Information processing method, information processing device, control system, article manufacturing method, program, and recording medium

Abstract

Provided is an information processing method that enables an operator to execute a risk reduction measure.　When it is determined, on the basis of a detection result of a detection unit (106) that detects a state of an object (2), that the risk reduction measure is insufficient, a notification unit (107) notifies a method for executing the risk reduction measure, and the object (2) is prevented from using a device (202) controllable by a computer (108). When it is determined, on the basis of the detection result of the detection unit (106), that the risk reduction measure is satisfied, the object (2) is permitted to use the device (202).

Description

Information Processing Method, Information Processing Apparatus, Control System, Method for Manufacturing an Article, Program, and Recording Medium

[0001] The present disclosure relates to an information processing method for processing information, an information processing apparatus, a control system, a method for manufacturing an article, a program, and a recording medium.

[0002] For example, a system for ensuring the safety of workers working in a factory, a construction site, a work site, etc. has been proposed (Patent Document 1). In the safety monitoring system of Patent Document 1, it has been proposed to image workers, determine the safety equipment and safety behavior of the workers, and notify a warning when it is determined that there is a lack of safety equipment or non - fulfillment of safety behavior.

[0003] Japanese Patent Application Laid - Open No. 2023 - 48521

[0004] However, in the safety monitoring system of the above - mentioned Patent Document 1, although a lack of safety equipment or non - fulfillment of safety behavior of workers is warned, the worker who has received the warning does not know what actions should be taken, and there is a risk that the lack of safety equipment or non - fulfillment of safety behavior cannot be eliminated. That is, there is a possibility that risk reduction measures are not executed by the worker and the risk in the work is not reduced.

[0005] The present disclosure provides an information processing method, an information processing apparatus, a control system, a method for manufacturing an article, a program, and a recording medium that enable a worker to execute risk reduction measures.

[0006] One aspect of the present invention is an information processing method executed by a computer. When it is determined based on the detection result of a detection unit that detects the state of a target that a risk reduction measure is insufficient, a method for executing the risk reduction measure is notified by a notification unit, and a device controllable by the computer is made unusable by the target. When it is determined based on the detection result of the detection unit that the risk reduction measure is sufficient, the device is made usable by the target. This is the information processing method.

[0007] One aspect of the present invention is an information processing apparatus having a processing unit, wherein, if the processing unit determines, based on the detection result of a detection unit that detects the state of the target, that the risk reduction measures are insufficient, the processing unit notifies the target of a method for executing the risk reduction measures via a notification unit and prevents the target from using a device that can be controlled by the processing unit, and if the processing unit determines, based on the detection result of the detection unit that the risk reduction measures are sufficient, the processing unit allows the target to use the device.

[0008] According to this disclosure, workers will be able to implement risk reduction measures, thereby reducing risks in their work.

[0009] Other features and advantages of this disclosure will become apparent from the following description with reference to the accompanying drawings. In the accompanying drawings, the same or similar components are given the same reference numeral.

[0010] This is a diagram showing the monitoring system according to the first embodiment. This is a block diagram showing the functions of the monitoring device according to the first embodiment. This is a diagram showing the risk assessment results according to the first embodiment. This is a block diagram showing the detailed functions of the judgment unit of the monitoring device according to the first embodiment. This is a block diagram showing the detailed functions of the video language conversion unit in the judgment unit of the monitoring device according to the first embodiment. This is a flowchart showing the control of the monitoring device according to the first embodiment. This is a flowchart showing the control of the monitoring device according to the second embodiment. This is a flowchart showing the control of the monitoring device according to the third embodiment. This is a flowchart showing the control of the monitoring device according to the fourth embodiment. This is a flowchart showing the control of the monitoring device according to the fifth embodiment. This is a flowchart showing the control of the monitoring device according to the sixth embodiment. This is a diagram showing the monitoring system according to the seventh embodiment. This is a block diagram showing the functions of the monitoring device according to the seventh embodiment. This is a diagram showing the risk assessment results according to the seventh embodiment. This is a flowchart showing the control of the monitoring device according to the seventh embodiment. This is a diagram showing the monitoring system according to the eighth embodiment. This is a block diagram showing the functions of the monitoring device according to the eighth embodiment. This is a diagram showing the risk assessment results according to the eighth embodiment. This is a flowchart showing the control of the monitoring device according to the eighth embodiment. This is a diagram showing the monitoring system according to the ninth embodiment. This is a block diagram showing the functions of the monitoring device according to the ninth embodiment. This is a flowchart showing the control of the monitoring device according to the ninth embodiment.

[0011] <First Embodiment> Hereinafter, a first embodiment for implementing the present disclosure will be described with reference to Figures 1 to 5. Figure 1 is a diagram showing a monitoring system according to the first embodiment. Figure 2 is a block diagram showing the functions of the monitoring device according to the first embodiment. Figure 3 is a diagram showing the risk assessment results according to the first embodiment. Figure 4A is a block diagram showing the detailed functions of the judgment unit of the monitoring device according to the first embodiment. Figure 4B is a block diagram showing the detailed functions of the video language conversion unit in the judgment unit of the monitoring device according to the first embodiment. Figure 5 is a flowchart showing the control of the monitoring device according to the first embodiment. Note that the embodiments shown below are merely examples, and for example, the detailed configuration can be appropriately modified by those skilled in the art without departing from the spirit of the present disclosure. Also, the numerical values ​​discussed in this embodiment are reference values ​​and do not limit the present invention.

[0012] [Configuration of the monitoring system] First, the configuration of the monitoring system 1 according to the first embodiment will be explained using Figure 1. As shown in Figure 1, the monitoring system 1 is a control system comprising a monitoring device 100 as an information processing device and a work device 200 arranged in the monitoring area AR1, which is the area to be monitored.

[0013] [Configuration of the work apparatus] The work apparatus 200 in the first embodiment includes an electric saw (circular saw) 202 as a work tool, a workbench 203, and a clamp 205 for fixing a workpiece 204 such as wood to the workbench. The worker 2 approaches the electric saw 202 and workbench 203 to prepare for and perform the work, and actually performs the work.

[0014] [Configuration of the monitoring device] (Hardware configuration of the monitoring device) Next, the hardware configuration of the monitoring device 100 according to the first embodiment will be described. The monitoring device 100 includes a control device 108 as an information processing device, a camera 106 connected to the control device 108 as an imaging device that captures images, and a speaker 107 connected to the control device 108 that outputs sound. The camera 106, together with the detection unit 113 which will be described in detail later, detects the state of the target by photographing the target, and therefore constitutes a detection unit in a broad sense. The speaker 107, together with the notification unit 114 which will be described in detail later, provides notification and therefore constitutes a notification unit in a broad sense.

[0015] The control device 108 is a computer equipped with a CPU (Central Processing Unit) 108a, which is an example of a processor. The CPU 108a is an example of a processing unit (control unit). The control device 108 also includes a ROM (Read Only Memory) 108b, a RAM (Random Access Memory) 108c, and an HDD (Hard Disk Drive) 108d as storage units. Furthermore, the control device 108 includes an interface (I / F) 108e that serves as an input or output unit. A camera 106 and a speaker 107 are connected to the interface 108e, allowing image information (image data) as a detection result from the detection unit to be input from the camera 106, and allowing audio information (audio data) to be broadcast from the speaker 107 to be output.

[0016] Although not shown in the diagram, the control device 108 is also connected via interface 108e to a display device such as a display that shows the displayed image as a screen, and to external devices such as a keyboard and mouse that accept operation inputs. Furthermore, although not shown in the diagram, the control device 108 can also be connected to an external computer and the internet via interface 108e. The CPU 108a, ROM 108b, RAM 108c, HDD 108d, and interface 108e are connected to each other by a bus so that they can communicate information. Through interface 108e, the camera 106, speaker 107, display device, external devices, external computer, and the internet are all connected so that they can communicate information to each part of the control device 108.

[0017] (Configuration of Monitoring Device Functions) Next, the configuration of the monitoring device 100 functions will be explained using Figure 2. As shown in Figure 2, the monitoring device 100 has a configuration in which the various hardware parts described above function as an input unit 111, a judgment unit 112, a detection unit 113, and a notification unit 114, which function as function-achieving units. These function-achieving units function, for example, when the CPU 108a executes various programs stored in the ROM 108b. In other words, a program for executing the information processing method performed by the monitoring device 100, which is an information processing device, is stored in a computer-readable recording medium such as the ROM 108b, and each function (each process) is realized when that program is executed. Note that the recording medium is not limited to the ROM 108b mentioned above, but can also be a recording medium such as a CD or DVD, or a rewritable flash memory, as long as the monitoring device 100 is equipped with a device that can read these recording media.

[0018] Specifically, the input unit 111 operates as a function that allows the interface 108e to input information. The input unit 111 receives the risk assessment results 300, which will be described in more detail later, from an external computer or server via the Internet or by direct communication. The detection unit 113 operates as a function that acquires the video captured by the camera 106 of the monitoring area AR1 as a detection result. The notification unit 114 operates as a function (notification process) that provides notification by having the speaker 107 output sound to the monitoring area AR1. The judgment unit 112 operates as a function (judgment process) that allows the CPU 108a to determine whether or not risk reduction measures are insufficient based on the risk assessment results 300 and the video captured by the camera 106 of the monitoring area AR1. In other words, the judgment unit 112 detects whether or not risk reduction measures are insufficient based on the risk assessment results 300 and the video captured by the camera 106 of the monitoring area AR1.

[0019] (Details of Risk Assessment Results) Next, the details of the risk assessment results 300 will be explained using Figure 3. In this specification, risk assessment refers to the process of identifying, analyzing, and evaluating risks, and risk assessment results refer to the results evaluated by the risk assessment. That is, for example, a manager or the like may identify, analyze, and evaluate the risks that may occur in the monitoring area AR1 for the work equipment 200 and the worker 2 through a risk assessment, and as a result, the risk assessment results 300 are created. In this embodiment, the risk assessment results 300 are described as being generated by a human, but are not limited to this, and may be generated by artificial intelligence, etc. For example, the risk assessment results 300 that have already been generated by a human may be subjected to machine learning by artificial intelligence. Then, when a task that requires consideration of risks arises, the task to be considered is input into the artificial intelligence that has been trained using the previous risk assessment results 300. Then, information on the risks that are likely to occur in the task to be considered and risk reduction measures may be output as the risk assessment results 300. In other words, information related to risk assessment can be expressed verbally using machine learning models.

[0020] As shown in Figure 3, the risk assessment results 300 in this embodiment record the results of a risk assessment for, for example, four items 301 to 304 (four disaster risks) related to cutting work using an electric saw (circular saw) 202. The risk assessment results 300 records the item, work name, anticipated disaster, risk estimate, risk reduction measures, risk estimate after reduction measures, implementation timing of reduction measures, and residual risk, all associated with each item.

