Safety supporting device, inference device, machine learning device, safe supporting method, inference method, and, machine learning method

Abstract

To provide a safe supporting device capable of improving work efficiency of a worker while appropriately securing safety of the worker.SOLUTION: A safety supporting device 5 supports safety of a worker U who performs predetermined work onto a processing device, which performs predetermined processing by moving a movable part having a predetermined movable range, while putting at least a part of the body into a movable range. The safety supporting device 5 comprises: image data acquisition section 500 for acquiring image data 110 picked up by an image pickup unit 63 which is disposed toward the front of the worker U when the worker U wears a worker device 6; and an interlock information generation section 501 for generating interlock information indicating an interlock state of the movable part on the basis of the image data 110 acquired by the image data acquisition section 500.SELECTED DRAWING: Figure 13

Description

[Technical field]

[0001] The present invention relates to a safety support device, an inference device, a machine learning device, a safety support method, an inference method, and a machine learning method. [Background technology]

[0002] In substrate processing apparatuses that perform various processes on substrates such as semiconductor wafers, for example, an interlock function is provided for a movable part having a predetermined range of movement in order to ensure the safety of an operator. For example, Patent Document 1 discloses a substrate processing apparatus that, when a sensor detects that a cover provided on each module of the substrate processing apparatus is open, activates an interlock to stop the module (movable part). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-077053 A Summary of the Invention [Problem to be solved by the invention]

[0004] In various processing apparatuses including the substrate processing apparatus disclosed in Patent Document 1, various tasks are performed by an operator with part of the operator's body within the movable range of a movable part. For this reason, the processing apparatus requires an interlock function as described above, but for example, when performing adjustment or maintenance work, the interlock function can be temporarily disabled to improve work efficiency. However, when the interlock function is temporarily disabled, the safety of the operator is not ensured by the processing apparatus, and there is a problem in terms of operator safety.

[0005] In view of the above problems, the present invention aims to provide a safety support device, an inference device, a machine learning device, a safety support method, an inference method, and a machine learning method that enable improving the work efficiency of a worker while appropriately ensuring the safety of the worker. [Means for solving the problem]

[0006] In order to achieve the above object, a safety support device according to one aspect of the present invention comprises: A safety support device that supports the safety of a worker who performs a predetermined operation by placing at least a part of his / her body within a predetermined movable range of a processing device that moves a movable part having a predetermined movable range and performs a predetermined process, an image data acquisition unit that acquires image data captured by the image capture unit that is disposed facing a front of the worker when the worker wears the worker device having the image capture unit; and an interlock information generating unit that generates interlock information indicating an interlock state of the movable part based on the image data acquired by the image data acquiring unit. Effect of the Invention

[0007] According to the safety support device of one aspect of the present invention, interlock information indicating the interlock state of the movable part is generated based on image data captured by an image capturing unit disposed facing the front of the worker, so that the interlock of the movable part can be controlled according to the positional relationship between the worker and the movable part. Therefore, the safety of the worker can be appropriately ensured while controlling the interlock state of the movable part. This can improve the work efficiency.

[0008] Other objects, configurations and effects will become apparent from the detailed description of the invention described below. [Brief description of the drawings]

[0009] [Figure 1] 1 is an overall configuration diagram showing an example of a substrate processing system 1. FIG. [Diagram 2]FIG. 2 is a plan view showing an example of a substrate processing apparatus 2. [Diagram 3] 2 is a perspective view showing an example of first to fourth polishing sections 22A to 22D. FIG. [Figure 4] 2 is a perspective view showing an example of first and second roll sponge cleaning units 24A and 24B. FIG. [Diagram 5] 11 is a perspective view showing an example of first and second pen sponge cleaning units 24C and 24D. FIG. [Figure 6] FIG. 2 is a perspective view showing an example of first and second drying sections 24E and 24F. [Figure 7] FIG. 2 is a block diagram showing an example of a substrate processing apparatus 2. [Figure 8] FIG. 9 is a hardware configuration diagram showing an example of a computer 900. [Figure 9] FIG. 2 is a block diagram showing an example of a machine learning device 4. [Figure 10] 1 is a diagram showing an example of a learning model 10 and learning data 11. FIG. [Figure 11] 4 is a flowchart showing an example of a machine learning method performed by the machine learning device 4. [Figure 12] FIG. 2 is a block diagram showing an example of a safety support device 5. [Figure 13] 2 is a functional explanatory diagram showing an example of a safety support device 5. FIG. [Figure 14] FIG. 2 is a block diagram showing an example of a user terminal device 6. [Figure 15] 13 is a flowchart showing an example of a safety support method performed by the substrate processing apparatus 2, the safety support device 5, and the user terminal device 6. [Figure 16] 13 is a diagram showing a first example in which interlock information 111A is superimposed on a movable part in real space. FIG. [Figure 17] FIG. 13 is a diagram showing a second example in which interlock information 111B is superimposed on a movable part in real space. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings. In the following, the scope necessary for the explanation to achieve the object of the present invention will be shown in a schematic manner, and the scope necessary for the explanation of the relevant part of the present invention will be mainly explained, and the parts that are omitted from the explanation will be based on the publicly known technology.

