data acquisition means acquires motion data from a plurality of experts each having different attribute data,
The motion data correcting means performs correction by selecting motion data matching or approximating data relating to a learner's attribute from among a plurality of motion data.
2. The skilled motion teaching system according to claim 1 , wherein: The motion data acquired by the motion data acquisition means is created by combining three-dimensional data and time data relating to the motion of the expert.
3. The skilled motion teaching system according to claim 1 or 2 , characterized in that: The motion data acquired by the motion data acquisition means is myoelectric information when a skilled person moves,
The motion reproducing means directly inputs motion data into the muscles of the learner to reproduce the motion of the expert.
The skilled movement teaching system according to any one of claims 1 to 3 , characterized in that: the learner is a robot;
The skilled movement teaching system according to any one of claims 1 to 3 , characterized in that:

Images