[0021] In detail, for the work named "circular saw cutting," the anticipated accidents are item 301 "saw chips getting into the eyes," item 302 "clothing getting caught," item 303 "workpiece bouncing back," and item 304 "incorrect cutting." As a risk estimate before implementing risk reduction measures for these items 301 to 304, the severity, probability, evaluation, and priority values ​​are stored as shown in Figure 3. For item 301, "saw chips getting into the eyes," the risk reduction measure stored is "wearing goggles," and for item 302, "clothing getting caught," the risk reduction measure stored is "wearing clothing correctly." Furthermore, for item 303, "workpiece bouncing back," the risk reduction measure stored is "clamping the workpiece," and for item 304, "incorrect cutting," the risk reduction measure stored is "safety check before setting up the circular saw."

[0022] Furthermore, the risk estimate after implementing risk reduction measures for items 301-304 stores the values ​​for severity, probability, assessment, and priority, as shown in Figure 3. In other words, the risk estimate after implementing risk reduction measures stores the severity, probability, assessment, and priority of the residual risk. Severity refers to the degree of severity that can be inflicted on workers, for example, or the degree of harm that can be inflicted on workers. Probability refers to the likelihood of the risk occurring, or its frequency.

[0023] Furthermore, the timing for implementing mitigation measures for items 301 and 302 is stored as "before work / during work," and the timing for implementing mitigation measures for items 303 and 304 is stored as "during work." Regarding item 301, "chips getting into eyes," the residual risk after implementing the risk mitigation measure of "wearing goggles" is stored as "pay attention to goggles shifting during work." Similarly, regarding item 302, "clothing getting caught," the residual risk after implementing the risk mitigation measure of "wearing clothing correctly" is stored as "pay attention to clothing becoming disheveled during work." Additionally, regarding item 303, "workpiece bouncing back," the residual risk after implementing the risk mitigation measure of "clamping the workpiece" is stored as "pay attention to ensuring the clamp does not loosen during cutting." Finally, regarding item 304, "miscutting," the residual risk after implementing the risk mitigation measure of "safety check before setting up the circular saw" is stored as "pay attention to ensuring that anything other than the workpiece does not come into contact with the blade during cutting."

[0024] As described above, the risk assessment result 300 can be said to include information on the risks in worker 2's work and information on worker 2's condition or work condition as risk reduction measures to mitigate those risks.

[0025] (Functions in the Judgment Unit of the Monitoring Device) Next, the configuration of the functions of the judgment unit 112 of the monitoring device 100 will be explained using Figures 4A and 4B. As shown in Figure 4A, the judgment unit 112 shown in Figure 2 has a configuration of a risk assessment processing unit 401, an image language processing unit 402, and a comparison unit 403 as function-achieving units, through the functioning of each of the hardware parts described above. Furthermore, as shown in Figure 4B, the image language processing unit 402 has a configuration of an image analysis unit 411, a skeletal analysis unit 412, an motion encoder 413, and an motion analysis unit 414 as function-achieving units, through the functioning of each of the hardware parts described above. Similarly, these function-achieving units function, for example, when the CPU 108a executes various programs stored in the ROM 108b.

[0026] Specifically, as shown in Figure 4A, the risk assessment processing unit 401 operates as a function that extracts necessary information from the risk assessment result 300 input from the input unit 111 and verbalizes it. One of the pieces of information to be extracted and verbalized is the risk reduction measures in the risk assessment result 300 shown in Figure 3. The extracted and verbalized information is generated as second information D2 and input to the comparison unit 403. If the risk assessment result 300 is already verbalized information, the risk assessment processing unit 401 extracts the necessary information and generates it as second information D2.

[0027] Although the input unit 111 has been described as primarily for inputting information from the risk assessment results 300, it may also be used to input information related to the risk assessment results 300 from external devices such as an external computer or external server. Examples of information related to the risk assessment results 300 include temperature, humidity, time, and date. This allows for the notification of warnings about long working hours or heatstroke prevention as risk reduction measures.

[0028] Meanwhile, the video language conversion unit 402 operates as a function to convert video information input from the detection unit 113 into language. For example, the HDD 108d has a pre-trained machine learning model recorded on it, and the video language conversion unit 402 recognizes the worker's equipment and work from the video information, converts it into language, generates it as first information D1, and inputs it to the comparison unit 403.

[0029] Here, the specific functions of the video language processing unit 402 will be explained using Figure 4B. The video language processing unit 402 has the configuration of a video analysis unit 411, a skeletal analysis unit 412, a motion encoder 413, and a motion analysis unit 414, which are function-achieving units through the operation of each hardware part described above. The video analysis unit 411 receives and analyzes the video as the imaging result (detection result) from the detection unit 113. The skeletal analysis unit 412 extracts the person who is the worker from the video analyzed by the video analysis unit 411 and obtains the person's skeletal information. The motion encoder 413 extracts the movement characteristics from the skeletal information obtained by the skeletal analysis unit 412. The motion analysis unit 414 classifies the type of movement from the movement characteristics extracted by the motion encoder 413, analyzes it using the machine learning model described above, and generates it as the first information D1. In other words, the detection result of the detection unit 113 is verbalized using the machine learning model. The verbalized first information D1 is then output to the comparison unit 403. The machine learning models used here are generally accepted behavioral analysis models, and for example, models generated using machine learning with artificial intelligence can be used.

[0030] [Control of the Monitoring Device] Next, the control of the monitoring device according to the first embodiment will be explained using Figure 5. As shown in Figure 5, the CPU 108a of the control device 108 of the monitoring device 100 starts this control when, for example, an administrator commands the start of monitoring. This control is performed at predetermined intervals (for example, the processing cycle of the CPU 108a) in the state before the worker 2 starts work (pre-work) and in the state when the worker 2 is performing work (work in progress). Furthermore, in this control, for example, all items (items 301 to 304) of the risk assessment result 300 shown in Figure 3 are judged simultaneously.

[0031] More specifically, as shown in Figure 5, when this control is started, the CPU 108a first receives the risk assessment result 300 shown in Figure 3 from the input unit 111 (I / F 108e) (S101). The input risk assessment result 300 is output to the comparison unit 403 as second information D2, which is verbalized by the risk assessment processing unit 401 as described above. The second information D2 is managed to be temporarily stored in the RAM 108c or HDD 108d.

[0032] Next, the CPU 108a acquires the image captured by the camera 106 (hereinafter also referred to as "video") as surveillance information via the detection unit 113 (S102). The acquired video is output to the comparison unit 403 as first information D1, which has been verbalized by the video language conversion unit 402 as described above. The first information D1 is also managed to be temporarily stored in the RAM 108c and HDD 108d.

[0033] Next, the CPU 108a uses the comparison unit 403 to compare the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 300 is verbalized). Then, it determines whether or not there are any problems with the result of this comparison (S103). In other words, the comparison unit 403 determines whether or not there are any inconsistencies between the first information D1 and the second information D2, and specifically, if there is information in the video captured by the camera 106 that matches all of the risk reduction measures, it determines that there are no problems (Yes in S103). In short, all of the risk reduction measures are satisfied. In this case, it determines whether or not there has been a command to end monitoring (S105), and if there has not been a command to end monitoring (No in S105), it returns to step S102, and if there has been a command to end monitoring (Yes in S105), this control is terminated.

[0034] On the other hand, in step S103, if even one piece of information in the video captured by camera 106 is inconsistent with the risk reduction measures (i.e., a risk reduction measure that has not been implemented), it is determined that there is a problem (No. in S103). In other words, it is determined that there is a shortage of risk reduction measures. In this case, CPU 108a instructs notification unit 114 to output the content of the inconsistent risk reduction measures as audio, that is, the risk reduction measures are notified by notification unit 114 (S104). After that, similarly, it is determined whether or not there has been a command to end monitoring (S105), and if there has been no command to end monitoring (No. in S105), it returns to step S102, and if there has been a command to end monitoring (Yes in S105), this control is terminated.

[0035] (Specific Examples of Control of the Monitoring Device) Next, a specific example of the control of the monitoring device 100 will be explained. In this first embodiment, as shown in Figure 1, an example will be described in which worker 2 cuts a workpiece 204 on a workbench 203 using an electric saw 202. In this work, items 301 to 304 of the risk assessment result 300 shown in Figure 3 are listed as accident risks, and the monitoring device 100 determines whether or not risk reduction measures have been taken for these items. Note that since monitoring of items 301 and 302 concerns the equipment of worker 2, monitoring is performed to detect whether or not risk reduction measures have been taken both before and during the work. In addition, since monitoring of items 303 and 304 concerns the actions of the worker, monitoring is performed to detect whether or not risk reduction measures have been taken during the work.

[0036] (In the case of item 301) For example, let's explain the case where the judgment unit 112 of the monitoring device 100 determines that the risk reduction measures for item 301 shown in Figure 3 have not been implemented. As shown in Figure 5, the risk assessment result 300 is input by the input unit 111 (S101) and temporarily stored in the RAM 108c or HDD 108d as verbalized second information D2. Here, when worker 2 enters, for example, a workroom and enters the monitoring area AR1, worker 2 is imaged by the detection unit 113 (camera 106) (S102), and the image is input to the comparison unit 403 as verbalized first information D1. At this time, for example, if worker 2 is not wearing goggles, there is a risk of "metal shavings getting into the eyes". The comparison unit 403 then determines a discrepancy between the first information D1 and the risk reduction measure "wearing goggles" in the second information D2 (see Figure 3). In other words, the monitoring information (video from camera 106) indicates that the risk reduction measure of "wearing goggles" has not been implemented, a deficiency in risk reduction measures is determined, and the comparison results indicate a problem (No. in S103). In response, the CPU 108a outputs audio information regarding the risk reduction measure of "wearing goggles" to the notification unit 114, that is, it notifies the worker 2 from speaker 107 with an audio message such as "Please wear goggles."

[0037] In the above example, we described a case where worker 2 enters the monitoring area AR1, that is, a notification regarding risk reduction measures is given before work begins, but this is not the only example. The control shown in Figure 5 continues to run until monitoring ends, so for example, if worker 2 removes their goggles during work, that state (action) is detected as monitoring information from the image of camera 106. Therefore, the comparison unit 403 determines that there is a discrepancy between the first information D1 and the risk reduction measure "wearing goggles" in the second information D2 (see Figure 3). Accordingly, even during such work, the speaker 107 will notify worker 2 with an audio message such as "Please wear your goggles." Furthermore, if worker 2 performs the risk reduction measure of "wearing goggles," the judgment unit 112 (comparison unit 403) determines that the risk reduction measure has been met, and the comparison result is judged to be no problem (Yes in S103). As a result, the audio notification regarding the risk reduction measure by the notification unit 114 (speaker 107) ends.

