System for monitoring crash inhibition apparatus

Abstract

To provide a system for monitoring a crash inhibition apparatus with a reduced introduction cost, not requiring reformation or the like of an existing crash inhibition apparatus.SOLUTION: A system for monitoring a crash inhibition apparatus 10 including a hook holder 40 which holds a hook 32 of a lanyard 30 comprises: a microphone 110; a learned model 120 learned to output a use state of the crash inhibition apparatus from input sound by defining as learning data, sound data including a first sound to be generated when the hook 32 of the lanyard 30 is held by a holding member of the hook holder 40, a second sound to be generated when the hook 32 is removed from the hook holder 40, a third sound to be generated when the hook 32 is hooked from an attachment facility 2, a fourth sound to be generated when the hook 32 is removed from the attachment facility; and an output section 130 for outputting at least one of the use state of the crash inhibition apparatus or warning based on an output result obtained by inputting the sound collected by the microphone 110 into the learned model 120.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a system for monitoring the usage status of a fall prevention device for workers at high work sites.

Background Art

[0002] When working at heights such as at construction sites, workers are obliged to wear fall prevention devices. The fall prevention device includes a body belt which is a fall prevention device body worn on the worker's body, and a lanyard having one end connected to the body belt and a hook provided at the other end. Prior to performing high-altitude work, the worker wears the fall prevention device, and at the high-altitude work site, the hook is hung on a main rope, a handrail of a scaffold, or a mounting facility such as a building, and then the high-altitude work is performed. Note that the fall prevention device is also called a safety belt.

[0003] In order to prevent a fall accident, it is important that the worker correctly and surely uses the fall prevention device. Therefore, a system for prompting the worker to wear the fall prevention device and managing the usage status has been proposed (see Patent Document 1). In the device described in Patent Document 1, sensors such as a pressure sensor and an open / close sensor, and a transmission circuit for wirelessly transmitting a sensor signal are provided on the hook, and a control box is provided on the body belt. The control box includes a sensor signal receiving unit, a speaker, an LED, and a control device. The control device of the control box performs various notification controls using the speaker and the LED based on the sensor signal received from the hook, and transmits the usage status of the fall prevention device to a personal computer by wireless communication. The personal computer manages the usage status received from the control box and performs notification control as necessary. The administrator monitors the usage status of the worker's fall prevention device by operating the personal computer.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the device described in Patent Document 1, a dedicated hook with sensors attached thereto is used as a fall prevention device, and there is a problem in that the lanyard of an existing fall prevention device cannot be used and the introduction cost is high. Further, in the device described in Patent Document 1, since the control box is integrally provided on the body belt, there is a problem in that an existing body belt cannot be used and the introduction cost is high.

[0006] The present invention has been made in view of the above circumstances, and an object thereof is to provide a monitoring system for a fall prevention device that does not require modification of an existing fall prevention device and has a low introduction cost.

Means for Solving the Problems

[0007] To achieve the above object, the invention of the present application is a monitoring system for monitoring the usage status of a fall prevention device, wherein the fall prevention device includes a fall prevention device main body worn by a user, and a lanyard having one end connected to the fall prevention device main body and a hook provided at the other end, and a hook holding member for holding the hook is provided on the fall prevention device main body, and the usage status monitoring system includes a microphone for collecting sound around the user, sound generated during the start operation of using the lanyard in which the hook is removed from the hook holding member and then hung on the mounting equipment, and sound generated during the end operation of using the lanyard in which the hook is removed from the mounting equipment and then held by the hook holding member as learning data, and a learned model learned to output whether the usage status of the fall prevention device is in use or not in use from the input sound, and an output unit for outputting at least one of the usage status of the fall prevention device or a warning based on the usage status based on the output result obtained by inputting the sound collected by the microphone into the learned model.

Advantages of the Invention

[0008] According to the present invention, using a learned model learned from sound data including sounds generated during the use of a fall prevention device, specifically, the sound generated during the start operation of using a lanyard that is removed from a hook holding member and then hooked onto a mounting facility, and the sound generated during the end operation of using a lanyard that is removed from the mounting facility and then held by the hook holding member, the usage status of the fall prevention device is detected based on the sound around the user collected by a microphone. Therefore, it is not necessary to modify the fall prevention device, etc., and the introduction cost is low.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0010] (First Embodiment) A monitoring system for a fall prevention device according to a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a configuration diagram of the monitoring system, and FIG. 2 is a functional block diagram of a mobile terminal.

