Load recognition method and apparatus therefor, and work support system

Abstract

To prevent an occurrence of decline / deterioration in work efficiency by appropriately detecting a state of an overload accumulated in an operator due to fatigue or work in an instable posture, and urging him or her to make an improvement.SOLUTION: A load recognition device includes a communication unit for receiving / transmitting signals and a processing unit for processing the signal received by the communication unit. The processing unit processes the signals from a plurality of posture sensors mounted on work clothes worn by an operator received by the communication unit, estimates the posture of the operator, and determines a state of a heavy load of the operator from a change over time in the estimated posture. The communication unit is configured so as to receive the signals from the plurality of posture sensors mounted on the work clothes worn by the operator, and transmit information on the state of a heady load of the operator determined by the processing unit.SELECTED DRAWING: Figure 1

Description

Technical Field

[0004] , , , , , ,

[0006] , , , ,

[0005] , , , , , ,

[0001] The present invention relates to a load recognition method for measuring the work load of a worker and reducing the load of the worker, an apparatus therefor, and a work support system.

Background Art

[0002] In order for a worker to continue working stably and safely on a production line, it is necessary to estimate the degree of fatigue of the worker caused by the repeatedly applied load, and when it is estimated that the fatigue has reached a certain level, the worker should be made to take measures for fatigue recovery.

[0003] As a method for determining the degree of the worker's load and taking measures, Patent Document 1 describes controlling a drive unit of a production line according to the load of the worker.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In Patent Document 1, when a worker performs an operation of transporting a work W in cooperation with a robot, a control device including a distance image sensor as a hardware circuit and a microcomputer that executes a software program for estimating a human body posture from the output of the distance image sensor is used. The distance image acquired from the distance image sensor is compared with a pre-registered pattern image to detect a change over time in the posture of the worker during work, and it is determined whether to change the control amount of the drive unit provided in the production line.

[0006] However, since each worker has a different body shape and build, and the changes in posture that occur when fatigue accumulates due to the workload during work differ from worker to worker, it is not possible to estimate the actual degree of fatigue of a worker by comparing it with pre-registered pattern images, and there is a possibility of overlooking a worker who is in a state of overwork.

[0007] The present invention solves the problems of the conventional technology described above and provides a load recognition method, apparatus, and work support system that can prevent a decrease in work efficiency and the occurrence of defects by accurately detecting and promoting improvements in the state of overload accumulated in workers due to fatigue and working in unstable postures. [Means for solving the problem]

[0008] To solve the above-mentioned problems, the present invention provides a load recognition device comprising a communication unit that transmits and receives signals, and a processing unit that processes the signals received by the communication unit, wherein the processing unit processes signals from a plurality of posture sensors attached to the work clothes worn by the worker received by the communication unit to estimate the worker's posture and the estimated posture time The system determines whether the worker is under high load based on the changes in the system. The communication unit is configured to receive signals from multiple posture sensors attached to the worker's work clothes and to transmit information regarding the worker's high load status determined by the processing unit.

[0009] Furthermore, in order to solve the above-mentioned problems, the present invention provides a method for recognizing the state of a worker's load using a load recognition device equipped with a communication unit and a processing unit, wherein the communication unit receives signals from multiple posture sensors attached to the work clothes worn by the worker, and the processing unit processes these received signals from the multiple posture sensors to estimate the worker's posture, and this estimated posture time The system determines if the worker is under high load based on the changes, and transmits information regarding the worker's high load status from the communication unit based on the results determined by the processing unit. It was decided .

[0010] Furthermore, in order to solve the above-mentioned problems, the present invention provides a work support system comprising: a plurality of posture sensors attached to the work clothes worn by the worker; a load recognition device unit that receives output signals from the plurality of posture sensors, determines the load on the worker wearing the work clothes, and transmits the result of this determination; and a receiving unit that receives the result of the determination transmitted from the load recognition device unit and notifies the worker wearing the work clothes. [Effects of the Invention]

[0011] According to the present invention, it is possible to prevent a decrease in work efficiency and the occurrence of defects caused by worker fatigue and working in unstable postures. Furthermore, by preventing the accumulation of worker fatigue, the working environment can also be improved. [Brief explanation of the drawing]

[0012] [Figure 1] This is a block diagram showing the schematic configuration of a work support system according to Embodiment 1 of the present invention. [Figure 2] This is a front view of a worker wearing work clothes equipped with multiple posture sensors and a communication unit. [Figure 3] This figure shows that in the work support system according to Embodiment 1 of the present invention, signals from multiple posture sensors attached to work clothes are input to the upper arm state estimation unit or the waist state estimation unit of the posture estimation unit, and a graph of the time change of the rotation angle of the upper arm as shown in (a) or a graph of the time change of the bending angle of the waist as shown in (b) is obtained. [Figure 4] (a) is a side view of a worker with their knees extended and their waist high while working, and (b) is a side view of a worker with their knees bent and their waist low while working. [Figure 5] This table shows an example of the relationship between a worker's body parts, length, and weight. [Figure 6](a) is a side view of a worker showing the horizontal distance from the waist to the center of gravity of the torso, arms, and head when the worker is working with their knees extended and their waist high, and (b) is a side view of a worker showing the horizontal distance from the waist to the center of gravity of the torso, arms, and head when the worker is working with their knees bent and their waist low. [Figure 7] This graph shows the time-dependent change in the moment acting on the worker's waist due to changes in the worker's posture when the same task is repeated in Embodiment 1 of the present invention. [Figure 8] This flowchart shows the data processing flow using the load recognition system of the work support system in Embodiment 1 of the present invention. [Figure 9] This is a front view of a display scene showing an example of warning information displayed on the display screen of the output unit based on notification information in the work support system according to Embodiment 1 of the present invention. [Figure 10] This figure shows that in the work support system according to Embodiment 2 of the present invention, signals from multiple posture sensors attached to work clothes are input to the waist state estimation unit or the knee state estimation unit of the posture estimation unit, and a graph of the time change of the waist bending angle as shown in (a) or a graph of the time change of the knee bending angle as shown in (b) is obtained. [Figure 11] (a) is a side view of a worker showing the relationship between the projected position of the worker's center of gravity on the floor and the midpoint of the contact surface between the worker's feet and the floor when the worker is working with their knees extended and their hips high, and (b) is a side view of a worker showing the relationship between the projected position of the worker's center of gravity on the floor and the midpoint of the contact surface between the worker's feet and the floor when the worker is working with their knees bent and their hips low. [Figure 12] This graph shows the time change in the distance between the midpoint of the worker's feet and the position of the worker's center of gravity projected onto the floor surface, when the same work is repeated in Embodiment 2 of the present invention, due to changes in the worker's posture. [Figure 13]It is a flowchart showing the data processing flow using the load recognition system of the work support system in Example 2 of the present invention. [Figure 14] In the work support system according to Example 3 of the present invention, in addition to the configuration described using FIG. 3 in Example 1, among the signals from a plurality of attitude sensors worn by the worker, the signal from the pressure sensor installed on the insole of the shoe is input to the load data processing unit of the attitude estimation unit, and it is a diagram showing that a graph of the time change of the load related to the foot as shown in (a) can be obtained. [Figure 15] (a) is a side view of the worker showing the horizontal distance from the waist to the center of gravity positions of the trunk, arms, head, and the load held by the hands when the worker is working with the knees extended and the waist position high, and (b) is a side view of the worker showing the horizontal distance from the waist to the center of gravity positions of the trunk, arms, head, and the load held by the hands when the worker is working with the knees bent and the waist position low.