[0038] (In the case of item 302) For example, let's explain the case where the judgment unit 112 of the monitoring device 100 determines that the risk reduction measure for item 302 shown in Figure 3 has not been implemented. Similarly to the above, when worker 2 enters, for example, a work room and enters the monitoring area AR1, worker 2 is imaged by the detection unit 113 (camera 106) (S102), and the image is input to the comparison unit 403 as verbalized first information D1. At this time, for example, if worker 2 is not wearing their clothes correctly, there is a risk of "clothing getting caught." The comparison unit 403 then determines that there is a discrepancy between the first information D1 and the risk reduction measure "correct wearing of clothing" in the second information D2 (see Figure 3). In other words, it is determined that the risk reduction measure "correct wearing of clothing" has not been implemented in the monitoring information (image from camera 106), a deficiency in risk reduction measures is determined, and the result of the comparison is determined to be problematic (No. in S103). In response, the CPU 108a outputs audio information regarding the risk reduction measure of "wearing clothes correctly" to the notification unit 114, that is, it notifies the worker 2 from the speaker 107 with an audio message such as "Please wear your clothes correctly."

[0039] In this specific example, we have explained the case where worker 2 enters the monitoring area AR1, that is, when notification regarding risk reduction measures is given before work begins, but this is not the only case. The control shown in Figure 5 continues to run until monitoring ends, so for example, if worker 2's clothing becomes disheveled during work, this condition is detected as monitoring information from the image of camera 106. Therefore, the comparison unit 403 determines that there is a discrepancy between the first information D1 and the risk reduction measure "wear your clothes correctly" (see Figure 3) in the second information D2. Accordingly, even during such work, the speaker 107 will notify worker 2 with an audio message such as "Please wear your clothes correctly." Furthermore, if worker 2 performs the risk reduction measure of "wearing clothes correctly," the judgment unit 112 (comparison unit 403) determines that the risk reduction measure has been met, and the comparison result is judged to be no problem (Yes in S103). As a result, the audio notification regarding the risk reduction measure by the notification unit 114 (speaker 107) ends.

[0040] (In the case of item 303) For example, let's explain the case where the judgment unit 112 of the monitoring device 100 determines that the risk reduction measure for item 303 shown in Figure 3 has not been implemented. When the worker 2 is cutting the workpiece 204 with the electric saw 202, the detection unit 113 (camera 106) captures an image of the monitoring area AR1 (S102), and this image is input to the comparison unit 403 as verbalized first information D1. At this time, for example, if the workpiece 204 is not clamped and secured by the clamp 205, there is a risk of "workpiece bouncing back." The comparison unit 403 then determines that there is a discrepancy between the first information D1 and the risk reduction measure "clamping and securing the workpiece" in the second information D2 (see Figure 3). In other words, the monitoring information (image from camera 106) determines that the risk reduction measure "clamping and securing the workpiece" has not been implemented, a deficiency in risk reduction measures is determined, and the result of the comparison is determined to be problematic (No. in S103). In response, the CPU 108a outputs audio information to the notification unit 114 regarding the risk reduction measure of "clamping the workpiece," that is, it notifies the worker 2 from the speaker 107 with an audio message such as "Please secure the workpiece with a clamp."

[0041] Furthermore, the control shown in Figure 5 continues to run until monitoring is terminated. Therefore, when operator 2 performs the risk reduction measure of "clamping and securing the workpiece," the judgment unit 112 (comparison unit 403) determines that the risk reduction measure has been met. As a result, the comparison result is determined to be without problems (Yes in S103). Consequently, the audio notification regarding the risk reduction measure by the notification unit 114 (speaker 107) ends.

[0042] (In the case of item 304) For example, let's explain the case where the judgment unit 112 of the monitoring device 100 determines that the risk reduction measures for item 304 shown in Figure 3 have not been implemented. When worker 2 is cutting the workpiece 204 with the electric saw 202, the detection unit 113 (camera 106) captures an image of the monitoring area AR1 (S102), and this image is input to the comparison unit 403 as verbalized first information D1. At this time, for example, if the safety check before setting up the electric saw (circular saw) 202 has not been performed, there is a risk of "miscutting". The comparison unit 403 then determines that there is a discrepancy between the first information D1 and the risk reduction measure "safety check before setting up the circular saw" (see Figure 3) in the second information D2. In other words, it is determined that the risk reduction measure "safety check before setting up the circular saw" has not been implemented in the monitoring information (image from camera 106), a deficiency in risk reduction measures is determined, and the result of the comparison is determined to be problematic (No. in S103). In response, the CPU 108a outputs audio information regarding risk reduction measures for "safety check before setting up the circular saw" to the notification unit 114, that is, it notifies the worker 2 from the speaker 107 with an audio message such as "Please check for safety before setting up the circular saw."

[0043] Furthermore, the control shown in Figure 5 continues to run until monitoring is terminated. Therefore, when worker 2 performs the risk reduction measure of "safety check before setting up the circular saw," the judgment unit 112 (comparison unit 403) determines that the risk reduction measure has been met. As a result, the comparison result is determined to be no problem (Yes in S103). Consequently, the audio notification regarding the risk reduction measure by the notification unit 114 (speaker 107) ends.

[0044] [Summary of the First Embodiment] As described above, in this first embodiment, for example, the camera 106 monitors whether the risk reduction measures such as items 301 to 304 are correctly executed by the operator 2. And when this risk reduction measure is not correctly executed, the speaker 107 notifies the method of executing the correct risk reduction measure. That is, for example, if a risk assessment is performed for each site and risk reduction measures are provided independently, the equipment and actions required for the operator will differ for each site. However, in this embodiment, the work of the operator 2 (for example, handling of work equipment, etc.) is monitored, and based on the risk assessment result 300, when the work is incorrect, the correct work (how to handle the correct work equipment, etc.) is notified. In short, when it is determined that the safety equipment and safety actions of the operator are not fulfilled, the correct safety equipment and safety actions are notified. As a result, the operator 2 can understand the actions to be taken as risk reduction measures, can reduce the risk (can take safety actions), and in short, can reduce the risk estimated by the risk assessment. In other words, as shown in FIG. 3, for example, by ensuring that the operator 2 executes the risk reduction measures of items 301 to 304, the evaluation calculated by the sum of severity and probability is reduced from "6" to "2", or from "5" to "2". As a result, the priority of items 301 to 304 can also be reduced from "IV" to "I".

[0045] By the way, in the case of the publication of Japanese Patent Application Laid-Open No. 2023-48521, since it does not perform such operations as verbalizing the captured video and comparing it with the risk assessment result 300, it is impossible to recognize the occurrence of work. Therefore, it is impossible to distinguish the necessary equipment and actions to be taken for the risk reduction measures set in detail for each work, and there is a possibility of false detection and false notification, such as notifying non-workers who do not require equipment or actions. On the other hand, the monitoring device 100 according to this embodiment can recognize the occurrence of work and can reduce the occurrence of false detection and false notification.

[0046] <Second Embodiment>Next, a second embodiment in which the above first embodiment is partially modified will be described with reference to FIG. 6. FIG. 6 is a flowchart showing the control of the monitoring device according to the second embodiment. In the description of this second embodiment, the same reference numerals are used for the same parts as in the above first embodiment, and the description thereof will be omitted.

[0047] In the control of the monitoring device 100 according to this second embodiment, compared with the above first embodiment, the level of the notification intensity when notifying the risk reduction measures is changed according to the risk level when the risk reduction measures have not been executed.

[0048] Specifically, as shown in FIG. 6, when this control is started, first, the CPU 108a inputs the risk assessment result 300 shown in FIG. 3 from the input unit 111 (I / F 108e) (S201). The input risk assessment result 300 is output to the comparison unit 403 as the second information D2 verbalized by the risk assessment processing unit 401 as described above. Subsequently, the CPU 108a acquires the video imaged by the camera 106 as monitoring information by the detection unit 113 (S202). The acquired video is output to the comparison unit 403 as the first information D1 verbalized by the video verbalization unit 402 as described above.

[0049] Next, the CPU 108a compares the monitoring information (the first information D1 in which the video is verbalized) with the risk reduction measures (the second information in which the risk assessment result 300 is verbalized) by the comparison unit 403. Then, it is determined whether or not there is a problem with the comparison result (S203). That is, the comparison unit 403 determines whether there is a part that does not match between the first information D1 and the second information D2. Specifically, if there is information that matches all the risk reduction measures in the video imaged by the camera 106, it is determined that there is no problem (Yes in S203). In short, all the risk reduction measures are satisfied. In this case, it is determined whether there is a command to end the monitoring (S207). If there is no command to end the monitoring (No in S207), the process returns to step S202. If there is a command to end the monitoring (Yes in S207), this control is ended.

[0050] On the other hand, in step S203, if even one piece of information in the image captured by camera 106 is inconsistent with the risk reduction measures (i.e., a risk reduction measure that has not been implemented), it is determined that there is a problem (No. in S203). In other words, it is determined that there is a shortage of risk reduction measures.

[0051] In this second embodiment, first, the CPU 108a obtains the risk level from the input risk assessment result 300 in a state where no risk reduction measures have been implemented (when no risk reduction measures have been implemented) (S204). This risk level refers to, for example, the evaluation of the risk estimate shown in Figure 3, and will be "6" if the risk reduction measures for items 301, 302, and 304 have not been implemented, and will be "5" if the risk reduction measure for item 303 has not been implemented.

[0052] Next, the CPU 108a sets the notification intensity level according to the risk level during non-risk reduction measures (i.e., "6" or "5" in the example shown in Figure 3) (S205). In short, for example, if the risk level during non-risk reduction measures is "6", the notification intensity level is set to "6", and for example, if the risk level during non-risk reduction measures is "5", the notification intensity level is set to "5". In this embodiment, the notification intensity level means that, for example, a higher level means a louder volume, but it is not limited to this, and the content of the voice when making the notification may be in a stronger tone, etc. In other words, it is sufficient if the notification intensity received by the worker who receives the notification is different.

[0053] The CPU 108a then instructs the notification unit 114 to output the content of the mismatched risk reduction measures as an audio at the set notification intensity level (volume), that is, the risk reduction measures are notified by the notification unit 114 with a volume according to the level (S206). After that, it similarly determines whether or not there has been a command to end monitoring (S207). If there has been no command to end monitoring (No in S207), it returns to step S202. If there has been a command to end monitoring (Yes in S207), this control is terminated.

