Abnormal condition monitoring system and abnormal condition monitoring method

Abstract

To improve the efficiency and quality of work.SOLUTION: An abnormal state monitoring system includes sensor information acquiring means for acquiring sensor information indicating a detection result from a sensor for detecting a worker's state, work check data holding means for holding in advance a worker's behavior assumed from a work procedure as work check data, and abnormality detection means for detecting an abnormality in the work by comparing the sensor information and the work check data. Furthermore, the abnormal state monitoring system includes control means for creating flag data including a flag indicating an abnormality level and state information indicating an abnormality state when the abnormality detection means detects an abnormality and transmitting the flag data to a management device. The management device transmits instruction information on the basis of the flag data.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an abnormal state monitoring system and an abnormal state monitoring method for detecting abnormalities in work performed by workers. [Background technology]

[0002] Conventionally, there is a technology that enables remote support of on-site maintenance work when an abnormality occurs during on-site maintenance work. For example, WO2019 / 116485 states, "A local server 10 for equipment management includes a proxy work management unit 42 that is connected to a proxy management center server 12. When an error code is output during maintenance work in a building to be maintained, in which a setting value stored in a setting value storage unit 46 is changed, the proxy work management unit 42 enables proxy maintenance work to be performed remotely from a maintenance worker's terminal device 26 that is connected to the center server 12. Furthermore, the proxy work management unit 42 can display, on a display unit 50, 72 provided in the building to be maintained, a proxy screen that includes an input field for the error code and an input field for inputting the details of the maintenance work to be performed in the building to be maintained when the error code is output." [Prior art documents] [Patent documents]

[0003] [Patent Document 1] WO2019 / 116485 Summary of the Invention [Problem to be solved by the invention]

[0004] According to the above-described conventional technology, when an error code is output, it is possible to perform maintenance work by remote control. However, if, for example, the operator makes a mistake in the work or is unfamiliar with the work, it is preferable to provide assistance to the operator before the situation in which an error code is output occurs. The present invention has been made to solve the above-mentioned problems, and aims to provide an abnormal condition monitoring system that improves work efficiency and quality by detecting abnormalities in work at an early stage and supporting workers. [Means for solving the problem]

[0005] In order to achieve the above-mentioned object, a typical abnormal condition monitoring system of the present invention is characterized by comprising a sensor information acquisition means for acquiring sensor information indicating the detection results from a sensor that detects the state of a worker, a work check data storage means for storing in advance the actions of the worker that are expected from the work procedure as work check data, and an abnormality detection means for detecting abnormalities in the work by comparing the sensor information with the work check data. Furthermore, one representative abnormal state monitoring method of the present invention is characterized by including a work check data retention step of retaining the worker's actions assumed from the work procedure as work check data, a sensor information acquisition step of acquiring sensor information indicating the detection results from a sensor that detects the worker's condition, and an abnormality detection step of comparing the sensor information with the work check data to detect an abnormality in the work. Other aspects of the present invention will be described in the following embodiments. [Effects of the Invention]

[0006] According to the present invention, it is possible to improve the efficiency and quality of work. Objects, configurations and effects other than those described above will become apparent from the following description of the preferred embodiments of the invention. [Brief explanation of the drawings]

[0007] [Figure 1] 1 is a diagram illustrating an overview of an abnormal state monitoring system for a measurement object according to an embodiment; [Figure 2] FIG. 1 is a diagram illustrating a configuration of a monitoring system for an abnormal state of a measurement object. [Figure 3] FIG. 10 is a diagram illustrating an example of a data structure of flag data. [Figure 4]FIG. 10 is a diagram illustrating an example of a data structure of notification data from a management device. [Figure 5] FIG. 10 is a diagram showing an example of an acquired data determination table for process S1. [Figure 6] FIG. 10 is a diagram showing an example of an instruction determination table for process S2. [Figure 7] 1 is a flowchart showing the overall processing of a system for monitoring an abnormal state of a measurement target. [Figure 8] 10 is a flowchart showing a flag generation process in a measurement target. [Figure 9] 10 is a flowchart illustrating a process of determining the priority of multiple flags in the management device. [Figure 10] 10 is a flowchart showing a sensing data selection process in a process S1 in the management device. [Figure 11] 10 is a flowchart showing a sensing data analysis process of a process S2 in the management device. [Figure 12] FIG. 10 is a diagram showing instruction contents based on flag data and sensing data in process S2. [Figure 13] FIG. 10 is a diagram showing an example of a work instruction table for process S10. [Figure 14] 10 is a flowchart showing a process for determining whether an object to be measured is in an unsteady state. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The examples are illustrative of the present invention, and for clarity of explanation, appropriate omissions and simplifications have been made. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural. In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings. Examples of various types of information may be described using expressions such as "table," "list," and "queue," but the various types of information may also be expressed using data structures other than these. For example, various types of information such as "XX table," "XX list," and "XX queue" may also be expressed as "XX information." When describing identification information, expressions such as "identification information," "identifier," "name," "ID," and "number" are used, but these are interchangeable. When there are multiple components with the same or similar functions, they may be described using the same reference numeral with different subscripts. When there is no need to distinguish between these multiple components, the subscripts may be omitted. In the embodiments, processing performed by executing a program may be described. Here, a computer executes the program using a processor (e.g., a CPU or a GPU) and performs processing defined by the program using storage resources (e.g., a memory) and interface devices (e.g., a communication port). Therefore, the entity performing the processing by executing the program may be the processor. Similarly, the entity performing the processing by executing the program may be a controller, device, system, computer, or node having a processor. The entity performing the processing by executing the program may be any computing unit, and may include a dedicated circuit that performs specific processing. Here, the dedicated circuit may be, for example, an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), or a CPLD (Complex Programmable Logic Device). A program may be installed on a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server may include a processor and a storage resource for storing the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. In addition, in an embodiment, two or more programs may be realized as one program, or one program may be realized as two or more programs. [Example]

[0009] <Outline of the abnormal state monitoring system 100 for the measurement target> FIG. 1 is a diagram showing an overview of an abnormal state monitoring system 100 for a measurement object according to an embodiment. FIG. 2 is a diagram showing the configuration of a system for monitoring abnormal states of measurement targets. The system 100 for monitoring abnormal states of measurement targets includes a management device 30 (remote monitoring center) that monitors abnormal states from multiple measurement targets, which are composed of numerous infrastructure facilities operating at various locations or personnel working at those locations, and a measurement target device 10 provided at the measurement target. When an abnormal state (abnormal state) occurs, the measurement target device 10 transmits flag data 21 indicating the abnormal state (described below) to the management device 30, which is the remote monitoring center. The management device 30 analyzes the flag data 21 and requests the required sensing data from the measurement target device 10. The measurement target device 10 transmits the requested sensing data to the management device 30, and the management device 30 transmits instruction information to the measurement target device 10 based on the flag data and sensing data.

