Work detection system, device and method

Abstract

To support learning of skill of work in need of prescribed operation including movement of a hand in a state where an object is pressurized by a finger.SOLUTION: A sensor set is prepared including a pressure sensor for measuring pressure applied by a finger and an acceleration sensor for measuring acceleration of a hand for a first worker who performs work requiring predetermined action. A computer detects operation based on sensing data including pressure and acceleration measured by the sensor set. In order to evaluate the first worker, the computer compares first operation information, which is operation information obtained from the sensing data of the first worker, with second operation information with respect to the detected operation.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates generally to task detection, for example, to assisting in skill acquisition. [Background technology]

[0002] There is a shortage of workers trained in the joint work of extra-high voltage CV (Cross-linked polyethylene insulated vinyl sheath) cables (hereinafter referred to as cable joint work).One of the reasons for this is the increase in plans for underground power transmission work, etc., and the concentration of replacement periods for extra-high voltage CV cables nearing the end of their lifespan.

[0003] Cable joint work includes the scraping of the outer semiconductive layer (hereinafter simply referred to as "scraping work"). The scraping work involves scraping off a certain length of the outer semiconductive layer around the entire circumference of the cable tip to expose an insulating layer of specified dimensions. Once the cable has been scraped to the specified dimensions and installation conditions, it has a precision that is close to a perfect circle and a smooth surface without any horizontal scratches, which also contributes to improving the durability of the equipment. The scraping work involves repetitive actions, such as multiple scraping strokes using a cutting tool. The scraping is done in one direction with a width measured in millimeters.

[0004] Acquiring cable jointing skills, especially the scraping skills, requires considerable experience. For example, it takes a young person several years to acquire this skill.

[0005] For instructors (e.g., experienced workers), it is not easy to allocate time to skill training due to an increase in construction work (or other reasons). For trainees (e.g., young workers), it is difficult to acquire skills through only verbal instruction from an instructor.

[0006] Therefore, there is a need for a technology to support skill acquisition. Examples of this type of technology include those disclosed in Patent Documents 1 and 2.

[0007] According to Patent Document 1, the motion data of a first worker and the motion data of a second worker are compared to determine the similarity between the motion data, and data indicating areas for improvement are presented to the first worker according to the result of the determination.

[0008] According to Patent Document 2, a repetitive motion is extracted from sensor data, and the extracted repetitive motion is compared with other repetitive motions. Based on the result of the comparison, it is determined whether the extracted repetitive motion is an outlier. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Japanese Patent Application Publication No. 2019-20913 [Patent Document 2] US2021 / 0177307 Summary of the Invention [Problem to be solved by the invention]

[0010] According to Patent Documents 1 and 2, data from the sensors is operational data.

[0011] Therefore, one possible method for evaluating the movements of a trainee in a scraping task is to have an expert and a trainee wearing or using the sensors disclosed in Patent Document 1 or 2 perform the scraping task, and compare the data obtained from the expert's sensor with the data obtained from the trainee's sensor.

[0012] The sensors disclosed in Patent Document 1 are a wearable sensor worn on the arm or waist and a different type of pressure sensor, and the data is data from these sensors.The sensor disclosed in Patent Document 2 is a wearable glove motion capture device, and the data is motion capture data captured by the device.

[0013] However, even if such data is used as motion data for scraping work, it is difficult to help trainees acquire the skills of scraping work. This is because scraping work requires moving the hands while the fingers are pressed against the ultra-high voltage CV cable, which scrapes the workpiece to a width of millimeters or less. Such movements are difficult to adequately capture in the data described above. Furthermore, when performing tasks using such tools, if only the worker's movements are captured and analyzed, the relationship between the tools and the work data alone becomes unclear, making it difficult for trainees to acquire the skills of the work using motion data alone. Furthermore, while it is possible to evaluate the worker's work and the tools and the work together using a camera alone, it may not be possible to evaluate the work primarily using the camera because the work area and tools change position for each task, the camera cannot be placed around the work area, or there may be obstructions between the camera and the work area.

[0014] Such a problem may also arise in assisting the acquisition of skills for other types of work that require specific movements including moving the hand while the fingers are pressed against an object. [Means for solving the problem]

[0015] A sensor set including a pressure sensor for measuring the pressure applied by the fingers and an acceleration sensor for measuring the acceleration of the hand is provided for a first worker performing a task requiring a predetermined movement including movement of the hand while the fingers are pressed against an object. A computer detects the movement based on sensing data including the pressure and acceleration measured by the sensor set. For evaluation of the first worker, the computer compares first movement information, which is movement information obtained from the sensing data of the first worker, with second movement information regarding the detected movement. [Effects of the Invention]

[0016] According to the present invention, it is possible to assist in the acquisition of skills for a task that requires a predetermined movement including movement of the hand while the fingers are pressed against an object. [Brief explanation of the drawings]

[0017] [Figure 1A] An example of the site is shown before scraping work began. [Figure 1B] An example of a site where scraping work is being carried out is shown. [Figure 1C] An example of the site after scraping work was completed is shown below. [Figure 2A] This is an explanatory diagram of step 1 cutting. [Figure 2B] This is an explanatory diagram of step 2 cutting. [Figure 2C] This is an explanatory diagram of step 3 cutting. [Figure 3A] FIG. 2 is a rear view of the sensor glove. [Figure 3B] FIG. 2 is a front view of the sensor glove. [Figure 4] The overall configuration of the work detection system, including the mobile PC, is shown below. [Figure 5] 1 shows the configuration of detection target data. [Figure 6A] An example of expected acceleration time series data for an expert is shown below. [Figure 6B] An example of acceleration time series data expected for beginners is shown below. [Figure 7] An example of the evaluation method is shown below. [Figure 8] An example of an evaluation UI is shown below. DETAILED DESCRIPTION OF THE INVENTION

[0018] In the following description, an "interface apparatus" may refer to one or more interface devices. The one or more interface devices may be at least one of the following: An I / O interface device is one or more I / O (Input / Output) interface devices. The I / O (Input / Output) interface devices are interface devices for at least one of an I / O device and a remote display computer. The I / O interface device for the display computer may be a communications interface device. The at least one I / O device may be a user interface device, for example, either an input device such as a keyboard and a pointing device, or an output device such as a display device. A communication interface apparatus that is one or more communication interface devices. The one or more communication interface devices may be one or more homogeneous communication interface devices (e.g., one or more NICs (Network Interface Cards)) or two or more heterogeneous communication interface devices (e.g., an NIC and an HBA (Host Bus Adapter)).

[0019] In the following description, "memory" refers to one or more memory devices, which are an example of one or more storage devices, and may typically be a primary storage device. At least one memory device in the memory may be a volatile memory device or a non-volatile memory device.

[0020] In the following description, a "persistent storage device" may refer to one or more persistent storage devices, which are an example of one or more storage devices. A persistent storage device may typically be a non-volatile storage device (e.g., an auxiliary storage device), and specifically may be, for example, a hard disk drive (HDD), a solid state drive (SSD), a non-volatile memory express (NVME) drive, or a storage class memory (SCM).

