Installation support device and installation support system

Abstract

This technology provides an improvement in the efficiency of installation work for remote devices used for remote monitoring or inspection of elevators. [Solution] The data acquisition unit 151 acquires input information including a nameplate image taken of the nameplate of the control panel 210 which contains information on the elevator manufacturer and model, and an image of the inside of the control panel 210 before the control panel is connected. The inference unit 152 inputs the input X11 acquired by the data acquisition unit 151 into the trained model 753 to obtain an output X13 which includes information on the connection of signal lines corresponding to the manufacturer and model.

Description

Technical Field

[0001] The present disclosure relates to an installation work support device and an installation work support system for assisting the installation work of a remote device for remotely monitoring or inspecting an elevator.

Background Art

[0002] There are various manufacturers of elevators, and even for the same manufacturer, there are various models. In recent years, there has been an increasing need for a remote device that can be connected to elevators of various manufacturers and models to remotely monitor or inspect the elevators.

[0003] When installing such a remote device on-site, it is necessary to perform installation work corresponding to elevators of each manufacturer and each model with different specifications. For example, in order for the remote device to acquire various signals for monitoring and inspection from the elevator control panel, operations such as branching signal lines from devices such as the control board installed inside the control panel and taking them into the remote device are required.

[0004] For example, as a system for assisting in the arrangement work of elevator equipment, there is a support system disclosed in International Publication No. 2024 / 009389 (Patent Document 1). This support system uses a set of a photographed image and the target arrangement of equipment that was input by an operator in the past for the image as learning data to learn the target arrangement in the image photographed by the photographing device, and guides the optimal target arrangement of the equipment from the image photographed by the photographing device.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, the aforementioned support system guides the placement of elevator equipment based on specific manufacturers and models, but it does not address the connection work (wiring, connection work) of signal lines in remote devices that can connect to various manufacturers and models.

[0007] In the installation of remote devices, the procedure for connecting signal lines varies depending on the manufacturer and model. The destinations of each signal line are also different, and some wiring methods involve, for example, cutting the original signal line and connecting a new branched signal line to it. As a result, even when referring to the manual, the installation process was time-consuming.

[0008] In particular, when elevators had special specifications, the layout of equipment within the control panel also changed, making it impossible to address all aspects using standard manuals. Because of these circumstances, the installation of remote control devices depended heavily on the skills of the workers, resulting in significant variations in work efficiency and time among them.

[0009] This disclosure has been made to solve the above-mentioned problems, and its purpose is to provide an installation support device and an installation support system that can improve the efficiency of installation work for remote devices used for remote monitoring or remote inspection of elevators. [Means for solving the problem]

[0010] The installation support device relating to this disclosure is a device that assists in the installation of a remote device for remote monitoring or remote inspection of an elevator. The installation work includes connecting signal lines that connect equipment installed inside the control panel that controls the elevator to the remote device. The installation support device includes a data acquisition unit and an inference unit. The data acquisition unit acquires input information including a first image of a label on the control panel that contains information on the elevator manufacturer and model, and a second image of the inside of the control panel before the connection work. The inference unit inputs the input information acquired by the data acquisition unit into artificial intelligence to obtain output information that includes information on the connection of signal lines corresponding to the manufacturer and model.

[0011] The installation support system disclosed herein comprises an installation support device and a terminal device. The terminal device is capable of communicating with the installation support device and includes a camera and a display unit. The data acquisition unit acquires a first image and a second image captured by the camera as input information. The display unit displays the output information obtained by the inference unit. [Effects of the Invention]

[0012] According to this disclosure, the efficiency of installation work for remote devices for remote monitoring or remote inspection of elevators can be improved. [Brief explanation of the drawing]

[0013] [Figure 1] This figure shows an example of the overall configuration of a work support system, an elevator system, and a remote inspection system. [Figure 2] This figure shows an example of the hardware configuration of an elevator system. [Figure 3] This is a diagram illustrating the signals used in a remote inspection system. [Figure 4] This is a diagram showing an example of the inside of a control panel. [Figure 5] This is a table showing information about each signal. [Figure 6] This is a table showing the information for each nameplate. [Figure 7] This is a diagram for explaining a display example on a terminal. [Figure 8] This is a diagram for explaining a display example on a terminal. [Figure 9] This is a flowchart of the processing executed by the work support system. [Figure 10] This is a configuration diagram of the learning device. [Figure 11] This is a flowchart showing the learning process of the learning device. [Figure 12] This is a configuration diagram of the work support device. [Figure 13] This is a flowchart showing the inference procedure by the work support device. [Figure 14] This is a configuration diagram of the work support device and the external server according to the modification example. [Figure 15] This is a flowchart showing the inference procedure by the work support device and the external server according to the modification example. [Figure 16] This is a configuration diagram of the work support device and the external server according to the modification example. [Figure 17] This is a flowchart showing the inference procedure by the work support device and the external server according to the modification example.

Embodiments for Carrying Out the Invention

[0014] Hereinafter, embodiments will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated.

[0015] FIG. 1 is a diagram showing an example of the overall configuration of an installation work support system (also referred to as "work support system") 1, an elevator system 200, and a remote inspection system 3. The work support system 1 includes an installation work support device (also referred to as "work support device") 100, a terminal device (also referred to as "terminal")​​​The remote inspection system 3 comprises a remote inspection device 500, a management server 300, and a terminal 400. The remote inspection device 500 is an example of a remote device for remote monitoring or remote inspection of an elevator. The work support device 100 is a device that assists in the installation of the remote inspection device 500. Further details will be described later.

[0017] If a building has elevators, the building owner is required to enter into a maintenance contract with an elevator maintenance company. The maintenance company's personnel (workers) perform maintenance inspections and periodic checks of the elevators based on the maintenance contract. The building owner can also include remote inspection or remote monitoring as an option when entering into the maintenance contract.

[0018] Remote monitoring refers to the maintenance company's monitoring center (information center) or similar facility constantly monitoring for any abnormalities or malfunctions in the elevator using communication lines, etc. Remote inspection, in addition to remote monitoring, refers to the maintenance company's monitoring center or similar facility inspecting, using communication lines, etc., the elevator's operating status and the functioning status of each piece of equipment, focusing on areas necessary for normal elevator operation, to determine whether they are functioning correctly.

[0019] During remote inspections, various inspection items are checked, including the elevator car's starting state, acceleration state, constant speed state, deceleration state, landing state, state of destination floor buttons inside the car, door opening / closing state, landing button state, door switch state, presence or absence of electromagnetic brake abnormalities, car travel distance, travel time or number of starts, and number of door openings / closings.