[0010] The abnormal state monitoring system 100 for the measurement target has M sensors installed at N locations and acquires and uses P types of sensor information. The M, N, and P sensors refer to the sensors installed in the infrastructure equipment and on-site workers shown in FIG. 1 and the sensors installed in their surroundings. In other words, the M, N, and P sensors refer to the sensors installed in the measurement target device 10 and the sensors installed in its surroundings. The P types of sensors include, for example, an image distance sensor, a sound sensor, a vibration sensor, a temperature sensor, and a vital sign sensor. In addition, in the case of a person, the sensors may be installed in a smart device carried by the person, or they may be wearable sensors in which sensors are incorporated into clothing, glasses, or a wristband worn by the person.

[0011] As an example of measurement targets, we will explain the inspection work of infrastructure facilities such as water supply and sewage, power transmission, roads, railways, bridges, rivers, and city gas that operate indoors and outdoors. In a steady state, infrastructure equipment operates according to a pre-programmed schedule while constantly sensing changes in the surrounding environment and its own state using multiple sensors, continuously checking the output results from each sensor for abnormal values ​​above a threshold. If an abnormal value above a threshold is detected as a result of the check, the measurement target device 10 transitions from the steady state to an unsteady state. Inspection work by an inspection worker involves an inspection worker visiting the on-site infrastructure equipment periodically or in response to an abnormality notification. The inspection work is performed according to a prescribed work procedure manual. When an inspection worker detects whether an abnormal state has occurred, the inspection worker has a smart device or smartwatch, etc., equipped with various sensors, processing units, etc., as the measurement target device 10 (either worn by the worker or brought and placed nearby by the worker). The status of the inspection work is constantly measured using multiple sensors, and the output results from each sensor are continuously checked for abnormal values ​​above a threshold specified in the work procedure manual. If an abnormal value above a threshold is detected as a result of the check, the measurement target device 10 transitions from the steady state to an unsteady state.

[0012] When the state transitions to an unsteady state, the measurement target device 10 first executes a preset emergency measure in response to the abnormal sensor values. Next, it generates flag data 21 based on the output results from each sensor. It then notifies the management device 30 of the generated flag data 21 via the network NW. After notifying the flag data 21, it waits to receive instructions from the management device 30. The instructions from the management device 30 include instructions from process S1 and instructions from process S2, which will be described later. While the above explanation concerns the inspection of infrastructure facilities as the measurement target, the same applies to on-site workers at manufacturing sites. In the case of on-site workers at production sites, the production facilities are the measurement target instead of the infrastructure facilities, and the on-site workers are the measurement targets instead of the inspection workers.

[0013] <Device Configuration of the Abnormal State Monitoring System 100 for Measurement Objects> Next, the configuration of the device will be described with reference to FIG. In process S1, when the management device 30 receives the flag data 21, it issues an instruction to acquire the required sensing data. In process S2, when the management device 30 receives the sensing data, it analyzes the flag data and the sensing data, and in accordance with the analysis results, issues an instruction to the on-site infrastructure equipment to take the best possible action via the network NW.

[0014] The measurement target device 10 includes a processing unit 11 that monitors the unsteady state of the measurement target, a memory unit 20, a communication unit 25 that communicates with an external sensor 27 and a management device 30, and multiple internal sensors 26. The processing unit 11 includes a sensor information acquisition unit 12 (sensor information acquisition means) that acquires sensor information from the multiple sensors 26, 27, an abnormality detection unit 13 (abnormality detection means) that determines whether the sensor information indicates an abnormality, an abnormality processing unit 14 (control means) that, when the abnormality detection means determines that an abnormality is present, creates flag data 21 including a flag indicating the abnormality level and status information indicating the abnormal state, and transmits the flag data 21 to the management device 30, and a flag generation unit 15 that generates the flag. The memory unit 20 stores the flag data 21, a first aid determination table 22, a work procedure manual table 23, etc.

[0015] Here, the work procedure manual table 23 is data created based on a work procedure manual that shows the procedures for work to be performed by workers (inspection workers, field workers, etc.), and shows the worker's actions that are expected from the work procedures. At least some of the sensors 26 and 27 are installed to detect the worker's state. If the worker's state detected by the sensors 26 and 27 is consistent with the actions shown in the work procedure manual table 23, the anomaly detection unit 13 determines that the worker is in a steady state. On the other hand, if the detected worker's state is inconsistent with the work procedure manual table 23, the anomaly detection unit 13 determines that the worker is in a non-steady state, that is, that an abnormality has occurred. In this way, the work procedure manual table 23 functions as work check data for detecting abnormalities in work.

[0016] As a specific example, a case where a worker operates a control panel will be described. The work procedure manual table 23 stores the number of times and timing at which the worker reaches out to the control panel. These number of times and timing can be derived from the work procedure. Also, the left and right hands may be specified. The measurement target device 10 is a portable device that is installed near the control panel before the worker starts work. At this time, the measurement target device 10 is installed so that the sensor 26 or the like can acquire information that can at least identify whether the worker is reaching out for the work target. The abnormality detection unit 13 identifies which hand the worker reached out to the control panel and at what timing based on the detection results of the sensor 26, etc., and compares this with the work procedure manual table 23 to detect an abnormality. This makes it possible to detect errors or delays in work as abnormalities. Furthermore, the accuracy of the sensor detection results is low for information such as whether or not the hand is being extended and the timing of the extension, so this information can be obtained sufficiently even if the portable measurement target device 10 is appropriately installed.

[0017] 3 is a diagram showing an example of the data structure of flag data 21. Flag data 21 consists of ID 211 for identifying the measurement target device 10 that generated the flag and the notification event, timestamp 212 indicating the time when the flag data was generated, status 213 indicating steady / unsteady state, responsiveness information 214 determined by abnormal values ​​from sensors, importance information 215, and intervention information 216. Responsiveness information 214, importance information 215, and intervention information 216 are each composed of levels 214L, 215L, and 216L that are flag levels for the respective information, and status information 214S, 215S, and 216S that indicate the status of that level.

[0018] Responsiveness information 214 is information for determining whether there is an abnormality in a person (on-site worker) or whether there is an impact on a person. If it is determined that there is an abnormality in a person or that there is an impact on a person, the responsiveness level is set to "1." Next, it is determined whether changes in the surrounding environment or the state of the equipment are rapid. If it is determined that they are rapid, the responsiveness level is set to "2." In all other cases, the responsiveness level is set to "3."

[0019] The importance information 215 is information that determines whether or not there will be a large impact on the surrounding environment. If it is determined that there will be a large impact, the importance level is set to "1." Next, it is determined whether or not there will be damage to infrastructure facilities or facilities in use. If it is determined that there will be damage, the responsiveness level is set to "2." In all other cases, the importance level is set to "3."

[0020] Intervention information 216 is information for determining whether expert intervention is required among remote instructions. If it is determined that expert intervention is required, the intervention level is set to "1." Next, it is determined whether intervention by artificial intelligence is required. If it is determined that intervention is required, the intervention level is set to "2." In all other cases, the intervention level is set to "3."

[0021] Returning to FIG. 2, the configuration of the management device 30 will be described. The management device 30 has a processing unit 31, a memory unit 40, an input unit 45, a display unit 46, and a communication unit 47. The processing unit 31 has a priority determination unit 32 that determines the priority when multiple flag data are received, a measurement target state monitoring unit 33 that monitors the state of the measurement target and the surrounding environment, a sensing data selection unit 34 that generates instruction information for sensing data to grasp the state of the measurement target, and a sensing data analysis unit 35 that generates response instructions for on-site workers.