[0021] In the following description, the term "storage device" may refer to at least one of memory and persistent storage device.

[0022] Furthermore, in the following description, a "processor" may refer to one or more processor devices. The at least one processor device may typically be a microprocessor device such as a CPU (Central Processing Unit), but may also be another type of processor device such as a GPU (Graphics Processing Unit). The at least one processor device may be a single-core or multi-core. The at least one processor device may also be a processor core. The at least one processor device may also be a processor device in a broader sense, such as a circuit that is a collection of gate arrays written in a hardware description language that performs some or all of the processing (for example, an FPGA (Field-Programmable Gate Array), a CPLD (Complex Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit)).

[0023] In the following description, information that provides an output in response to an input may be described using expressions such as "xxx database." However, this information may be data of any structure (for example, structured data or unstructured data), or may be a neural network that generates an output in response to an input, or a learning model such as a genetic algorithm or random forest. Therefore, the "xxx database" may be referred to as "xxx information." In the following description, the configuration of each database is an example, and one database may be divided into two or more databases, or all or part of two or more databases may be one database.

[0024] In the following description, functions are sometimes described using the expression "yyy unit." However, the functions may be realized by one or more computer programs executed by a processor, by one or more hardware circuits (e.g., FPGAs or ASICs), or by a combination thereof. When a function is realized by a program executed by a processor, the specified processing is performed using a storage device and / or an interface device, etc., as appropriate, and therefore the function may be considered to be at least a part of the processor. Processing described using a function as the subject may be processing performed by a processor or a device having the processor. A program may be installed from a program source. The program source may be, for example, a program distribution computer or a computer-readable storage medium (e.g., a non-transitory storage medium). The description of each function is merely an example; multiple functions may be combined into one function, or one function may be divided into multiple functions.

[0025] In the following description, when elements of the same type are described without distinction, common reference symbols are used, and when elements of the same type are described with distinction, reference symbols are used.

[0026] An embodiment will be described below.

[0027] As shown in FIGS. 1A to 1C, in this embodiment, the "task requiring a predetermined action including moving a hand while applying pressure to an object with a finger" is the task of scraping off an outer semiconductive layer from CV cable 115 to expose an insulating layer of a specified size from CV cable 115, and the "object" is CV cable 115. Note that FIG. 1A shows an example of the site before the scraping work begins. FIG. 1B shows an example of the site during the scraping work. FIG. 1C shows an example of the site after the scraping work has been completed.

[0028] Here, we will explain the relationship between workers and specified tasks. Depending on the specified task, there are parameters that evaluate the results of the task, such as good quality of the workpiece, short processing time, and low wear on tools used in processing. Workers who have good results in these parameters are said to be skilled or experienced in the specified task, and are accustomed to the specified task, and are called experts or instructors. Meanwhile, those who are less skilled or have less experience in the specified task than experts or instructors are called trainees or trainees. Trainees and trainees are simply called trainees and trainees.

[0029] In this embodiment, an example of the first worker is the trainee 110. An example of the second worker is an expert. One or more experts may be employed for the entire scraping work of the outer semiconductive layer. One or more experts may be employed for each of the tasks constituting the scraping work (specifically, step 1 scraping, step 2 scraping, and step 3 scraping, which will be described later).

[0030] The scraping operation of the outer semiconductive layer requires delicate operations, such as scraping with a width in the order of millimeters, and in order to ensure that such operations are properly reflected in the data, this embodiment employs a sensor glove 102 equipped with a sensor set. A mobile PC (Personal Computer) 150 is also provided, which receives data including values ​​measured by the sensor glove 102 and detects and evaluates the operations.

[0031] The sensor set includes multiple sensors, including a pressure sensor that measures pressure applied by a finger and an acceleration sensor that measures hand acceleration. As shown in FIG. 1B , when trainee 110 performs a scraping task while wearing such a sensor glove 102 on his / her hand, data including values ​​measured by sensor glove 102 (pressure, angular velocity, acceleration, etc.) is obtained. Based on the sensing data including such data, mobile PC 150 detects and evaluates the trainee's movement. Sensor glove 102 may be worn on only one hand (e.g., dominant hand) or on both hands. Alternatively, instead of sensor glove 102, each sensor in the sensor set may be worn directly on the hand (e.g., on the fingers or back of the hand) as an example of a wearable sensor.

[0032] The mobile PC 150 is an example of an information processing device. The mobile PC 150 includes a motion detection unit and a motion comparison unit. A general-purpose computer can be used as the mobile PC 150. If the information acquired by the sensor glove 102 requires a large amount of calculation, a computer using a GPU (Graphics Processing Unit) or FPGA, etc., may be used. Alternatively, the data acquired by the sensor glove 102 may be transmitted to a cloud service for data analysis. The motion detection unit detects motion based on sensing data including values ​​measured by the sensor glove 102. The motion comparison unit compares first motion information (motion information obtained from the sensing data of the trainee 110) with second motion information regarding the detected motion to evaluate the trainee 110, and performs evaluation based on the comparison result. The "motion information" will be described later. In the mobile PC 150, the motion detection unit detects motion in response to data from the sensor glove 102 (e.g., in real time), and the motion comparison unit determines whether or not the detected motion is abnormal, depending on the detected motion. If an abnormality is detected, an alert may be output. 1B, the alert is output as a display and sound output by the mobile PC 150, and as an LED light on the sensor glove 102. In particular, the sound alert output allows the worker to know whether the work is going well while continuing the work. Also, outputting different sounds depending on the quality of the worker's work is effective because it allows the worker to adjust the work they are doing to improve it.

[0033] In this embodiment, in addition to the data obtained by the sensor glove 102, data on the photographing operation by the camera 101 is input to the mobile PC 150. The photographing may be performed either stationary or mobile. If stationary photographing is employed, the camera 101 may be provided on a tower frame as shown in FIG. 1A. If mobile photographing is employed, the camera 101 may be provided on the trainee 110 (for example, a helmet worn by the trainee 110). Alternatively, a camera with an adjustable angle and magnification to change the angle of view, or a camera mounted on a rail or equipped with a self-propelled means, may be used. The angle of view of the captured video may be at least a portion of the trainee 110's field of view, typically a range that includes all or part of the trainee's 110's hand (for example, the fingers). Although video is employed in this embodiment, still images may be used instead of or in addition to the video.

[0034] The mobile PC 150 displays output information including information indicating the evaluation of the trainee 110. The output information may include at least one captured image. As shown in FIG. 1C, after the scraping work is completed (or after each of the multiple tasks constituting the scraping work is completed), the instructor 111 provides instruction to the trainee 110 while viewing the output information displayed on the mobile PC 150 together with the trainee 110. The output information may be displayed on a remote instructor's PC (instructor's PC) via a communication network (details will be described later).

[0035] The scraping work is made up of multiple operations: scraping in step 1, scraping in step 2, and scraping in step 3. The number of steps in the scraping work is explained as a representative example, and the number of steps may be appropriate for the tool and the object on which the scraping work is performed.