[0054] [Summary of the Second Embodiment] As described above, in this second embodiment, if risk reduction measures are not properly implemented, the speaker 107 will notify the operator of the method for implementing the correct risk reduction measures. Furthermore, in addition to this, the level of notification intensity when notifying the operator of the method for implementing the correct risk reduction measures can be changed according to the risk level at the time of non-risk reduction measures. This allows for stronger notification to be given to the operator 2 when there is a shortage of risk reduction measures that pose a higher risk, thereby improving safety.

[0055] <Third Embodiment> Next, a third embodiment, which is a modified version of the first and second embodiments described above, will be explained with reference to Figure 7. Figure 7 is a flowchart showing the control of the monitoring device according to the third embodiment. In this explanation of the third embodiment, the same reference numerals are used for parts that are the same as those in the first and second embodiments, and their explanations are omitted.

[0056] In the control of the monitoring device 100 according to this third embodiment, compared to the first and second embodiments, if risk reduction measures are not being implemented, the device notifies the user of a method for implementing the correct risk reduction measures and turns off the power to the electric saw 202, which is a work device.

[0057] More specifically, as shown in Figure 7, when this control is started, first the CPU 108a receives the risk assessment result 300 shown in Figure 3 from the input unit 111 (I / F 108e) (S301). The input risk assessment result 300 is output to the comparison unit 403 as second information D2, which has been verbalized by the risk assessment processing unit 401 as described above. Next, the CPU 108a acquires the video captured by the camera 106 as monitoring information using the detection unit 113 (S302). The acquired video is output to the comparison unit 403 as first information D1, which has been verbalized by the video verbalization unit 402 as described above.

[0058] Next, the CPU 108a compares the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 300 is verbalized) using the comparison unit 403. Then, it determines whether or not there are any problems with the result of this comparison (S303). In other words, the comparison unit 403 determines whether or not there are any inconsistencies between the first information D1 and the second information D2, and specifically, if there is information in the video captured by the camera 106 that matches all of the risk reduction measures, it determines that there are no problems (Yes in S303). In short, all of the risk reduction measures are met and satisfied. In this case, the power to the electric saw 202, which is the work equipment, is turned on (and kept on) (S304), that is, the electric saw 202 is made usable. Then, it is determined whether or not a command to terminate monitoring has been received (S307). If there is no command to terminate monitoring (No in S307), the process returns to step S302. If there is a command to terminate monitoring (Yes in S307), this control process ends.

[0059] On the other hand, in step S303, if even one piece of information in the image captured by camera 106 is inconsistent with the risk reduction measures (i.e., a risk reduction measure that has not been implemented), it is determined that there is a problem (No. in S303). In other words, it is determined that there is a shortage of risk reduction measures.

[0060] The CPU 108a then turns off the power to the electric saw 202, which is the work equipment (S305), making the electric saw 202 unusable. The CPU 108a then instructs the notification unit 114 to output the content of the mismatched risk reduction measures as an audio, that is, the risk reduction measures are notified by the notification unit 114 (S306). After that, it similarly determines whether or not there has been a command to end monitoring (S307), and if there has been no command to end monitoring (No in S307), it returns to step S302. After this, for example, if the worker 2 has executed the risk reduction measures and all of the risk reduction measures have been satisfied (Yes in S303), the power to the electric saw 202, which is the work equipment, is turned on (S304). If there has been a command to end monitoring (Yes in S307), this control is terminated.

[0061] [Summary of the Third Embodiment] As described above, in this third embodiment, if risk reduction measures are not properly implemented, the speaker 107 announces the method for implementing the correct risk reduction measures, and in addition, the power to the electric saw 202, which is the work equipment, is turned off. This prevents the worker 2 from operating the electric saw 202 when risk reduction measures are insufficient, thereby improving safety. In this embodiment, the power to the electric saw 202, which is the work equipment, is turned on and off, but it is not limited to this, and the power to the entire device, i.e., the work device 200, may also be turned on and off.

[0062] <Fourth Embodiment> Next, a fourth embodiment, which is a modified version of the first to third embodiments described above, will be explained with reference to Figure 8. Figure 8 is a flowchart showing the control of the monitoring device according to the fourth embodiment. In this explanation of the fourth embodiment, the same reference numerals are used for parts that are the same as those in the first to third embodiments, and their explanations are omitted.

[0063] In the control of the monitoring device 100 according to this fourth embodiment, compared to the first to third embodiments, residual risk is notified when risk reduction measures have been implemented.

[0064] More specifically, as shown in Figure 8, when this control is started, first the CPU 108a receives the risk assessment result 300 shown in Figure 3 from the input unit 111 (I / F 108e) (S401). The input risk assessment result 300 is output to the comparison unit 403 as second information D2, which has been verbalized by the risk assessment processing unit 401 as described above. Next, the CPU 108a acquires the video captured by the camera 106 as monitoring information using the detection unit 113 (S402). The acquired video is output to the comparison unit 403 as first information D1, which has been verbalized by the video verbalization unit 402 as described above.

[0065] Next, the CPU 108a uses the comparison unit 403 to compare the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 300 is verbalized). Then, it determines whether or not there are any problems with the result of this comparison (S403). In other words, the comparison unit 403 determines whether or not there are any inconsistencies between the first information D1 and the second information D2, and specifically, if there is information in the video captured by the camera 106 that matches all of the risk reduction measures, it determines that there are no problems (Yes in S403). In short, all of the risk reduction measures are met and satisfied.

[0066] In this fourth embodiment, the CPU 108a instructs the notification unit 114 to output the content of the residual risk shown in Figure 3 as an audio, that is, the residual risk is notified by the notification unit 114 (S404). When notifying the content of the residual risk, for example, the residual risk of items with a high evaluation in the risk estimate after the reduction measures may be notified, or multiple residual risks may be notified repeatedly in order. Then, it is determined whether or not there has been a command to end monitoring (S406), and if there has been no command to end monitoring (No in S406), the process returns to step S402, and if there has been a command to end monitoring (Yes in S406), this control is terminated.

[0067] Residual risk refers to risks that cannot be completely eliminated even if worker 2 implements all risk reduction measures. For example, even if the risk reduction measure in item 301, "wearing goggles," is implemented, there is a risk that the goggles may shift during work and cutting dust may get into the eyes, so worker 2 is informed of the residual risk of "paying attention to goggle shifting during work." Similarly, even if the risk reduction measure in item 302, "wearing clothing correctly," is implemented, there is a risk that clothing may become disheveled during work and get caught in the electric saw 202, so worker 2 is informed of the residual risk of "paying attention to clothing becoming disheveled during work." Furthermore, even if the risk reduction measure in item 303, "clamping the workpiece," is implemented, there is a risk that the clamp may loosen during work and the workpiece may bounce back, so worker 2 is informed of the residual risk of "paying attention to ensuring the clamp does not loosen during cutting." Furthermore, even if the risk reduction measure in item 304, "safety check before setting up the circular saw," is carried out, there is still a risk of miscutting of the workpiece. Therefore, the worker 2 is informed of the residual risk, "be careful not to let anything other than the workpiece come into contact with the blade during cutting."

[0068] On the other hand, in step S403, if even one piece of information in the image captured by camera 106 is inconsistent with the risk reduction measures (i.e., a risk reduction measure that has not been implemented), it is determined that there is a problem (No. in S403). In other words, it is determined that there is a shortage of risk reduction measures.

[0069] The CPU 108a then instructs the notification unit 114 to output the content of the mismatched risk reduction measures as an audio signal, that is, the risk reduction measures are notified by the notification unit 114 (S405). After that, it similarly determines whether or not there has been a command to end monitoring (S406), and if there has been no command to end monitoring (No in S406), it returns to step S402. If there has been a command to end monitoring (Yes in S406), this control is terminated.

[0070] [Summary of the Fourth Embodiment] As described above, in this fourth embodiment, the speaker 107 notifies the details of the residual risk when the risk reduction measures are properly implemented. This makes it possible to reduce the risk related to residual risk even when the implementation of risk reduction measures is satisfied, thereby further improving safety.

[0071] <Fifth Embodiment> Next, a fifth embodiment, which is a modified version of the first to fourth embodiments described above, will be explained with reference to Figure 9. Figure 9 is a flowchart showing the control of the monitoring device according to the fifth embodiment. In this explanation of the fifth embodiment, the same reference numerals are used for parts that are the same as those in the first to fourth embodiments, and their explanations are omitted.

[0072] In the control of the monitoring device 100 according to this fifth embodiment, compared to the first to fourth embodiments, it determines whether or not the work environment is tidy and encourages tidying of the work environment.

[0073] More specifically, as shown in Figure 9, when this control is started, first the CPU 108a receives the risk assessment result 300 shown in Figure 3 from the input unit 111 (I / F 108e) (S501). The input risk assessment result 300 is output to the comparison unit 403 as second information D2, which has been verbalized by the risk assessment processing unit 401 as described above. Next, the CPU 108a acquires the video captured by the camera 106 as monitoring information using the detection unit 113 (S502). The acquired video is output to the comparison unit 403 as first information D1, which has been verbalized by the video verbalization unit 402 as described above.

[0074] Next, the CPU 108a uses the comparison unit 403 to compare the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 300 is verbalized). Then, it determines whether or not there are any problems with the result of this comparison (S503). In other words, the comparison unit 403 determines whether or not there are any inconsistencies between the first information D1 and the second information D2, and specifically, if there is information in the video captured by the camera 106 that matches all of the risk reduction measures, it determines that there are no problems (Yes in S503). In short, all of the risk reduction measures are met and satisfied.

[0075] Next, the CPU 108a determines from the monitoring information (first information D1 in which the video is verbalized) whether or not the work environment in the monitoring area AR1 (see Figure 1) is tidy (S505). For example, the tidy state is determined by setting an indicator such as whether unnecessary items appear in the monitoring information for a certain period of time relative to the reference image. In short, a disorganized work environment is equivalent to determining that the work environment has deteriorated. In this fifth embodiment, a determination of the tidyness of the work environment is described, but it is not limited to this, and it is also acceptable to monitor and determine the organization, cleaning, cleanliness, and discipline of workers (so-called 5S) of the work environment. Furthermore, instead of determining the work environment of the entire monitoring area AR1, it is also acceptable to set a work area and determine whether or not the work environment is tidy in the work area as shown in the monitoring information.

[0076] If the CPU 108a determines that the work environment is tidy (improved) (Yes in S505), it then determines whether or not a command to terminate monitoring has been received (S507). If there is no command to terminate monitoring (No in S507), it returns to step S502. If there is a command to terminate monitoring (Yes in S507), this control is terminated.

[0077] On the other hand, in step S503 above, if even one piece of information in the image captured by the camera 106 is inconsistent with the risk reduction measures (i.e., a risk reduction measure that has not been implemented), it is determined that there is a problem (No. in S503). In other words, it is determined that there is a shortage of risk reduction measures.