[0011] 1 is an overall configuration diagram showing an example of a substrate processing system 1. The substrate processing system 1 according to this embodiment functions as a system for managing a series of substrate processing including a chemical mechanical polishing process (hereinafter referred to as a "polishing process") in which a substrate (hereinafter referred to as a "wafer") W such as a semiconductor wafer is pressed against a polishing pad supplied with a polishing fluid to polish the surface of the wafer W flat, a cleaning process in which a cleaning fluid is supplied to the wafer W after the polishing process and the surface of the wafer W is cleaned by bringing the wafer W into contact with a cleaning tool, and a drying process in which the surface of the wafer W after the cleaning process is dried.

[0012] The substrate processing system 1 mainly comprises a substrate processing apparatus 2, a database device 3, a machine learning device 4, a safety support device 5, and a user terminal device 6. Each of the devices 2 to 6 is, for example, configured as a general-purpose or dedicated computer (see FIG. 8 described later), and is connected to a wired or wireless network 7 so as to be able to transmit and receive various data to and from each other (FIG. 1 shows transmission and reception of some data by dashed arrows). The number of the devices 2 to 6 and the connection configuration of the network 7 are not limited to the example in FIG. 1, and may be changed as appropriate.

[0013] The substrate processing apparatus 2 is made up of a plurality of units each having a movable part with a predetermined movable range. and performs, for example, a loading process, a polishing process, a cleaning process, a drying process, an unloading process, etc., as a series of substrate processes on one or a plurality of wafers W. In this regard, the substrate processing apparatus 2 operates each unit by moving each movable part while referring to apparatus setting information 265 consisting of a plurality of apparatus parameters set for each unit, and substrate recipe information 266 that defines the operating conditions of the polishing process, the cleaning process, the drying process, etc.

[0014] The movable part is configured to be movable linearly or rotatably using, for example, electric power or fluid pressure as a driving source, and when a predetermined interlock condition is established, an interlock control for restricting the movement of the movable part is performed by the substrate processing apparatus 2. The interlock condition is treated as being established, for example, when an interlock state of the movable part is detected by various interlock sensors provided in the substrate processing apparatus 2, or when interlock information 111 indicating the interlock state of the movable part is received from the safety support device 5. Note that, as for the interlock control, the movement of the movable part may be restricted with the driving source not cut off, and the restriction on the movement may be released when the interlock condition is resolved, or the movement of the movable part may be restricted with the driving source cut off, and the restriction on the movement may be released when the interlock condition is resolved and an operation for releasing the cut-off of the driving source is performed.

[0015] The substrate processing apparatus 2 transmits various reports R to the database device 3, the user terminal device 6, etc. in response to the movement of each movable part under interlock control. The various reports R include, for example, process information identifying the target wafer W when the substrate processing is performed, apparatus status information indicating the status of each unit when each process is performed, event information detected by the substrate processing apparatus 2, operation information of the operator U on the substrate processing apparatus 2, etc.

[0016] The database device 3 is a device that manages production history information 30 relating to the history of substrate processing performed using the wafers W, processing members, and processing fluids for this production. When the substrate processing is performed by the substrate processing apparatus 2, the database device 3 receives various reports R from the substrate processing apparatus 2 as needed, and registers the reports R in the production history information 30, whereby the reports R relating to the substrate processing are accumulated in the production history information 30. In addition to the above, the database device 3 may also store apparatus setting information 255 and substrate recipe information 256, and in this case, the substrate processing apparatus 2 may refer to these pieces of information.

[0017] The machine learning device 4 is a device that operates as a main part of the learning phase of machine learning. The machine learning device 4 generates a learning model 10 used in the safety support device 5 by machine learning, for example, based on a plurality of learning data 11. The trained learning model 10 is provided to the safety support device 5 via the network 7, a recording medium, or the like.

[0018] The safety support device 5 is a device that operates as a subject of the inference phase of machine learning and supports the safety of an operator U of the substrate processing apparatus 2 when the operator U performs a predetermined operation by putting at least a part of his / her body within the movable range of a movable part of the substrate processing apparatus 2. Examples of the operation of the operator U include various operations such as replacement of consumables used in the substrate processing apparatus 2, adjustment of the apparatus setting information 265, abnormality recovery when an abnormality occurs in the substrate processing apparatus 2, and maintenance of the substrate processing apparatus 2 (including preventive maintenance).