Mode for Carrying Out the Invention

[0013] The present invention estimates the magnitude of physical quantities such as the moment applied to the body part targeted by the worker, estimates the state of the load applied to the worker by observing the change in the physical quantity, and when it is determined that the state is a high-load state, attempts to reduce the load applied to the worker through notification to the worker or control of the control device.

[0014] That is, in the present invention, paying attention to the fact that the fatigue accumulated in the worker due to continuous work on the production line appears as a change in the relative positions of multiple parts of the worker's body, the degree of fatigue of the worker is judged from the degree of change in the relative positions of multiple parts of the worker's body, and by prompting the worker to pay attention, it is possible to prevent accidents that could lead to industrial accidents.

[0015] Embodiments of the present invention will be described in detail below with reference to the drawings. In all the drawings used to illustrate these embodiments, components having the same function will be denoted by the same reference numerals, and repeated explanations will be omitted in principle.

[0016] However, the present invention shall not be construed as being limited to the embodiments described below. It will be readily apparent to those skilled in the art that the specific configuration may be modified without departing from the spirit or intent of the present invention. [Examples]

[0017] This invention Example 1 To illustrate this, Figures 1 to 9 illustrate the case where a worker repeatedly performs relatively low-impact tasks while standing, such as tightening screws, wiring, or carrying lightweight objects.

[0018] Figure 1 shows the configuration of the work support system 100 according to this embodiment. The work support system 100 according to this embodiment consists of a sensor unit 110 attached to the worker's work clothes 10, a load recognition system 130 that receives operation data 120 from the sensor unit 110 and determines the state of the worker's load, and a receiving unit 150 attached to the worker's work clothes 10 that receives notification information 140 generated by the load recognition system 130.

[0019] The sensor unit 110, which is attached to the worker's work clothes 10, consists of a plurality of posture sensors 111, a communication unit 112 that transmits operation data 120 received from the output signals of the plurality of posture sensors 111 to the load recognition system 130, and wiring 113 that connects the plurality of posture sensors 111 and the communication unit 112. Instead of wiring 113, the plurality of posture sensors 111 and the communication unit 112 may be connected by wireless communication using Bluetooth.

[0020] Figure 2 shows the worker's work clothes 10 with multiple posture sensors 111 and a communication unit 112 of the sensor unit 110 attached. The posture sensors 111 are attached to multiple locations on the worker's shoulders, arms, waist, and lower limbs, as well as to the hat 20 and shoes 30. In Figure 2, the communication unit 112 is integrated with the receiving unit 150.

[0021] The posture sensor 111 is comprised of multiple sensors, including an accelerometer that detects the movement of the worker's shoulders, arms, waist, lower limbs, and head, a gyroscope that detects tilt, and a geomagnetic sensor that detects the direction of movement.

[0022] Signals transmitted from multiple attitude sensors 111 are received by the communication unit 112 and transmitted from the communication unit 112 to the load recognition system 130 as operation data 120. The operation data 120 transmitted from the communication unit 112 includes acceleration data detected by the accelerometer, tilt data detected by the gyro sensor, and geomagnetic information detected by the geomagnetic sensor for each of the multiple attitude sensors 111. The communication unit 112 and the load recognition system 130 are connected by wireless communication.

[0023] As shown in Figure 1, the load recognition system 130 has a communication unit 131 that receives operation data 120 transmitted from the communication unit 112 of the sensor unit 110 and transmits notification information 140 to the receiving unit 150. 、 The communication unit 131 receives the operation data 120. worker A posture estimation unit 132 analyzes the motion data of posture sensors 111 attached to each part of the work clothes 10 to estimate the worker's posture. 、 Data storage unit 133 、 Estimated by the posture estimation unit 132 worker Based on the worker's posture and the data stored in the memory unit 133, the load estimation unit 134 estimates, for example, the load on the worker's waist. 、 The high-load determination unit 135 determines whether the load on the worker's waist, as estimated by the load estimation unit 134, is causing a high load on the worker. 、 Information generation unit 136 generates information based on the results determined by the high load determination unit 135. 、It includes a control unit 137 that controls the entire system, and these are connected by a communication line 138.

[0024] The notification information 140 transmitted from the communication unit 131 to the receiving unit 150 of the load recognition system 130 includes high load notification information that notifies the worker that the load on the worker's waist is in a high load state, and work posture information that prompts the worker to correct their posture.

[0025] The receiving unit 150 includes a communication unit 151 that receives information transmitted from the communication unit 131 of the load recognition system 130, and a control unit 152 that controls the output unit 153 based on the signal received by the communication unit 151 to display text and / or images on the output unit 153, or to emit voice or alarm sounds. Alternatively, the output unit 153 may use augmented reality (AR) glasses or the like.