[0020] In this embodiment, the elevator system 200 and the remote inspection device 500 are installed within building 2. The remote inspection device 500 is connected to the elevator system 200 and performs remote inspection or remote monitoring of the elevator. The remote inspection device 500 may also be configured as a device that performs only remote monitoring. The remote inspection device 500 is configured to include, for example, a PLC (Programmable Logic Controller).

[0021] The management server 300 is installed, for example, in the maintenance company's information center (monitoring center). The terminal 400 may be installed in the maintenance company's information center or in any other location. The terminal 400 and the remote inspection device 500 can connect to the management server 300 via a communication line.

[0022] The management server 300 manages various data such as customer information for each building that has concluded an elevator maintenance contract, building information, information on the elevators installed in the building, and remote inspection results. The management server 300 is a device that manages the remote inspection device 500, and transmits commands to the remote inspection device 500 to perform remote inspections, and also acquires the inspection results of remote inspections performed by the remote inspection device 500.

[0023] Terminal 400 is, for example, a PC (Personal Computer), a smartphone, or a tablet. Terminal 400 can obtain elevator monitoring information from the management server 300 and perform remote monitoring, and can also have a maintenance worker perform a remote inspection on the remote inspection device 500 via the management server 300. In addition, terminal 400 can display the inspection results of the remote inspection performed by the remote inspection device 500.

[0024] The elevator system 200 comprises a control panel 210 for controlling the elevator and an elevator equipment group 220. The elevator equipment group 220 consists of various equipment including the elevator and elevator landing equipment. The control panel 210 controls the various equipment of the elevator equipment group 220.

[0025] The control panel 210 inputs and outputs signals to and from the elevator equipment group 220 via multiple signal lines. The signals transmitted and received between the elevator equipment group 220 and the control panel 210 include those transmitted and received by parallel transmission (parallel communication) and those transmitted and received by serial transmission (serial communication). The former (parallel transmission) are signals obtained directly from, for example, various switches or sensors of the elevator equipment group 220. The latter (serial transmission) are signals transmitted and received by the control panel 210 via serial communication between a control board provided in a device installed on the elevator landing side or car side.

[0026] In this embodiment, a portion of the signal lines connecting the control panel 210 and the elevator equipment group 220 that transmit and receive signals via parallel transmission is branched and connected to terminals on the remote inspection device 500. This allows a portion of the signals transmitted and received between the control panel 210 and the elevator equipment group 220 via parallel transmission to be input and output on the remote inspection device 500 side. The remote inspection device 500 is configured to be connectable to the elevator system 200 regardless of the manufacturer of the elevator system 200.

[0027] As described above, the work support system 1 comprises a work support device 100, a terminal 110, and a learning device 700. The work support device 100 can communicate with the learning device 700 and the terminal 110 via the Internet. Furthermore, the work support device 100 is configured to communicate with an external server 800 via the Internet.

[0028] The work support device 100 outputs information to assist in the installation of the remote inspection device 500, and maintenance personnel (workers) can check this information on the terminal 110. This information is generated by the artificial intelligence (trained model) of the learning device 700 or the external server 800. Details of the operation of the learning device 700 and the external server 800 will be described later.

[0029] The work support device 100 comprises a processor 101, memory 102, and a communication interface (IF) 103. The processor 101 is, for example, a CPU (Central Processing Unit). The memory 102 may be configured to include, for example, ROM (Read Only Memory), RAM (Random Access Memory), and a storage unit. The storage unit is a non-volatile storage device. The storage unit may be, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive). These are connected to each other via a bus so that they can communicate with one another.

[0030] The CPU loads programs stored in ROM or memory into RAM and executes them. RAM is the working area where the CPU executes programs, and it temporarily stores programs and data used to execute them. The work support device 100 is configured to communicate with the learning device 700, terminal 110, or external server 800 via the communication IF 103.

[0031] The learning device 700, like the work support device 100, includes a processor 701, memory 702, and communication IF 703. The terminal 110, like the work support device 100, includes a processor 111, memory 112, and communication IF 113.

[0032] Terminal 110 is, for example, a portable device such as a tablet or smartphone, and further comprises a display unit 121, an input unit 122, a speaker 123, a microphone 124, and a camera 125. The display unit 121 displays various information. The display unit 121 is, for example, a liquid crystal display or a touch panel. The input unit 122 receives input from the user. The input unit 122 is, for example, a touch panel, but may also be a keyboard, mouse, etc.

[0033] The worker (maintenance worker) performing the installation of the remote inspection device 500 carries a terminal 110. The terminal 110 has dedicated application software (hereinafter referred to as "dedicated software") pre-installed to support the installation of the remote inspection device 500. The worker starts the dedicated software and uses the camera 125 of the terminal 110 to take pictures of the nameplate attached to the control panel 210 and the inside of the control panel 210. When the dedicated software transmits the captured information from the terminal 110 to the work support device 100, the work support device 100 transmits various display information to the terminal 110 to support the installation of the remote inspection device 500. The worker can perform the installation of the remote inspection device 500 while viewing the various display information displayed on the display screen of the dedicated software (display unit 121 of the terminal 110).

[0034] Furthermore, the worker uses terminal 110 to record video of the installation process. Terminal 110 is equipped with a speaker 123 and a microphone 124, and audio is input from the microphone 124 when recording video. In addition, during the installation of the remote inspection device 500, the worker can also listen to work information via audio from speaker 123.

[0035] Figure 2 shows an example of the hardware configuration of the elevator system 200. In this embodiment, building 2, where the elevator system 200 is installed, is a five-story building. In addition, there is one elevator installed in building 2 (this elevator is referred to as "Unit 1").

[0036] The control panel 210 includes a module group M (Figure 3) which contains a control board for controlling the elevator equipment group 220. The elevator equipment group 220 includes landing devices 230 installed at the landings of each floor from the 1st floor (1F) to the 5th floor (5F), various sensors and switches used in the elevator system 200, and the hoisting machine 250 and car device 240 of the first elevator.

[0037] The hoisting machine 250 is a motor that drives the elevator car up and down. The car device 240 consists of various devices installed in the car, including a destination floor button for registering the destination floor. The landing device 230 consists of various devices installed at the landings on each floor, including a landing button for registering a landing call. Module group M is connected to the landing devices 230 on each floor, various sensors, and various switches, etc., via a control cable 21 which bundles multiple signal lines. Module group M is also connected to the hoisting machine 250 and car device 240 of Unit 1, etc., via a control cable 22 which bundles multiple signal lines. Like the work support device 100, module group M also includes a processor, memory, and a communication interface. Module group M is configured to communicate with the elevator equipment group 220, including the landing devices 230, hoisting machine 250, and car device 240, via serial or parallel communication through the communication interface.