[0019] The safety support device 5 receives image data 110 (details will be described later) at any time from a user terminal device 6 worn by a worker U performing work on the substrate processing apparatus 2, and inputs the image data 110 into a learning model 10 provided by the machine learning device 4 to generate interlock information 111 indicating the interlock state of the moving parts when the worker U performs work on the substrate processing apparatus 2, and transmits it to the substrate processing apparatus 2, the user terminal device 6, etc. at any time.

[0020] The user terminal device 6 is an operation device used when an operator U performs an operation on the substrate processing apparatus 2. The user terminal device 6 is a portable device that the worker U can wear on the head or the like, and is configured with a wearable device such as smart glasses or a see-through head mounted display. The user terminal device 6 includes an image capturing unit 63 that can capture images of the real space, and an output unit 65 that can display object information 112 superimposed on an object in the real space. Other specific configurations will be described later.

[0021] The image capturing unit 63 is composed of a camera (image sensor) such as a CMOS sensor or a CCD sensor having a predetermined resolution (number of pixels), and is arranged facing the front of the worker U when the worker U wears the user terminal device 6. When the worker U works on the substrate processing apparatus 2, the user terminal device 6 transmits image data 110 captured by the image capturing unit 63 arranged facing the front of the worker U to the safety support device 5 at any time. When the worker U wears the user terminal device 6, the output unit 65 is arranged, for example, in front of one or both eyes of the worker U. The user terminal device 6 receives interlock information 111 or object information 112 for the image data 110 from the safety support device 5 at any time, and displays the interlock information 111 or object information 112 so as to be superimposed on an object existing in the real space within the field of view of the worker U, for example.

[0022] (Substrate processing device 2) 2 is a plan view showing an example of the substrate processing apparatus 2. The substrate processing apparatus 2 is configured to include a load / unload unit 21, a polishing unit 22, a substrate transport unit 23, a finishing unit 24, and a control unit 25 inside a housing 20 that is substantially rectangular in plan view. The load / unload unit 21 is partitioned from the polishing unit 22, the substrate transport unit 23, and the finishing unit 24 by a first partition 200A, and the substrate transport unit 23 is partitioned from the finishing unit 24 by a second partition 200B. An openable or detachable cover or door (neither shown) is attached to the side wall of the housing 20 so that at least a part of the body of a worker U can enter inside during work.

[0023] (Load / Unload Unit) The load / unload unit 21 includes first to fourth front load sections 210A to 210D on which wafer cassettes (FOUPs, etc.) capable of storing a large number of wafers W in the vertical direction are placed, a transport robot 211 that can move up and down along the storage direction (vertical direction) of the wafers W stored in the wafer cassettes, and a horizontal movement mechanism section 212 that moves the transport robot 211 along the arrangement direction of the first to fourth front load sections 210A to 210D (the short side direction of the housing 20).

[0024] The transfer robot 211 is configured to be accessible to the wafer cassettes placed on each of the first to fourth front load sections 210A to 210D, the substrate transfer unit 23 (specifically, the lifter 232 described below), and the finishing unit 24 (specifically, the first and second drying sections 24E and 24F described below), and is provided with upper and lower two-stage hands (not shown) for transferring the wafer W between them. The lower hand is used when transferring the wafer W before processing, and the upper hand is used when transferring the wafer W after processing. When transferring the wafer W to the substrate transfer unit 23 or the finishing unit 24, a shutter (not shown) provided on the first partition 200A is opened and closed.

[0025] In addition, in FIG. 2, the specific configuration of the transport robot 211 and the horizontal movement mechanism 212, which function as movable parts, is omitted, but for example, they are configured by appropriately combining AC devices such as servo motors, driving force transmission mechanisms such as linear guides, ball screws, gears, belts, couplings, bearings, output devices such as fluid pressure cylinders, valves, and input devices such as linear sensors, encoder sensors, limit sensors, torque sensors, etc.

[0026] (Polishing unit) The polishing unit 22 includes first to fourth polishing parts 22A to 22D that respectively perform a polishing process (flattening) of the wafer W. The first to fourth polishing parts 22A to 22D are arranged side by side along the longitudinal direction of the housing 20.

[0027] 3 is a perspective view showing an example of the first to fourth polishing units 22A to 22D. The first to fourth polishing units 22A to 22D have the same basic configuration and functions.

[0028] Each of the first to fourth polishing units 22A to 22D includes a polishing table (processing member support unit) 220 that rotatably supports a polishing pad 2200 having a polishing surface, a top ring (substrate holding unit) 221 that rotatably holds a wafer W and polishes the wafer W while pressing it against the polishing pad 2200 on the polishing table 220, a polishing fluid supply unit 222 that supplies a polishing fluid to the polishing pad 2200, a dresser 223 that rotatably supports a dresser disk 2230 and brings the dresser disk 2230 into contact with the polishing surface of the polishing pad 2200 to dress the polishing pad 2200, and an atomizer 224 that sprays a pad cleaning fluid onto the polishing pad 2200.