[0026] As shown in Figure 3, among the multiple posture sensors 111 attached to the worker's work clothes 10, the posture sensor 111-1 attached to the shoulder portion of the worker's work clothes 10, the posture sensor 111-2 attached to the upper arm portion, and the posture sensor 111-3 attached to the forearm portion receive signals from these sensors, and the upper arm state estimation unit 301 in the posture estimation unit 132 detects the time change of the rotation angle of the upper arm as shown by the curve 311 in graph (a).

[0027] Meanwhile, among the multiple posture sensors 111 attached to the worker's work clothes 10, the posture sensor 111-1 attached to the shoulder area of ​​the worker's work clothes 10, the posture sensor 111-4 attached to the waist area, and the posture sensor 111-5 attached to the thigh area are received by the waist state estimation unit 302 of the posture estimation unit 132, which detects the time change of the waist bending angle as shown by the curve 312 in graph (b).

[0028] Additionally, the shoes 30 worn by worker 210 have posture sensors. 111-7、 210 workers but The hat 20 has a posture sensor. 111-8 Each of these is attached.

[0029] In this way, by using data from multiple posture sensors 111 attached to the worker's work clothes 10, it is possible to detect changes in the state (posture) of various parts of the worker's body, such as their position and tilt angle.

[0030] Figure 4 shows examples of worker postures during work, specifically a state where the worker is bent forward with knees 401 extended (a) and a state where the worker is bent forward with knees 401 bent (b), both while working at the same height P.

[0031] In the state where the knees 401 are extended and the person is leaning forward (a), the angle of bending A1 from the waist 402 to the torso 403 with respect to the direction perpendicular to the floor and the angle B1 of the upper arm 404 relative to the torso 403 are compared. In the state where the knees 401 are bent forward (b), the angle of bending A2 from the waist 402 to the torso 403 is smaller than A1, while the angle B2 of the upper arm 404 relative to the torso 403 is larger than B1.

[0032] Thus, when working at the same height P, the bending angle from the waist 402 to the torso 403 and the angle of the upper arm 404 relative to the torso 403 differ depending on whether the knees 401 are extended and the person is bent forward (a) or whether the knees 401 are bent and the person is bent forward (b). Consequently, the load (moment) on the waist 402 and the load (moment) on the shoulder 407 due to the weight of the head 406, torso 403, upper arm 404, and forearm 405 will differ.

[0033] Table 500 in Figure 5 shows an example of a dataset of length 520 and weight 530 for each body part 510. In Figure 5, examples of body parts 510 are shown as the torso, upper arm, forearm, thigh, and lower leg, but other parts such as the head and neck are also included.

[0034] Thus, since the length 520 and weight 530 differ for each body part 510, it can be seen that the moment acting on the waist 402 differs depending on the worker's posture.

[0035] These data are stored in the memory unit 133 of the load recognition system 130 shown in Figure 1, and are used by the load estimation unit 134 to estimate, for example, the load on a worker's waist.

[0036] Figure 5 shows an example of a dataset of lengths 520 and weights 530 of each body part 510 of a worker, as shown in Table 500. However, multiple datasets with different lengths 520 and weights 530 of each body part 510 may be stored in the storage unit 133, and a dataset that closely matches the body shape of the worker 210 may be selected from among them, and the load estimation unit 134 may estimate, for example, the load on the worker's waist.

[0037] Figure 6 shows examples of moments acting on the waist 402 depending on the worker's posture. Figure 6, similar to Figure 4, shows examples of worker postures during work, specifically a state where the worker is bent forward with knees 401 extended (a) and a state where the worker is bent forward with knees 401 bent (b), both performed at the same height P. Lines 610 and 611 represent horizontal lines drawn at the height of the waist 402, and lines 620 and 621 represent vertical lines drawn at the waist 402.

[0038] The load on the worker's waist 402 can be attributed to the weights of the torso 403, the upper arms 404, the forearms 405, and the head 406 including the neck. In Figures 6(a) and (b), the weight of the torso 403 is M1 and its center of gravity is 601, the weight of the upper arms 404 and forearms 405 is M2 and its center of gravity is 602, and the weight of the head 406 is M3 and its center of gravity is 603.

[0039] In the state shown in Figure 6(a), if the distance from the line 620 to the center of gravity 601 is L1, the distance from the line 620 to the center of gravity 602 is L2, and the distance from the line 620 to the center of gravity 603 is L3, then the load (moment) F1 acting on the worker's waist 402 is: F1 = M1 × L1 + M2 × L2 + M3 × L3 (Math 1) It is expressed as follows.

[0040] On the other hand, in the state shown in Figure 6(b), if the distance from the line 621 to the center of gravity 601 is L1', the distance from the line 621 to the center of gravity 602 is L2', and the distance from the line 621 to the center of gravity 603 is L3', then the load (moment) F2 acting on the worker's waist 402 is: F2 = M1 × L1' + M2 × L2' + M3 × L3' (Math 2) It is expressed as follows.

[0041] As explained in Figure 4, the bending angle A1 from the waist 402 to the torso 403 in the state where the knee 401 is extended and the body is bent forward (a) is greater than the angle A2 in the state where the knee 401 is bent forward (b). Therefore, L1, L2, and L3 in Figure 6 (a) are greater than L1', L2', and L3' in (b), respectively, and as a result, F1 is greater than F2.

[0042] In other words, the moment on the hip 402 is greater when the knee 401 is extended and the person is bent forward as shown in Figure 6(a), compared to when the knee 401 is bent and the person is bent forward as shown in Figure 6(b), resulting in a greater load on the hip 402.

[0043] Graph 700 in Figure 7 shows the time change of the moment acting on the worker's waist 402 due to changes in the worker's posture when the same task is performed repeatedly.

[0044] In the graph in Figure 7, the moment 701 acting on the worker's waist 402 is relatively large in the initial stages, but gradually decreases over time. This is presumed to be because fatigue accumulates due to the load on the worker's waist from continuing the same work, causing the worker to change their posture in order to reduce the moment 701 acting on the waist 402. In other words, the degree of fatigue of the worker can be estimated from the change in the moment 701 acting on the worker's waist 402.