[0038] Figure 3 is a diagram illustrating the signals used in the remote inspection system 3. The installation work for the remote inspection device 500 includes connecting signal lines (wiring and connection work) that connect the equipment (module group M including modules M1 to M5, etc.) installed inside the control panel 210 to the remote inspection device 500.

[0039] As described above, the elevator system 200 comprises a control panel 210 and an elevator equipment group 220. The control panel 210 and the elevator equipment group 220 are connected by multiple signal lines, thereby enabling the transmission and reception of multiple signals. These multiple signals include signals A1 to A4, etc. All of the signals exemplified here are transmitted and received by parallel transmission.

[0040] For example, signal A1 is the DZ signal. The DZ signal is detected by the door zone detection device. The DZ signal is ON when the elevator car is located within the door zone (within the range of positions where the doors can be opened) on each floor, and OFF when it is located outside the door zone. For example, signal A2 is the LB signal. The LB signal is ON when the brake is released due to power being supplied to the brake coil of the hoisting machine 250. The LB signal is OFF when the brake is activated due to the cessation of power supply to the brake coil of the hoisting machine 250.

[0041] The remote inspection device 500 includes a control device, an input interface (IF), a communication interface, etc. (none of which are shown in the diagram). The input interface is a circuit board for inputting a portion of the signals that are input and output between the control panel 210 and the elevator equipment group 220 via parallel transmission. Each signal line of the signal input to the control panel 210 is branched, and each branched signal line is connected to a terminal on the input interface. Each signal input to the input interface is further transmitted to the control device. These signals are transmitted to the management server 300 via the communication interface. The status of these signals can be monitored on the terminal 400.

[0042] Figure 4 shows an example of the interior of the control panel 210. When the remote inspection device 500 is installed, the remote inspection device 500 and the control panel 210 are connected by a signal line (also referred to as "wiring") L0. This signal line L0 is a bundle of signal lines (also referred to as "wiring") L1 to L5, etc. Inside the control panel 210, the signal lines L1 to L5 etc. are separated and connected to the module group M.

[0043] Module group M includes modules M1, M2, etc., as shown in Figure 4. Module M1 is a control board that controls the elevator equipment group 220. The other modules are other control boards that control the elevator equipment group 220, other electrical equipment, terminal blocks, etc. When installing the remote inspection device 500, the worker connects wiring L1 to L5, etc., to modules M1, M2, etc.

[0044] Signal line (wiring) L1 is a signal line for receiving signal A1 into the remote inspection device 500. For example, signal line LX1, which receives signal A1 input from the elevator equipment group 220 to module M1, is cut, and a connector C1 is attached to connect both ends of the cut signal line LX1 and the end of signal line L1. By making these connections, signal A1 received by module M1 via signal line LX1 is also received by signal line L1 and transmitted to the remote inspection device 500.

[0045] Similarly, wiring (signal line) L2 is connected to module M1 via connector C2. Wiring (signal line) L3 is connected to module M2 via connector C3. Although not shown in the diagram, wirings L4, L5, etc. are also connected to one of the modules in module group M. If signal lines for receiving signals from elevator equipment group 220 to control panel 210 are connected to the terminal block, the signal lines from the remote inspection device 500 should be connected to the same terminal.

[0046] Figure 5 shows Table T1, which displays information for each signal. Table T1 summarizes information such as the wiring, wiring color, wiring length, destination module, connector, and connection information corresponding to each signal.

[0047] For example, signal A1 is received by the remote inspection device 500 via wiring L1, the color of wiring L1 is B1 (for example, yellow), the length of wiring L1 is D1, and wiring L1 is connected to module M1 via connector C1, which is color B6.

[0048] Wiring information is information that indicates how to connect signal lines. For example, in the above example, the wiring information would describe how to identify signal line LX1 that receives signal A1 input to module M1, how to cut signal line LX1, and how to connect both ends of signal line LX1 and the end of signal line L1 by attaching connector C1. The color information of signal line LX1 may also be included in the wiring information. Color coding makes wiring and connection work easier for the worker. Wiring information may also include information such as pin numbers and terminal numbers.

[0049] Signal A2 is received by the remote inspection device 500 via wiring L2, the color of wiring L2 is B2, the length of wiring L2 is D2, and wiring L2 is connected to module M1 via connector C2, which is color B7.

[0050] Figure 6 shows Table T2, which displays information for each nameplate. The control panel 210 is fitted with labels (nameplates) that contain information such as the elevator manufacturer and model. Table T2 shows the manufacturer, model, catalog information, etc., corresponding to each nameplate.

[0051] For example, the nameplate P1 attached to the control panel 210 indicates that the elevator manufacturer is manufacturer X1 and the elevator model is model Y1. It also indicates that the catalog for manufacturer X1's model Y1 is catalog Z1. Catalog Z1 contains information about the specifications of manufacturer X1's model Y1, including information such as the production period of manufacturer X1's model Y1.

[0052] The nameplate P2 attached to the control panel 210 indicates that the elevator manufacturer is manufacturer X1 and the elevator model is model Y2. It also indicates that the catalog for model Y2 from manufacturer X1 is catalog Z2.

[0053] Figures 7 and 8 illustrate an example of display on terminal 110. The worker launches dedicated software and uses the camera 125 of terminal 110 to photograph the nameplate attached to the control panel 210 and the inside of the control panel 210. The image of the nameplate is referred to as the "nameplate image," and the image of the inside of the control panel 210 is referred to as the "image before control panel connection." At that time, the dedicated software provides instructions on the appropriate shooting position and angle for photographing the inside of the control panel 210. In addition, multiple shooting positions and angles may be used to easily show wiring routes, connection locations, etc.

[0054] When terminal 110 transmits captured images, etc. (referred to as "input X11") as input information to the work support device 100, the work support device 100 generates display images, etc. (referred to as "output X13") as output information based on the captured images. Terminal 110 acquires the output information from the work support device 100 and displays it on the display unit 121.

[0055] As will be explained in more detail using Figure 12, the work support device 100 inputs the input information, input X11, into the trained model 753 and obtains the output information, output X13. Also, as will be explained in more detail using Figure 10, the trained model 753 is a predictive model (artificial intelligence) that has been trained by the learning device 700.