[0036] Step 1 (scraping) is the process of roughly scraping away outer semiconductive layer 202 from the entire circumference of the tip of the CV cable, as shown in FIG. 2A. This process involves repeatedly pressing cutting tool 201 against outer semiconductive layer 202 and moving it repeatedly a certain distance in one direction 203 indicated by an arrow. Each stroke of cutting tool 201 scrapes away a section of outer semiconductive layer 202 over a width of several millimeters. This process exposes insulating layer 204 over the entire tip of the CV cable, as shown in FIG. 2B. It is desirable that the hardness of cutting tool 201 be higher than the Vickers hardness of the material of outer semiconductive layer 202.

[0037] As shown in FIG. 2B, step 2 scraping is the process of scraping the base end of the distal end portion of the CV cable to a specified dimension (for example, several millimeters in the longitudinal direction) of outer semiconductive layer 202. Step 2 scraping is a process of scraping using finer strokes than step 1 scraping. In this process, a cutting tool 201 is pressed against the base end of the distal end portion of the CV cable while being repeatedly moved back and forth along arrow reciprocation direction 205. With each stroke of cutting tool 201 from the distal end side to the proximal end side, a width of several millimeters is scraped off of outer semiconductive layer 202.

[0038] 2C, step 3 cutting is performed on the surface of insulating layer 204. In this operation, cutting tool 201 is pressed against insulating layer 204 and repeatedly moved a certain distance in one direction 206 indicated by an arrow. Each stroke of cutting tool 201 cuts off a width of several millimeters of insulating layer 204.

[0039] The sensor glove 102 described above is employed to accurately detect the movements during the scraping work, which consists of the above-described step 1 scraping, step 2 scraping, and step 3 scraping. Because it is difficult to capture the angle, movement distance, and pressure of the worker's hand movements and the cutting tool 201 movements using a camera alone, using the sensor glove 102 is effective. Furthermore, the smoothness of the surface of the outer semiconductive layer 202 is affected by the quality of the stroke in which the worker presses the cutting tool 201 from the tip end to the base end to scrape away. Therefore, obtaining the pressure applied by the worker's hand to the cutting tool 201 and the inertia of the hand movement are important evaluation parameters when evaluating the worker's work. A detailed evaluation method will be described later.

[0040] As shown in FIGS. 3A and 3B, the sensor glove 102 includes a pressure sensor 302, a microphone 303, and a module device 301.

[0041] Pressure sensors 302 are provided on the fingertips of sensor glove 102 and measure pressure applied by the fingers. For example, pressure sensors 302A to 302C are provided on the fingertips of the thumb, index finger, and middle finger of sensor glove 102.

[0042] The microphones 303 are provided at predetermined positions. For example, microphone 303A is provided near the tip of the thumb, and microphone 303B is provided near the wrist. Microphones 303 collect sounds around the hand. For example, if an unusual sound is emitted from cutting tool 201 during scraping work, an alert can be sent to an experienced worker or a different flag can be set in the data as a checkpoint for the worker's work, making it possible to later confirm that there was an abnormality in the work operation.

[0043] The module device 301 is provided on the back of the sensor glove 102. The module device 301 includes an inertial sensor and a communication device. The inertial sensor includes at least an acceleration sensor. In this embodiment, the inertial sensors include a gyro sensor and a geomagnetic sensor in addition to the acceleration sensor. The acceleration sensor measures acceleration related to hand movement. The gyro sensor detects angular velocity related to hand movement. The gyro sensor detects the orientation of the hand (direction of hand movement). The communication device transmits data including data (values) measured by the pressure sensor 302, microphone 303, and each inertial sensor to an external device (in this embodiment, the mobile PC 150). Note that in this embodiment, the orientation of the inertial sensor includes a first direction from the fingertip to the wrist, a second direction from the little finger to the thumb that is perpendicular to the first direction, and a third direction from the palm to the back of the hand that is perpendicular to the first and second directions.

[0044] The mobile PC 150 detects and evaluates the movement based on the sensing data including the data received from the communication device of the sensor glove 102. In this embodiment, the sensor glove 102 and the mobile PC 150 communicate wirelessly, but wired communication may be used instead. The wireless communication may be any method such as radio waves, infrared rays, visible light, or sound waves, and the most suitable method can be selected depending on the work site.

[0045] FIG. 4 shows the overall configuration of the work detection system including the mobile PC 150.

[0046] The mobile PC 150 has a communication interface device (communication I / F) 401, an input / output interface device (I / OI / F) 402, an input device 403, a display device 404, a storage device 405, and a processor 406. The interface devices (communication I / F 401 and I / OI / F 402) and the storage device 405 are connected to the processor 406.

[0047] The communication I / F 401 communicates with devices external to the mobile PC 150, such as the camera 101, the sensor glove 102, and the external system 450. In the communication I / F 401, the communication interface device for communication with the camera 101, the communication interface device for communication with the sensor glove 102, and the communication interface device for communication with the external system 450 may be different, or a communication interface device may be common to two or more devices among the camera 101, the sensor glove 102, and the external system 450. The external system 450 may be a computer system external to the mobile PC 150, such as a remote server (e.g., a server as a cloud computing service) or the instructor PC described above.

[0048] The I / OI / F 402 mediates input and output to and from the processor 406 via the input device 403 and the display device 404 .

[0049] The input device 403 is a user interface device such as a keyboard or a pointing device. The display device 404 is an example of an output device and is a user interface device such as a liquid crystal display. The input device 403 and the display device 404 may be integrated into one device, such as a touch panel.

[0050] The storage device 405 stores information input and output by the processor 406 and computer programs executed by the processor 406. The information includes, for example, accumulated data 410, detection target data 411, evaluation method data 412, evaluation data 413, proficiency management data 414, trainee management data 415, action evaluation data 416, quality evaluation data 417, and integrated data 418.

[0051] When the processor 406 executes the computer program, functions such as a data receiving unit 420, a motion detecting unit 421, a motion comparing unit 422, a UI (User Interface) control unit 423, and an integration unit 424 are realized.

[0052] The data 410 to 418 and the functions 420 to 424 will be explained below.

[0053] The data receiving unit 420 receives captured video data from the camera 101 and stores the received captured video data in the storage device 405. The data receiving unit 420 also receives data from the sensor glove 102 and stores the received data in the storage device 405. Data including the stored captured video data from the camera 101 and the stored data from the sensor glove 102 is accumulated data 410. That is, the accumulated data 410 includes the captured video data from the camera 101 and sensing data including data from the sensor glove 102. The sensing data includes data of one or more values ​​(specifically, a time series of measured values) measured by each of the pressure sensor 302, microphone 303, acceleration sensor, gyro sensor, and geomagnetic sensor.