[0022] The display unit 46 is a display or the like, and displays the execution status and execution results of processing by the management device 30. The input unit 45 is a device for inputting instructions to a computer, such as a keyboard or mouse, and inputs instructions such as program startup. The processing unit 31 is a central processing unit (CPU) and executes various programs stored in the memory unit 40 or the like. The communication unit 47 exchanges various data and commands with other devices via the network NW.

[0023] The memory unit 40 stores an acquisition data judgment table 41 to be acquired based on flag data 21 from the measurement target device 10 used in process S1, an instruction judgment table 42 for response instructions to the measurement target device 10 used in process S2, instruction content 43 based on the flag data 21 and sensing data, notification data 44 to the measurement target, etc.

[0024] 4 is a diagram showing an example of the data structure of notification data 44 from the management device 30. The notification data 44 includes an ID 441 for identifying the measurement target device 10 that generated the flag and the notification event, a timestamp 442 that is the time when the notification was generated, instruction content 443, etc.

[0025] 5 is a diagram showing an example of the acquired data determination table 41 for process S1. The acquired data determination table 41 includes flag levels, status information, acquired data, etc. The flag levels are levels 214L, 215L, and 216L shown in FIG. 3, and the status information is status information 214S, 215S, and 216S shown in FIG. 3.

[0026] A specific example of the data to be acquired based on the flag level and status information will be described below. If the responsiveness level in row 411 is "1" and the status information indicates that the person is walking slowly and has a pale complexion, the person's position coordinates, image data of the person, distance data of the person, temperature and humidity around the person, and vital signs of the person are acquired.

[0027] When the responsiveness level in row 412 is "2" and the status information indicates that the equipment and the person are approaching each other rapidly, the position coordinates of the person and the equipment are obtained. When the responsiveness level in row 413 is "3" and the status information is "guerrilla heavy rain," the amount of rainfall and image data of the surrounding area are acquired.

[0028] If the importance level of row 414 is "1" and the status information is notified as environmental destruction, surrounding environmental data such as the location coordinates of people, the location coordinates of infrastructure facilities, surrounding image data, temperature, humidity, and gas concentration are obtained. If the importance level of line 415 is "2" and the status information is notified as self-destruction, the sound, vibration, and temperature of the equipment are acquired.

[0029] If the importance level of row 416 is "3" and the status information indicates that maintenance is required, the equipment sound, equipment temperature, continuous operating time, surrounding image data, surrounding environmental data, and location coordinates of the equipment are obtained. If the intervention level of row 417 is "1" and the status information indicates that expert judgment is required, the person's position coordinates, image data of the person, distance data of the person, temperature and humidity around the person, position coordinates of the equipment, image data of the surroundings, distance data of the surroundings, temperature and humidity around the person, rainfall, sound of the equipment, and temperature of the equipment are obtained.

[0030] When the intervention level in row 418 is "2" and the status information indicates that artificial intelligence instruction is required, the same data group as when the intervention level is "1" is acquired. If the intervention level in row 419 is "3" and the status information indicates that only emergency measures are required, there is no need to acquire sensing data.

[0031] In the example of Figure 5, three examples of each of the levels of responsiveness, importance, and intervention are shown, but the present invention is not limited to these. For example, there may be multiple status information items for level "1." 6 is a diagram showing an example of the instruction determination table 42 for process S2. The instruction determination table 42 includes flag levels, state information, determination criteria, etc. The flag levels are levels 214L, 215L, and 216L shown in FIG. 3, and the state information is state information 214S, 215S, and 216S shown in FIG. 3.

[0032] A specific example of the judgment criteria based on the flag level, status information, and sensing data will be explained below. If the responsiveness level of row 421 is "1" and the status information indicates that the person is walking slowly and has a poor complexion, the walking speed is determined from the person's position coordinates, vital signs such as complexion, pulse rate, and fatigue level are determined from the person's image data, the posture and fatigue level are determined from the person's distance data, and the comfort level of the working environment is determined from the temperature and humidity around the person.

[0033] The criteria for assessment are: if fatigue level is 80% or higher, a rest of at least 1 hour is required; if the person is crouching or collapsed, rescue is required; if the walking speed is unsteady or the person has trouble walking, a rest of at least 1 hour is required; if the temperature / humidity is 39°C / 90%, a rest of 0.5 hours is required. Note that hour is a unit of time.

[0034] If the responsiveness level of row 422 is "2" and the status information indicates that a person is rapidly approaching an unmanned transport vehicle (drone) or a moving part of equipment with moving parts, the predicted closest approach distance and predicted closest approach time are determined from the position coordinates of the person and the position coordinates of the equipment. The criteria for judgment are: if the closest distance is 3 m / predicted time is 15 seconds later, the equipment is stopped; otherwise, an alarm is issued to personnel and the equipment's movement speed is reduced.

[0035] If the responsiveness level of row 423 is "3" and the status information indicates a sudden downpour, the system determines whether to continue operations based on the amount of rainfall and the rainfall conditions in the surrounding image data. The criteria for this determination are: if the rainfall rate is 500 mm / h for 0.5 hours based on the results of the rain gauge and image recognition, the equipment is stopped; if the rate is 500 mm / h for 1 hour, the equipment is evacuated.

[0036] If the importance level of line 424 is "1" and the status information indicates signs of explosion, heat generation, or rupture, or environmental damage such as the generation of harmful gas or electrical leakage, the closest approach distance is determined from the position coordinates of the person and the equipment, and the need for evacuation is determined from the surrounding image data and surrounding distance data. The judgment criteria are as follows: if the closest approach distance is 3 m, the equipment is stopped, and if an unrecognized object or an object that should not be approached is detected as a result of image recognition of the surrounding image data and distance data, it is determined that evacuation is necessary.

[0037] If the importance level of line 425 is "2" and the status information is self-destruct, a specific frequency that can predict component failure or deterioration is detected from the sound of the equipment, and a location that has reached a higher temperature than expected is detected from the temperature. The judgment criteria are: if a specific frequency is detected as a result of analyzing the sound of the equipment and it continues for one minute, the equipment is stopped; if a temperature analysis detects a location that is 90 degrees or higher and it continues for one minute, the equipment is stopped.

[0038] If the importance level of row 426 is "3" and the status information indicates that maintenance is required, specific frequencies that can predict component failure or deterioration are detected from the sound of the equipment, areas that have reached higher temperatures than expected are detected from the temperature of the equipment, the need for maintenance work is determined from the continuous operating time, and the travel time from the surrounding image data and distance data of the surrounding area, as well as the location coordinates of the equipment, is estimated. The criteria for this are as follows: if a specific frequency is detected as a result of analyzing the equipment's sound, it is determined that maintenance is required; if a temperature of 90 degrees or higher is detected as a result of analyzing the equipment's temperature, it is determined that maintenance is required; if the continuous operating time is 50 hours or more, it is determined that maintenance is required; and the required travel time and remaining operating time are calculated from the equipment's own position, surrounding images, and distance data.