[0078] The CPU 108a then instructs the notification unit 114 to output the details of the mismatched risk reduction measures as an audio signal, that is, the risk reduction measures are notified by the notification unit 114 (S504). After that, it similarly determines whether or not there has been a command to end monitoring (S507), and if there has been no command to end monitoring (No in S507), it returns to step S502. If there has been a command to end monitoring (Yes in S507), this control is terminated.

[0079] Furthermore, in step S505, if it is determined that the work environment is not tidy (has deteriorated) (No. in S505), the notification unit 114 is instructed to output an audio message instructing the worker to tidy up the work environment. In other words, the notification unit 114 notifies the worker 2 of a message urging them to tidy up (improve) the work environment (S506). After that, it is determined whether or not there has been a command to end monitoring (S507), and if there has been no command to end monitoring (No. in S507), the process returns to step S502. If there has been a command to end monitoring (Yes in S507), this control is terminated.

[0080] [Summary of the Fifth Embodiment] As described above, in this fifth embodiment, if it is determined that the work environment is not tidy, the speaker 107 broadcasts a message prompting the worker to tidy up the work environment. This reduces the risks that may arise from a disorganized work environment, specifically the risks of errors and accidents, and thus improves safety.

[0081] <Sixth Embodiment> Next, a sixth embodiment, which is a modified version of the first to fifth embodiments described above, will be explained with reference to Figure 10. Figure 10 is a flowchart showing the control of the monitoring device according to the sixth embodiment. In this explanation of the sixth embodiment, the same reference numerals are used for parts that are the same as those in the first to fifth embodiments, and their explanations are omitted.

[0082] In the control of the monitoring device 100 according to this sixth embodiment, compared to the first to fifth embodiments, the device predicts the occurrence of work related to the items in the risk assessment result 300 and determines whether risk reduction measures are being implemented for the predicted work.

[0083] More specifically, as shown in Figure 10, when this control is started, first the CPU 108a receives the risk assessment result 300 shown in Figure 3 from the input unit 111 (I / F 108e) (S601). The input risk assessment result 300 is output to the comparison unit 403 as second information D2, which has been verbalized by the risk assessment processing unit 401 as described above. Next, the CPU 108a acquires the video captured by the camera 106 as monitoring information using the detection unit 113 (S602). The acquired video is output to the comparison unit 403 as first information D1, which has been verbalized by the video verbalization unit 402 as described above.

[0084] Next, the CPU 108a uses the comparison unit 403 to predict the occurrence of tasks related to items 301 to 304 in the risk assessment result 300 from the monitoring information (first information D1, in which the video is verbalized), and generates information about the predicted tasks. Subsequently, it compares the information about the predicted tasks with the risk reduction measures (second information, in which the risk assessment result 300 is verbalized). Then, it determines whether or not there is a problem with the result of this comparison (S603). In other words, the comparison unit 403 determines whether or not there is a discrepancy between the tasks predicted from the first information D1 and the second information D2. Specifically, if there is information that matches the risk reduction measures in the tasks predicted from the video captured by the camera 106, it determines that there is no problem (Yes in S603). In short, if the risk reduction measures have been implemented in the predicted tasks, it means that all the risk reduction measures have been met and satisfied. In this case, it is determined whether or not a command to terminate monitoring has been issued (S605). If there is no command to terminate monitoring (No in S605), the process returns to step S602. If there is a command to terminate monitoring (Yes in S605), this control process is terminated.

[0085] On the other hand, in step S603 above, if there is any information in the work predicted from the image captured by the camera 106 that does not match the risk reduction measures (i.e., risk reduction measures that have not been implemented), it is determined that there is a problem (No. in S603). In other words, it is determined that there is a shortage of risk reduction measures.

[0086] The CPU 108a then instructs the notification unit 114 to output the content of the mismatched risk reduction measures as an audio signal, that is, the risk reduction measures are notified by the notification unit 114 (S604). After that, it similarly determines whether or not there has been a command to end monitoring (S605), and if there has been no command to end monitoring (No in S605), it returns to step S602. If there has been a command to end monitoring (Yes in S605), this control is terminated.

[0087] [Summary of the Sixth Embodiment] As described above, in this sixth embodiment, based on the detection results of the detection unit 113, the occurrence of work related to the items in the risk assessment result 300 is predicted, and if the execution of risk reduction measures does not coincide with the predicted work, the method for executing the risk reduction measures is notified. In other words, it is predicted that risk reduction measures will not be executed, and in that case, the method for executing the risk reduction measures is notified before the work occurs. This makes it possible to prevent the failure to execute risk reduction measures in future work and improve safety.

[0088] <Seventh Embodiment> Next, a seventh embodiment, which is a modified version of the first to sixth embodiments described above, will be explained with reference to Figures 11 to 14. Figure 11 is a diagram showing the monitoring system according to the seventh embodiment. Figure 12 is a block diagram showing the functions of the monitoring device according to the seventh embodiment. Figure 13 is a diagram showing the risk assessment results according to the seventh embodiment. Figure 14 is a flowchart showing the control of the monitoring device according to the seventh embodiment. In this explanation of the seventh embodiment, the same reference numerals are used for parts similar to those in the first to sixth embodiments, and their explanations are omitted.

[0089] [Configuration of the monitoring system according to the seventh embodiment] The monitoring system 1001 according to the seventh embodiment includes a monitoring device 1100 as an information processing device that monitors the monitoring areas AR11 and AR12, as shown in Figure 11. That is, the monitoring system 1001 monitors forklifts 601 and 602 and box pallets 603 and 604 as objects placed in the monitoring areas AR11 and AR12. The monitoring system 1001 may be connected to the forklifts 601 and 602 and box pallets 603 and 604 in a communicative manner, and in a broader sense, it may include forklifts 601 and 602 and box pallets 603 and 604.

[0090] Forklifts 601 and 602 are self-propelled moving objects that can, for example, lift and transport box pallets 603 and 604. In this seventh embodiment of the monitoring system 1001, a system that monitors two monitoring areas AR11 and AR12 is described, but it is not limited to two areas; it may be capable of monitoring three or more monitoring areas. Also, two forklifts 601 and 602 and two box pallets 603 and 604 are described as an example, but it is not limited to these; there may be one or more forklifts or box pallets.

[0091] The monitoring device 1100 includes, as multiple detection units connected to the control device 108 shown in Figure 1, a camera 1106A that images the monitoring area AR11 and a camera 1106B that images the monitoring area AR12. The monitoring device 1100 also includes, as multiple notification units connected to the control device 108 shown in Figure 1, a speaker 1107A that outputs (notifies) sound to the monitoring area AR11 and a speaker 1107B that outputs (notifies) sound to the monitoring area AR12.

[0092] Next, the configuration of the functions of the monitoring device 1100 will be explained using Figure 12. As shown in Figure 12, the monitoring device 1100 has a configuration in which the various parts of the hardware of the control device 108 function as an input unit 1111, a determination unit 1112, a first detection unit 1113A, and a second detection unit 113B, which function as function-achieving units. The monitoring device 1100 also has a configuration in which the various parts of the hardware of the control device 108 function as a first notification unit 1114A and a second notification unit 1114B, which function as function-achieving units. These function-achieving units function, for example, when the CPU 108a executes various programs stored in the ROM 108b.

[0093] Specifically, the input unit 1111 operates as a function that allows the interface 108e to input information. The input unit 1111 inputs, for example, the risk assessment result 1300 shown in Figure 13. The first detection unit 1113A is capable of detecting forklifts 601 and 602 and operates as a function that acquires the image captured by the camera 1106A of the monitoring area AR11 as a detection result. The second detection unit 1113B is capable of detecting forklifts 601 and 602 and operates as a function that acquires the image captured by the camera 1106B of the monitoring area AR12 as a detection result. The first notification unit 1114A operates as a function that provides notification that the speaker 1107A outputs sound to the monitoring area AR11. The second notification unit 1114B operates as a function that provides notification that the speaker 1107B outputs sound to the monitoring area AR12. The determination unit 1112 operates as a function that determines whether or not risk reduction measures are insufficient, based on the risk assessment result 1300 and the images captured by cameras 1106A and 1106B of the monitoring areas AR11 and AR12, as determined by the CPU 108a.

[0094] [Details of the Risk Assessment Results According to the Seventh Embodiment] Next, the details of the risk assessment results 1300 according to the seventh embodiment will be explained using Figures 11 and 13. In the monitoring system 1001 according to this seventh embodiment, as shown in Figure 11, forklifts 601 and 602 perform the work of transporting box pallets 603 and 604, respectively. The box pallets 603 and 604 may be at a height that obstructs the view of the operator of the forklifts 601 and 602. In this case, if the forklifts 601 and 602 are driven forward, visibility will be poor and there is a risk of collision accidents, etc. Therefore, when transporting tall box pallets 603 and 604, it is necessary to take measures to move (drive) the forklifts 601 and 602 backward.

[0095] Therefore, in the risk assessment result 1300, under the item for the work named "Moving tall box pallets," the anticipated accident is "Contact accident due to poor visibility," and the risk reduction measure is "Move in reverse." In addition, under this item, the severity, probability, evaluation, and priority values ​​are stored as shown in Figure 13, representing the risk estimate before the implementation of the risk reduction measures. Furthermore, under this item, the severity, probability, evaluation, and priority values ​​are stored as shown in Figure 13, representing the risk estimate after the implementation of the risk reduction measures.

[0096] [Control of Monitoring Device] Next, the control of the monitoring device according to the seventh embodiment will be explained using Figure 14. In the first embodiment described above, the camera 106 monitored the work that occurred at a specific location. However, in this seventh embodiment, since the forklifts 601 and 602, as movable objects, can be moved to any location within the site where work is performed, for example, a situation in which risk reduction measures are insufficient may occur at any location within the site. Furthermore, because the forklifts 601 and 602 are moved, it is difficult to monitor all locations with just one camera. Therefore, in this seventh embodiment, multiple cameras 1106A and 1106B monitor whether the risk reduction measures are being operated correctly. If the risk reduction measures are not being operated correctly, speakers 1107A and 1107B will announce the method for implementing the correct risk reduction measures.

[0097] More specifically, as shown in Figure 14, when this control is started, the CPU 108a first receives the risk assessment result 1300 shown in Figure 13 from the input unit 1111 (I / F 108e) (S701). The input risk assessment result 1300 is output to the comparison unit 403 as second information D2, which is verbalized by the risk assessment processing unit 401 as described above.

[0098] First, let's explain the processing of the monitoring area AR11. The CPU 108a acquires video footage captured by the camera 1106A as monitoring information for the monitoring area AR11 using the first detection unit 1113A (S702). The acquired video footage is output to the comparison unit 403 as first information D1, which has been verbalized by the video language conversion unit 402 as described above.