[0054] The motion detection unit 421 detects the motion of the trainee 110 based on the accumulated data 410. At least sensing data is used for motion detection. The motion to be detected is specified in the detection target data 411. As shown in FIG. 5, the detection target data 411 indicates, for each task constituting the scraping operation, the motion to be detected and a flag indicating whether the motion is subject to real-time detection. Some motions are subject to detection for two or more tasks constituting the scraping operation. For example, the motion “(8) Change in the contact angle of the cutting tool” is subject to detection for all of step 1 scraping, step 2 scraping, and step 3 scraping. Furthermore, among the motions shown in FIG. 5, examples of repetitive motions involving repeated hand movement are “(1) Constant speed change in scraping stroke,” “(4) Regular motion,” and “(7) Application of constant pressure.” Furthermore, motions for which the real-time detection flag is on in the detection target data 411 are motions for which the presence or absence of an abnormality is determined in response to data received from the sensor glove 102. The presence or absence of an abnormality in the detected motion is determined by the motion comparison unit 422. For example, the motion comparison unit 422 compares one or more predetermined measurement values ​​in the received data with thresholds for the various measurement values. If at least one measurement value is determined to be abnormal based on its relationship with the threshold, the motion comparison unit 422 may output an alert via the UI control unit 423. For example, the UI control unit 423 may display the alert on the display device 404, output the alert as audio from a speaker (an example of an output device) (not shown), or output the alert to the external system 450. As a modified example of real-time detection, a processor in the sensor glove 102 may perform simple data processing to detect abnormalities or signs of abnormalities in the motion of the worker or the motion of a tool, etc., and then emit a sound from a microphone separately provided in the sensor glove 102. In this case, the sensor glove 102 can be operated standalone without using a PC for analysis and evaluation.

[0055] The action comparison unit 422 compares the first action information with the second action information regarding the detected action, performs an evaluation based on the comparison result, and stores evaluation data 413 representing the evaluation in the storage device 405. In this embodiment, evaluation includes scoring.

[0056] Here, "motion information" refers to information obtained from sensing data. The motion information may depend on the type of detected motion, and may be, for example, a waveform as a time series of measurement values, a part of the waveform, or information including various values ​​obtained from the time series of measurement values ​​(e.g., the number of strokes).

[0057] The "first movement information" is information obtained from the sensing data of the trainee 110. The "second movement information" to be compared with the first movement information is information obtained from the sensing data of the skilled worker (e.g., information prepared based on information obtained from the sensing data of each of multiple skilled workers). For example, the second movement information may be prepared based on sensing data including data obtained from the sensor glove 102 of the skilled worker performing the scraping work. The second movement information may be information including a waveform (measurement value time series) to be compared, a portion of the waveform, or thresholds of various values ​​(e.g., the number of sensing operations). The second movement information may be identified, for example, from the evaluation method data 412. For example, for each task constituting the scraping work, the evaluation method data 412 includes second movement information and evaluation method information for each detection target movement belonging to the task. The evaluation method information indicates an evaluation method.

[0058] The motion detection method, first motion information, second motion information, and evaluation method information depend on the motion to be detected (and the task to which the motion belongs). As described above, some motions (e.g., motion "(8) Change in cutting tool contact angle") are detection targets for two or more tasks among the multiple tasks that make up the scraping operation, but even for the same motion, at least some of the motion detection method, first motion information, second motion information, and evaluation method information may differ depending on which task it belongs to. Furthermore, if different cutting tools are used by different workers or if the tip of the cutting tool has different characteristics (sharpness, taper, etc.) for each cutting tool, the type of cutting tool, the tip state of the cutting tool, and the angle of the cutting tool may be detected separately.

[0059] Below, examples of the first motion information, the second motion information, and the evaluation method will be described using the motion "(1) constant speed change of the cutting stroke" as an example.

[0060] When detecting the action of step 1 scraping in the scraping work, the action detection unit 421 identifies the action of the detection target corresponding to step 1 scraping from the detection target data 411. The action detection unit 421 detects the action "(1) a constant speed change in the scraping stroke" from the acceleration time-series data in the sensing data, among the actions of the identified detection target. Note that which part of the scraping work is being performed may be specified from the input device 403, may be specified from the sensor glove 102, or may be detected based on the sensing data. Also, as described above, the actions of the detection target may be narrowed down depending on which part of the scraping work is being performed.

[0061] The action comparison unit 422 divides the time-series data (waveform data) of the trainee 110's acceleration into multiple partial waveforms based on the peak values ​​of the acceleration. Each partial waveform corresponds to one stroke in the step 1 cutting. The partial waveform has one peak value and two minimum values ​​before and after the peak (i.e., two minimum values). For each partial waveform, the action comparison unit 422 compares the partial waveform with a reference waveform (triangle) and determines whether the difference between the partial waveform and the reference waveform is equal to or less than the difference threshold. The first action information may be information including multiple partial waveforms, and the second action information may be information including the reference waveform and a difference threshold. The action comparison unit 422 may assign a score for the action "(1) constant speed change of the cutting stroke" based on the total number of partial waveforms and the comparison result. The "comparison result" here may be the percentage or number of passing waveforms (partial waveforms whose difference is equal to or less than the difference threshold).

[0062] If the operator is an expert, it is expected that the ratio of acceptable waveforms to the total number of partial waveforms for operation "(1) constant speed change of cutting stroke" will be high, as shown in Figure 6A. On the other hand, if the operator is a beginner, it is expected that the ratio of acceptable waveforms to the total number of partial waveforms will be low, as shown in Figure 6B. In Figures 6A and 6B, the symbol 600 represents a time period during which the ratio of acceptable waveforms to the total number of partial waveforms is below a certain percentage.

[0063] The evaluation method information corresponding to the action "(1) Constant speed change of the scraping stroke" includes a graph as shown in FIG. 7. The graph shows the relationship between the proportion of passing waveforms and the score. The action comparison unit 422 identifies a score corresponding to the proportion of passing waveforms of the trainee 110 from the graph, and assigns the identified score as the score for the action "(1) Constant speed change of the scraping stroke."

[0064] The above are examples of the first motion information, second motion information, and evaluation method for the motion "(1) Constant speed change of the scraping stroke." Depending on the motion to be detected, data including values ​​measured by a sensor other than an acceleration sensor may be used. For example, the motion "(5) Scrapings removal motion" in step 2 scraping may be detected based on time-series data from a geomagnetic sensor.

[0065] The movement detection unit 421 detects all detection target movements belonging to each scraping operation based on at least the sensing data in the accumulated data 410. The movement comparison unit 422 assigns a score to each detected movement based on the first movement information, second movement information, and evaluation method information obtained for that movement. The evaluation data 413 includes the score for each detected movement.