[0039] If the intervention level in row 427 is "1" and the status information indicates that an expert's judgment is required, the expert determines how to respond when multiple abnormalities occur based on the person's location coordinates, the person's image data, the person's distance data, the temperature and humidity around the person, the equipment's location coordinates, the surrounding image data, the surrounding distance data, the surrounding temperature and humidity, the amount of rainfall, the sound of the equipment, and the temperature of the equipment. For example, this may involve determining priority when multiple flags occur simultaneously, or issuing evacuation instructions when there is conflict over the evacuation / relocation destinations for people / equipment. Another example is when the difficulty level of a task specified in a work procedure manual is high and expert intervention is required. Specifically, if it is detected that the standard work time for each work process specified in the work procedure manual is being used as a threshold, and it is determined that the task is taking significantly longer than that, it is possible that the task is not being completed smoothly due to its high level of difficulty, resulting in stagnation or delays, and therefore it is determined that expert intervention is necessary.

[0040] When the intervention level in row 428 is "2" and the status information indicates that instructions from the artificial intelligence are required, the AI ​​determines from the database how to respond when an isolated abnormality not included in the emergency action list occurs, based on the same data group as when the intervention level is "1." For example, this is how to respond when an unknown object not on the list is recognized as a result of image recognition, or when an abnormal sound with a frequency not on the list is detected. If the intervention level in line 429 is "3" and the status information indicates that only emergency measures are required, no judgment process is required.

[0041] <Effects of the abnormal state monitoring system 100 for the measurement target> The effects of the abnormal state monitoring system 100 for the measurement target are as follows. (1) Since the size of the notification data (flag data 21 (see FIG. 3)) from the measurement target device 10 to the management device 30 is small, even if many notifications are generated simultaneously from many locations, the notifications can be sent to the management device 30 in real time. (2) The urgency and importance of a measurement are determined before the measurement target device 10 notifies the management device 30, so the processing load on the management device 30 is small. In addition, the management device 30 can easily determine priority and can handle a large number of measurement targets. (3) When the measurement target device 10 detects abnormal data, it first performs emergency measures at the measurement target, so that no delay occurs in response. (4) By combining a wide variety of sensors, the situation on-site can be grasped accurately. For example, in human sensing, by combining coordinates (position), posture, movement, and vital information, the condition of on-site workers (health, fatigue, safety of surrounding equipment, and comfort of the work environment) can be grasped more accurately. (5) By analyzing detailed sensing data in the management device 30, it is possible to create more accurate improvement measures than by analysis in the measurement target device 10. (6) By identifying the actions of workers and determining their consistency with the actions derived from the work procedure manual, it is possible to detect work errors and stagnation early and provide support to workers.

[0042] <Processing of the abnormal state monitoring system 100 for the measurement target> The processing of the abnormal state monitoring system 100 for the measurement target will be described below. FIG. 7 is a flowchart showing the overall processing of the abnormal state monitoring system 100 for a measurement target. The description will be made with reference to FIGS. 2 and 3 as appropriate. The abnormality detection unit 13 of the measurement target device 10 determines whether the sensor information acquired by the sensor information acquisition unit 12 is abnormal (whether it is abnormal) (Process S10). If it is not abnormal (Process S10, No), the process returns to Process S10. If it is abnormal (Process S10, Yes), emergency measures are taken (Process S11). The emergency measures are taken based on the emergency measures determination table 22 stored in the storage unit 20. Then, the abnormality processing unit 14 obtains the responsiveness information 214, importance information 215, and intervention information 216 shown in FIG. 3 via the flag generation unit 15 (Process S13: flag generation process). Next, the abnormality processing unit 14 notifies the management device 30 of the flag data 21 (see FIG. 3) (Process S14).

[0043] When the abnormality processing unit 14 of the measurement target device 10 receives the instruction from the management device 30 (process S15), it executes the instruction process for collecting sensing data (process S16). Then, the abnormality processing unit 14 transmits the sensing data to the management device 30 (process S17).

[0044] When the abnormality processing unit 14 of the measurement target device 10 receives the instruction from the management device 30 (process S18), it executes the instructed process (process S19) and returns to process S10. On the other hand, when the measurement target status monitoring unit 33 of the management device 30 receives flag data 21 from the measurement target device 10 (process S31), it obtains the items of data to be acquired shown in Figure 5 via the sensing data selection unit 34 (process S32) and instructs the measurement target device 10 to acquire sensing data (process S33).

[0045] Thereafter, when the measurement target status monitoring unit 33 of the management device 30 receives sensing data from the measurement target device 10 (process S34), it obtains response instructions based on the judgment criteria shown in Figure 6 via the sensing data analysis unit 35 (process S35) and issues response instructions to the measurement target device 10 (process S36).

[0046] 8 is a flowchart showing the flag generation process (process S13) in the measurement target device 10. The description will be made with reference to FIGS. 2 and 3 as appropriate. The flag generation unit 15 of the measurement target device 10 determines whether there is an abnormality in a person (worker) or whether there is an impact on a person (process S131). If it is determined that there is an abnormality in a person or an impact on a person (process S131, Yes), it sets the responsiveness level 214L to "1" and sets the status information 214S (process S133), and proceeds to process S136. If it is determined that there is no abnormality in a person or an impact on a person (process S131, No), the flag generation unit 15 proceeds to process S132.

[0047] Next, the flag generation unit 15 determines whether the change in the surrounding environment or the change in the equipment state is fast (whether the change is large) (process S132), and if it is determined that the change is fast (process S132, Yes), it sets the responsiveness level 214L to "2" and sets the status information 214S (process S134), and proceeds to process S136. Otherwise (process S132, No), the flag generation unit 15 sets the responsiveness level 214L to "3" and sets the status information 214S (process S135), and proceeds to process S136.

[0048] The flag generation unit 15 determines whether or not the event will have a large impact on the surrounding environment (whether or not the event will cause environmental destruction) (process S136), and if it is determined that the impact will be large (process S136, Yes), it sets the importance level 215L to "1" and sets the status information 215S (process S138), and proceeds to process S141. If it is not determined that the impact will be large (process S136, No), the flag generation unit 15 proceeds to process S137.

[0049] Next, the flag generation unit 15 determines whether or not the event will lead to damage to infrastructure facilities or facilities in use (process S137), and if it is determined that the event will cause damage (process S137, Yes), it sets the importance level 215L to "2" and sets the status information 215S (process S139), and proceeds to process S141. Otherwise (process S137, No), it sets the importance level 215L to "3" and sets the status information 215S (process S140), and proceeds to process S141.

[0050] Furthermore, the flag generation unit 15 determines whether emergency processing alone is sufficient (process S141), and if it is determined that emergency processing alone is sufficient (process S141, Yes), it sets the intervention level 216L to "3", sets the status information 216S (process S143), and ends the flag generation process (process S13). If emergency processing alone is not sufficient (process S141, No), the flag generation unit 15 proceeds to process S142.

[0051] The flag generation unit 15 determines whether or not it is necessary to respond using artificial intelligence (process S142), and if it is determined that it is necessary to respond using artificial intelligence (process S142, Yes), it sets the intervention level 216L to "2", sets the status information 216S (process S144), and ends the flag generation process (process S13). If it is not necessary to respond using artificial intelligence (process S142, No), the flag generation unit 15 sets the intervention level 216L to "1", sets the status information 216S (process S145), and ends the flag generation process (process S13).