[0029] The polishing table 220 is supported by a polishing table shaft 220a and includes a rotational movement mechanism 220b that rotates the polishing table 220 about its axis, and a temperature adjustment mechanism 220c that adjusts the surface temperature of the polishing pad 2200.

[0030] The top ring 221 is supported by a top ring shaft 221a that is movable in the vertical direction, and includes a rotational movement mechanism 221c that rotates the top ring 221 about its axis, a vertical movement mechanism 221d that moves the top ring 221 in the vertical direction, and a swinging movement mechanism 221e that swings (swings) the top ring 221 about the support shaft 221b. The rotational movement mechanism 221c, the vertical movement mechanism 221d, and the swinging movement mechanism 221e function as a substrate moving mechanism that moves the relative positions of the polishing pad 2200 and the polished surface of the wafer W.

[0031] The polishing fluid supply unit 222 includes a polishing fluid supply nozzle 222a that supplies a polishing fluid to the polishing surface of the polishing pad 2200, a swinging movement mechanism 222c that is supported by a support shaft 222b and moves the polishing fluid supply nozzle 222a in a swinging motion around the support shaft 222b, a flow rate adjustment unit 222d that adjusts the flow rate of the polishing fluid, and a temperature adjustment mechanism 222e that adjusts the temperature of the polishing fluid. The polishing fluid is a polishing liquid (slurry) or pure water, and may further include a chemical solution or may be a polishing liquid to which a dispersant has been added.

[0032] The dresser 223 is supported by a dresser shaft 223a that can move in the vertical direction, and is equipped with a rotational movement mechanism 223c that rotates the dresser 223 around its axis, a vertical movement mechanism 223d that moves the dresser 223 in the vertical direction, and a swinging movement mechanism 223e that moves the dresser 223 in a swinging motion around the support shaft 223b.

[0033] The atomizer 224 includes a swinging mechanism 224b that is supported by a support shaft 224a and swings the atomizer 224 around the support shaft 224a, and a flow rate regulator 224c that regulates the flow rate of the pad cleaning fluid. The pad cleaning fluid is a mixture of a liquid (e.g., pure water) and a gas (e.g., nitrogen gas) or a liquid (e.g., pure water).

[0034] The wafer W is adsorbed and held on the underside of the top ring 221 and moved to a predetermined polishing position on the polishing table 220, and then polished by being pressed by the top ring 221 against the polishing surface of the polishing pad 2200 to which polishing fluid is supplied from the polishing fluid supply nozzle 222a.

[0035] 3, the specific configurations of the rotational movement mechanism 220b, 221c, 223c, the vertical movement mechanism 221d, 223d, and the swinging movement mechanism 221e, 222c, 223e, 224b, which function as movable parts, are omitted, but for example, they are configured by appropriately combining AC devices such as servo motors, driving force transmission mechanisms such as linear guides, ball screws, gears, belts, couplings, and bearings, output devices such as fluid pressure cylinders and valves, and input devices such as linear sensors, encoder sensors, limit sensors, and torque sensors. In FIG. 3, the specific configurations of the flow rate adjustment units 222d and 224c are omitted, but for example, they are configured by appropriately combining output devices for fluid adjustment such as pumps, valves, and regulators, and input devices such as flow rate sensors, pressure sensors, liquid level sensors, temperature sensors, fluid concentration sensors, fluid physical properties sensors, and fluid particle sensors. Although FIG. 3 omits the specific configuration of the temperature adjustment mechanisms 220c and 222e, they are configured by appropriately combining, for example, AC devices such as contact or non-contact heaters and input devices such as temperature sensors and current sensors.

[0036] (Substrate transport unit) As shown in FIG. 2, the substrate transport unit 23 includes first and second linear transporters 230A, 230B that can move horizontally along the arrangement direction of the first to fourth polishing sections 22A-22D (the longitudinal direction of the housing 20), a swing transporter 231 arranged between the first and second linear transporters 230A, 230B, a lifter 232 arranged on the load / unload unit 21 side, and a temporary placement table 233 for wafer W arranged on the finishing unit 24 side.

[0037] The first linear transporter 230A is disposed adjacent to the first and second polishing units 22A and 22B, and is a mechanism for transporting the wafer W between four transfer positions (first to fourth transfer positions TP1 to TP4, in order from the load / unload unit 21 side). The second transfer position TP2 is a position where the wafer W is delivered to the first polishing unit 22A, and the third transfer position TP3 is a position where the wafer W is delivered to the second polishing unit 22B.