[0056] As shown in Figure 7, the display unit 121 of the terminal 110 shows that the manufacturer of the installed elevator is manufacturer X1 and the model is model Y1. Furthermore, the display unit 121 also displays the image after connection to the control panel, connection information, and other information.

[0057] The image after control panel connection is an image of the inside of control panel 210 (image before control panel connection) taken by a worker, with images showing the wiring added. The image after control panel connection shows images of the wires to be connected, L1 to L5, etc., along with information to identify the wires, such as "L1" to "L5", along with arrows. "L1" indicates wire L1. The color of the image showing wire L1 is displayed as B1, which is the wire color.

[0058] Furthermore, information identifying the modules and connectors to which wiring L1~L5 etc. are connected is provided along with arrows. For example, "M1" to identify module M1, "C1" to identify connector C1, etc. are displayed along with arrows.

[0059] The connection information is the same as the information shown in Figure 5. For example, information such as that signal A1 is acquired by wiring L1, and that wiring L1 is a signal line with wiring length D1 and wiring color B1 is shown along with the wiring information.

[0060] Additionally, following the "Connection Image" link from this screen will display an enlarged image of the connection point when connecting each wire (for example, an enlarged image of the connection point between wire L1 and module M1 (around connector C1)).

[0061] Furthermore, following the "Installation and Construction Images" link from this display screen will show the installation and construction images. The installation and construction images are general images of the control panel 210 and the remote inspection device 500, and include, for example, images showing the installation status of the remote inspection device 500.

[0062] In addition, other information such as installation conditions, installation precautions, necessary tools, and other details are provided. These will be explained later using the diagrams from Figure 10 onwards.

[0063] Workers can efficiently perform wiring and connection work by referring to this information. Furthermore, this information can be output as audio from speaker 123. For example, during the wiring and connection work for wiring L1, information regarding wiring L1 can be output as audio.

[0064] Figure 8 shows an example of the display at terminal 110 when the position of the connection point differs significantly from the usual position. Module M1 is typically located in the upper left position of the control panel 210, as shown in Figure 7. On the other hand, according to an actual internal image of the control panel 210 taken by an operator, it is located in the upper right position of the control panel 210, as shown in Figure 8.

[0065] In this case, a warning message will be displayed stating, "Please check the wiring positions C1 and C2 of module M1." Additionally, module M1 and connectors C1 and C2 are highlighted in a more conspicuous color to emphasize their importance.

[0066] This warning message was displayed to draw attention because the standard position of module M1 during learning by the learning device 700 (described later) differs significantly from the position of module M1 in the image taken by the operator before connecting to the control panel. In this case, the connection position of wiring L1 and L2 during learning (near connectors C1 and C2) differs significantly from the actual connection position of wiring L1 and L2 (near connectors C1 and C2), prompting the operator to pay attention. Upon seeing the warning message, the operator should check the manual or other documentation to reconfirm the connection position, or perform the work carefully and cautiously to avoid any connection errors.

[0067] Furthermore, the installation process is recorded as a video of the work in progress. The video of the work in progress may be transmitted in real time to the work support device 151, which may be configured to analyze the work in real time. For example, if there is a wiring error in the video of the work in progress, a warning message to that effect may be displayed in real time. Alternatively, the system may be configured to take a picture of the inside of the control panel 210 after the work is completed and compare that image with the image taken after the control panel is connected to check for any wiring errors.

[0068] Figure 9 is a flowchart of the processes performed by the work support system 1. This flowchart shows the processing content of the dedicated software when the dedicated software described above is launched on terminal 110.

[0069] The worker launches dedicated software and uses terminal 110 to capture images and videos of the nameplate, control panel 210, etc. In S1, terminal 110 acquires the captured images and videos of the nameplate, control panel 210, etc. In S2, terminal 110 transmits the acquired information (input X11) to the work support device 100. Video recording and acquisition continue from S2 onward until the installation work is completed.

[0070] The work support device 100 acquires input X11, inputs the acquired input X11 to the trained model 753, and obtains output X13. Details will be described later using Figure 12. The work support device 100 transmits output X13 to the terminal 110.

[0071] In S3, terminal 110 acquires the information (output 13) transmitted by the work support device 100. Output 13 includes text information such as "connection information" and "other information" shown in Figures 7 and 8, image information such as "image after control panel connection", and warning messages shown in Figure 8. Terminal 110 displays the received text information (connection information, etc.), image information (image after control panel connection, etc.), and warning messages, respectively (see S4-S6, Figures 7 and 8).

[0072] Next, the learning process performed by the learning device 700 and the inference process performed by the work support device 100 will be described. Figure 10 is a diagram of the configuration of the learning device 700. The learning device 700 includes a data acquisition unit 751, a model generation unit 752, and a learned model storage unit 753.

[0073] Here, input X11 includes X11A and X11B. X11A includes nameplate image and image before control panel connection. X11B includes installation / construction area image and work status video. Input X12 includes X12A and X12B. X12A includes model information, image after control panel connection, and connection information. X12B includes installation conditions, installation precautions, work precautions, etc.

[0074] The nameplate image is a photograph of the nameplate (label) attached to the control panel 210. The nameplate contains information that identifies the manufacturer and model of the elevator connected to the remote inspection device 500 being installed. The worker is photographing the nameplate to obtain information that identifies the manufacturer and model.

[0075] The pre-control panel connection images are images taken of the inside of the control panel 210 before the installation of the remote inspection device 500, that is, before the signal line connection work (wiring work, connection work). The pre-control panel connection images consist of multiple images, including the entire interior of the control panel and the main parts.

[0076] Model information refers to the manufacturer and model of the elevator. This model information may also include catalog data (such as generation information) for the relevant model.

[0077] The images after the control panel connection show the inside of the control panel 210 after the installation of the remote inspection device 500, that is, after the signal line connection work (wiring work, connection work). The images after the control panel connection consist of multiple images, including the entire interior of the control panel and the main parts.

[0078] The images after the control panel connection (see Figures 7 and 8) include wiring images taken to visually confirm the wiring status (wiring path) of the signal lines inside the control panel connecting the remote inspection device 500 and the module group M, and connection images taken to visually confirm the connection status (connection parts such as around connectors) between the signal lines and the module group M. The images after the control panel connection clearly show the connection location, connection status, connector connection status, wiring path, and wiring length. After the connection work (wiring work, connection work), the worker takes pictures of the inside of the control panel 210 so that the wiring status and connection status inside the control panel 210 can be identified.