[0052] Fig. 9 is a flowchart showing the priority determination process for multiple flags in the management server. The description will be made with reference to Figs. 2 and 3 as appropriate. The priority determination process is a more specific illustration of processes S31 to S36 of the measurement target state monitoring unit 33 shown in Fig. 7. Processes S31 and S34 in Fig. 7 correspond to process S328 in Fig. 9, and processes S32 and S35 in Fig. 7 correspond to process S327 in Fig. 9. Furthermore, processes S33 and S36 in Fig. 7 correspond to process S327 in Fig. 9.

[0053] The priority determination unit 32 of the management device 30 determines whether or not there are any remaining reception queues sent from measurement targets in various locations (process S321), and if there are any reception queues (process S321, Yes), it generates a response deadline from the timestamp 212 and responsiveness level 214L of the flag data 21 (process S322) and returns to process S321. If there are no remaining reception queues (process S321, No), the priority determination unit 32 proceeds to process S323.

[0054] The priority determination unit 32 sorts the receiving queues in order of earliest response deadline (process S323), determines whether there are any receiving queues remaining that have not been processed (process S324), and if there are any receiving queues remaining (process S324, Yes), proceeds to process S325, and if there are no receiving queues remaining (process S324, No), proceeds to process S328.

[0055] Then, the priority determination unit 32 extracts the head of the reception queue (process S325), performs analysis of process S1 or process S2 (process S326), transmits instruction information to the measurement target (process S327), and returns to process S324.

[0056] In process S328, the priority determination unit 32 determines whether or not flag data 21 has been received from the measurement target device 10. If flag data 21 has not been received (process S328, No), the process returns to process S328; if flag data 21 has been received (process S328, Yes), the process returns to process S321.

[0057] 10 is a flowchart showing the sensing data selection process (process S32) of process S1 in the management device 30. Figures 2 and 3 are referenced as appropriate. The sensing data selection unit 34 selects sensing data required for analysis based on the acquired data determination table 41.

[0058] The sensing data selection unit 34 determines whether the responsiveness level is "1" (process S341), and if the responsiveness level is "1" and the status information indicates a slow walking posture and a pale complexion (process S341, Yes), it selects the person's position coordinates, image data of the person, distance data of the person, and the temperature and humidity around the person as acquired data (process S343), and proceeds to process S346. If the responsiveness level is not "1" (process S341, No), the sensing data selection unit 34 proceeds to process S342.

[0059] The sensing data selection unit 34 determines whether the responsiveness level is "2" (process S342), and if the responsiveness level is "2" and the status information indicates that the equipment and the person are approaching each other rapidly (process S342, Yes), it selects the position coordinates of the person and the position coordinates of the equipment as acquired data (process S344), and proceeds to process S346. If the responsiveness level is not "2" (process S342, No), the sensing data selection unit 34 proceeds to process S345.

[0060] In step S345, if the response level is "3" and the status information indicates a sudden downpour, the sensing data selection unit 34 selects the amount of rainfall and surrounding image data as acquired data, and proceeds to step S346.

[0061] In process S346, the sensing data selection unit 34 determines whether the importance level is "1" or not, and if the importance level is "1" and the status information is environmental destruction (process S346, Yes), it selects the position coordinates of the person, the position coordinates of the facility, the image data of the surroundings, and the distance data of the surroundings as the acquired data (process S348), and proceeds to process S34B. If the importance level is not "1" (process S346, No), the sensing data selection unit 34 proceeds to process S347.

[0062] If the importance level is "2" and the status information is "self-destruct" (Yes in process S347), the sensing data selection unit 34 selects the equipment sound and machine temperature as the acquired data (process S349) and proceeds to process S34B. If the importance level is not "2" (No in process S347), the sensing data selection unit 34 proceeds to process S34A.

[0063] In process S34A, if the importance level is "3" and the status information indicates that maintenance is required, the sensing data selection unit 34 selects the sound of the equipment, the temperature of the equipment, the continuous operating time, the surrounding image data, the surrounding distance data, and the location coordinates of the equipment as the acquired data, and proceeds to process S34B.

[0064] In process S34B, the sensing data selection unit 34 determines whether the intervention level is "1" or "2." If the status information indicates that expert judgment or artificial intelligence instruction is required (process S34B, Yes), the sensing data selection unit 34 selects the person's position coordinates, image data of the person, distance data of the person, the temperature and humidity around the person, the equipment's position coordinates, image data of the surroundings, distance data of the surroundings, the temperature and humidity around the person, the amount of rainfall, the sound of the equipment, and the temperature of the equipment as acquired data (process S34C), and terminates the sensing data selection process of process S1 (process S32). If the intervention level is not "1" or "2" (process S34B, No), the sensing data selection process of process S1 (process S32) terminates.

[0065] 11 is a flowchart showing the sensing data analysis process (process S35) of process S2 in the management device 30. Figures 2 and 3 are referenced as appropriate. The sensing data selection unit 34 selects a response instruction for the measurement target based on the flag data 21, the sensing data, and the instruction determination table 42.

[0066] The sensing data analysis unit 35 determines whether the measurement target is a person (process S351), and if it is a person (process S351, Yes), analyzes the person's coordinates from the acquired sensing data (process S353), analyzes posture and movement (process S354), analyzes vital signs (process S355), and analyzes the person's surrounding environment (process S356), and then generates instruction information (process S357), ending the sensing data analysis process of process S2 (process S35). If the measurement target is not a person (process S351, No), the sensing data analysis unit 35 proceeds to process S352. Vitals is an abbreviation for vital signs. They are the most basic information about a patient's life, and can also be translated as signs of life. Specifically, they often refer to four things: pulse or heart rate, respiratory rate, blood pressure, and body temperature. These numerical values ​​are used to understand and express a person's current condition.

[0067] In processes S353 to S356, if the responsiveness level is "1" and the status information indicates that the person is walking slowly and has a poor complexion, the sensing data analysis unit 35 determines the walking speed from the person's position coordinates, determines vital signs such as complexion, pulse rate, and fatigue level from the person's image data, determines the person's posture and fatigue level from the person's distance data, and determines the comfort level of the working environment from the temperature and humidity around the person.

[0068] In process S357, the criteria for judgment are that if the fatigue level is 80% or more, a rest of at least 1 hour is required; if the person is crouching or collapsed, rescue is required; if the walking speed is unsteady or the person has trouble walking, a rest of at least 1 hour is required; and if the temperature / humidity is 39 degrees / 90% for 1 hour, a rest of 0.5 hours is required.

[0069] In processes S353 to S356, if the responsiveness level is "2" and the status information indicates that the equipment and a person are approaching each other rapidly, the sensing data analysis unit 35 determines the predicted closest approach distance and the predicted closest approach time from the position coordinates of the person and the position coordinates of the equipment.

[0070] In process S357, the criteria for judgment are that if the closest approach distance is 3 m / predicted time is 15 seconds later, the equipment is stopped; otherwise, an alarm is issued to personnel, the equipment movement speed is reduced, and the operation schedule is changed.