[0011] The monitoring system for the fall prevention device according to this embodiment monitors the usage status of the fall prevention device 10 worn by the worker 1 who is the user. As shown in FIG. 1, this monitoring system includes a mobile terminal 100 carried by the worker 1 wearing the fall prevention device 10, a management server 200, and a manager terminal 300 used by the manager.

[0012] The fall prevention device 10 is a conventionally well-known full harness type fall prevention device. This fall prevention device 10 includes a body belt 20 which is a fall prevention device body worn on the body of the worker 1, and two left and right lanyards 30, one end of which is connected to the body belt 20 and the other end of which is provided with a hook 32.

[0013] The body belt 20 is what is called a so-called full harness type, and includes a pair of left and right shoulder belts 21 that cross on the back side and are hung on both shoulders of the body, a leg belt 22 that is connected to the shoulder belt 21 and is worn on the legs of the body, and a chest belt 23 that connects the pair of shoulder belts 21 in the vicinity of the chest of the body. On the left and right shoulder belts 21, a hook holder 40 for holding the hook 32 of the lanyard 30 is attached slightly above the connecting portion of the chest belt 23. The hook holder 40 is composed of a substantially horizontally long rectangular annular member. The upper long side of the hook holder 40 is sewn to the shoulder belt 21 so that the hook holder 40 can rotate around the upper long side as an axis. The material of the hook holder 40 is not limited as long as it has a predetermined strength, and it may be made of resin or metal.

[0014] The lanyard 30 includes a rope 31, a connecting tool for connecting to the body belt 20 provided on one end side of the rope 31, and the hook 32 provided on the other end side of the rope 31. The lanyard 30 may further include a shock absorber. The hook 32 includes a hook body with a hook-shaped tip, an opening / closing lever rotatably attached to the ventral side of the hook body to open and close the hook of the hook body from the inside, and a safety lever rotatably attached to the dorsal side of the hook body to hold the state where the opening / closing lever closes the hook. The material of the hook 32 is not limited as long as it has a predetermined strength, and it may be made of resin or metal.

[0015] As shown in FIG. 2, the mobile terminal 100 includes a microphone 110 that converts the sound around the worker 1 into an electrical signal, a learned model 120, an output control unit 130, a position detection unit 140, a moving speed detection unit 150, a communication unit 160, and a display unit 170. The mobile terminal 100 is a well-known computer with a communication function carried by the worker 1. Typically, the mobile terminal 100 is composed of a high-function mobile communication terminal called a smartphone with a built-in microphone 110, a communication unit 160, a display unit 170, etc. The learned model 120, the output control unit 130, the position detection unit 140, and the moving speed detection unit 150 are configured by installing a program in a computer via a network or from a predetermined storage medium. Also, the mobile terminal 100 may be configured as a dedicated terminal with each part 110 to 170 implemented in hardware. The mobile terminal 100 may be housed in the pocket of the worker 1's clothes, or may be fixed to the clothes or the fall arrest device 10 via a predetermined fixture. Also, the mobile terminal 100 may be a wearable computer such as a wristwatch type or glasses type. From the viewpoint of the stability of recognition accuracy, it is preferable that the mounting position of the mobile terminal 100, particularly the microphone 110, with respect to the worker 1 is fixed.

[0016] The learned model 120 is a learning device that is machine-learned to output whether the usage status of the fall arrest device 10 is "in use" or "not in use" from the input sound data, using the sound data including the sound generated according to the usage status of the fall arrest device 10 as learning data. Here, the usage status of the fall arrest device 10 being "in use" means a state where the hook 32 of at least one of the two lanyards 30 is hung on the mounting facility 2. On the other hand, the usage status of the fall arrest device 10 being "not in use" means a state other than the above "in use" state. Also, the mounting facility 2 includes structures such as a main rope or a single pipe installed for work, or structures such as a building that is the object of work, and means those that can be used as the object of use of the fall arrest device.

[0017] The sound data used as the learning data includes the sound data of the sound generated during the start operation of using the lanyard 30, which is to be hung on the mounting facility 2 after removing the hook 32 from the hook holder 40, and the sound generated during the end operation of using the lanyard 30, which is to be held by the hook holder 40 after removing the hook 32 from the mounting facility 2. More specifically, it includes the first sound generated when removing the hook 32 from the hook holder 40, the second sound generated when hanging the hook 32 on the mounting facility 2, the third sound generated when removing the hook 32 from the mounting facility 2, and the fourth sound generated when holding the hook 32 by the hook holder 40.