[0038] The second linear transporter 230B is disposed adjacent to the third and fourth polishing units 22C, 22D, and is a mechanism for transporting the wafer W between three transfer positions (fifth to seventh transfer positions TP5 to TP7, in order from the load / unload unit 21 side). The sixth transfer position TP6 is a position where the wafer W is delivered to the third polishing unit 22C, and the seventh transfer position TP7 is a position where the wafer W is delivered to the fourth polishing unit 22D.

[0039] The swing transporter 231 is disposed adjacent to the fourth and fifth transfer positions TP4, TP5, and has a hand movable between the fourth and fifth transfer positions TP4, TP5. The swing transporter 231 is a mechanism for transferring the wafer W between the first and second linear transporters 230A, 230B, and for temporarily placing the wafer W on the temporary placement table 233. The lifter 232 is disposed adjacent to the first transfer position TP1, and is a mechanism for transferring the wafer W between the lifter 232 and the transfer robot 211 of the load / unload unit 21. When transferring the wafer W, a shutter (not shown) provided on the first partition 200A is opened and closed.

[0040] In addition, in FIG. 2, the specific configuration of the first and second linear transporters 230A, 230B, the swing transporter 231, and the lifter 232, which each function as a movable part, is omitted, but for example, they are configured by appropriately combining AC devices such as servo motors, driving force transmission mechanisms such as linear guides, ball screws, gears, belts, couplings, and bearings, output devices such as fluid pressure cylinders and valves, and input devices such as linear sensors, encoder sensors, limit sensors, and torque sensors.

[0041] (Finishing unit) 2, the finishing unit 24 includes first and second roll sponge cleaning units 24A and 24B arranged in two vertical stages as a substrate cleaning device using a roll sponge 2400, first and second pen sponge cleaning units 24C and 24D arranged in two vertical stages as a substrate cleaning device using a pen sponge 2401, first and second drying units 24E and 24F arranged in two vertical stages as a substrate drying device for drying the cleaned wafer W, and first and second transport units 24G and 24H for transporting the wafer W. The number and arrangement of the roll sponge cleaning units 24A and 24B, the pen sponge cleaning units 24C and 24D, the drying units 24E and 24F, and the transport units 24G and 24H are not limited to the example in FIG. 2 and may be changed as appropriate.

[0042] The parts 24A to 24H of the finishing unit 24 are arranged along the first and second linear transporters 230A and 230B in a partitioned state, for example, in the order of the first and second roll sponge cleaning parts 24A and 24B, the first transport part 24G, the first and second pen sponge cleaning parts 24C and 24D, the second transport part 24H, and the first and second drying parts 24E and 24F (in order of furthest from the load / unload unit 21). The finishing unit 24 sequentially performs a primary cleaning process by one of the first and second roll sponge cleaning parts 24A and 24B, a secondary cleaning process by one of the first and second pen sponge cleaning parts 24C and 24D, and a drying process by one of the first and second drying parts 24E and 24F on the wafer W after the polishing process. The order of the processes performed by the parts 24A-24H of the finishing unit 24 may be changed as appropriate, or some of the processes may be omitted. For example, the cleaning process by the roll sponge cleaning parts 24A and 24B may be omitted, and the cleaning process by the pen sponge cleaning parts 24C and 24D may be started. The finishing unit 24 may also be provided with a buff cleaning part (not shown) in place of or in addition to any of the roll sponge cleaning parts 24A and 24B and the pen sponge cleaning parts 24C and 24D to perform a buff cleaning process. In addition, although the parts 24A-24H of the finishing unit 24 hold the wafer W horizontally (horizontally held) in this embodiment, they may hold the wafer W vertically or at an angle.

[0043] The roll sponge 2400 and the pen sponge 2401 are made of synthetic resin such as PVA or nylon, and have a porous structure. The roll sponge 2400 and the pen sponge 2401 function as cleaning tools for scrubbing the wafer W, and are replaceably attached to the first and second roll sponge cleaning units 24A and 24B and the first and second pen sponge cleaning units 24C and 24D, respectively.

[0044] The first transfer section 24G includes a first transfer robot 246A that is movable in the vertical direction. The first transfer robot 246A is configured to be accessible to the temporary placement table 233 of the substrate transfer unit 23, the first and second roll sponge cleaning units 24A and 24B, and the first and second pen sponge cleaning units 24C and 24D, and includes upper and lower two-stage hands for transferring the wafer W therebetween. For example, the lower hand is used when transferring the wafer W before cleaning, and the upper hand is used when transferring the wafer W after cleaning. When transferring the wafer W to the temporary placement table 233, a shutter (not shown) provided on the second partition 200B is opened and closed.

[0045] The second transfer unit 24H includes a second transfer robot 246B that is movable in the vertical direction. The second transfer robot 246B is configured to be accessible to the first and second pen sponge cleaning units 24C and 24D and the first and second drying units 24E and 24F, and includes a hand for transferring the wafer W between them.