[0071] In process S357, if the responsiveness level is "3" and the status information indicates a sudden downpour, the sensing data analysis unit 35 determines whether to continue activities based on the amount of rainfall and the rainfall conditions in the surrounding image data. The criteria for this determination are as follows: if the rainfall rate is 500 mm / h for 0.5 hours based on the results of the rain gauge and image recognition, stop the equipment; if the rainfall rate is 500 mm / h for 1 hour, evacuate the equipment.

[0072] The sensing data analysis unit 35 determines whether the measurement target is infrastructure equipment (process S352), and if it is infrastructure equipment (process S352, Yes), analyzes the coordinates of the equipment from the acquired sensing data (process S360), analyzes the surrounding environment (process S361), analyzes the equipment status (process S362), and then generates instruction information (process S363), ending the sensing data analysis process of process S2 (process S35). If the measurement target is not infrastructure equipment (process S352, No), the sensing data analysis unit 35 proceeds to process S365.

[0073] In processes S360 to S362, if the importance level is "1" and the status information is that a building has been destroyed, the sensing data analysis unit 35 determines the closest approach distance from the position coordinates and movement line of the person and the position coordinates of the equipment, and determines whether evacuation is necessary from the surrounding image data and surrounding distance data.

[0074] In process S363, the criteria for judgment are that if the closest distance is 3 m, the equipment is stopped, and if an unrecognized object or an object that should not be approached is detected as a result of image recognition of the surrounding image data and distance data, evacuation is required.

[0075] In processes S360 to S362, if the importance level is "2" and the status information is notified as self-destruction, the sensing data analysis unit 35 detects specific frequencies from the sounds of the equipment that can predict component failure or deterioration, and detects areas that have reached higher temperatures than expected from the temperatures.

[0076] In process S363, the criteria are as follows: if a specific frequency is detected as a result of analyzing the sound from the equipment and continues for one minute, the equipment is stopped; if a temperature analysis detects a part that is 90 degrees or higher and continues for one minute, the equipment is stopped.

[0077] In processes S360 to S362, if the importance level is "3" and the status information indicates that maintenance is required, the sensing data analysis unit 35 detects specific frequencies from the sounds of the equipment that can predict component failure or deterioration, detects areas that have reached higher temperatures than expected from the temperature, determines whether maintenance work is necessary from the continuous operating time, and estimates the travel time from the surrounding image data and surrounding distance data, as well as the location coordinates of the equipment, to travel from the current location to the maintenance location.

[0078] In process S363, the criteria for judgment are as follows: if a specific frequency is detected as a result of analyzing the equipment's sound, it is judged that maintenance is required; if a temperature of 90 degrees or higher is detected as a result of analyzing the temperature, it is judged that maintenance is required; if the continuous operating time is 50 hours or more, it is judged that maintenance is required; and the required travel time and remaining operating time are calculated from the equipment's own position, surrounding images, and distance data.

[0079] In process S365, the sensing data analysis unit 35 analyzes the surrounding environment and generates instruction information (process S366), and ends the sensing data analysis process of process S2 (process S35).

[0080] In process S365, when the intervention level is "1" and the status information indicates that expert judgment is required, the sensing data analysis unit 35 determines how to respond when multiple expert abnormalities occur, based on the person's position coordinates, image data of the person, distance data of the person, temperature and humidity around the person, position coordinates of the equipment, image data of the surroundings, distance data of the surroundings, temperature and humidity around the equipment, rainfall, sound from the equipment, and temperature of the equipment. For example, this may include determining priority when multiple flags occur simultaneously, or issuing evacuation instructions when there is competition for evacuation destinations / relocation destinations for people / equipment.

[0081] In process S365, when the intervention level is "2" and the status information indicates that an instruction from an artificial intelligence is required, the sensing data analysis unit 35 determines from the database, based on the same data group as when the intervention level is "1," how to respond when an isolated abnormality not included in the emergency action list occurs. For example, this is how to respond when an unknown object not included in the list is recognized as a result of image recognition, or when an abnormal sound with a frequency not included in the list is detected.

[0082] In the process S365, if the intervention level is "3" and the status information indicates that only emergency treatment is required, the sensing data analysis unit 35 does not need to perform the determination process.

[0083] 12 is a diagram showing the instruction contents based on the flag data and sensing data in process S2. In line 431, when the responsiveness level is "1" and the status information indicates that the person is walking slowly and has a bad complexion, the distance data indicates that the walking speed is 2 km / h and that the person is walking unsteadily, the complexion indicates that the person's vital signs indicate that the fatigue level is 90%, and the temperature and humidity are 37 degrees and 90%, respectively, and the working environment is determined to be uncomfortable, so the instruction is to take a break of at least 1 hour.

[0084] In line 432, when the responsiveness level is "2" and the status information indicates that the equipment and a person are approaching each other rapidly, it is determined that the person's movement speed is 4 km / h, the equipment's movement speed is 40 km / h, the predicted closest approach distance between the person and the equipment is 1 m, and the predicted closest approach time is 20 seconds later, resulting in an instruction to reduce the equipment's speed.

[0085] In line 433, when the responsiveness level is "3" and the status information indicates a sudden downpour, the rainfall is 100m and has continued for 5 hours, and rainfall is confirmed from the rainfall conditions in the surrounding image data, and it is determined that activity cannot continue, so an instruction is given to stop the equipment.

[0086] In line 434, when the importance level is "1" and information indicating environmental destruction is notified as status information, the closest distance between people and equipment is 5 m, and the analysis results of the image and distance data indicate that there are other equipment in the vicinity but no people, so an instruction is given to evacuate the equipment.

[0087] Line 435 indicates that when the importance level is "2" and the status information indicates that the equipment is self-destructing, a specific frequency of 90 Hz is detected from the sound of the equipment for one minute or more, and a temperature of 90 degrees or higher is detected at a point where this temperature continues for one minute or more, and therefore the equipment is to be shut down.

[0088] Line 436 indicates that when the importance level is "3" and the status information indicates that maintenance is required, it will detect a specific frequency (90Hz), detect a part with a temperature of 90 degrees or higher, and determine that the continuous operating time is 50 hours or more, and based on its own position, surrounding images, and distance data, it will determine that the required travel time is 2 hours and the remaining operating time is 6 hours, and therefore will instruct the robot to finish work in a maximum of 4 hours and move on to maintenance work.

[0089] Line 437 indicates that when the intervention level is "1" and the status information indicates that expert judgment is required, the expert will recognize, based on the person's position coordinates, image data of the person, distance data of the person, temperature and humidity around the person, position coordinates of the equipment, image data of the surroundings, distance data of the surroundings, temperature and humidity around the person, rainfall, sound of the equipment, and temperature of the equipment, that two pieces of equipment are approaching each other, that multiple abnormalities have occurred, and that a person is present near the evacuation destination, and therefore will instruct the expert to stop three pieces of equipment and issue an alert to the people.

[0090] In line 438, when the intervention level is "2" and the status information indicates that an artificial intelligence instruction is required, the artificial intelligence determines from the analysis results of the image data database based on the same data group as when the intervention level is "1" that the unknown object is an abnormally intruding vehicle, and therefore instructs the system to stop operation and to issue an alert to people and abnormally intruding vehicles in the vicinity. In line 439, when the intervention level is "3" and the status information indicates that only first aid is required, no instructions are given.