[0018] Here, as shown in FIG. 3, the sound generated during the start operation of using the lanyard 30 is a series of sounds from the first sound generated when removing the hook 32 from the hook holder 40 to the second sound generated when hanging the hook 32 on the mounting facility 2, which occurs after a certain period of time from the generation of the first sound. Similarly, as shown in FIG. 4, the sound generated during the end operation of using the lanyard 30 is a series of sounds from the third sound generated when removing the hook 32 from the mounting facility 2 to the fourth sound generated when hanging the hook 32 on the hook holder 40, which occurs after a certain period of time from the generation of the third sound. These sounds vary depending on conditions such as the structure and material of the hook 32, the structure and material of the hook holder 40 and the mounting facility 2, and the habits of the operator 1. Therefore, it is preferable to use a large number of sound data collected under various conditions as the learning data. Also, it is preferable to use the sound generated within a predetermined range (for example, a range of radius 1 m) from the subject as the learning data.

[0019] The learned model 120 extracts, from the input sound, for example, the frequency, frequency distribution, sound pressure, generation interval of the series of sounds described above, sound pressure difference of the series of sounds, etc. for each of the first to fourth sounds as feature quantities, and is configured to classify and output whether the usage status of the fall prevention device 10 is "in use" or "not in use" based on these feature quantities. Here, the above-described feature quantities are merely examples, and any combination of feature quantities can be used to obtain a desired output result. Also, various classification algorithms of the learned model 120 can be used. For example, Decision Tree and k-NN can be mentioned.

[0020] The output control unit 130 controls to output to the outside at least one of the usage status of the fall prevention device 10 or a warning based on the usage status based on the detection result obtained by inputting the sound collected by the microphone 110 into the learned model 120.

[0021] The output control unit 130 can be configured to output a predetermined warning when the state where the output of the learned model 120 is not in use continues for a predetermined time. Also, the output control unit 130 can be configured to output a predetermined warning when the current position of the worker 1 detected by the position detection unit 140 described later is within a preset warning target area and the output of the learned model 120 is not in use. Also, the output control unit 130 can be configured to output a predetermined warning when the moving speed of the worker 1 detected by the moving speed detection unit 150 described later is equal to or lower than a predetermined threshold value and the output of the learned model 120 is not in use. Also, the output control unit 130 can be configured by appropriately combining the above-described control based on the duration of the output of the learned model 120, control based on the current position, and control based on the moving speed.

[0022] In this embodiment, the output control unit 130 transmits the usage status of the fall prevention device 10 to the management server 200 via a network using the communication unit 160 of the mobile terminal 100. Here, the output control unit 130 can be configured to transmit the current position of the worker 1 detected by the position detection unit 140 described later and the moving speed of the worker 1 detected by the moving speed detection unit 150 described later to the management server 200 as additional information. Here, the transmission process of the usage status and the additional information to the management server 200 may be performed periodically (for example, every 10 seconds), may be performed at the timing when a change occurs in the output of the learned model 120, or may be a combination of both.

[0023] Also, in this embodiment, the output control unit 130 is configured to output a warning to the display unit 170 of the mobile terminal 100 using the notification function of the mobile terminal 100. In other embodiments, the output control unit 130 can be configured to output a warning to a display worn by the worker 1. Here, the display is, for example, attached to the fall prevention device 10 or the helmet and is composed of a display element such as an LED. Also, the output control unit 130 can be configured to cancel the reported warning by a predetermined operation using a predetermined input means of the mobile terminal 100.

[0024] The position detection unit 140 is a position detection means for detecting the current position of the mobile terminal 100, that is, the current position of the worker 1. The position detection unit 140 can acquire the position information (latitude, longitude, altitude) of the worker 1 on the earth using the position detection function of the mobile terminal 100 and use this position information as the detection result. The position detection function of the mobile terminal 100 can be configured by appropriately combining any sensing technologies such as a satellite positioning system, a barometric pressure sensor, an acceleration sensor, and an angular velocity sensor.