[0079] The connection information includes the signal, wiring information (signal lines), wiring color, wiring length, destination information (module), connector information, and connection information, as shown in Table T1 of Figure 5.

[0080] The installation and construction images are general images related to the installation work of the control panel 210 and the remote inspection device 500. The installation and construction images are not limited to internal images of the control panel 210, but also include, for example, images showing the installation status of the remote inspection device 500, images showing the installation status of the control panel 210, etc.

[0081] The work progress video is a continuous recording of the installation process of the Remote Inspection Device 500. The workers film the process so that they can visually confirm the wiring, connection, and other related tasks.

[0082] The installation conditions and precautions section provides general information relating to the installation of the control panel 210 and the remote inspection device 500. These conditions may include installation conditions and precautions related to the wiring and connections within the control panel 210, as well as, for example, where the remote inspection device 500 will be installed (e.g., next to the control panel 210, on the wall of the elevator shaft), and any installation precautions in that case. The installation conditions and precautions may be extracted and compiled from a pre-prepared installation manual, or they may be extracted (readable) from installation and construction images.

[0083] Work precautions and other points are extracted from work status videos. For example, in connector wiring work, connectors that are frequently connected incorrectly are learned from work status videos. For example, in the situation shown in Figure 7, if an error is detected from multiple work status videos where connector C1 of wiring L1 is mistakenly connected to the location of connector C2 (an image of connector C1 of wiring L1 being connected to the location of connector C2 is detected), then connector C1 is learned to be a connector that is frequently connected incorrectly. Alternatively, wiring work that takes a long time can be extracted from multiple images in a time series contained in the video, and this can be learned to be a work with a high difficulty level. In addition, work points that are prone to becoming unsafe during work can be learned from work status videos, and necessary tools can also be learned from work status videos. Necessary tools can be extracted from images of tools that are filmed in the work status video.

[0084] The data acquisition unit 751 acquires the datasets of input X11 and input X12 as training data. Input X12 is the correct answer data for input X11.

[0085] The model generation unit 752 learns input X12 based on training data created from combinations of ground truth data for input X11 and input X12. In other words, the model generation unit 752 generates a trained model that estimates the optimal input X12 based on the ground truth data for input X11 and input X12. The training data is data obtained by relating the ground truth data for input X11 and input X12 to each other.

[0086] The learning device 700 is used to learn input X12, and may, for example, be connected to the work support device 100 via a network, and may be a separate device from the work support device 100, or it may be the same device as the work support device 100. The learning device 700 may also reside on a cloud server.

[0087] The model generation unit 752 can use any known learning algorithm, such as supervised learning, unsupervised learning, or reinforcement learning. As an example, the case where a neural network is applied will be described.

[0088] The model generation unit 752 learns the input X12 by supervised learning, for example, according to a neural network model. Supervised learning is a method in which pairs of input and result data are provided to the learning device 700, which learns the features in the training data and estimates the result from the input.

[0089] A neural network consists of an input layer with multiple neurons, an intermediate layer (e.g., a hidden layer) with multiple neurons, and an output layer with multiple neurons. The intermediate layer can be one or more layers.

[0090] For example, in a three-layer neural network, when multiple inputs are input to the input layer, the values ​​are multiplied by weight W1 and input to the hidden layer, and the result is further multiplied by weight W2 and output from the output layer. This output result changes depending on the values ​​of weights W1 and W2. The neural network learns an estimate of input X12 through supervised learning, according to training data created based on combinations of the correct data for input X11 and input X12. In other words, the neural network learns by adjusting weights W1 and W2 so that the result output from the output layer when input X11 is input to the input layer approaches the correct data for input X12.

[0091] The model generation unit 752 generates and outputs a trained model by performing the training described above. The trained model storage unit 753 stores the trained model (referred to as "trained model 753") output from the model generation unit 752.

[0092] In the above training, a trained model was generated using the datasets of input X11 (input 11A, input 11B) and input X12 (input 12A, input 12B) as training data. In this case, if nameplate images, images before control panel connection, installation / construction section images, and work status videos are input to the trained model, the output information will include model information, images after control panel connection, connection information, installation conditions / precautions, and work precautions. Alternatively, nameplate images and images before control panel connection may be input to the trained model, and model information, images before control panel connection, connection information, installation conditions / precautions, and work precautions may be output.

[0093] However, this is not the only option; for example, a trained model may be generated using a dataset of nameplate images and model information (ground truth data) as training data. In this case, when a nameplate image is input to the trained model, model information will be obtained as output information.

[0094] Model information (ground truth data) can be recorded by an operator visually inspecting the nameplate image during training (by reading and recording the manufacturer name and model name written on the nameplate image), or by using the Generative Artificial Intelligence (AI) described later. In this case, the nameplate image is input into the Generative AI to output the model information.

[0095] Alternatively, a trained model may be generated using the datasets of input 11A (nameplate image and image before control panel connection) and input 12A (model information, image after control panel connection, connection information: ground truth data) as training data. In this case, if the nameplate image and image before control panel connection are input to the trained model, the model information, image after control panel connection, and connection information will be obtained as output information. Alternatively, the model information may be input to the trained model, and the image before control panel connection and connection information may be output.

[0096] For elevators of the same model from the same manufacturer and with standard specifications, the images and connection information after connecting to the control panel will be almost identical. For example, in the example in Figure 7, the module group M is arranged in almost the same layout within the same size control panel 210. Therefore, the wiring routes and wiring lengths will be almost the same, and other information will be the same. However, if, for example, the position of module M2 is slightly shifted, the positions of connector C3 and wiring L3 in the image after connecting to the control panel will also shift accordingly, and the wiring length will also be slightly changed.

[0097] Even with the same model from the same manufacturer, if the elevator has special specifications, even if the connected modules, connectors, and wiring methods are the same, the differences in wiring routes, wiring lengths, and connection locations will be greater compared to standard elevators. For example, if the specifications are special, such as having a large number of floors, a large car capacity, or being a bed elevator, the size of the control panel 210 may increase or the number of module groups M to be placed inside the control panel 210 may differ due to reasons such as a larger inverter capacity or a larger number of module groups M to be placed inside the control panel 210. As a result, the differences in wiring routes, wiring lengths, and connection locations will be greater compared to standard elevators in the image after the control panel is connected.

[0098] Therefore, if a less skilled worker performs connection work while referring to a standard manual, the work time will be longer and errors are more likely to occur. On the other hand, the work support system 1 of this embodiment correctly presents the wiring route, wiring length, connection position, etc. to the worker even if the elevator specifications are special and the positions of the module group M are different, so even less skilled workers can perform the work efficiently.