[0091] FIG. 13 is a diagram showing the contents of the work procedure manual table 23. As shown in FIG. Line 1311 indicates the judgment condition for the first operation process A1 in process A, where the operation ID is "A". The determination specified by line 1311 is as follows: - Regarding the number of times the left hand is detected, if it is between a and b times, it is determined that there is no abnormality. - Regarding the number of times the right hand is detected, if it is between c and d times, it is determined that there is no abnormality. Regarding the timing at which detection of the left hand begins, if it occurs between e and f seconds after the start of work process A1, it is determined that there is no abnormality. Regarding the timing at which the right hand detection begins, if it occurs between g and h seconds after the start of work process A1, it is determined that there is no abnormality. Regarding the timing at which the detection of the left hand ends, if it is between i and j seconds after the start of work process A1, it is determined that there is no abnormality. Regarding the timing at which the detection of the right hand ends, if it is between k and m seconds after the start of work process A1, it is determined that there is no abnormality. Regarding the detection time length for the left hand, if it is between n and p seconds per time, it is determined to be normal. Regarding the detection time length for the right hand, if it is between q and r seconds per time, it is determined to be normal. Regarding the presence or absence of a next process, if there is a work process A2 and the interval between the work process A2 is between s and t seconds, it is determined that there is no abnormality.

[0092] For example, if it is necessary to determine the detection start timing and detection end timing as separate timings for each hand detection, this can be achieved by assigning each hand detection to one work process (e.g., A1, A2, etc.).

[0093] Line 1312 indicates the judgment condition for operation process A2, which has operation ID "A" and is the second operation in process A. The determination specified by line 1312 is as follows: Regarding the number of times the left hand is detected, if it is between a' and b' times, it is determined that there is no abnormality. - Regarding the number of times the right hand is detected, if it is between c' and d', it is determined that there is no abnormality. Regarding the timing at which detection of the left hand begins, if it occurs between e' and f' seconds after the start of work process A2, it is determined that there is no abnormality. Regarding the timing at which detection of the right hand begins, if it occurs between g' and h' seconds after the start of work process A2, it is determined that there is no abnormality. Regarding the timing at which detection of the left hand ends, if it is between i' and j' seconds after the start of work process A2, it is determined that there is no abnormality. Regarding the timing at which the detection of the right hand ends, if it is between k' and m' seconds after the start of work process A2, it is determined that there is no abnormality. - Regarding the detection time length for the left hand, if it is between n' and p' seconds per time, it is determined to be normal. - Regarding the detection time length for the right hand, if it is between q' and r' seconds per time, it is determined that there is no abnormality. Regarding the presence or absence of a next process, if there is a work process A3 and the interval between the work process A3 is between s' and t' seconds, it is determined that there is no abnormality.

[0094] Line 1313 shows the judgment conditions for task step A3, which has task ID "A" and is the third task step in step A. The number of times the left hand is detected and the length of time the right hand is detected are the same as in lines 1311 and 1312, so they are omitted here, but the presence or absence of a next task is indicated as "none." Line 1314 indicates the judgment conditions for the operation with operation ID "B", but since it is the same as lines 1311 to 1313, it is omitted here.

[0095] 14 is a flowchart showing the unsteady state determination process (process S10) of process S2 in the measurement target device 10. Refer to FIGS. 2 and 13 as appropriate. When the communication unit 25 of the measurement target device 10 receives the job ID, it transfers the job ID to the processing unit 11, and the processing unit 11 acquires the job ID (process S1411). The processing unit 11 refers to the work procedure manual table 23 and checks the job ID in the table (process S1412).

[0096] If the target task ID corresponds to the task ID in the table (Yes in step S1413), the sensor information acquisition unit 12 of the processing unit 11 acquires the sensor information (step S1414). If not (No in step S1413), the processing unit 11 determines that the task is not subject to measurement and terminates abnormally (step S1424).

[0097] The sensor information acquisition unit 12 acquires information from the sensor to be measured (process S1414) and associates the sensor information with the task ID (process S1415). When the task starts, the processing unit 11 starts sensing (process S1416) and waits for hand information to be received from the sensor (process S1417). Upon receiving the hand information, the abnormality detection unit 13 refers to the judgment conditions in the work procedure manual table 1310 described in FIG. 13, performs judgment on the judgment conditions of the corresponding task ID (process S1418), and determines whether the judgment result is abnormal (process S1419). If the judgment result indicates an abnormality (process S1419, No), the abnormality detection unit 13 issues an abnormality notification (process S1422) and terminates the process with an abnormality (process S1423). If the judgment result indicates no abnormality (process S1419, Yes), the processing unit 11 checks whether there are any next judgment conditions or next task steps (process S1420). If there is a next process (Yes in process S1420), the processing unit 11 waits for reception of hand information from the sensor (process S1417). If there is no next determination process (No in process S1420), the processing unit 11 ends the process (process S1421).

[0098] As described above, the system disclosed in the embodiment comprises a sensor information acquisition unit 12 as sensor information acquisition means that acquires sensor information indicating detection results from sensors 26, 27 that detect the state of a worker, a memory unit 20 as work check data storage means that stores in advance the actions of the worker that are expected from the work procedure as work check data, and anomaly detection unit 13 as anomaly detection means that compares the sensor information with the work check data to detect anomalies in the work. This configuration makes it possible to detect abnormalities in work at an early stage and support the worker, thereby improving the efficiency and quality of the work.

[0099] The disclosed system also includes a measurement target device 10 as a target device equipped with the sensor information acquisition means, the work check data storage means, and the anomaly detection means and provided near the work target that is the object of the work and / or at the work target itself, and a management device 30 capable of communicating with the target device. The target device further includes control means (e.g., anomaly processing unit 14, flag generation unit 15) that, when the anomaly detection means detects an abnormality, creates flag data including a flag indicating an abnormality level and status information indicating an abnormal state, and transmits the flag data to the management device, and the management device transmits instruction information to at least one of the target device, the worker, and the work target based on the flag data received from the target device. As a result, since the size of the notification data (flag data 21 (see Figure 3)) from the measurement target device 10 to the management device 30 is small, even if a large number of notifications are generated simultaneously from a large number of locations, the notifications can be sent to the management device 30 in real time.

[0100] The sensor information includes information that can at least identify whether the worker is reaching out for the work object, and the work check data indicates the number of times and timing at which the worker reaches out for the work object. Furthermore, the sensor is a portable measuring device, the work check data storage means stores multiple work check data corresponding to multiple work objects, and the abnormality detection means selects the work check data based on information identifying the work object and compares the selected work check data with the sensor information to detect abnormalities. With this configuration, abnormalities in the work of a worker can be detected easily and inexpensively.

[0101] In addition, the management device 30 has a memory unit 40 that stores an acquired data judgment table 41 that associates abnormality levels, status information, and acquired data, and when flag data 21 is received from the measurement target device 10, it can send an instruction to the measurement target device 10 to acquire the necessary sensing data depending on the abnormality level of the flag data 21.