[0066] As described above, the evaluation data 413 is data including scores assigned to each detection target action for each task. As described below, the evaluation data 413 also includes data representing task proficiency determined based on the scores assigned to each detection target action for each task. The action comparison unit 422 may identify, from the accumulated data 410, a video captured during a time period corresponding to the time-series data (part of the sensing data) on which the scores are based, and associate the video with the evaluation data 413. Based on at least a portion of the evaluation data 413, the UI control unit 423 displays an evaluation UI on the display device 404 (or the instructor's PC). The evaluation UI is a UI (typically a GUI (Graphical User Interface)) that displays the evaluation results. The evaluation UI may be prepared for the entire scraping task, for each task constituting the scraping task, or for each pair of task and detection target action. As illustrated in FIG. 8 , the evaluation UI 800 represents a time axis 801 and a video 802 captured at a specified time point on the time axis 801. The evaluation UI 800 may also display, at reference numeral 803, a comment offering advice regarding a corresponding point on the time axis 801 for an evaluation item or the like for which the score is lower than a certain value. The evaluation UI 800 also displays, as indicated by reference numeral 804, the assigned score and a time series of measured values ​​(such as acceleration) for each detection target action for the entire task targeted by the evaluation UI 800. The evaluation UI 800 also displays, as indicated by reference numeral 805, the total score. In addition to the total score, the specified task proficiency level (e.g., "beginner") may be displayed as setting X, and the total score evaluation for the task proficiency level (e.g., "pass") may also be displayed as setting Y.

[0067] Referring again to FIG.

[0068] The proficiency management data 414 is data that represents the task proficiency corresponding to each of a plurality of different score ranges. The action comparison unit 422 may store the assigned score in the storage device 405, identify the task proficiency corresponding to the score range to which the assigned score belongs, and include the identified task proficiency in the evaluation data 413. The "score range" referred to here may be prepared for the entire scraping operation, for each operation that constitutes the scraping operation, or for each pair of an operation and an action to be detected. Furthermore, the "score" compared with the score range may be a score statistic such as the cumulative value or average value of the assigned scores.

[0069] The trainee management data 415 is data including information for each trainee (for example, the trainee's ID). The evaluation data 413 may include at least a part of the information of the trainee 110 (for example, the ID).

[0070] The movement evaluation data 416 is data in a predetermined format obtained by the integration unit 424 from the evaluation data 413. The "predetermined format" referred to here may be a format that includes, as items, the ID of the trainee 110 and each detection target movement for each task that constitutes the scraping work. The movement evaluation data 416 may include the ID of the trainee 110 and a movement evaluation result (e.g., score) of each detection target movement for each task that constitutes the scraping work. The movement evaluation data 416 may exist for each trainee.

[0071] The quality evaluation data 417 may be data that represents quality in the same format as the above-mentioned predetermined format. Specifically, for example, the quality evaluation data 417 may include the ID of the trainee 110 and the quality evaluation result (e.g., three levels: A, B, and C) of each detection target action for each task that constitutes the scraping task. The quality evaluation result may be input by the trainer 111 or may be automatically input based on the analysis result of the captured video. The quality evaluation data 417 may exist for each trainee.

[0072] The integrated data 418 is data obtained by integrating the action evaluation data 416 and the quality evaluation data 417 by the integration unit 424. For example, the integration unit 424 identifies the action evaluation data 416 and the quality evaluation data 417 using the trainee ID input from the input device 403 as a key, and integrates the identified action evaluation data 416 and quality evaluation data 417. As a result, for each detection target action for each task constituting the scraping task, the action evaluation result is associated with the quality evaluation result in one integrated data 418, making it possible to manage both the action evaluation result and the quality evaluation result.

[0073] The above description can be summarized, for example, as follows: The following summary may include supplementary and modified explanations of the above description.

[0074] A sensor set including a pressure sensor 302 that measures pressure applied by a finger and an acceleration sensor that measures the acceleration of the worker's hand is worn on the worker's hand. In an embodiment, a sensor glove 102 equipped with the sensor set and a communication device is worn on the worker's hand. The worker may be a trainee 110 (an example of a first worker) or an expert (an example of a second worker). If the latter is a worker, second motion information may be prepared based on sensing data of one or more experts, and the second motion information may be compared with the motion information of the first worker.

[0075] The activity detection system includes at least the action detection unit 421 and the action comparison unit 422 among the above-described functions 420 to 424. In the embodiment, the activity detection system includes a sensor set (one example is the sensor glove 102) and an information processing device (one example is the mobile PC 150), but the sensor set may also be the activity detection system (for example, the sensor glove 102 may store the accumulated data 410 and include the action detection unit 421 and the action comparison unit 422). Furthermore, at least a part of the action detection unit 421 and the action comparison unit 422 may be realized by an information processing device, which may be a local or remote physical computer system (one or more physical computers) capable of short-range wireless communication or other types of communication with the sensor set, or may be a logical computer system (for example, a cloud computing service) based on a physical computer system (for example, a cloud platform).

[0076] The movement detection unit 421 detects the movement of the trainee 110 during work based on sensing data including pressure and acceleration measured by the sensor set. The movement comparison unit 422 compares first movement information (movement information obtained from the sensing data of the trainee 110) with second movement information regarding the detected movement in order to evaluate the trainee 110.

[0077] This makes it possible to assist the trainee in acquiring skills for a task that requires a predetermined motion including movement of the hand while the finger is pressed against an object (an example is the CV cable 115). The motion detection may be performed, for example, as follows: For each motion to be detected, motion meta-information such as a pattern or characteristics of a time series of measurement values ​​in the sensing data is prepared (for example, the motion meta-information for each motion to be detected is included in the evaluation method data 412), and the motion detection unit 421 may detect a motion to be detected that corresponds to motion meta-information whose similarity to the motion meta-information obtained from the sensing data of the trainee 110 is equal to or greater than a predetermined value.

[0078] When the difference between the first motion information and the second motion information exceeds a predetermined range, the motion comparing unit 422 may detect that the difference exceeds the predetermined range. This makes it possible to determine whether the work of the trainee 110 is good or bad.

[0079] The predetermined motion can be a repetitive motion involving repeated hand movements. Therefore, it is expected that the repetitive motion (for example, a delicate repetitive motion such as scraping with a width in the millimeter) can be accurately detected and the work that includes the motion (for example, scraping work) can be appropriately evaluated.

[0080] The motion detection unit 421 may detect a predetermined motion based on at least one of pressure, speed, acceleration, angle, angular velocity, movement distance, and geomagnetism identified from the sensing data, which is expected to improve the accuracy of detecting the predetermined motion.

[0081] The task detection system may include a UI control unit 423. The UI control unit 423 outputs output information, which is information including information representing an evaluation based on the result of the comparison, to the trainee 110 or to the trainee of the trainee 110. When output information (e.g., an evaluation UI) for the trainee 110 is output, the trainee 110 can self-train. When output information for the trainee 110 is output, the trainee 110 can understand the situation of the trainee 110 (e.g., tasks for improving proficiency) from the output information and identify what instruction to provide to the trainee 110. For example, in response to the output information for the trainee 110 being output for the trainee 110, the UI control unit 423 may accept input of instruction information, which is information representing instruction to the trainee 110, and output output information for the trainee 110 including the input instruction information to the trainee 110. In this way, the trainer can provide support for instruction to the trainee 110.

[0082] The second movement information may be movement information of a second worker who is more skilled in the repetitive movement than the trainee 110. The movement comparison unit 422 may identify first movement information of the repetitive movement from the sensing data. The movement comparison unit 422 may assign a score to the trainee 110 based on a result of comparing the first movement information with the second movement information and a predetermined evaluation method. The evaluation of the work of the trainee 110 may be based on the assigned score. This enables a quantitative skill check of the work.