[0045] Furthermore, when the moment 701 acting on the worker's waist 402 falls below a certain value (the level indicated by the dotted line 702 in the example shown in Figure 7), the load recognition system 130 issues a warning to the receiving unit 150, thereby preventing a decrease in work efficiency and the occurrence of defects due to worker fatigue or working in an unstable posture. In addition, by preventing the accumulation of worker fatigue, it also leads to an improvement in the working environment.

[0046] In the example described above, a warning is issued from the load recognition system 130 to the receiving unit 150 when the level falls below the level indicated by the dotted line 702 in Figure 7. However, this is not limited to this, and for example, the change in the moment 701 acting on the waist 402 in Figure 7 could be represented by a curved graph, and a warning could be issued from the load recognition system 130 to the receiving unit 150 if the slope of that curve continuously falls outside a predetermined range for a certain period of time.

[0047] Furthermore, in the example described above, a warning is issued from the load recognition system 130 to the receiving unit 150 when the load falls below the level indicated by the dotted line 702 in Figure 7. However, conversely, a warning may also be issued from the load recognition system 130 to the receiving unit 150 when the moment 701 acting on the waist 402 becomes greater than a certain level compared to the initial state.

[0048] Furthermore, the moment 701 acting on the waist 402 is in its initial state Moment The load recognition system 130 may be configured to issue a warning to the receiving unit 150 if the value becomes significantly larger or smaller than a certain level (i.e., if the moment 701 acting on the waist 402 changes by a certain level or more from the initial state).

[0049] Figure 8 shows the data processing flow using the load recognition system 130 in the work support system 100 of this embodiment. As a premise for this processing flow, as shown in Figure 2, the worker 210 performs the work while wearing work clothes 10 to which the sensor unit 110 has multiple posture sensors 111, a communication unit 112, and a receiving unit 150 attached.

[0050] First, the movement data 120, such as acceleration, angular velocity, and geomagnetic field, detected by the posture sensor 111 attached to the work clothes 10 worn by the worker and transmitted from the communication unit 112, is received by the communication unit 131 of the load recognition system 130 (S801).

[0051] The data received by the communication unit 131 is sent to the posture estimation unit 132 via the communication line 138, where the posture estimation unit 132 extracts feature quantities that represent the posture of the worker's waist 402 and knees 401, which change moment by moment (S802).

[0052] Next, the feature quantities representing the posture of the worker's waist 402 and knees 401 extracted by the posture estimation unit 132 are sent to the load estimation unit 134. The load estimation unit 134 then selects a dataset that closely matches the worker's body shape from among multiple datasets of length 520 and weight 530 of each body part 510, as shown in Table 500 of Figure 5, stored in the memory unit 133, and uses that dataset to estimate the moment acting on the worker's waist 402, which changes moment by moment (S803).

[0053] The moment data on the worker's waist 402 at each moment, estimated by the load estimation unit 134, is sent to the high load determination unit 135 to determine whether the estimated moment has fallen below a reference value (for example, the dotted line 702 shown in the graph in Figure 7) (S804).

[0054] If the high load determination unit 135 determines that the moment estimated by the load estimation unit 134 is not below the reference value (i.e., NO in S804), then processing of the operation data received in S801 is terminated.

[0055] On the other hand, if the high load determination unit 135 determines that the moment estimated by the load estimation unit 134 has fallen below a standard value (if the answer is YES in S804), that information is sent to the information generation unit 136 to create warning information indicating a high load condition, which is then transmitted from the communication unit 131 as notification information 140 (S805). This completes the processing performed internally by the load recognition system 130.

[0056] The notification information 140 transmitted from the communication unit 131 is received by the communication unit 151 of the receiving unit 150 attached to the work clothes 10 worn by the worker 210, and warning information indicating a high load condition based on the notification information 140 is output from the output unit 153 (S806).

[0057] Figure 9 shows an example of cautionary information 910 based on notification information 140 displayed on the display screen 900 of the output unit 153. Figure 9 shows a case where the display screen 900 displays information about a change in posture, such as "You are taking a posture to avoid strain on your lower back," and information encouraging recovery, such as "If you feel fatigue in your lower back, take a break to recover." The cautionary information displayed on the display screen 900 may be other information than the example shown in Figure 9.

[0058] As an example of outputting notification information 140 received by the communication unit 151 of the receiving unit 150 from the output unit 153, Figure 9 shows an example where it is displayed on the display screen 900. However, it is not limited to this, and for example, the information may be transmitted as sound from a speaker (not shown) provided in the output unit 153. Alternatively, the information may be transmitted as vibration from a vibrator (not shown) provided in the output unit 153. Furthermore, a combination of these may be transmitted from the output unit 153.

[0059] In the above-described embodiment, a method for estimating the degree of fatigue of a worker 210 from the change in moment applied to the worker's waist 402 was explained. However, the method is not limited to this, and the degree of fatigue may also be estimated from the change in moment applied to the worker's back, neck, or knees, or from the change in moment at multiple locations.

[0060] Furthermore, in S805, in addition to the receiving unit 150, the communication unit 131 may also output to a control unit (not shown) that controls an equipment (not shown) on which the worker 210 is performing work.

[0061] According to this embodiment, it is possible to prevent a decrease in work efficiency and the occurrence of defects caused by worker fatigue and working in unstable postures. Furthermore, by preventing the accumulation of worker fatigue, it also leads to an improvement in the working environment. [Examples]

[0062] In Example 1, a method for estimating the degree of worker fatigue from the change in the moment acting on the worker's waist 402 was described. In this example, a method for estimating worker fatigue from the change in the distance between the center of both feet and the projected position of the center of gravity of the upper body will be explained using Figures 10 to 13.

[0063] The configuration of the load recognition system 130 used in this embodiment is basically the same as the configuration described with reference to Figure 1 in Embodiment 1, except that the posture estimation unit 132 in Figure 1 is replaced with the posture estimation unit 232 shown in Figure 10. In other words, the posture estimation unit 132 in Embodiment 1 includes the upper arm state estimation unit 301 and the waist state estimation unit 302 described in Figure 3, whereas the posture estimation unit 232 in this embodiment is configured to include the waist state estimation unit 302 and the knee state estimation unit 303 shown in Figure 10.