[0046] FIG. 4 is a perspective view showing an example of the first and second roll sponge cleaning units 24A and 24B. The first and second roll sponge cleaning units 24A and 24B have the same basic configuration and function. 4, the first and second roll sponge cleaning units 24A and 24B have a pair of roll sponges 2400 arranged above and below so as to sandwich the surface to be cleaned (front and back surfaces) of the wafer W therebetween.

[0047] Each of the first and second roll sponge cleaning units 24A, 24B includes a substrate holding unit 241 that holds the wafer W, a cleaning fluid supply unit 242 that supplies cleaning fluid to the wafer W, a substrate cleaning unit (processing member support unit) 240 that rotatably supports the roll sponge 2400 and brings the roll sponge 2400 into contact with the wafer W to clean the wafer W, and a cleaning tool cleaning unit 243 that cleans (self-clean) the roll sponge 2400 with cleaning tool cleaning fluid.

[0048] The substrate holding unit 241 includes a substrate holding mechanism 241a that holds a plurality of portions of the side edge of the wafer W, and a substrate rotating mechanism 241b that rotates the wafer W around a third rotation axis perpendicular to the surface of the wafer W to be cleaned. In the example of FIG. 4, the substrate holding mechanism 241a is configured with four rollers, and at least one roller is configured to be movable so as to hold or release the side edge of the wafer W. In addition, in the example of FIG. 4, the substrate rotating mechanism 241b is configured with two drive rollers, and the drive roller also serves as the substrate holding mechanism 241a that holds the wafer W. The substrate holding unit 241 may be configured with the substrate holding mechanism 241a that is configured with a plurality of rollers, and the substrate rotating mechanism 241b that is configured with at least one drive roller. In addition, the substrate holding mechanism 241a may be configured with a chuck instead of a roller.

[0049] The cleaning fluid supply unit 242 includes a cleaning fluid supply nozzle 242a for supplying a cleaning fluid to the surface to be cleaned of the wafer W, a swinging movement mechanism 242b for turning and moving the cleaning fluid supply nozzle 242a, a flow rate adjustment unit 242c for adjusting the flow rate and pressure of the cleaning fluid, and a temperature adjustment mechanism 242d for adjusting the temperature of the cleaning fluid. The cleaning fluid may be either pure water (rinse liquid) or a chemical liquid, and the cleaning fluid supply nozzle 242a may be provided with a nozzle for pure water and a nozzle for the chemical liquid separately, as shown in Fig. 4. The cleaning fluid may be a liquid, a two-fluid mixture of liquid and gas, or a fluid containing a solid such as dry ice.

[0050] The substrate cleaning unit 240 includes a cleaning tool rotation mechanism 240a that rotates the roll sponge 2400 around a first rotation axis parallel to the surface to be cleaned of the wafer W, a vertical movement mechanism 240b that moves at least one of the pair of roll sponges 2400 in the vertical direction in order to change the height of the pair of roll sponges 2400 and the distance between them, and a linear movement mechanism 240c that moves the pair of roll sponges 2400 linearly in the horizontal direction. The vertical movement mechanism 240b and the linear movement mechanism 240c function as a cleaning tool movement mechanism that moves the relative position between the roll sponge 2400 and the surface to be cleaned of the wafer W.

[0051] The cleaning tool cleaning unit 243 includes a cleaning tool cleaning tank 243a arranged at a position not interfering with the wafer W and capable of storing and discharging a cleaning tool cleaning fluid, a cleaning tool cleaning plate 243b accommodated in the cleaning tool cleaning tank 243a and against which a roll sponge 2400 is pressed, a flow rate regulator 243c for regulating the flow rate and pressure of the cleaning tool cleaning fluid supplied to the cleaning tool cleaning tank 243a, and a flow rate regulator 243d for regulating the flow rate and pressure of the cleaning tool cleaning fluid flowing inside the roll sponge 2400 and discharged to the outside from the outer circumferential surface of the roll sponge 2400. The cleaning tool cleaning fluid may be either pure water (rinse liquid) or a chemical liquid.

[0052] In the primary cleaning process by the first and second roll sponge cleaning units 24A and 24B, the wafer W is rotated by the substrate rotation mechanism 241b while being held by the substrate holding mechanism 241a. Then, while a cleaning fluid is supplied from the cleaning fluid supply nozzle 242a to the surface of the wafer W to be cleaned, the roll sponge 2400 rotated about its axis by the cleaning tool rotation mechanism 240a comes into sliding contact with the surface of the wafer W to be cleaned, thereby cleaning the wafer W. Thereafter, the substrate cleaning The cleaning unit 240 moves the roll sponge 2400 to a cleaning tool cleaning tank 243a, and the roll sponge 2400 is cleaned, for example, by rotating the roll sponge 2400, pressing it against a cleaning tool cleaning plate 243b, or supplying a cleaning tool cleaning fluid to the roll sponge 2400 by the flow rate adjustment unit 243d.