[0083] The action comparison unit 422 may assign a score for each of one or more evaluation items. If there is an evaluation item for which the assigned score is lower than a predetermined score, the output information output by the UI control unit 423 may include information representing the evaluation item and information representing the factor for which the assigned score is lower than the predetermined score. This is expected to appropriately improve the actions of the trainee 110, thereby promoting task proficiency. Note that the output of the information representing the factor may be realized, for example, as follows. The evaluation method data 412 may include, for each detection target action (or for each task constituting the scraping task), a predetermined score (score threshold), information representing one or more cases in which the score is lower than the predetermined score (e.g., information representing characteristics of the measurement value time series), and information representing possible factors for the score being lower than the predetermined score for each case (e.g., a text message). The action comparison unit 422 may identify possible factors from the evaluation method data 412 based on the first action information and the results of the comparison, and output information including information representing the identified factors may be output. The factor represented by the output information may be a primary factor, which may be the factor that accounts for the largest proportion of one or more applicable factors. For example, the primary factor may be the factor corresponding to the evaluation item that has been assigned a score that is the largest deviation from the target score among one or more evaluation items. Information representing the target score of the evaluation item for each detection target behavior may be included in the evaluation method data 412.

[0084] The action detection unit 421 may detect actions other than the repetitive actions based on the sensing data. This is expected to improve the accuracy of the evaluation by the action comparison unit 422. The different actions may be actions between the repetitive actions. For example, in a series of repetitive actions, if a repetitive action is interrupted, a different action is performed, and then the repetitive action is resumed, the different action performed between the repetitive actions may be detected. Specific examples of different actions may include replacing the cutting tool 201 (an example of a processing tool) or wiping sweat. The different actions may be detected, for example, by preparing information representing the characteristics of the measurement value time series of the different actions and identifying a measurement value time series that has a high similarity to such characteristics from the sensing data. The different actions may also be actions corresponding to a portion of the sensing data whose correlation with the portion of the sensing data corresponding to the repetitive action is lower than a predetermined value. This is expected to enable accurate detection of the different actions.

[0085] The action comparison unit 422 may assign a score to the trainee 110 based on the rate or number of times different actions occur within a predetermined time. This is expected to improve the accuracy of the assigned score. For example, the score may be based on at least one of the quality of the repetitive action, the rate of occurrence of non-repetitive actions (actions that do not repeat), the number of occurrences, and the duration of occurrence.

[0086] Each of the different score ranges is associated with a task proficiency corresponding to the score range, and the output information output by the UI control unit 423 may include information indicating the task proficiency corresponding to the score range to which the assigned score belongs among the multiple score ranges. This makes it possible to estimate the task proficiency of the trainee 110.

[0087] Furthermore, for each of a plurality of different task proficiencies, there may be one or more evaluation items corresponding to the task proficiency. The predetermined evaluation method may include assigning a score for each of the one or more evaluation items corresponding to the task proficiency of the trainee 110. The task proficiency of the trainee 110 may be a task proficiency corresponding to a score range to which the assigned score belongs. This is expected to provide evaluation and task proficiency support according to the task proficiency of the trainee 110.

[0088] The motion detection unit 421 may use data of one or more captured images of the work performed by the trainee 110 to identify the motion. This is expected to improve the accuracy of motion identification.

[0089] The action comparison unit 422 may compare the first action information with the second action information, and may also compare the first status information with the second status information. The first status information may be information representing the status of an object as a result of the trainee 110's repetitive action. The second status information may be information representing the status of the object and being compared with the first status information. The action result (in other words, quality) of the object's status is also considered as one of the evaluation criteria for the trainee 110, and therefore, improvement in evaluation accuracy is expected. Both the first and second status information may be manually input. The status represented by the first status information may be a state identified based on data of one or more captured images of the work performed by the trainee 110. This reduces the burden of manually inputting the first status information.

[0090] The task detection system may include an integrating unit 424. The integrating unit 424 may associate the assigned score with a score representing the state of an object as a result of the task performed by the trainee 110. The output information may include information representing the score assigned to the trainee 110 and the score representing the state of the object as a result of the task performed by the trainee 110. This enables unified management of tasks and their resulting quality. For example, the task evaluation data 416 may include information representing task scores, and the quality evaluation data 417 may include information representing quality scores (e.g., a three-level score of A, B, and C). The integrating unit 424 may generate integrated data 418 that associates the task scores represented by the task evaluation data 416 with the quality scores represented by the quality evaluation data 417, and the integrated data 418 may be displayed by the UI control unit 423.

[0091] The output information output by the UI control unit 423 may include information representing advice for improvement for items of the first motion information that deviate from the second motion information by a certain degree or more. This provides further support for work mastery. Note that the information representing advice may be included in the evaluation method data 412 for each detection target motion, for example.

[0092] The predetermined motion may be a repetitive motion including repeated movement of the hand with the fingers pressing the cutting tool 201 against the CV cable 115. The different motion may be a motion of changing the orientation of the cutting tool 201. The motion comparison unit 422 may evaluate the trainee 110 based on the elapsed time from the start of the predetermined motion and the ratio or number of times that the different motion occurs within the predetermined time. This is expected to appropriately evaluate the task proficiency of the trainee 110 in terms of the frequency of the motion of changing the orientation of the cutting tool 201. For example, if the motion of changing the orientation of the cutting tool 201 is frequently performed within a short time after the start of the repetitive motion, the evaluation may be low, whereas if the motion of changing the orientation of the cutting tool 201 is frequently performed after a long time has passed since the start of the repetitive motion, the evaluation may not be low. This is because, when the time elapsed since the start of the repetitive motion is short, the degree of wear on the cutting tool 201 is small, and therefore the frequency of the action of changing the orientation of the cutting tool 201 is low, whereas, when the time elapsed since the start of the repetitive motion is long, the cutting tool 201 wears out, and the frequency of the action of changing the orientation of the cutting tool 201 increases. Note that the cutting tool 201 may be an example of a processing tool that is subject to wear and tear. Furthermore, when there are many different motions, the time spent on cutting work is relatively small out of the time the worker is engaged in work, and therefore the evaluation may be lowered as it is considered that the time contributed to the worker's work is small.

[0093] The task of scraping off the outer semiconductive layer is composed of multiple tasks, and the motion detection unit 421 may detect the motion of the trainee 110 based on information indicating which of the multiple tasks the trainee 110 is performing. The same motion may be detected in two or more of the multiple tasks. The motion comparison unit 422 may identify to which task the detected motion belongs based on information indicating which of the multiple tasks the trainee 110 is performing. The motion comparison unit 422 may compare first motion information regarding the detected motion with second motion information corresponding to the task to which the detected motion belongs. This enables appropriate evaluation according to the task to which the detected motion belongs.

[0094] Although one embodiment has been described above, this is merely an example for explaining the present invention, and the scope of the present invention is not limited to only these embodiments. The present invention can be implemented in various other forms.