[0099] Next, the CPU 108a uses the comparison unit 403 to compare the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 1300 is verbalized). Then, it determines whether or not there are any problems with the result of this comparison (S703). In other words, the comparison unit 403 determines whether or not there are any inconsistencies between the first information D1 and the second information D2, and specifically, if there is information in the video captured by the camera 1106A that matches the risk reduction measures, it determines that there are no problems (Yes in S703). In short, all risk reduction measures are met and satisfied. In this case, it determines whether or not there has been a command to end monitoring (S708), and if there has not been a command to end monitoring (No in S708), it returns to step S702, and if there has been a command to end monitoring (Yes in S708), this control is terminated.

[0100] On the other hand, in step S703, if there is information in the image captured by camera 1106A that does not match the risk reduction measures (i.e., risk reduction measures that have not been implemented), it is determined that there is a problem (No. in S703). In other words, it is determined that there is a shortage of risk reduction measures. The CPU 108a then instructs the first notification unit 1114A to output the content of the inconsistent risk reduction measures as audio, that is, the first notification unit 1114A notifies the monitoring area AR11 of the risk reduction measures (S704). After that, it is similarly determined whether or not there has been a command to end monitoring (S708), and if there has been no command to end monitoring (No. in S708), it returns to step S702, and if there has been a command to end monitoring (Yes in S708), this control is terminated.

[0101] Next, the processing of the monitoring area AR12 will be explained. Simultaneously with the processing of the monitoring area AR11 described above, the CPU 108a acquires images captured by the camera 1106B as monitoring information for the monitoring area AR12 using the second detection unit 1113B (S705). Each of the acquired images is output to the comparison unit 403 as first information D1, which has been verbalized by the image language conversion unit 402 as described above.

[0102] Next, the CPU 108a uses the comparison unit 403 to compare the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 1300 is verbalized). It then determines whether or not there are any problems with the result of this comparison (S706). In other words, the comparison unit 403 determines whether or not there are any inconsistencies between the first information D1 and the second information D2. Specifically, if there is information in the video captured by the camera 1106B that matches the risk reduction measures, it determines that there are no problems (Yes in S706). In short, all risk reduction measures are met and satisfied. In this case, it determines whether or not there has been a command to end monitoring (S708). If there has not been a command to end monitoring (No in S708), it returns to step S702. If there has been a command to end monitoring (Yes in S708), this control is terminated.

[0103] On the other hand, in step S706, if there is information in the image captured by camera 1106B that does not match the risk reduction measures (i.e., risk reduction measures that have not been implemented), it is determined that there is a problem (No in S706). In other words, it is determined that there is a shortage of risk reduction measures. The CPU 108a then instructs the second notification unit 1114B to output the content of the inconsistent risk reduction measures as audio, that is, the second notification unit 1114B broadcasts the risk reduction measures to the monitoring area AR12 (S707). After that, it is similarly determined whether or not there has been a command to end monitoring (S708). If there has not been a command to end monitoring (No in S708), it returns to step S702, and if there has been a command to end monitoring (Yes in S708), this control is terminated.

[0104] [Summary of the Seventh Embodiment] As described above, in this seventh embodiment, a deficiency in risk reduction measures is determined in each of the monitoring areas AR11 and AR12, and a method for implementing the correct risk reduction measures is announced via speaker in the monitoring area where the deficiency was determined. As a result, even if, for example, the forklifts 601 and 602, which are moving objects, move between the monitoring areas AR11 and AR12, a deficiency in risk reduction measures can be determined, and a method for implementing the correct risk reduction measures at that location can be announced.

[0105] In this seventh embodiment, we have described a system for monitoring moving objects that can move between multiple monitoring areas, such as a forklift, but the system is not limited to this. For example, if the area where a worker is working or a work device such as a robot cannot be captured within the field of view of a single camera, it is conceivable that multiple cameras could be used to monitor the worker's work or the range of motion of the work device. In this case, the worker or work device can be considered a moving object that moves between the imaging ranges (fields of view) of multiple cameras.

[0106] <Eighth Embodiment> Next, an eighth embodiment, which is a modified version of the first to seventh embodiments described above, will be explained with reference to Figures 15 to 18. Figure 15 is a diagram showing the monitoring system according to the eighth embodiment. Figure 16 is a block diagram showing the functions of the monitoring device according to the eighth embodiment. Figure 17 is a diagram showing the risk assessment results according to the eighth embodiment. Figure 18 is a flowchart showing the control of the monitoring device according to the eighth embodiment. In this explanation of the eighth embodiment, the same reference numerals are used for parts similar to those in the first to seventh embodiments, and their explanations are omitted.

[0107] [Configuration of the monitoring system according to the eighth embodiment] As shown in Figure 15, the monitoring system 2001 includes a monitoring device 2100 as an information processing device that monitors the monitoring area AR 21. The monitoring device 210 also monitors a work device 2200 having a robot 2201 as an object placed in the monitoring area AR 21. Furthermore, the monitoring system 2001 according to this embodiment is communicatively connected to a control device (not shown) of the robot 2201 of the work device 2200, and in a broad sense, includes the work device 2200. In this case, the monitoring system 2001 constitutes a control system comprising the monitoring device 1100 and the robot 2201. This monitoring system 2001 as a control system can execute a method for manufacturing articles using the robot 2201 of the work device 2200.

[0108] Robot 2201 is equipped with monitoring sensors (e.g., cameras, infrared sensors, ultrasonic sensors, etc.) not shown in the diagram, and monitors the robot's operating range to prevent workers from approaching. However, the robot's posture and trajectory create blind spots for the monitoring sensors, resulting in unmonitored areas AR22, AR22 within the robot's operating range that cannot be monitored. Monitoring device 2100 monitors the monitoring area AR21, which includes these unmonitored areas AR22, AR22.

[0109] Furthermore, the monitoring device 2100 includes a camera 2106 that captures images of the monitoring area AR21, as a detection unit connected to the control device 108 shown in Figure 1. The monitoring device 2100 also includes a speaker 2107 that outputs (notifies) sound to the monitoring area AR21, as a notification unit connected to the control device 108 shown in Figure 1.

[0110] Next, the configuration of the monitoring device 2100's functions will be explained using Figure 16. As shown in Figure 16, the monitoring device 2100 has a configuration in which the various parts of the hardware of the control device 108 function as an input unit 2111, a device status input unit 2115, a judgment unit 2112, a detection unit 2113, and a notification unit 2114, which function as function-achieving units. These function-achieving units function, for example, when the CPU 108a executes various programs stored in the ROM 108b.

[0111] Specifically, the input unit 2111 and the device status input unit 2115 function as the interface 108e inputting information. The input unit 2111 inputs, for example, the risk assessment result 2300 shown in Figure 17. The device status input unit 2115 communicates with a control device (not shown) of the work device 2200 to input the status of the work device 2200 and the information monitored by the robot 2201's monitoring sensor. The detection unit 2113 functions to acquire the image captured by the camera 2106 of the monitoring area AR 21 as a detection result. The notification unit 2114 functions to provide notification by having the speaker 2107 output sound to the monitoring area AR 21. The judgment unit 2112 functions to have the CPU 108a determine whether or not risk reduction measures are insufficient based on the risk assessment result 2300, the status of the robot 2201 and the information from the monitoring sensor, and the image captured by the camera 2106. Furthermore, the device status input unit 2115 can be broadly described as a detection unit that acquires first information, since it detects the status of the target work device 2200.

[0112] In this eighth embodiment, the determination unit 2112 does not verbalize the video acquired from the camera 2106 as monitoring information. That is, it determines whether or not to perform risk reduction measures based on the video captured by the camera 2106 as the input unit 2111 and the information from the robot 2201's monitoring sensor from the robot 2201's control device as the device status input unit 2115. In short, determining whether or not to perform risk reduction measures in the risk assessment result 2300 described later only requires determining whether or not a worker is approaching the robot 2201, so there is no need to verbalize various types of information. It should be noted that whether or not collaborative work is in progress can also be determined from the state of the robot 2201 (work state or work phase), etc., which are set in advance. For example, if the robot is in a state where it is receiving parts from a worker, it can be determined that collaborative work is in progress.

[0113] [Details of the Risk Assessment Results According to the Eighth Embodiment] Next, the details of the risk assessment results 2300 according to the eighth embodiment will be explained using Figures 15 and 17. In the monitoring system 2001 according to this eighth embodiment, as shown in Figure 15, the robot 2201 of the work device 2200 performs work to manufacture goods, such as assembling parts. In this case, the worker may cooperate with the robot 2201 to manufacture goods, for example, by transporting and supplying parts. Whether or not the worker and the robot 2201 are working together can be determined by detecting the state (work phase) of the robot 2201 or by detecting from the video of the camera 2106 that the worker is in the manufacturing line where the robot 2201 is located. In other words, the CPU 108a determines that a worker is in the monitoring area AR21 and working together by analyzing the signals transmitted from the control device of the robot 2201 and the video of the camera 2106.

[0114] However, during this collaborative work, since the robot 2201 is in operation, there is a risk of contact accidents occurring where it comes into contact with workers. Therefore, in the risk assessment result 2300, under the work item for the work name "Collaborative robot approaching during non-collaborative area work," the expected accident is "contact accident," and the risk reduction measure is a warning such as "Do not approach the robot while it is working." Furthermore, in the risk assessment result 2300, under the work item for the work name "Collaborative robot approaching during collaborative area work," the expected accident is "contact accident," and the risk reduction measure is a warning such as "Do not approach the robot while it is working." In addition, under this item, the severity, probability, evaluation, and priority values ​​are stored as shown in Figure 17, respectively, as a risk estimate before the implementation of risk reduction measures. Furthermore, under this item, the severity, probability, evaluation, and priority values ​​are stored as shown in Figure 17, respectively, as a risk estimate after the implementation of risk reduction measures.

[0115] "Collaborative robot approaching during non-collaborative work" refers to a situation where robot 2201 moves to an area outside the collaborative area where it works in cooperation with a worker. In other words, it refers to a situation where robot 2201 moves outside the area where the worker is supposed to be working. In this case, robot 2201 may perform actions not intended by the worker. On the other hand, "Collaborative robot approaching during collaborative work" refers to a situation where robot 2201 moves within the collaborative area where it works in cooperation with a worker. In other words, it refers to a situation where robot 2201 performs its normal work actions as performed by the worker. In this case, robot 2201 will perform the actions intended by the worker. Therefore, in the risk assessment result 23000, the item "Collaborative robot approaching during non-collaborative work" is rated as more severe and has a higher evaluation than the item "Collaborative robot approaching during collaborative work".