[0025] The moving speed detection unit 150 is a moving speed detection means for detecting the moving speed of the mobile terminal 100, that is, the moving speed of the worker 1. The moving speed detection unit 150 can obtain the position information (latitude, longitude, altitude) of the worker 1 on the earth using the position detection function of the mobile terminal 100, and use the change over time of this position information in the horizontal direction as the detection result of the calculated moving speed. Also, the moving speed detection unit 150 can calculate the moving speed in the horizontal direction using the acceleration sensor of the mobile terminal 100, and use this moving speed as the detection result.

[0026] The management server 200 is a well-known computer arranged on the network. In the present embodiment, the management server 200 is arranged on the Internet. That is, the management server 200 is implemented as a so-called cloud server. In another embodiment, the management server 200 is arranged within the mobile communication network to which the mobile terminal 100 is connected. That is, the management server 200 is implemented as a so-called edge server.

[0027] As shown in FIG. 5, the management server 200 has a receiving unit 210 that receives the usage status and additional information of the fall prevention device 10 from a plurality of mobile terminals 100 and stores and holds them in a predetermined storage unit 220, and a user interface unit 230 that provides various information regarding the usage status of the fall prevention device 10 to the administrator terminal 300.

[0028] The user interface unit 230 provides various data such as the usage status and position information of the fall prevention device 10 for each worker of the fall prevention device 10 to the administrator terminal 300. The user interface unit 230 can be configured to provide the administrator terminal 300 with a set of the positions of each worker and the usage status of the fall prevention device 10 as a 3D image. Also, the user interface unit 230 can be configured to provide the administrator terminal 300 with the appropriateness of the usage status of the fall prevention device 10. The appropriateness of the usage status of the fall prevention device 10 can use the duration of the usage status, the position information and moving speed information received as additional information, similar to the warning process in the above-described mobile terminal 100.

[0029] The administrator terminal 300 is a well-known computer used by an administrator. The administrator terminal 300 may be a mobile terminal carried by the administrator or a computer installed at a predetermined location such as an office. The administrator terminal 300 is configured to be communicable with the management server 200 via a network and functions as a client of the management server 200. The administrator terminal 300 can refer to various information such as the usage status and location information of the fall arrest device 10 for each worker in real time via the user interface unit 230 of the management server 200. Further, the administrator terminal 300 can set the operation of the management server 200 via the user interface unit 230 of the management server 200, such as setting a warning target area, for example.

[0030] An example of the management screen displayed on the administrator terminal 300 is shown in FIG. 6. The management screen is composed of information provided from the user interface unit 230 of the management server 200. In the example of FIG. 6, the management screen displays an image of the work site and the usage status of the fall arrest device 10 for each worker. The icon indicating the worker on the management screen is mapped and displayed on the image of the work site based on the position information of the worker. Further, the icon indicating the worker is displayed in a mode (color in the example of FIG. 6) corresponding to the usage status. Further, the image of the work site is displayed so as to be able to identify a preset warning target area. With such a management screen, the administrator can grasp the position of the worker and the usage status of the fall arrest device 10 in real time.

[0031] According to the monitoring system for the fall arrest device according to the present embodiment, the usage status of the fall arrest device 10 is detected based on the sound around the worker 1 collected by the microphone 110 using the learned model 120 learned from the sound data of the sound generated when the fall arrest device 10 is used. Therefore, it is not necessary to modify the fall arrest device 10 and the introduction cost is low.

[0032] (Second Embodiment) A monitoring system for a fall prevention device according to a second embodiment of the present invention will be described. The difference between this embodiment and the first embodiment lies in the configuration of the learned model. Since the other configurations, operations, learning data, etc. are the same as those in the first embodiment, only the differences will be described here.

[0033] In this embodiment, as shown in FIG. 7, the learned model 120 includes a first learned model 121 that outputs, as a detection result, the start operation or end operation of the lanyard 30 from the input sound data, and a second learned model 122 that outputs whether the usage status of the fall prevention device 10 is "in use" or "not in use" from the output of the first learned model 121.

[0034] The first learned model 121 is a learning device that is machine-learned to output "usage operation detection" and "usage end detection" of the lanyard 30, using, as learning data, the sound data of the start sound of use generated during the start operation of the lanyard 30 when the hook 32 is removed from the hook holder 40 and then hung on the mounting facility 2, and the sound data of the end sound of use generated during the end operation of the lanyard 30 when the hook 32 is removed from the mounting facility 2 and then held by the hook holder 40.

[0035] The first learned model 121 is learned to output "usage operation detection" and "usage end detection" separately for each of the left and right lanyards 30. That is, the first learned model 121 is learned to identify the left and right lanyards 30.