[0099] Alternatively, a trained model may be generated using a dataset of model information, installation / construction images, and installation conditions / precautions (ground truth data) as training data. In this case, if model information and installation / construction images are input to the trained model, installation conditions / precautions will be obtained as output information. Alternatively, model information may be input to the trained model, and installation conditions / precautions may be output.

[0100] The installation conditions and precautions (correct data) can be determined during training by having a worker visually inspect images of the installation / construction area and write down the installation conditions and precautions that can be gleaned from them. Alternatively, the installation conditions and precautions compiled by the worker for each manufacturer and model (installation conditions and precautions listed in the remote inspection device 500 installation manual) can be set as the installation conditions and precautions (correct data). Alternatively, model information and images of the installation / construction area can be input into the generating AI to output the installation conditions and precautions. In this case, fine tuning can be used to train the generating AI to recognize the installation conditions and precautions corresponding to the model information.

[0101] Alternatively, a trained model may be generated using a dataset of machine information, work status videos, and work precautions (ground truth data) as training data. In this case, if machine information and work status videos are input to the trained model, work precautions will be obtained as output information. Alternatively, machine information may be input to the trained model and work precautions may be output.

[0102] During training, the correct data for work precautions may be obtained by having an operator visually review a video of the work situation and write down the points to be careful about. Alternatively, the correct data for work precautions may be set as a compilation of work precautions for each manufacturer and model by the operator (such as the work precautions listed in the remote inspection device 500 installation manual). Alternatively, the model information and video of the work situation may be input into the generating AI to output work precautions. In this case, fine tuning may be used to train the generating AI to produce work precautions corresponding to the model information.

[0103] Once multiple trained models are obtained as described above, one or more of these trained models can be used to perform the inference shown in Figure 12, which will be described later.

[0104] Figure 11 is a flowchart of the learning process of the learning device 700. The steps in Figure 11 are typically performed by the processor 701. The learning device 700 acquires input X11 and input X12 (i.e., ground truth data for input X12) (S101). The learning device 700 learns an estimate of input X12 by supervised learning according to the training data created based on the combination of ground truth data for input X11 and input X12, and generates a trained model (S102). The learning device 700 stores the generated trained model 753 (step S103).

[0105] Figure 12 is a diagram of the configuration of the work support device 100. The work support device 100 includes a data acquisition unit 151 and an inference unit 152. The trained model 753, as described using Figure 10, is a predictive model trained using training data that includes input information (input X11) and output information (input X12), and functions as artificial intelligence for inferring output information from input information.

[0106] The data acquisition unit 151 acquires input information (input X11) which includes a nameplate image as a first image of the label (nameplate) of the control panel 210, and a pre-control panel connection image as a second image of the inside of the control panel 210. The nameplate image and the pre-control panel connection image are images taken by the camera 125. The input information (input X11) further includes installation images and construction images (installation and construction images), and a video of the work status captured during the connection work.

[0107] The inference unit 152 inputs the input information (input X11) acquired by the data acquisition unit 151 into the trained model 753, which is an artificial intelligence, to obtain output information (output X13) that includes information on signal line connections corresponding to the manufacturer and model.

[0108] Information regarding signal line connections includes model information, a third image (Figure 8) showing the signal lines (wiring) added to the pre-connection image (second image) of the control panel, and connection information. The connection information includes the length of the signal lines (wiring lengths D1, D2, etc.), information indicating the connection method between the signal lines and equipment (modules M1 to M5, etc.) (wiring information), the color of the signal lines (colors B1 to B5, etc.), the type of connector for connecting the signal lines and equipment (connectors C1 to C3, etc.), and the color of the connector (colors B6 to B8, etc.). Output information (output X13) further includes installation conditions, precautions, and precautions during connection work (work precautions).

[0109] The inference unit 152 transmits output information (output X13) to the terminal 110. The display unit 121 of the terminal 110 displays the output information (output X13) obtained by the inference unit 152.

[0110] In this way, the inference unit 152 outputs learned elevator model information, an image display of the wiring section, the corresponding connector information, the wiring location and connection method, the wiring length, the wiring route, the installation conditions and installation precautions for the corresponding model, and the terminal 110 displays this information.

[0111] As mentioned above, the input information may be part or all of the input X11, and the output information may be part or all of the input X13. Furthermore, the trained model 753 may be composed of one or two of the multiple trained models described above.

[0112] For example, if a nameplate image and an image before connecting the control panel are input to the trained model 753, the output information could include model information, an image after connecting the control panel, and connection information. In addition to these, installation conditions, precautions, and work precautions could also be output.

[0113] The work progress video may be recorded solely for educational purposes, or it may be possible to input a real-time video of the work progress as input information, and generate output information that overlays wiring and connection images onto the real-time video of the work progress, which is then displayed on terminal 110. In this case, if the worker makes a wiring or connection error, the system may be able to detect this and display a warning message to that effect.

[0114] Furthermore, the work support device 100 pre-stores standard images of the control panel after connection for each manufacturer and model. It then calculates the difference between the image of the control panel after connection output by the learned model 753 and the standard image of the control panel after connection, and is configured to generate a warning message if the difference in the position of modules, connectors (wiring parts), etc. is greater than a certain size (see Figure 8).

[0115] Figure 13 is a flowchart showing the inference procedure by the work support device 100. The data acquisition unit 151 acquires the input X11 (see S2) transmitted by the terminal 110 (S201). The inference unit 152 inputs the input X11 into the trained model 753 stored in the trained model storage unit 753 (S202), thereby obtaining an estimated value of the output X13 (S203). The inference unit 152 transmits the output X13 to the terminal 110 (S204). As a result, the information of the output X13 is displayed on the terminal 110 (S4-S6, see Figures 7 and 8).

[0116] In this embodiment, the case in which supervised learning is applied to the learning algorithm used by the model generation unit 752 has been described, but it is not limited to this. In addition to supervised learning, reinforcement learning, unsupervised learning, or semi-supervised learning can also be applied to the learning algorithm. Furthermore, as the learning algorithm used by the model generation unit 752, deep learning, which learns to extract the features themselves, can be used, or machine learning may be performed according to other known methods, such as genetic programming, functional logic programming, or support vector machines.

[0117] Alternatively, as shown below, estimation processing may be performed using an external server 800. Figure 14 is a configuration diagram of the work support device 100 and external server 800 according to a modified example. The work support device 100 includes a data acquisition unit 151 and a processing unit 153.