[0102] Furthermore, the management device 30 stores in the memory unit 40 a countermeasure method determination table that determines a countermeasure method based on the abnormality level, status information, and sensing data, and when sensing data is received from the measurement target side device 10, it can send instruction information, which is a countermeasure method for the measurement target, to the measurement target side device 10 based on the countermeasure method determination table (e.g., instruction determination table 42).

[0103] The flag data 21 includes responsiveness information for determining whether there is a human abnormality or whether there is an impact on people, importance information for determining whether there is a significant impact on the surrounding environment, and intervention information for determining whether expert intervention is required among remote instructions. This reduces the processing load on the management device 30, as the urgency and importance are determined before notification from the measurement target device 10 to the management device 30. In addition, the management device 30 has the advantage of being able to easily determine priority and respond to multiple measurement targets at the same time. In this embodiment, the measurement targets are described as infrastructure equipment and on-site workers, but are not limited to this. For example, the measurement targets may be manufacturing equipment and on-site workers at a manufacturing site.

[0104] As such, the disclosed invention has a wide range of applications, and contributes to solving problems in the fields of manufacturing and infrastructure maintenance, for example. In the fields of manufacturing and infrastructure maintenance, labor shortages are a challenge, especially in countries and regions experiencing population decline and an aging population. To address this issue, various measures have been implemented, including streamlining business processes, encouraging greater participation in the workforce by women, the elderly, and foreigners, and reducing the number of personnel required for operations and promoting automation. Information and communication technology (ICT) has been utilized to improve operational efficiency, reduce labor, and automate operations. For example, tablet PCs and AR / VR technologies have been applied to support on-site workers. However, a challenge with ICT-based measures is how to provide support for non-routine or highly difficult tasks that are difficult for unskilled personnel to handle. When a limited number of skilled personnel are required to support unskilled workers on-site, a shortage of skilled personnel limits the number of projects, resulting in inefficiency. Therefore, by using the disclosed system, it is possible to provide unskilled workers on-site with support from skilled workers at an early stage. [Explanation of symbols]

[0105] 10: Measurement target device, 11: Processing unit, 12: Sensor information acquisition unit (sensor information acquisition means), 13: Abnormality detection unit (abnormality detection means), 14: Abnormality processing unit (control means), 15: Flag generation unit (control means), 20: Storage unit, 21: Flag data, 22: Emergency treatment determination table, 23: Work procedure manual table, 25: Communication unit (communication means), 26, 27: Sensor, 30: Management device, 31: Processing unit, 32: Priority determination unit, 33: Calculation Measurement target state monitoring unit, 34: Sensing data selection unit, 35: Sensing data analysis unit, 40: Memory unit, 41: Acquired data judgment table, 42: Instruction judgment table (countermeasure judgment table), 43: Instruction content, 44: Notification data, 100: Abnormal state monitoring system, 214: Responsiveness information, 215: Importance information, 216: Intervention information, 214L, 215L, 216L: Level, 214S, 215S, 216S: State information

Claims

1. a target-side device provided in the vicinity of a work object that is a target of work and / or in the work object itself, and a management device capable of communicating with the target-side device; The target device is a sensor information acquisition means for acquiring sensor information indicating a detection result from a sensor that detects the state of the worker; a work check data storage means for storing in advance the operations expected from the work procedure as work check data; an abnormality detection means for detecting an abnormality in the work by comparing the sensor information with the work check data; a control means for, when the abnormality detection means detects an abnormality, creating flag data including a flag indicating an abnormality level and status information indicating an abnormal state, and transmitting the flag data to the management device; the management device transmits instruction information to at least one of the target device, the worker, and the work target based on flag data received from the target device; the sensor information includes information that can identify at least whether the worker is reaching out his / her left hand or right hand toward the work object, The work check data indicates, for each of the left hand and the right hand, a detection count range which is a range of the number of times that the hand is detected reaching out to the work object, a start timing range which is a range of timings at which hand detection starts, an end timing range which is a range of timings at which hand detection ends, and a detection time range which is a range of length of hand detection time per hand detection, The anomaly detection means detects an anomaly in the work based on a comparison between the sensor information and the work check data to determine, for each of the worker's left and right hands, whether the number of hand detections falls within the detection number range, whether the timing of hand detection start falls within the start timing range, whether the timing of hand detection end falls within the end timing range, and whether the length of hand detection time per hand detection falls within a detection time range. An abnormal condition monitoring system characterized by:

2. the sensor is a portable measuring device; The work check data storage means stores a plurality of work check data corresponding to a plurality of work objects, The abnormal state monitoring system according to claim 1, characterized in that the abnormality detection means selects the work check data based on information identifying the work object, and detects an abnormality by comparing the selected work check data with the sensor information.

3. The management device a storage unit that stores a data acquisition determination table that associates the abnormality level, the status information, and sensing data that needs to be acquired; When the flag data is received from the target device, an instruction to acquire necessary sensing data is transmitted to the target device according to the abnormality level of the flag data.

2. The abnormal condition monitoring system according to claim 1.

4. The management device a countermeasure determination table for determining a countermeasure based on the abnormality level, the state information, and the sensing data is stored in the storage unit; When the sensing data is received from the target device, a countermeasure method identified based on the countermeasure method determination table is transmitted to the target device as instruction information.

4. The abnormal condition monitoring system according to claim 3.

5. The flag data includes: Responsiveness information for determining whether a person is abnormal or has an effect on the person; Importance information for determining whether or not the information has a significant impact on the surrounding environment; and intervention information for determining whether expert intervention is required in the remote instruction.

2. The abnormal condition monitoring system according to claim 1.

6. 2. The abnormal condition monitoring system according to claim 1, wherein the work target is infrastructure equipment at a construction site, and the worker is a site inspector.

7. a work check data storage step in which a target-side device provided near the work object that is the target of the work and / or at the work object itself stores operations expected from the work procedure as work check data; a sensor information acquisition step in which the target side device acquires sensor information indicating a detection result from a sensor that detects the state of the worker; an anomaly detection step in which the target side device compares the sensor information with the work check data to detect an anomaly in the work; a step of, when the target side device detects an abnormality in the abnormality detection step, creating flag data including a flag indicating an abnormality level and status information indicating an abnormal state, and transmitting the flag data to a management device that can communicate with the target side device; Including, the management device transmits instruction information to at least one of the target device, the worker, and the work target based on flag data received from the target device; the sensor information includes information that can identify at least whether the worker is reaching out his / her left hand or right hand toward the work object, The work check data indicates, for each of the left hand and the right hand, a detection count range which is a range of the number of times the worker has detected reaching out to the work object, a start timing range which is a range of timings at which hand detection starts, an end timing range which is a range of timings at which hand detection ends, and a detection time range which is a range of length of hand detection time per hand detection, In the anomaly detection step, a comparison between the sensor information and the work check data detects an anomaly in the work based on whether, for each of the worker's left hand and right hand, the number of hand detections falls within the detection number range, whether the timing of hand detection start falls within the start timing range, whether the timing of hand detection end falls within the end timing range, and whether the length of hand detection time per hand detection falls within a detection time range. An abnormal state detection method comprising:

Images