[0036] The first pre-trained model 121 extracts, as feature quantities, from the input sound, for example, the frequency, frequency distribution, sound pressure, generation interval of the series of sounds described above, sound pressure difference of the series of sounds, etc. for each of the first to fourth sounds, and is configured to classify and output the lanyard 30 into "usage operation detection" and "usage end detection" based on this feature quantity. Here, the above-described feature quantities are merely examples, and any combination of feature quantities can be used so as to obtain a desired output result. Also, various classification algorithms can be used for the first pre-trained model 121. For example, Decision Tree and k-NN can be mentioned.

[0037] The second pre-trained model 122 is a learning device that has been learned using, as learning data, the combination pattern of the results most recently output from the pre-trained model 121 for the left and right lanyards 30. Specifically, when the output result of the first pre-trained model 121 for at least one of the lanyards 30 is "usage operation detection", the second pre-trained model 122 outputs that the usage status of the fall arrest device 10 is "in use". On the other hand, when the output results of the first pre-trained model 121 for both lanyards 30 are "usage end detection", the second pre-trained model 122 outputs that the usage status of the fall arrest device 10 is "not in use".

[0038] According to the monitoring system for the fall arrest device according to the present embodiment, similarly to the first embodiment, the pre-trained model 120 learned from the sound data of the sound generated when the fall arrest device 10 is used is used, and the usage status of the fall arrest device 10 is detected based on the sound around the worker 1 collected by the microphone 110. Therefore, modification of the fall arrest device 10 or the like is not required, and the introduction cost is low.

[0039] (Third Embodiment) A monitoring system for a fall prevention device according to a third embodiment of the present invention will be described. The difference between this embodiment and the second embodiment lies in the configuration of the learned model. Since other configurations, operations, learning data, etc. are the same as those in the second embodiment, only the differences will be described here.

[0040] The learned model 120 according to the third embodiment includes a first learned model 121 that is learned to detect the first to fourth sounds from the input sound data and output the detection results, and a second learned model 122 that is learned to output whether the usage status of the fall prevention device 10 is "in use" or "not in use" based on the output of the first learned model 121 and the change over time of the output.

[0041] The first learned model 121 is configured to extract, from the input sound, for example, the frequency, frequency distribution, sound pressure, etc. of each of the first to fourth sounds as feature amounts, and classify and output each sound based on these feature amounts. Here, the above-mentioned feature amounts are merely examples, and any combination of feature amounts can be used to obtain a desired output result. Also, various classification algorithms can be used for the first learned model 120. For example, Decision Tree and k-NN can be mentioned. Further, the first learned model 121 is learned to recognize which of the left and right lanyards 30 the detected first to fourth sounds belong to, and add and output the recognition result.

[0042] The second trained model 122 is a learning device that is machine-learned to output whether the usage status of the fall prevention device 10 is "in use" or "not in use" using, as learning data, the sound generation pattern that occurs over time during the start-up operation of the lanyard 30 that hooks the hook 32 onto the attachment equipment 2 after removing the hook 32 from the hook holder 40, and the sound generation pattern that occurs over time during the end-of-use operation of the lanyard 30 that removes the hook 32 from the attachment equipment 2 and then holds it with the hook holder 40. Here, as shown in FIGS. 3 and 4, the sound generation pattern includes at least the type of the first sound, the type of the sound that subsequently occurs, and the time from the first sound until the subsequently occurring sound occurs.

[0043] According to the monitoring system for the fall prevention device according to the present embodiment, similar to the first embodiment, the trained model 120 learned from the sound data of the sound generated during the use of the fall prevention device 10 is used, and the usage status of the fall prevention device 10 is detected based on the sound around the worker 1 collected by the microphone 110. Therefore, modification of the fall prevention device 10 is not required and the introduction cost is low.

[0044] As described above in detail for the first to third embodiments of the present invention, the present invention is not limited to the above embodiments, and various improvements and modifications may be made without departing from the gist of the present invention.

[0045] For example, in the above embodiment, the trained model 120 was learned so as to be able to recognize which lanyard 30 of the left and right lanyards 30 the sound related to the hook 32 occurred in, but it may be learned so as not to recognize the distinction between left and right. In this case, the trained model 120 may be learned to output the usage status of the fall prevention device 10 based on the temporal sound generation pattern of the sound related to the hooks 32 of the left and right lanyards 30. For example, when the sound of hooking the hook 32 onto the hook holder 40 is detected continuously twice, the trained model 120 is learned to output that the fall prevention device 10 is not in use.