[0118] The data acquisition unit 151 acquires the input X11. This is the same process as in Figure 12. The processing unit 153 obtains the output X13 from the input X11 using the external server 800. Specifically, it inputs the input X11 to the inference unit 851, which is artificial intelligence on the external server 800, and obtains the output X13. The display unit 121 of the terminal 110 displays the output X13 transmitted by the processing unit 153.

[0119] The artificial intelligence described above may be a generative AI (Generative Artificial Intelligence) that includes a Large Language Model (LLM), or it may be a multimodal AI that handles input and output of text, images, audio, and video. Examples of generative AIs include GPT, Gemini, and Claude. Furthermore, fine-tuning as described above may be performed to improve the output accuracy of output X13 for input X11.

[0120] Figure 15 is a flowchart showing the inference procedure by the work support device 100 and external server 800 according to a modified example. The data acquisition unit 151 acquires the input X11 (see S2) transmitted by the terminal 110 (S301). The processing unit 153 transmits the input X11 to the external server 800 (S302). As a result, the artificial intelligence (inference unit 851) obtains the output X13 using the trained model 852. The processing unit 153 transmits the output X13 to the terminal 110 (S304). As a result, the information of the output X13 is displayed on the terminal 110 (S4-S6, see Figures 7 and 8).

[0121] Furthermore, for some inferences, the trained model of the learning device 700 may be used, and for some inferences, the trained model (generating AI) of the external server 800 may be used to generate the output X13. Figure 16 is a configuration diagram of the work support device 100 and the external server 800 according to a modified example.

[0122] The data acquisition unit 151 of the work support device 100 acquires the input X11. This is the same process as shown in Figures 12 and 14. Next, the work support device 100 has the external server 800 execute inferences 1 to 4 and acquires the results from the external server 800.

[0123] First, in inference 1, the work support device 100 sends the input X "nameplate image" to the external server 800, and obtains the output Y "model information," which is the result of inference by the inference unit 851 of the external server 800. For example, if the nameplate image contains the "manufacturer name" and "model name," the generating AI extracts the "manufacturer name" and "model name" as text from the image.

[0124] Next, in inference 2, the work support device 100 transmits the input X "model information, image before control panel connection" to the external server 800, and obtains the output Y "image after control panel connection, connection information," which is the result of inference by the inference unit 851 of the external server 800. For example, the generating AI generates wiring images etc. corresponding to the model information and generates the image after control panel connection by superimposing the wiring images etc. onto the image before control panel connection. Fine tuning as described above may be performed in advance.

[0125] Next, in inference 3, the work support device 100 transmits the input X "model information, installation / construction part images" to the external server 800, and obtains the output Y "installation conditions / points to note" which is the result of inference by the inference unit 851 of the external server 800. For example, the generating AI reads the model information and installation / construction part images and generates installation conditions / points to note that are appropriate for them. Fine tuning as described above may be performed in advance.

[0126] Next, in inference 4, the work support device 100 transmits the input X "machine information, work status video" to the external server 800, and obtains the output Y "work precautions, etc." which is the result of inference by the inference unit 851 of the external server 800. The generating AI reads the machine information and work status video and generates work precautions, etc. that are appropriate for them. Fine tuning as described above may be performed in advance.

[0127] Based on the inferences 1-4 performed by the external server 800, the output Z set includes "model information" and "image after control panel connection, connection information, installation conditions / precautions, work precautions, etc.". In other words, when using the external server 800, the output Z set can be obtained from input X11.

[0128] Here, input α is defined as "model information," and output β is defined as "image after control panel connection, connection information, installation conditions / points to note, work precautions, etc." The learning device 700 pre-trains the learned model in the learned model storage unit 753 so that output X13 (=output β) can be obtained by inputting input α. Specifically, in the learning shown in Figure 10, the learned model is trained using the set of correct data for input α "model information" and output β "image after control panel connection, connection information, installation conditions / points to note, work precautions, etc."

[0129] Then, when the calculation unit 153 receives input α "model information" from the external server 800, it inputs input α to the learned model to obtain output X13 (=output β) "image after control panel connection, connection information, installation conditions / precautions, work precautions, etc.". The calculation unit 153 outputs output X13 to the terminal 110, which displays it.

[0130] Figure 17 is a flowchart showing the inference procedure by the work support device 100 and external server 800 according to a modified example. The data acquisition unit 151 acquires the input X11 (see S2) transmitted by the terminal 110 (S401). The calculation unit 153 transmits the input Y to the external server 800 and obtains the output Z (execution of inference 1-4: S402). The calculation unit 153 inputs the input α to the trained model 753 to obtain the output β (output X13) (S403). The calculation unit 153 transmits the output X13 to the terminal 110 (S304). As a result, the information of the output X13 is displayed on the terminal 110 (S4-6, see Figures 7 and 8).

[0131] The above-mentioned set of input Y and output Z is merely an example, and the parts of the inference process that are executed by the generating AI and the parts that are executed by the trained model 753 generated by the learning device 700 may be changed as appropriate. In addition, "installation conditions, precautions, work precautions, etc." corresponding to "model information" may be maintained as a database. In this case, the "installation conditions, precautions, work precautions, etc." corresponding to "model information" can be obtained by searching the database.

[0132] As described above, in this embodiment, the installation support device (also referred to as the "work support device") 100 is a device that supports the installation work of a remote inspection device 500 for remote monitoring or remote inspection of an elevator. The installation work includes connecting signal lines that connect equipment (modules M1 to M5, etc.) installed inside the control panel 210 that controls the elevator to the remote inspection device 500. The work support device 100 includes a data acquisition unit 151 and an inference unit 152. The data acquisition unit 151 acquires input information including a first image (nameplate image) of the label (nameplate) of the control panel 210 which contains information on the elevator manufacturer and model, and a second image (image before control panel connection) of the inside of the control panel 210. The inference unit 152 inputs the input information (input X11) acquired by the data acquisition unit 151 into artificial intelligence (trained model 753) to obtain output information (output X13) which includes information on the connection of signal lines corresponding to the manufacturer and model.

[0133] The installation support system 1 comprises a work support device 100 and a terminal device (also referred to as "terminal") 110. The terminal 110 is capable of communicating with the work support device 100 and includes a camera 125 and a display unit 121. The data acquisition unit 151 acquires a first image (nameplate image) and a second image (image before control panel connection) captured by the camera 125 as input information. The display unit 121 displays the output information obtained by the inference unit 152.