[0095] The above-described embodiments can also be implemented in tasks in which a worker processes a workpiece using a cutting tool, abrasive, or the like. For example, the present invention can be applied to carpentry work such as planing the surface of wood. This can detect tasks that differ from the repetitive task of a worker moving the surface of wood from a first position to a second position, and the task of replacing the plane blade. It can also be applied to so-called spatula squeezing, in which a tool is pressed against a rotating metal to change its shape. That is, it can detect tasks that differ from the repetitive task of a worker moving a spatula from a first position to a second position on the metal to be processed while pressing it against the metal, such as changing the angle of the spatula or pressing it against the metal to check changes in the spatula's surface shape or the shape of the metal to be processed. It can also be applied to polishing work to increase the sphericity of the spherical surface of a glass lens. When polishing a rotating glass lens, a worker moves his or her hand in a certain amount while pressing a cloth or the like against it, which is a repetitive action. Typical different tasks include changing the position of the cloth that comes into contact with the glass lens, changing to a cloth with a different surface roughness or hardness, impregnating the cloth with abrasives, applying cutting oil, and visually checking the condition of the glass lens.

[0096] The above-described embodiment can be implemented for tasks other than the scraping-off of an external semiconductor layer as described above, and can be used to evaluate processes and detect work conditions in which different actions are performed between tasks in which an operator repeatedly uses a tool or implement on a workpiece. [Explanation of symbols]

[0097] 102: Sensor Glove 150: Mobile PC

Claims

1. A sensor set including a pressure sensor that measures pressure applied by the fingers of a first worker performing a task that requires a predetermined movement including hand movement while the fingers are pressed against an object, and an acceleration sensor that measures the acceleration of the hand of the first worker; a motion detection unit that detects a motion during the work based on sensing data including pressure and acceleration measured by the sensor set; a motion comparison unit that compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; Equipped with the work is a work of scraping off an outer semiconductive layer from the CV cable to expose an insulating layer of a specified size from the CV cable; the object is a CV cable, the sensor set is mounted in a glove that is worn on a hand; the predetermined motion is a repetitive motion including repeated hand movements with the fingers pressing the cutting tool against the CV cable; the motion detection unit detects a motion different from the repetitive motion based on the sensing data; the different motion is a motion of changing the orientation of the cutting tool; the action comparison unit evaluates the first worker based on the elapsed time from the start of the predetermined action and the rate or number of times the different action occurs within the predetermined time. A work detection system characterized by:

2. A sensor set including a pressure sensor that measures the pressure applied by the fingers of a first worker performing a task that requires a predetermined movement including the movement of the hand while the fingers are pressed against an object, and an acceleration sensor that measures the acceleration of the hand of the first worker; a motion detection unit that detects a motion during the work based on sensing data including pressure and acceleration measured by the sensor set; a motion comparison unit that compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; Equipped with the work is a work of scraping off an outer semiconductive layer from the CV cable to expose an insulating layer of a specified size from the CV cable; the object is a CV cable, the sensor set is mounted in a glove that is worn on a hand; The process of scraping off the outer semiconductive layer consists of several steps: the movement detection unit detects a movement of the first worker based on information indicating which of the plurality of tasks the first worker is performing; and The same motion is detected in two or more of the plurality of tasks, The operation comparison unit Identifying which of the plurality of tasks the detected movement belongs to based on information indicating which of the plurality of tasks the first worker is performing; and comparing first motion information relating to the detected motion with second motion information corresponding to the type of task to which the detected motion belongs; A work detection system characterized by:

3. a sensor set including a pressure sensor that measures pressure applied by a finger of a first worker performing a task that requires a predetermined movement including movement of a hand while the finger is pressed against an object, and an acceleration sensor that measures acceleration of the hand of the first worker; a motion detection unit that detects a motion during the work based on sensing data including pressure and acceleration measured by the sensor set; a motion comparison unit that compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; Equipped with the motion detection unit detects a motion different from a repetitive motion based on the sensing data, the action comparison unit assigns a score to the first worker based on a rate or number of times the different actions occur within a predetermined time period; the different action is an action corresponding to a sensing data portion of the sensing data, the correlation of which with the sensing data portion corresponding to the repetitive action is lower than a predetermined value; A work detection system characterized by:

4. The operation detection system according to claim 3, When a difference between the first motion information and the second motion information exceeds a predetermined range, the motion comparison unit detects that the difference exceeds the predetermined range. A work detection system characterized by:

5. The operation detection system according to claim 3, The predetermined motion is a repetitive motion in which the movement of the hand is repeated. A work detection system characterized by:

6. The operation detection system according to claim 5, the motion detection unit detects the predetermined motion based on at least one of pressure, speed, acceleration, angle, angular velocity, movement distance, sound, and geomagnetism identified from the sensing data; A work detection system characterized by:

7. The operation detection system according to claim 3, a user interface control unit that outputs output information, which is information including information representing an evaluation based on a result of the comparison, to the first worker or to an instructor of the first worker; The work detection system further comprises:

8. The operation detection system according to claim 5, the second motion information is motion information of a second worker who is more skilled in the repetitive motion than the first worker, The operation comparison unit Identifying first motion information of the repetitive motion from the sensing data; assigning a score to the first worker based on a result of comparing the identified first motion information with the second motion information and a predetermined evaluation method; The evaluation of the work of the first worker is based on the assigned score. A work detection system characterized by:

9. The operation detection system according to claim 8, a user interface control unit that outputs output information, which is information including information representing an evaluation based on a result of the comparison, to the first worker or to an instructor of the first worker; Further provided with the action comparison unit assigns a score for each of one or more evaluation items; If there is an evaluation item for which the assigned score is lower than a predetermined score, the output information includes information representing the evaluation item and information representing a factor for which the assigned score is lower than the predetermined score. A work detection system characterized by:

10. The operation detection system according to claim 8, a user interface control unit that outputs output information, which is information including information representing an evaluation based on a result of the comparison, to the first worker or to an instructor of the first worker; Further provided with A task proficiency level corresponding to each of the different score ranges is associated with the score range; the output information includes information representing a task proficiency level corresponding to a score range to which the assigned score belongs among the plurality of score ranges; A work detection system characterized by:

11. The operation detection system according to claim 5, the motion detection unit uses data of one or more captured images of the work performed by the first worker to identify the motion; A work detection system characterized by:

12. The operation detection system according to claim 5, the operation comparison unit compares the first operation information with the second operation information and also compares the first status information with the second status information; the first state information is information representing a state of the object as a result of the repetitive action of the first worker, the second status information is information that represents a status of the object and is compared with the first status information; A work detection system characterized by:

13. The operation detection system according to claim 12, the state represented by the first state information is a state identified based on data of one or more captured images of the work performed by the first worker; A work detection system characterized by:

14. The operation detection system according to claim 7, The user interface control unit receiving input of guidance information representing guidance to the first worker in response to output of output information intended for the instructor of the first worker; outputting output information for the first worker, including the input guidance information, to the first worker; A work detection system characterized by:

15. The operation detection system according to claim 3, a user interface control unit that outputs output information to the first worker or to an instructor of the first worker, the output information including information representing an evaluation based on a result of the comparison; Integration Department and Further provided with the motion comparison unit assigns a score to the first worker based on a result of comparing the first motion information with the second motion information and a predetermined evaluation method; the integrating unit associates the assigned score with a score representing a state of the object as a result of the work of the first worker; the output information includes information representing a score assigned to the first worker and a score representing a state of the object as a result of the work of the first worker; A work detection system characterized by:

16. The operation detection system according to claim 3, the action comparison unit assigns a score to the first worker based on a result of comparing the first action information with the second action information, the work proficiency of the first worker, and a predetermined evaluation method; A task proficiency level corresponding to each of the different score ranges is associated with the score range; For each of a plurality of different work proficiencies, there are one or more evaluation items corresponding to the work proficiency, the predetermined evaluation method includes assigning a score to each of one or more evaluation items corresponding to the work proficiency of the first worker; The task proficiency of the first worker is a task proficiency corresponding to a score range to which the score assigned to the first worker belongs. A work detection system characterized by:

17. The operation detection system according to claim 3, a user interface control unit that outputs output information to the first worker, the output information including information representing an evaluation based on the result of the comparison; Further provided with the output information includes information indicating advice for improvement for an item of the first motion information that deviates from the second motion information by a certain degree or more; A work detection system characterized by:

18. a motion detection unit that detects a motion in a task based on sensing data including pressure and acceleration measured by a sensor set including a pressure sensor that measures pressure applied by the fingers of a first worker performing a task that requires a predetermined motion including movement of a hand while the fingers are pressed against an object, and an acceleration sensor that measures acceleration of the hand of the first worker; a motion comparison unit that compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; Equipped with the work is a work of scraping off an outer semiconductive layer from the CV cable to expose an insulating layer of a specified size from the CV cable; the object is a CV cable, the sensor set is mounted in a glove that is worn on a hand; the predetermined motion is a repetitive motion including repeated hand movements with the fingers pressing the cutting tool against the CV cable; the motion detection unit detects a motion different from the repetitive motion based on the sensing data; the different motion is a motion of changing the orientation of the cutting tool; the action comparison unit evaluates the first worker based on the elapsed time from the start of the predetermined action and the rate or number of times the different action occurs within the predetermined time. A work detection device characterized by:

19. A motion detection unit that detects a motion in a task based on sensing data including pressure and acceleration measured by a sensor set including a pressure sensor that measures pressure applied by the fingers of a first worker performing a task that requires a predetermined motion including movement of the hand while the fingers are pressed against an object, and an acceleration sensor that measures the acceleration of the hand of the first worker; a motion comparison unit that compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; Equipped with the work is a work of scraping off an outer semiconductive layer from the CV cable to expose an insulating layer of a specified size from the CV cable; the object is a CV cable, the sensor set is mounted in a glove that is worn on a hand; The process of scraping off the outer semiconductive layer consists of several steps: the movement detection unit detects a movement of the first worker based on information indicating which of the plurality of tasks the first worker is performing; and The same motion is detected in two or more of the plurality of tasks, The operation comparison unit Identifying which of the plurality of tasks the detected movement belongs to based on information indicating which of the plurality of tasks the first worker is performing; and comparing first motion information relating to the detected motion with second motion information corresponding to the type of task to which the detected motion belongs; A work detection device characterized by:

20. A motion detection unit that detects a motion in a task based on sensing data including pressure and acceleration measured by a sensor set including a pressure sensor that measures pressure applied by the fingers of a first worker performing a task that requires a predetermined motion including movement of the hand while the fingers are pressed against an object, and an acceleration sensor that measures the acceleration of the hand of the first worker; a motion comparison unit that compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; Equipped with the motion detection unit detects a motion different from a repetitive motion based on the sensing data, the action comparison unit assigns a score to the first worker based on a rate or number of times the different actions occur within a predetermined time period; the different action is an action corresponding to a sensing data portion of the sensing data, the correlation of which with the sensing data portion corresponding to the repetitive action is lower than a predetermined value; A work detection device characterized by:

21. a motion detection step in which a computer detects a motion in a task based on sensing data including pressure and acceleration measured by a sensor set including a pressure sensor that measures pressure applied by the fingers of a first worker performing a task that requires a predetermined motion including movement of a hand while the fingers are pressed against an object, and an acceleration sensor that measures acceleration of the hand of the first worker; a motion comparison step in which a computer compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; and the work is a work of scraping off an outer semiconductive layer from the CV cable to expose an insulating layer of a specified size from the CV cable; the object is a CV cable, the sensor set is mounted in a glove that is worn on a hand; the predetermined motion is a repetitive motion including repeated hand movements with the fingers pressing the cutting tool against the CV cable; In the motion detection step, a motion different from the repetitive motion is detected based on the sensing data, the different motion is a motion of changing the orientation of the cutting tool; In the action comparison step, the first worker is evaluated based on the elapsed time from the start of the predetermined action and the rate or number of occurrences of the different action within the predetermined time. A work detection method characterized by:

22. A motion detection step in which a computer detects a motion in a task based on sensing data including pressure and acceleration measured by a sensor set including a pressure sensor that measures the pressure applied by the fingers of a first worker performing a task that requires a predetermined motion including movement of the hand while the fingers are pressed against an object, and an acceleration sensor that measures the acceleration of the hand of the first worker; a motion comparison step in which a computer compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; and the work is a work of scraping off an outer semiconductive layer from the CV cable to expose an insulating layer of a specified size from the CV cable; the object is a CV cable, the sensor set is mounted in a glove that is worn on a hand; The process of scraping off the outer semiconductive layer consists of several steps: In the movement detection step, a movement of the first worker is detected based on information indicating which of the plurality of tasks the first worker is performing; The same motion is detected in two or more of the plurality of tasks, In the operation comparison step, Identifying which of the plurality of tasks the detected movement belongs to based on information indicating which of the plurality of tasks the first worker is performing; and comparing first motion information relating to the detected motion with second motion information corresponding to the type of task to which the detected motion belongs; A work detection method characterized by:

23. A motion detection step in which a computer detects a motion in a task based on sensing data including pressure and acceleration measured by a sensor set including a pressure sensor that measures the pressure applied by the fingers of a first worker performing a task that requires a predetermined motion including movement of the hand while the fingers are pressed against an object, and an acceleration sensor that measures the acceleration of the hand of the first worker; a motion comparison step in which a computer compares first motion information, which is motion information obtained from sensing data of the first worker, with second motion information regarding the detected motion for evaluation of the first worker; and In the motion detection step, a motion different from a repetitive motion is detected based on the sensing data, In the action comparison step, a score is assigned to the first worker based on a rate or number of times the different actions occur within a predetermined time period; the different action is an action corresponding to a sensing data portion of the sensing data, the correlation of which with the sensing data portion corresponding to the repetitive action is lower than a predetermined value; A work detection method characterized by:

Images