[0116] [Control of the Monitoring Device] Next, the control of the monitoring device according to the eighth embodiment will be explained with reference to Figure 18. In the first embodiment described above, the presence or absence of risk reduction measures was determined by verbalizing the image captured by the camera 106 and comparing it with the risk assessment result 300. However, in this eighth embodiment, the presence or absence of risk reduction measures is determined by information on whether the robot 2201 and the worker are working together, and information on whether the robot 2201 is working in a non-collaboration area or a collaborative area.

[0117] More specifically, as shown in Figure 18, when this control is started, the CPU 108a first receives the risk assessment result 2300 shown in Figure 17 from the input unit 1111 (I / F 108e) (S801). The input risk assessment result 2300 is output to the decision unit 2112 as second information D2.

[0118] Next, the CPU 108a acquires video footage captured by the camera 2106 as monitoring information for the monitoring area AR21 via the detection unit 2113 (S802). The acquired video footage is output to the judgment unit 2112 as first information D1. The CPU 108a also acquires information about the area above the robot 2201 from the control device (not shown) of the robot 2201 via the device status input unit 2115 (I / F 108e). The acquired information is output to the judgment unit 2112 as first information D1.

[0119] Next, the CPU 108a compares the monitoring information (first information D1, which is video) with the status information of the robot 2201 (first information D1) and the risk reduction measures (second information) using the judgment unit 2112. Then, it determines whether or not there is a problem based on the result of this comparison (S803). In other words, the judgment unit 2112 determines whether or not the risk reduction measures have been implemented based on the first information D1 and the second information D2. Specifically, it determines whether or not a worker is approaching the robot 2201 based on the video captured by the camera 2106, the information from the monitoring sensor of the robot 2201, and the status of the robot 2201. In particular, even if the monitoring sensor of the robot 2201 does not detect a worker, it determines from the video captured by the camera 2106 whether or not there is a worker in the unmonitored area AR22, AR22. If a worker is not approaching the robot 2201, it determines that the risk reduction measures have been implemented and that there is no problem (Yes in S803). In this case, it is determined whether or not a command to terminate monitoring has been issued (S807). If there is no command to terminate monitoring (No in S807), the process returns to step S802. If there is a command to terminate monitoring (Yes in S807), this control process is terminated.

[0120] On the other hand, in step S803, it is determined from the video captured by the camera 2106 or the detection results of the robot 2201's monitoring sensor that risk reduction measures have not been implemented (the worker is approaching the robot 2201). In this case, the CPU 108a determines that there is a problem (No. in S803), meaning that it has determined that there is a lack of risk reduction measures.

[0121] In this case, the CPU 108a determines whether the robot 2201 and the worker are working together (S804). That is, it determines whether the robot 2201 and the worker are working together based on the state of the robot 2201 (e.g., work phase) input from the robot 2201's control device (not shown). If it determines that they are not working together (No. in S804), the CPU 108a sets the warning level to "low" and instructs the notification unit 2114 to output the content of the risk reduction measures as an audio message at a low volume. In other words, the warning intensity is set (changed) according to the state of the robot 2201 (work equipment). Then, as a risk reduction measure, the notification unit 2114 broadcasts a warning such as "Do not approach the robot while it is working" at a low volume towards the monitoring area AR21 (S805). Subsequently, the system determines whether or not a command to terminate monitoring has been issued (S807). If no such command has been issued (No in S807), the system returns to step S802. If a command to terminate monitoring has been issued (Yes in S807), the system terminates this control.

[0122] On the other hand, in step S804, if it is determined that collaborative work is in progress (Yes in S804), the CPU 108a sets the warning level to "high" and instructs the notification unit 2114 to output the content of the risk reduction measures in a loud volume. In other words, the intensity of the warning is set (changed) according to the state of the robot 2201 (work equipment). Then, as a risk reduction measure, the notification unit 2114 broadcasts a warning such as "Do not approach the robot while it is working" to the monitoring area AR21 at a louder volume than when collaborative work is not in progress (S806). After that, it is determined in the same way whether or not there has been a command to end monitoring (S807). If there has not been a command to end monitoring (No in S807), it returns to step S802, and if there has been a command to end monitoring (Yes in S807), this control is terminated.

[0123] [Summary of the Eighth Embodiment] As described above, in this eighth embodiment, the monitoring area AR21 determines whether there is a deficiency in risk reduction measures, and the method for executing the correct risk reduction measures is announced via speaker in the monitoring area where the deficiency was determined. As a result, even if a worker approaches the robot 2201, for example, the deficiency in risk reduction measures can be determined, and the method for executing the correct risk reduction measures at that location can be announced. This allows the worker to execute the correct risk reduction measures when there is a deficiency in risk reduction measures, thereby improving safety.

[0124] <Ninth Embodiment> Next, a ninth embodiment, which is a modified version of the first to eighth embodiments described above, will be explained with reference to Figures 19 to 21. Figure 19 is a diagram showing the monitoring system according to the ninth embodiment. Figure 20 is a block diagram showing the functions of the monitoring device according to the ninth embodiment. Figure 21 is a flowchart showing the control of the monitoring device according to the ninth embodiment. In this explanation of the ninth embodiment, the same reference numerals are used for parts that are the same as those in the first to eighth embodiments, and their explanations are omitted.

[0125] [Configuration of the monitoring system according to the ninth embodiment] As shown in Figure 19, the monitoring system 3001 includes a monitoring device 3100 as an information processing device that monitors the monitoring area AR31. The monitoring device 3100 also monitors the work equipment 200 and the worker 2 as targets located in the monitoring area AR31. The work equipment 200 is the same as that described in the first embodiment.

[0126] Furthermore, the monitoring device 3100 includes a camera 3106 that captures images of the monitoring area AR 31 as a detection unit connected to the control device 108 shown in Figure 1. The monitoring device 3100 also includes a speaker 3107 that outputs (notifies) sound to the monitoring area AR 31 as a first notification unit connected to the control device 108 shown in Figure 1. In addition, the monitoring device 3100 includes a speaker 3108 that outputs (notifies) sound to the notification area AR 32 as a second notification unit connected to the control device 108 shown in Figure 1. The notification area AR 32 is where the administrator 3 is located and is an external and separate location from where the work device 200 is installed. In other words, the speaker 3108 located in the notification area AR 32 is a pre-set contact and is in a different location from the monitoring area AR 31. The notification area AR32 is not limited to the location where administrator 3 is present; it can also be an emergency room where a nurse is present, or anywhere else as long as it can notify someone other than worker 2.

[0127] Next, the configuration of the monitoring device 3100's functions will be explained using Figure 20. As shown in Figure 20, the monitoring device 3100 has a configuration in which each part of the hardware of the control device 108 functions as an input unit 3111, a judgment unit 3112, a detection unit 3113, a first notification unit 3114A, and a second notification unit 3114B, which function as function-achieving units. These function-achieving units function, for example, when the CPU 108a executes various programs stored in the ROM 108b.

[0128] Specifically, the input unit 3111 operates as a function that allows the interface 108e to input information. The input unit 3111 inputs, for example, the risk assessment result 300 shown in Figure 3. The detection unit 3113 operates as a function that acquires the image captured by the camera 3106 of the monitoring area AR 31 as a detection result. The first notification unit 3114A operates as a function that causes the speaker 3107 to output sound to the monitoring area AR 31 as a notification. The second notification unit 3114B operates as a function that causes the speaker 3108 to output sound to the notification area AR 32 as a notification. The judgment unit 3112 operates as a function that allows the CPU 108a to determine whether or not risk reduction measures are insufficient based on the risk assessment result 300 and the image captured by the camera 3106 of the monitoring area AR 31. The risk assessment results 300 according to this ninth embodiment are the same as those of the first embodiment, as they are the same as those of the work device 200.

[0129] [Control of the Monitoring Device] Next, the control of the monitoring device according to the ninth embodiment will be explained using Figure 21. In the first embodiment described above, it was determined whether or not risk reduction measures were being taken based on the image captured by the camera 106, and if the risk reduction measures were insufficient, a notification was given by the speaker 107. In this ninth embodiment, in addition to the above, if an accident occurs in the monitoring area AR31, the accident is notified to the notification area AR32.

[0130] More specifically, as shown in Figure 21, when this control is started, the CPU 108a first receives the risk assessment result 300 shown in Figure 3 from the input unit 3111 (I / F 108e) (S901). The input risk assessment result 300 is output to the decision unit 3112 as verbalized second information D2.

[0131] Next, the CPU 108a acquires the video captured by the camera 3106 as monitoring information for the monitoring area AR 31 using the detection unit 3113 (S902). The acquired video is output to the judgment unit 3112 as the first information D1, which is verbalized as described above.

[0132] Next, the CPU 108a determines whether or not an accident has occurred in the monitoring area AR 31 based on the first information D1 (or the video itself) (S903). That is, the CPU 108a detects the occurrence of an accident based on the detection result of the detection unit. An accident here refers to, for example, when worker 2 is injured, or when worker 2 encounters a problem that they cannot resolve on their own. If it is determined that an accident has occurred (Yes in S903), a command to notify the occurrence of the accident is output to the first notification unit 3114A and the second notification unit 311B, and the first notification unit 3114A and the second notification unit 311B make the notification (S904). In other words, in the monitoring area AR 31, an audio message notifying that an accident has occurred is output from speaker 3107, and in the notification area AR 32, an audio message notifying that an accident has occurred is output from speaker 3108. Then, it is determined whether or not a command to terminate monitoring has been received (S907). If there is no command to terminate monitoring (No in S907), the process returns to step S902. If there is a command to terminate monitoring (Yes in S907), this control process ends.

[0133] On the other hand, in step S903, if it is determined that there was no accident (No in S903), the comparison unit 403 compares the monitoring information (first information D1, in which the video is verbalized) with the risk reduction measures (second information, in which the risk assessment result 300 is verbalized). Then, it determines whether or not there is a problem with the result of this comparison (S905). In other words, the comparison unit 403 determines whether or not there is a discrepancy between the first information D1 and the second information D2, and specifically, if there is information in the video captured by the camera 3106 that matches all of the risk reduction measures, it determines that there is no problem (Yes in S905). Then, it determines whether or not there has been a command to end monitoring (S907), and if there has been no command to end monitoring (No in S907), it returns to step S902, and if there has been a command to end monitoring (Yes in S907), this control is terminated.

[0134] On the other hand, in step S905, if even one piece of information in the video captured by camera 3106 is inconsistent with the risk reduction measures (i.e., a risk reduction measure that has not been implemented), it is determined that there is a problem (No. in S905). In other words, it is determined that there is a shortage of risk reduction measures. Then, CPU 108a instructs first notification unit 3114A to output the content of the inconsistent risk reduction measures as audio, that is, it notifies the risk reduction measures by first notification unit 114A (speaker 3107) (S906). In short, it notifies the monitoring area AR31 of the method for implementing the risk reduction measures. In this case, it is explained that no notification is made to notification area AR32, but it is also permissible to notify notification area AR32 and inform administrator 3. After that, similarly, it is determined whether or not there has been a command to end monitoring (S907), and if there has been no command to end monitoring (No. in S907), it returns to step S902. Then, if a command to terminate monitoring is received (Yes in S907), this control is terminated.