[0064] In this embodiment, as shown in Figure 10, the posture estimation unit 232 receives signals output from three posture sensors 111-1 attached to the shoulder portion of the worker's work clothes 10, a posture sensor 111-4 attached to the waist portion, and a posture sensor 111-5 attached to the thigh portion, among a plurality of posture sensors 111 attached to the worker's work clothes 10. The waist state estimation unit 302 in the posture estimation unit 232 then detects the time change in the waist bending angle as shown by the curve 1001 in graph (a).

[0065] Meanwhile, among the multiple posture sensors 111 attached to the worker's work clothes 10, the knee state estimation unit 303 in the posture estimation unit 232 receives signals output from posture sensor 111-4 attached to the waist of the worker's work clothes 10, posture sensor 111-5 attached to the thigh, and posture sensor 111-6 attached to the lower leg, and detects the time change of the knee bending angle as shown by curve 1002 in graph (b).

[0066] In this way, by using data from multiple posture sensors 111 attached to the worker's work clothes 10, it is possible to detect changes in the position of the center of gravity of the worker's upper body 210 from changes in the state (posture) of the worker, such as the position and tilt angle of the worker's waist and knees.

[0067] Figure 11 shows an example of how the distance between the center of both feet and the projected position of the center of gravity of the upper body changes depending on the worker's posture. Figure 11 shows examples of worker postures during work, illustrating the same height P while working in two positions: (a) with the right knee 1103 and left knee 1104 extended and leaning forward, and (b) with the right knee 1103 and left knee 1104 bent and leaning forward. 1101 represents the center of gravity of the worker 210's upper body, 1102 represents the worker 210's waist, 1103 is the right knee, 1104 is the left knee, 1105 is the heel of the right foot, 1106 is the heel of the left foot, and 1111 represents the center position between the right heel 1105 and the left heel 1106. In Figures 11(a) and (b), the positions of the right heel 1105 and the left heel 1106 may be the same or different.

[0068] Furthermore, 1110 in Figure 11(a) and 1120 in Figure 11(b) represent the projected positions of the center of gravity 1101 of the worker's upper body onto the surface (e.g., the floor) on which the heels 1105 of the right foot and 1106 of the left foot are placed.

[0069] The positions of worker 210's right heel 1105 and left heel 1106, and the respective postures of the right knee 1103 and left knee 1104 (the degree of bending of the right knee 1103 and left knee 1104) can be determined in the knee state estimation unit 303 of the posture estimation unit 232 using acceleration, angular velocity, and geomagnetic information obtained from signals output from posture sensors 111-4 attached to the waist, posture sensors 111-5 attached to the thigh, and posture sensors 111-6 attached to the lower leg of the work clothes 10, as well as data on the length 520 of each body part 510 as explained in Figure 5, which is stored in the memory unit 133. The positions of worker 210's right heel 1105 and left heel 1106 can also be directly determined from data from posture sensor 111-7 attached to the shoes 30 worn by worker 210.

[0070] Furthermore, the posture of the waist 1102 is estimated by the waist state estimation unit 302 of the posture estimation unit 232 using signals output from posture sensors 111-1 attached to the shoulder area of ​​the worker's work clothes 10, posture sensors 111-4 attached to the waist area, and posture sensors 111-5 attached to the thigh area.

[0071] In the load estimation unit 134, based on the information obtained by the posture estimation unit 232 regarding the positions of the right heel 1105 and the left heel 1106, the postures of the right knee 1103 and the left knee 1104, and the posture of the waist 1102, the unit determines the position of the center of gravity 1101 of the worker's upper body projected onto the floor on which the worker 210 is standing (1110 in Figure 11(a), 1120 in Figure 11(b)) and the center position 1111 of the right heel 1105 and the left heel 1106, and calculates the distance between them. In Figure 11(a), the calculated distance is shown as D1, and in Figure 11(b), it is shown as D2.

[0072] In Figure 11(a), the distance D1 when the right knee 1103 and left knee 1104 are extended and the person is bent forward is greater than the distance D2 when the right knee 1103 and left knee 1104 are bent forward (b). Therefore, state (a) places a greater load on the worker than state (b).

[0073] Graph 1200 in Figure 12 shows the time change in the distance 1201 (corresponding to D1 in Figure 11(a) or D2 in Figure 11(b)) between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor surface, when the same task is repeated, due to changes in the worker's posture.

[0074] In the graph in Figure 12, the distance 1201 between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor is relatively large in the initial stages, but gradually decreases over time. This is presumed to be due to the accumulation of fatigue from the load on the worker's lower back due to continuing the same work, causing the worker to change their posture. In other words, the degree of fatigue of the worker can be estimated from the change in the distance 1201 between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor.

[0075] Furthermore, when the distance 1201 between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor surface falls below a certain value (the level indicated by the dotted line 1202 in the example shown in Figure 12), the load recognition system 130 issues a warning to the receiving unit 150, thereby preventing a decrease in work efficiency and the occurrence of defects due to worker fatigue or working in an unstable posture. In addition, by preventing the accumulation of worker fatigue, it also leads to an improvement in the working environment.

[0076] In the example described above, a warning is issued from the load recognition system 130 to the receiving unit 150 when the load falls below the level indicated by the dotted line 1202 in Figure 12. However, this is not limited to this. For example, the change in the distance 1201 between the midpoint of the worker's feet and the center of gravity of their upper body projected onto the floor surface could be represented by a curved graph, and a warning could be issued from the load recognition system 130 to the receiving unit 150 if the slope of that curve continuously falls outside a predetermined range for a certain period of time.

[0077] Furthermore, in the example described above, the load recognition system 130 issues a warning to the receiving unit 150 when the load falls below the level indicated by the dotted line 1202 in Figure 12. Conversely, the load recognition system 130 may also issue a warning to the receiving unit 150 when the distance 1201 between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor surface becomes a certain level greater than the initial state.