[0134] The artificial intelligence is a pre-trained model 753 for inferring output information from input information. The pre-trained model 753 is a predictive model that has been machine-learned using training data that includes input information (input X11) and output information (input X12).

[0135] In this way, workers can take photographs of the inside of the control panel on-site to confirm information regarding signal line connections that is specific to the manufacturer and model. This allows even less skilled workers to reliably perform wiring and connection work according to the information regarding signal line connections. Furthermore, it eliminates the need to read manuals in advance. This improves the efficiency of installation work for remote devices used for remote monitoring or inspection of elevators.

[0136] The connection information includes a third image (Figure 8) which is a second image with an image showing the signal lines after the connection work has been completed, as well as information indicating the length of the signal lines (wiring lengths D1, D2, etc.) and the connection method between the signal lines and the equipment (wiring information). This clearly indicates the work location. Workers can perform the wiring and connection work of the signal lines while referring to the images, so installation work can be done reliably and quickly regardless of the worker's skill level. This allows for the provision of more consistent service to customers.

[0137] The connection information also includes the color of the signal wires, the type of connector used to connect the signal wires to the equipment, and the color of the connector. Color coding makes it easier to understand which wiring and connection work is being performed, thus reducing errors.

[0138] The input information (Input X11) further includes a video recording of the connection process (work progress video). The output information (Output X13) further includes precautions to take during the connection process (work precautions). By reviewing the precautions for the connection process while performing the work, errors can be reduced.

[0139] Furthermore, the signals input to and output from the remote inspection device 500 are not limited to signals acquired from the control panel 210. For example, signals transmitted by the remote inspection device 500 to simulate the ON state of the landing call button or car call button are also included. In this case, connection work of the signal lines between the landing side and the car side and the remote inspection device 500 is required. In this case, the inference unit 152 of the work support device 100 outputs information including the method of connection work of signal lines other than those of the control panel 210, such as landing side equipment and car side equipment.

[0140] Furthermore, although this installation support system 1 has been described as a device that supports the installation of remote inspection devices 500 compatible with multiple manufacturers and multiple models, it is not limited to this and can also function as a device that supports the installation of remote devices compatible with a single manufacturer and a single model.

[0141] [Note] The embodiments described above are specific examples of the following appendix.

[0142] (Note 1) An installation support device that assists in the installation work of a remote device for remote monitoring or remote inspection of an elevator, The installation work includes connecting signal lines that connect equipment installed inside the control panel for controlling the elevator to the remote device. A data acquisition unit that acquires input information including a first image of a label on the control panel containing information about the elevator's manufacturer and model, and a second image of the inside of the control panel before the connection work. Installation work support device comprising: an inference unit that inputs the input information acquired by the data acquisition unit into artificial intelligence to obtain output information including information regarding the connection of the signal lines, corresponding to the manufacturer and the model.

[0143] (Note 2) The installation work support device according to Appendix 1, wherein the information relating to the connection includes a third image in which an image showing the signal line after the connection work is added to the second image, the length of the signal line, and information indicating the method of connection between the signal line and the equipment.

[0144] (Note 3) The installation support device according to Appendix 1 or Appendix 2, further including the color of the signal line, the type of connector for connecting the signal line and the device, and the color of the connector, for the connection information.

[0145] (Note 4) The aforementioned input information further includes a video recording of the connection process, The output information further includes precautions to be taken during the connection work, as described in any of the appendices 1 to 3 of the installation work support device.

[0146] (Note 5) The artificial intelligence is a trained model for inferring the output information from the input information, The installation work support device according to any one of the appendices 1 to 4, wherein the trained model is a predictive model that has been machine-trained using training data including the input information and the output information.

[0147] (Note 6) An installation support device as described in one of the appendices 1 to 5, The aforementioned installation work support device is capable of communicating with the terminal device which includes a camera and a display unit, The data acquisition unit acquires the first image and the second image captured by the camera as input information. The display unit is an installation support system that displays the output information obtained by the inference unit.

[0148] The embodiments disclosed herein are illustrative and not limited to those described herein. The scope of the present invention is defined by the claims, and all modifications within the meaning and scope equivalent to the claims are intended to be included. [Explanation of Symbols]

[0149] 1 Work support system, 2 Building, 3 Remote inspection system, 21, 22 Control cable, 100 Work support device, 110 Terminal, 101, 111, 701 Processor, 102, 112, 702 Memory, 103, 113, 703 Communication IF, 121 Display unit, 122 Input unit, 123 Speaker, 124 Microphone, 125 Camera, 151 Data acquisition unit, 152 Inference unit, 153 Processing unit, 200 Elevator system, 210 Control panel, 220 Elevator equipment group, 230 Landing device, 240 Car device, 250 Hoisting machine, 300 Management server, 400 Terminal, 500 Remote inspection device, 700 Learning device, 751 Data acquisition unit, 752 Model generation unit, 753 Learned model storage unit, 800 External server, 851 Inference unit, 852 trained model memory unit, C1, C2 connectors, L0~L5 wiring, M module group, M1, M2 modules, T1, T2 tables.

Claims

1. An installation support device that assists in the installation work of a remote device for remote monitoring or remote inspection of an elevator, The installation work includes connecting signal lines that connect equipment installed inside the control panel for controlling the elevator to the remote device. A data acquisition unit that acquires input information including a first image of a label on the control panel containing information about the elevator's manufacturer and model, and a second image of the inside of the control panel before the connection work, Installation work support device comprising: an inference unit that inputs the input information acquired by the data acquisition unit into artificial intelligence to obtain output information including information regarding the connection of the signal lines, corresponding to the manufacturer and the model.

2. The installation work support device according to claim 1, wherein the information relating to the connection includes a third image in which an image showing the signal line after the connection work is added to the second image, the length of the signal line, and information indicating the method of connection between the signal line and the equipment.

3. The installation support device according to claim 2, wherein the information relating to the connection further includes the color of the signal line, the type of connector for connecting the signal line and the device, and the color of the connector.

4. The aforementioned input information further includes a video recording of the connection process, The installation work support device according to claim 1, further comprising the output information including precautions to be taken during the connection work.

5. The artificial intelligence is a trained model for inferring the output information from the input information, The installation work support device according to claim 1, wherein the trained model is a predictive model that has been machine-trained using training data including the input information and the output information.

6. An installation work support device according to any one of claims 1 to 5, The aforementioned installation work support device is capable of communicating with the terminal device which includes a camera and a display unit, The data acquisition unit acquires the first image and the second image captured by the camera as input information. The display unit is an installation support system that displays the output information obtained by the inference unit.

Images