[0046] In the above-described embodiment, as the sound data for training the trained model 120, in order to correspond to the structure and material of the hook 32, the structure and material of the hook holder 40 and the mounting equipment 2, and the habits of the worker 1, a large number of sound data collected under various conditions were used. However, the trained model 120 may be formed using training data with specific conditions fixed. For example, a trained model 120 that only corresponds to the resin hook holder 40 may be formed. In such a case, depending on the usage and environment as conditions, the trained model 120 corresponding to the conditions may be appropriately installed in the mobile terminal 100.

[0047] In the above-described embodiment, a smartphone was used as the mobile terminal 100, and the microphone 110 used was the one built into the mobile terminal 100. However, an external microphone may also be used.

Explanation of Reference Numerals

[0048] 1…Worker 2…Mounting Equipment 10…Fall Arrest Equipment 30…Lanyard 32…Hook 40…Hook Holder 100…Mobile Terminal 110…Microphone 120…Trained Model 121…First Trained Model 122…Second Trained Model

Claims

1. A monitoring system for monitoring the usage status of a fall prevention device, comprising: The fall prevention device includes a fall prevention device main body to be worn by a user, and a lanyard having one end connected to the fall prevention device main body and a hook provided at the other end. The fall prevention device main body is provided with a hook holding member for holding the hook. The monitoring system includes: A microphone for collecting sounds around the user; A learned model that is trained to output whether the usage status of the fall prevention device is in use or not in use from the input sound, using as learning data sound data including the sound generated during the start operation of using the lanyard, where the hook is removed from the hook holding member and then hooked onto the mounting equipment, and the sound generated during the end operation of using the lanyard, where the hook is removed from the mounting equipment and then held by the hook holding member; An output unit that outputs at least one of the usage status of the fall prevention device or a warning based on the usage status, based on the output result obtained by inputting the sound collected by the microphone into the learned model. A monitoring system for a fall prevention device, characterized by the above.

2. The sound data used in the training of the learned model includes a series of sounds from the generation of the first sound generated when the hook is removed from the hook holding member to the generation of the second sound generated when the hook is hooked onto the mounting equipment after a certain time from the generation of the first sound, and a series of sounds from the generation of the third sound generated when the hook is removed from the mounting equipment to the generation of the fourth sound generated when the hook is held by the hook holding member after a certain time from the generation of the third sound. The monitoring system for a fall prevention device according to Claim 1, characterized by the above.

3. The learned model includes: A first learned model that is trained to detect from the input sound data the start operation of using the lanyard, where the hook is removed from the hook holding member and then hooked onto the mounting equipment, and the end operation of using the lanyard, where the hook is removed from the mounting equipment and then held by the hook holding member, and output the detection result; A second learned model that is trained to output whether the usage status of the fall prevention device is in use or not in use from the output of the first learned model. The monitoring system for a fall prevention device according to Claim 1 or 2, characterized by the above.

4. The learned model includes: A first trained model that is trained to detect the first to fourth sounds from the input sound data and output the detection results, and a second trained model that is trained to output whether the usage status of the fall prevention device is in use or not based on the output of the first trained model and the change over time of the output. The monitoring system for a fall prevention device according to claim 2, characterized in that.

5. The fall prevention device includes two of the lanyards, and a pair of left and right hook holding members are provided on the fall prevention device body corresponding to the lanyards. The trained model is trained to identify the left and right of the hook and the hook holding member. The monitoring system for a fall prevention device according to claim 1 or 2, characterized in that.

6. The output unit outputs a predetermined warning when the state in which the output of the trained model is not in use continues for a predetermined time. The monitoring system for a fall prevention device according to claim 1 or 2, characterized in that.

7. It is provided with position detection means for detecting the current position of the user. The output unit outputs a predetermined warning when the current position of the user is within a preset warning target area and the output of the trained model is not in use. The monitoring system for a fall prevention device according to claim 1 or 2, characterized in that.

8. It is provided with moving speed detection means for detecting the moving speed of the user. The output unit outputs a predetermined warning when the moving speed of the user is equal to or lower than a predetermined threshold value and the output of the trained model is not in use. The monitoring system for a fall prevention device according to claim 1 or 2, characterized in that.

Images