[0078] Furthermore, the load recognition system 130 may issue a warning to the receiving unit 150 if the distance 1201 between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor surface becomes larger or smaller than the initial state by a certain level (or a certain percentage) or more (if the distance 1201 between the midpoint of the worker's feet and the projected position of the center of gravity of the upper body on the floor surface changes by a certain level (percentage) or more from the initial state).

[0079] Figure 13 shows the data processing flow using the load recognition system 130 in the work support system 100 of this embodiment. As a premise for this processing flow, as shown in Figure 10, the worker 210 performs the work while wearing work clothes 10 to which the sensor unit 110 has multiple posture sensors 111, a communication unit 112, and a receiving unit 150 attached.

[0080] First, the movement data 120, such as acceleration, angular velocity, and geomagnetic field, detected by the posture sensor 111 attached to the work clothes 10 worn by the worker and transmitted from the communication unit 112, is received by the communication unit 131 of the load recognition system 130 (S1301).

[0081] The data received by the communication unit 131 is sent to the attitude estimation unit via the communication line 138. 232 It is sent to the posture estimation unit. 232 In this process, feature quantities representing the constantly changing posture of the worker's waist 402 and knees 401 are extracted (S1302).

[0082] Next, the posture estimation unit 232The feature quantities representing the posture of the worker's waist 402 and knees 401 extracted by the system are sent to the load estimation unit 134. The load estimation unit 134 selects a dataset that closely matches the worker's body shape from among multiple datasets of length 520 and weight 530 of each body part 510, as shown in Table 500 of Figure 5, stored in the memory unit 133, and uses that dataset to estimate the center of gravity position 1101 of the worker's upper body projected onto the floor surface, which changes moment by moment. ( S1303 ) .

[0083] The load estimation unit 134 estimates the center of gravity position 1101 of the worker's upper body projected onto the floor surface at each moment, and this data is sent to the high load determination unit 135. The unit calculates the distance between the estimated center of gravity position 1101 of the worker's upper body projected onto the floor surface and the center position 1111 of the heels 1105 and 1106 of the worker's feet (S1304). It then determines whether this calculated distance is less than or equal to a reference value (for example, the value shown by the dotted line 1202 in the graph of Figure 12, or a certain percentage of the value at the start of measurement) (S1305).

[0084] If the high load determination unit 135 determines that the distance calculated in S1304 is not below the reference value (i.e., NO in S1305), it terminates processing of the operation data received in S1301.

[0085] On the other hand, if the high load determination unit 135 determines that the distance calculated in S1304 is below the standard value (if the answer is YES in S1305), that information is sent to the information generation unit 136 to create information indicating a high load state, and the communication unit 131 transmits it as notification information 140 (S1306).

[0086] The notification information 140 transmitted from the communication unit 131 is received by the communication unit 151 of the receiving unit 150 attached to the work clothes 10 worn by the worker 210, and information based on the notification information 140 is output from the output unit 153 to the display screen 900 as shown in Figure 9 in Embodiment 1 (S1307).

[0087] The method for outputting the notification information 140 received by the communication unit 151 of the receiving unit 150 from the output unit 153 is as described in Embodiment 1, for example, by transmitting the information as sound from a speaker (not shown) provided in the output unit 153. Alternatively, the information may be transmitted as vibration from a vibrator (not shown) provided in the output unit 153. Alternatively, a combination of these methods may be used to transmit the information from the output unit 153.

[0088] Alternatively, the posture sensor 111 may not be attached to the shoe 30, and the posture of the entire body may be estimated to estimate the position of both feet, or only the posture of the upper body may be estimated, and the position of both feet may be set to a predetermined value (e.g., directly below the waist). Furthermore, sensors such as pressure gauges may be placed on the floor, and the midpoint between the two feet may be estimated from the pressure distribution.

[0089] In the above-described embodiment, a method for estimating the degree of fatigue of a worker was explained based on the change in the position of the center of gravity 1101 of the upper body of the worker 210 projected onto the floor. However, the method is not limited to this, and the degree of fatigue of the worker may also be estimated from the change in the position of the center of gravity of the entire body or the center of gravity of the head projected onto the floor, or from the change in the projection positions of the centers of gravity of multiple parts of the body onto the floor.

[0090] According to this embodiment, it is possible to prevent a decrease in work efficiency and the occurrence of defects caused by worker fatigue and working in unstable postures. Furthermore, by preventing the accumulation of worker fatigue, it also leads to an improvement in the working environment. [Examples]

[0091] This invention Example 3 This section explains how to estimate the weight of the object being held by the worker and use this information to calculate the magnitude of the moment.

[0092] In this embodiment, as shown in Figure 14, a pressure sensor 1402 is installed in the insole 1401 of the worker's shoe 30. The load of the load lifted by the worker is estimated from the change in load detected by the pressure sensor, and the weight of the load held by the worker is taken into consideration when detecting the worker's fatigue state as described in Embodiment 1 or Embodiment 2.

[0093] The configuration of the load recognition system 130 used in this embodiment is basically the same as the configuration described with reference to Figure 1 in Embodiment 1. However, the posture estimation unit 332 shown in Figure 14 differs in that, in addition to the upper arm state estimation unit 301 and waist state estimation unit 302 described in the posture estimation unit 132 of Figure 1, a load data estimation unit 304 is added.

[0094] In other words, while the posture estimation unit 132 in Embodiment 1 detected the fatigue state of the worker 210 from changes in the worker 210's posture, including the influence of the load 220 held by the worker 210 during work, in this embodiment, the fatigue state of the worker 210 is detected by also considering the weight of the load 220.

[0095] In the configuration shown in Figure 14, the configuration is the same as in Embodiment 1, where signals output from the posture sensor 111-1 attached to the shoulder portion of the worker's work clothes 10, the posture sensor 111-2 attached to the upper arm portion, and the posture sensor 111-3 attached to the forearm portion are received by the upper arm state estimation unit 301 to detect the time change in the rotation angle of the upper arm, and signals output from the posture sensor 111-1 attached to the shoulder portion of the worker's work clothes 10, the posture sensor 111-4 attached to the waist portion, and the posture sensor 111-5 attached to the thigh portion are received by the waist state estimation unit 302 to detect the time change in the bending angle of the waist.