[0053] 5 is a perspective view showing an example of the first and second pen sponge cleaning units 24C and 24D. The first and second pen sponge cleaning units 24C and 24D have the same basic configuration and functions.

[0054] Each of the first and second pen sponge cleaning units 24C, 24D includes a substrate holding unit 241 that holds the wafer W, a cleaning fluid supply unit 242 that supplies cleaning fluid to the wafer W, a substrate cleaning unit (processing member support unit) 240 that rotatably supports the pen sponge 2401 and brings the pen sponge 2401 into contact with the wafer W to clean the wafer W, and a cleaning tool cleaning unit 243 that cleans (self-clean) the pen sponge 2401 with cleaning tool cleaning fluid. The following describes the pen sponge cleaning units 24C, 24D, focusing on the differences from the roll sponge cleaning units 24A, 24B.

[0055] The substrate holding unit 241 includes a substrate holding mechanism 241c that holds a plurality of portions of the side edge of the wafer W, and a substrate rotating mechanism 241d that rotates the wafer W around a third rotation axis perpendicular to the surface of the wafer W to be cleaned. In the example of FIG. 5, the substrate holding mechanism 241c is four rollers, and at least one roller is configured to be movable so as to hold or release the side edge of the wafer W. In addition, in the example of FIG. 5, the substrate rotating mechanism 241d is configured with two drive rollers, and the drive roller that constitutes the substrate rotating mechanism 241b also serves as the substrate holding mechanism 241a that holds the wafer W. The substrate holding unit 241 may be configured with the substrate holding mechanism 241c that is configured with a plurality of rollers, and the substrate rotating mechanism 241d that is configured with at least one drive roller. In addition, the substrate holding mechanism 241c may be configured with a chuck instead of a roller.

[0056] The cleaning fluid supply unit 242 is configured similarly to that in FIG. 4, and includes a cleaning fluid supply nozzle 242a, a swinging movement mechanism 242b, a flow rate regulator 242c, and a temperature adjustment mechanism 242d.

[0057] The substrate cleaning unit 240 includes a cleaning tool rotation mechanism 240d that rotates the pen sponge 2401 around a second rotation axis perpendicular to the surface to be cleaned of the wafer W, a vertical movement mechanism 240e that moves the pen sponge 2401 in the vertical direction, and a swing movement mechanism 240f that moves the pen sponge 2401 in a horizontal direction. The vertical movement mechanism 240e and the swing movement mechanism 240f function as a cleaning tool movement mechanism that moves the relative position between the pen sponge 2401 and the surface to be cleaned of the wafer W.

[0058] The cleaning tool cleaning section 243 is arranged in a position that does not interfere with the wafer W and includes a cleaning tool cleaning tank 243e that is capable of storing and discharging cleaning tool cleaning fluid, a cleaning tool cleaning plate 243f that is contained in the cleaning tool cleaning tank 243e and against which the pen sponge 2401 is pressed, a flow rate adjustment section 243g that adjusts the flow rate and pressure of the cleaning tool cleaning fluid supplied to the cleaning tool cleaning tank 243e, and a flow rate adjustment section 243h that adjusts the flow rate and pressure of the cleaning tool cleaning fluid that flows inside the pen sponge 2401 and is discharged to the outside from the outer surface of the pen sponge 2401.

[0059] In the secondary cleaning process by the first and second pen sponge cleaning units 24C and 24D, the wafer W is rotated by the substrate rotation mechanism 241d while being held by the substrate holding mechanism 241c. Then, while a cleaning fluid is supplied from the cleaning fluid supply nozzle 242a to the surface of the wafer W to be cleaned, the pen sponge 2401 rotated about its axis by the cleaning tool rotation mechanism 240d comes into sliding contact with the surface of the wafer W to be cleaned, thereby cleaning the wafer W. Thereafter, the substrate cleaning unit 240 moves the pen sponge 2401 to the cleaning tool cleaning tank 243e, and, for example, the pen sponge 2401 is rotated by the cleaning tool rotation mechanism 240d. The pen sponge 2401 is cleaned by rotating the sponge 2401, pressing it against the cleaning tool cleaning plate 243f, or supplying the cleaning tool cleaning fluid to the pen sponge 2401 by the flow rate regulator 243h.

[0060] 6 is a perspective view showing an example of the first and second drying sections 24E and 24F. The first and second drying sections 24E and 24F have the same basic configuration and functions.

[0061] Each of the first and second drying sections 24E and 24F includes a substrate holding section 241 that holds the wafer W, and a drying fluid supply section 245 that supplies a drying fluid to the wafer W.