[0135] [Summary of the Ninth Embodiment] As described above, in this ninth embodiment, the monitoring area AR31 determines whether there is a deficiency in risk reduction measures, and the method for executing the correct risk reduction measures is announced via speaker in the monitoring area where the deficiency was determined. This allows workers to execute the correct risk reduction measures when there is a deficiency, thereby improving safety. In addition, the monitoring area AR31 determines whether there is an accident, and if an accident is determined to have occurred, the occurrence of the accident is announced to the monitoring area AR31 and the announcement area AR32. This allows for prompt notification when an accident occurs at a work site where risk assessment is necessary, preventing the accident from worsening and enabling prompt relief for those involved in the accident.

[0136] <Possibility of other embodiments> In the first to ninth embodiments described above, the notification of risk reduction measures to workers was described by outputting sound from a speaker. However, the invention is not limited to this, and the notification may also be given to workers by displaying it on a display or the like.

[0137] Furthermore, the first to ninth embodiments described a system that detects risk reduction measures that have not been implemented by an operator and notifies the operator of this. However, it is also possible to determine that risk reduction measures are insufficient if, for example, the total value of the risk estimate evaluation (see, for example, Figure 3) is above a predetermined threshold. That is, the total value may be calculated so that the evaluation value of the risk estimate after the implementation of the risk reduction measures is adopted, and it may be determined that the risk reduction measures have been met when the total value falls below the threshold. In this case, it is also conceivable that all risk reduction measures would be notified, allowing the operator to select the risk reduction measures that have not been implemented from all of them and to correctly implement them.

[0138] Furthermore, in the first to seventh embodiments and the ninth embodiment, we described the images captured by the camera and the information that is verbalized in the risk assessment result information processing device. However, it is also possible to configure the device to verbalize the camera images and risk assessment results using another external computer or server, and to input the verbalized first and second information into the monitoring device.

[0139] Furthermore, in the first to ninth embodiments, we described a system in which cameras placed in the monitoring area capture images of workers and work equipment. However, the system is not limited to this, and for example, cameras may be placed at the entrance of a room where work equipment is located or in the passageway leading to that entrance, and the status of workers may be acquired as monitoring information before they enter the monitoring area. For example, if a worker attempting to enter a room has not taken risk reduction measures, it is conceivable that entry into the room may not be permitted (the door lock may not be released) until the risk reduction measures are taken.

[0140] Furthermore, while the first to ninth embodiments have provided examples such as cutting work with an electric saw, transportation work with a forklift, and manufacturing work of parts with a robot, the invention is not limited to these. For example, a monitoring device like that of this embodiment can be applied to any location where a risk assessment result has been created, such as chemical substance management work, medical work, work on ships, forestry work, care work in social welfare settings, and construction work at construction sites. In addition, the robot arm may be a vertical articulated robot, a horizontal articulated robot, a parallel link robot, or a Cartesian robot. Furthermore, a machine that can automatically perform movements such as extension and retraction, bending and straightening, vertical movement, horizontal movement, or rotation, or a combination thereof, based on information stored in a memory device provided in the control device may be used.

[0141] Furthermore, in the first to ninth embodiments, we explained how, if risk reduction measures do not match the information described in the risk assessment results, it is determined that the measures have not been implemented and that the risk reduction measures are insufficient. However, this is not the only example; even if the measures do not perfectly match the information described in the risk assessment results, if they are achieved to some extent, it may be determined that the risk reduction measures are satisfied. Conversely, even if the measures match the information described in the risk assessment results, if there is a discrepancy, it may be determined that the risk reduction measures are insufficient. Specifically, for example, in item 301 of Figure 3, which includes the risk reduction measure "wearing goggles," if the worker is wearing a full-face helmet, it can be determined that the risk reduction measures are satisfied even if they do not perfectly match. Conversely, even if the worker is wearing goggles, if the goggles are not properly fitted, it may be determined that the risk reduction measures are insufficient.

[0142] This disclosure can also be implemented by supplying a program that implements one or more of the functions of the above-described embodiments to a system or device via a network or storage medium, and by having one or more processors in the computer of that system or device read and execute the program. It can also be implemented by a circuit (e.g., ASIC) that implements one or more functions.

[0143] This disclosure can be used in information processing methods, information processing devices, control systems, methods for manufacturing articles, programs, and recording media for processing information.

[0144] The present invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the following claims are attached to make the scope of the invention public.

[0145] This application claims priority based on Japanese Patent Application No. 2024-191566, filed on 31 October 2024, and all of its contents are incorporated herein by reference.

[0146] 1... Monitoring system (control system) / 2... Worker / 106... Camera (detection unit, imaging device) / 107... Speaker (notification unit) / 108... Control device (information processing device) / 108a... CPU (processing unit) / 113... Detection unit / 114... Notification unit / 200... Work equipment / 202... Electric saw (work equipment) / 300... Risk assessment results (risk assessment) / 601... Forklift (moving object) / 602... Forklift (moving object) / 1106A... Camera (detection unit, imaging device) / 1106B... Camera (detection unit, imaging device) / 1107A... Speaker (notification unit) / 1107B... Speaker (notification unit) / 1113A... First detection unit (detection unit) / 1113B... Second detection unit (detection unit) / 1114A...First Notification Department (Notification Department) / 1114B...Second Notification Department (Notification Department) / 1300...Risk Assessment Results (Risk Assessment) / 2001...Monitoring System (Control System) / 2106...Camera (Detection Unit, Imaging Device) / 2107...Speaker (Notification Department) / 2113...Detection Unit / 2114...Notification Department / 2200...Work Equipment / 2201...Robot / 2300...Risk Assessment Results (Risk Assessment) / 3106...Camera (Detection Unit, Imaging Device) / 3107...Speaker (Notification Department) / 3108...Speaker (Notification Department, Contact Information) / 3113...Detection Unit / 3114A...First Notification Department (Notification Department) / 3114B...Second Notification Department (Notification Department) / D1...First Information / D2...Second Information

Claims

1. An information processing method performed by a computer, wherein, when it is determined that risk reduction measures are insufficient based on the detection results of a detection unit that detects the state of an object, a notification unit notifies the object of a method for performing the risk reduction measures and prevents the object from using a device that can be controlled by the computer; and when it is determined that the risk reduction measures are sufficient based on the detection results of the detection unit, the object allows the object to use the device.

2. The information processing method according to claim 1, wherein when the notification unit performs notification, the intensity of the notification is increased as the risk level increases when the risk reduction measures are insufficient.

3. The information processing method according to claim 1 or 2, wherein if it is detected that the risk reduction measures are insufficient, the power to the work equipment performing the work is turned off, and if it is detected that the risk reduction measures have been performed, the power to the work equipment is turned on.

4. The information processing method according to any one of claims 1 to 3, wherein the notification unit detects that the risk reduction measures have been implemented and determines that there is a residual risk, and then issues a notification including the existence of the residual risk.

5. The information processing method according to claim 4, wherein the residual risk information includes information on severity and likelihood of occurrence, and when it is determined that the residual risk exists, the notification unit provides notification including information on the content of the residual risk, the severity of the residual risk, and the likelihood of occurrence of the residual risk.

6. The information processing method according to any one of claims 1 to 5, wherein when an accident is detected in accordance with the detection result of the detection unit, information regarding the occurrence of the accident is output to a pre-set contact.

7. The information processing method according to claim 6, wherein the contact point is a location different from the location where the object whose state is detected by the detection unit is located.

8. The information processing method according to any one of claims 1 to 7, wherein it is determined whether or not the risk reduction measures are insufficient based on the detection results of the multiple detection units.

9. The information processing method according to claim 8, wherein the target includes a moving object, and when the moving object moves, the method detects that the risk reduction measures are insufficient based on the detection result of a detection unit among the plurality of detection units that is capable of detecting the moving object.

10. The information processing method according to any one of claims 1 to 9, wherein the processing unit, when it detects a deterioration in the worker's working environment, executes a notification to the notification unit prompting improvement of the working environment.

11. The information processing method according to any one of claims 1 to 10, wherein, based on the detection results of the detection unit, if it is predicted that the risk reduction measures are insufficient, the notification unit executes a notification including a method for determining that the risk reduction measures are insufficient and for executing the predicted risk reduction measures.

12. An information processing method according to any one of claims 1 to 11, which determines whether or not the risk reduction measures are insufficient based on information regarding the results of processing a risk assessment using a machine learning model.

13. The information processing method according to any one of claims 1 to 12, wherein the detection result of the detection unit is verbalized using a machine learning model, and the results are compared with the verbalized risk assessment information to determine whether or not the risk reduction measures are insufficient.

14. The information processing method according to any one of claims 1 to 13, wherein the detection result of the detection unit is verbalized information, and the information relating to the risk assessment is verbalized using a machine learning model, and the verbalized detection result of the detection unit is compared with it to determine whether or not the risk reduction measures are insufficient.

15. The information processing method according to any one of claims 1 to 14, wherein the notification includes a warning to the worker, and the processing unit changes the intensity of the warning according to the state of the work equipment used for the work.

16. The information processing method according to any one of claims 1 to 15, wherein the detection unit is an imaging device that captures images of the state of the target, and the detection result is an image input from the imaging device.

17. The information processing method according to any one of claims 1 to 16, wherein the subject includes a worker, and the information relating to the risk assessment includes information on the risks in the worker's work and information on the worker's condition or the worker's work status as a risk reduction measure to reduce the risks.

18. The information processing method according to any one of claims 1 to 17, wherein the subject includes a worker and a robot, and the information relating to the risk assessment includes information on risks in collaborative work between the worker and the robot, and information on warnings to the worker as risk reduction measures to reduce those risks.

19. Information processing apparatus having a processing unit, wherein the processing unit, when it determines that risk reduction measures are insufficient based on the detection result of a detection unit that detects the state of the target, notifies the target of a method for executing the risk reduction measures via a notification unit and prevents the target from using a device that can be controlled by the processing unit, and when it determines that the risk reduction measures are sufficient based on the detection result of the detection unit, enables the target to use the device.

20. A control system comprising the information processing device described in claim 19 and a work device used in the process of performing a risk assessment.

21. The control system according to claim 20, wherein the work apparatus includes a robot.

22. A method for manufacturing an article, comprising manufacturing an article using the control system described in claim 20.

23. A program for causing a computer to execute the information processing method described in any one of claims 1 to 18.

24. A computer-readable recording medium on which the program described in claim 23 is recorded.

Images