[0096] In this embodiment, the load data estimation unit 304 receives the output from the pressure sensor 1402 installed on the insole 1401 of the shoe 30 and detects a load 1400 as shown in graph (a) of Figure 14, which is the load on the worker's foot. Here, 1410 is the load on the worker's foot when the worker 210 is not carrying the load 220, and 1420 is the load on the worker's foot when the worker 210 is carrying the load 220. The difference between 1410 and 1420, M4, corresponds to the weight of the load 220.

[0097] Alternatively, instead of the pressure sensor 1402 in the insole 1401 of the shoe 30, a pressure sheet may be placed in the area where the worker 210 moves, and the load on the worker's feet may be measured using this pressure sheet.

[0098] Figure 15 shows an example of the moment acting on the waist 402 depending on the posture of the worker 210. Figure 15 shows examples of worker postures during work, illustrating the situation when the worker is bending forward with knees 401 extended (a) and bending forward with knees 401 bent (b), while holding a load 220 at the same height P. Lines 1510 and 1511 represent horizontal lines drawn at the height of the waist 402, and lines 1520 and 1521 represent vertical lines drawn at the waist 402.

[0099] At the worker's waist position 402 such The load consists of the torso 403 and the upper arm 404. 、 Forearm 405 、 The weights of the head, including the neck, and the luggage, as well as the weight of the cargo, can be considered. Figure 15 In (a) and (b), the weight of the fuselage section 403 is M11, and its center of gravity is 1501 Assuming the weight of the upper arm 404 and forearm 405 is M12, the position of its center of gravity is 1502 Assuming the weight of the head 406 is M13, its center of gravity is 1503 Assuming the weight of the 220 load is M14, its center of gravity is 1504. and do.

[0100] In the state shown in Figure 15(a), if the distance from the line 1520 to the center of gravity position 1501 is L11, the distance from the line 1520 to the center of gravity position 1502 is L12, the distance from the line 1520 to the center of gravity position 1503 is L13, and the distance from the line 1520 to the center of gravity position 1504 is L14, then the load (moment) F11 acting on the worker's waist 402 is: F11 = M11 × L11 + M12 × L12 + M13 × L13 + M14 × L × 14 (Math 3) It is expressed as follows.

[0101] On the other hand, in the state shown in Figure 15(b), the distance from the line 1521 to the center of gravity position 1501 is L11', the distance from the line 1521 to the center of gravity position 1502 is L12', and the distance from the line 1521 to the center of gravity position 1503 is L13´ If the distance from the line 1521 to the center of gravity position 1504 is L14', then the load (moment) F12 acting on the worker's waist position 402 is, F12=M11×L11´+M12×L12´+M13×L13´+M14×L14´ (Math 4) It is expressed as follows.

[0102] As explained in Figure 4 in Example 1, the bending angle A1 from the waist 402 to the torso 403 in the state where the knees 401 are extended and the body is bent forward (a) is greater than the angle A2 in the state where the knees 401 are bent forward (b). Therefore, L11, L12, and L13 in (a) in Figure 15 are greater than L11', L12', and L13' in (b), respectively, and as a result, F11 is greater than F12.

[0103] In other words, the moment on the hip 402 is greater when the knee 401 is extended and the person is bent forward as shown in Figure 15(a), compared to when the knee 401 is bent and the person is bent forward as shown in Figure 15(b), resulting in a greater load on the hip 402.

[0104] As a result, as explained in the graph of Figure 7 in Example 1, when the same task is repeated, fatigue accumulates due to the load on the worker's waist. The worker changes their posture to reduce the moment corresponding to the moment 701 on the waist 402. Therefore, the moment on the worker's waist 402 is relatively large in the initial stages, but gradually decreases over time. In other words, in this example as well, the degree of fatigue of the worker can be estimated from the change in the moment on the worker's waist 402.

[0105] Furthermore, similar to the first example, when the moment equivalent to the moment 701 acting on the worker's waist 402 falls below a certain value (the level indicated by the dotted line 702 in the example shown in Figure 7), the load recognition system 130 issues a warning to the receiving unit 150, thereby preventing a decrease in work efficiency and the occurrence of defects due to worker fatigue or working in an unstable posture. In addition, by preventing the accumulation of worker fatigue, it also leads to an improvement in the working environment.

[0106] Furthermore, in the example described above, the load recognition system 130 issues a warning to the receiving unit 150 when the load falls below the level indicated by the dotted line 702 in Figure 7. Conversely, the load recognition system 130 may also issue a warning to the receiving unit 150 when the moment corresponding to the moment 701 acting on the waist 402 in Figure 7 becomes greater than a certain level compared to the initial state.

[0107] Furthermore, the moment applied to the waist 402 is in its initial state Moment The load recognition system 130 may be configured to issue a warning to the receiving unit 150 if the value becomes significantly larger or smaller than a certain level (i.e., if the moment applied to the waist 402 changes by a certain level or more from the initial state).

[0108] The data processing flow using the load recognition system 130 in the work support system 100 of this embodiment, and the caution information 910 based on the notification information 140 displayed on the display screen 900 of the output unit 153, are the same as those described in Embodiment 1 using Figures 8 and 9, so their explanation will be omitted.

[0109] Furthermore, this embodiment may be combined with the method described in Embodiment 2 for estimating worker fatigue from the change in distance between the center of both feet and the projected position of the center of gravity of the upper body.

[0110] According to this embodiment, it is possible to prevent a decrease in work efficiency and the occurrence of defects caused by worker fatigue and working in unstable postures. Furthermore, by preventing the accumulation of worker fatigue, it also leads to an improvement in the working environment. [Explanation of Symbols]

[0111] 100 Work Support Systems 110 Sensor section 111 Attitude Sensor 112 Communications Department 130 Load Recognition System 131 Communications Department 132, 232, 332 Posture estimation section 133 Storage section 134 Load Estimation Unit 135 High load judgment section 136 Information generation section 137 Control Unit 150 Receiver 151 Communications Department 152 Control Unit 153 Output section 301 Upper arm condition estimation unit 302 Lumbar Condition Estimation Unit 303 Knee condition estimation unit 304 Load Data Estimation Unit

Claims

1. The communication unit that transmits and receives signals, A processing unit that processes the signal received by the communication unit. A load recognition device equipped with, The processing unit processes signals from multiple posture sensors attached to the work clothes worn by the worker, received by the communication unit, estimates the worker's posture, and determines the worker's high-load state from the time change of the estimated posture. The communication unit receives the signals from the multiple posture sensors attached to the work clothes worn by the worker, and transmits information regarding the worker's high-load state determined by the processing unit. The aforementioned processing unit, A posture estimation unit that processes the signals from the plurality of posture sensors received by the communication unit to estimate the posture of the worker, A load estimation unit estimates the worker's load from the worker's posture estimated by the posture estimation unit, A high-load determination unit determines that the worker is under high load when the load state of the worker estimated by the load estimation unit changes to a certain level or lower over time, The system further comprises: an information generation unit that generates information to be notified to the worker when the high load determination unit determines that the worker is in the high load state, The load estimation unit estimates the worker's load using the worker's posture estimated by the posture estimation unit and at least one of the length or weight of each body part of the worker. A load recognition device characterized by the following features.

2. A load recognition device according to claim 1, The processing unit further includes a storage unit that stores multiple datasets in which the length and weight of each body part differs, The load estimation unit selects a dataset from among the multiple datasets that is closest to the worker's body structure, and estimates the worker's load using the worker's posture and the selected dataset. A load recognition device characterized by the following features.

3. A load recognition device according to claim 1, The attitude estimation unit, An upper arm state estimation unit that detects the rotation angle of the worker's upper arm relative to the worker's shoulder, A waist state estimation unit that detects the bending angle of the worker's waist, A load recognition device characterized by comprising the following features.

4. A load recognition device according to claim 1, The attitude estimation unit, A waist state estimation unit that detects the bending angle of the worker's waist, A knee state estimation unit that detects the bending angle of the worker's knee, A load recognition device characterized by comprising the following features.

5. A load recognition device according to claim 1, The load recognition device is characterized in that the load estimation unit estimates the load on the worker's waist from the time change in the worker's posture estimated by the posture estimation unit.

6. A method for recognizing the state of a worker's load using a load recognition device equipped with a communication unit and a processing unit, The communication unit receives signals from multiple posture sensors attached to the work clothes worn by the worker. The processing unit processes the signals from the multiple posture sensors received by the communication unit to estimate the worker's posture and determines the worker's high-load state from the time change of the estimated posture. Based on the results determined by the processing unit, the communication unit transmits information regarding the worker's high-load state. In the processing of signals from the plurality of attitude sensors by the processing unit, The communication unit processes the signals from the multiple posture sensors received to estimate the worker's posture. Using the estimated posture of the worker and at least one of the length or weight of each body part of the worker, the worker's load is estimated. If the estimated workload of the worker changes to a certain level or lower over time, it is determined that the worker is in the high-load state. When the aforementioned worker determines that the high load condition is present, information is generated to notify the worker. A load recognition method characterized by the following:

7. A load recognition method according to claim 6, Multiple datasets with different lengths and weights for each of the aforementioned body parts are stored, In estimating the worker's workload, a dataset that closely matches the worker's physique is selected from among multiple datasets, and the worker's workload is estimated using the worker's posture and the selected dataset. A load recognition method characterized by the following:

8. A load recognition method according to claim 6, To estimate the posture of the worker, The rotation angle of the worker's upper arm relative to the worker's shoulder is detected, The bending angle of the worker's waist is detected, The worker's posture is estimated from the detected rotation angle of the worker's upper arm and the bending angle of the worker's waist. A load recognition method characterized by the following:

9. A load recognition method according to claim 6, To estimate the posture of the worker, The bending angle of the worker's waist is detected, The bending angle of the worker's knee is detected, The worker's posture is estimated from the detected bending angle of the worker's waist and the bending angle of the worker's knees. A load recognition method characterized by the following:

10. A load recognition method according to claim 8, A load recognition method characterized by estimating the load on the worker's waist from the estimated posture of the worker, as the load of the worker to be estimated.

11. A load recognition method according to claim 8, A load recognition method characterized in that the information transmitted from the communication unit regarding the worker's high-load state includes information that encourages the worker to recover from fatigue.

12. Multiple posture sensors attached to the work clothes worn by the worker, A load recognition device unit receives output signals from the plurality of posture sensors, determines the load state of the worker wearing the work clothes, and transmits the result of the determination. A receiving unit that receives the result of the determination transmitted from the load recognition unit and notifies the worker wearing the work clothes. A work support system equipped with, The aforementioned load recognition device unit is The communication unit that transmits and receives signals, The system comprises a processing unit that processes signals received by the communication unit, The processing unit processes signals from the multiple posture sensors attached to the work clothes worn by the worker, received by the communication unit, estimates the worker's posture, and determines the worker's high-load state from the time change of the estimated posture. The communication unit receives the signals from the multiple posture sensors attached to the work clothes worn by the worker, and transmits information regarding the worker's high-load state determined by the processing unit. The aforementioned processing unit, A posture estimation unit that processes signals from the plurality of posture sensors received by the communication unit to estimate the posture of the worker, A load estimation unit estimates the worker's load from the worker's posture estimated by the posture estimation unit, A high-load determination unit determines that the worker is in the high-load state if the worker's load state, estimated by the load estimation unit, changes to a certain level or below over time, The system further comprises: an information generation unit that generates information to be notified to the worker when the high load determination unit determines that the worker is in the high load state, The load estimation unit estimates the worker's load using the worker's posture estimated by the posture estimation unit and at least one of the length or weight of each body part of the worker. A work support system characterized by the following features.

13. A work support system according to claim 12, The processing unit further includes a storage unit that stores multiple datasets in which the length and weight of each body part differs, The load estimation unit selects a dataset from among the multiple datasets that is closest to the worker's body structure, and estimates the worker's load using the worker's posture and the selected dataset. A work support system characterized by the following features.

14. A work support system according to claim 12, The receiving unit receives the information transmitted from the communication unit and notifies the worker wearing the work clothes with sound, an image, or both. A work support system characterized by the following features.

Images