[0062] The substrate holding unit 241 includes a substrate holding mechanism 241e that holds a plurality of portions of the side edge of the wafer W, and a substrate rotating mechanism 241g that rotates the wafer W around a third rotation axis perpendicular to the surface to be cleaned of the wafer W. The substrate holding mechanism 241e is installed so as to rotate around a horizontal axis with respect to a vertical movement mechanism 241f that moves one end in the vertical direction, and the other end is configured as a chuck that can be brought into contact with and separated from the peripheral edge of the wafer W. The substrate holding mechanism 241e constitutes an umbrella mechanism in which a gripper moves in a direction to abut against or separate from the wafer W as the vertical movement mechanism 241f moves in the vertical direction. The substrate holding mechanism 241e may be configured as a roller instead of a chuck.

[0063] The dry fluid supply unit 245 includes a dry fluid supply nozzle 245a that supplies a dry fluid to the surface to be cleaned of the wafer W, a vertical movement mechanism 245b that moves the dry fluid supply nozzle 245a in the vertical direction, a swing movement mechanism 245c that rotates the dry fluid supply nozzle 245a in the horizontal direction, a flow rate adjustment unit 245d that adjusts the flow rate and pressure of the dry fluid, and a temperature adjustment mechanism 245e that adjusts the temperature of the dry fluid. The vertical movement mechanism 245b and the swing movement mechanism 245c function as a dry fluid supply nozzle movement mechanism that moves the relative position between the dry fluid supply nozzle 245a and the surface to be cleaned of the wafer W. The dry fluid is, for example, IPA vapor and pure water (rinsing liquid), and the dry fluid supply nozzle 245a may be provided with a nozzle for IPA vapor and a nozzle for pure water separately as shown in FIG. 6. The dry fluid may be a liquid, a two-fluid mixture of liquid and gas, or a solid such as dry ice.

[0064] In the drying process by the first and second drying parts 24E and 24F, the wafer W is rotated by the substrate rotating mechanism 241g while being held by the substrate holding mechanism 241e. Then, while a drying fluid is supplied from the drying fluid supply nozzle 245a to the surface to be cleaned of the wafer W, the drying fluid supply nozzle 245a is moved to the side edge side (radial outer side) of the wafer W. Thereafter, the wafer W is rotated at high speed by the substrate rotating mechanism 241g, whereby the wafer W is dried.

[0065] In addition, in Figures 4 to 6, the specific configurations of the substrate holding mechanism parts 241a, 241c, 241e, the substrate rotation mechanism parts 241b, 241d, 241g, the up and down movement mechanism parts 240b, 240e, 241f, 245b, the linear movement mechanism part 240c, the swing movement mechanism parts 240f, 242b, 245c, and the cleaning tool rotation mechanism parts 240a, 240d, which respectively function as movable parts, are omitted, but for example, they are configured by appropriately combining AC devices such as servo motors, driving force transmission mechanisms such as linear guides, ball screws, gears, belts, couplings, bearings, output devices such as fluid pressure cylinders and valves, and input devices such as linear sensors, encoder sensors, limit sensors, torque sensors, etc. 4 to 6, the specific configuration of the flow rate adjustment units 243c, 243d, 243g, 243h, and 245d is omitted, but for example, they are configured by appropriately combining output devices required for fluid adjustment, such as pumps, valves, and regulators, with input devices, such as flow rate sensors, pressure sensors, liquid level sensors, temperature sensors, fluid concentration sensors, fluid physical property sensors, and fluid particle sensors. Although detailed configuration is omitted, for example, the sensor may be configured by appropriately combining AC devices such as contact or non-contact heaters and input devices such as temperature sensors and current sensors.

[0066] (Control Unit) 7 is a block diagram showing an example of the substrate processing apparatus 2. The control unit 25 is electrically connected to each of the units 21 to 24 and functions as a control unit that comprehensively controls each of the units 21 to 24. In the following, the control systems (AC devices, input devices, output devices, control devices) of the polishing unit 22 and the finishing unit 24 will be described as examples, but the other units 21 and 23 have the same basic configurations and functions, so descriptions thereof will be omitted.

[0067] The polishing unit 22 is provided with a plurality of AC devices 227 and output devices 228B that are to be controlled and are arranged in each sub-unit (e.g., polishing table 220, top ring 221, polishing fluid supply unit 222, dresser 223, atomizer 224, etc.) of the polishing unit 22, a plurality of input devices 228A that detect data (detection values) necessary for controlling each sub-unit, and a control device 229 that controls the AC devices 227 and the output devices 228B based on the detection values ​​of each input device 228A.

[0068] The finishing unit 24 is provided with a plurality of AC devices 247 and output devices 248B to be controlled, which are arranged in each sub-unit (e.g., first and second roll sponge cleaning units 24A, 24B, first and second pen sponge cleaning units 24C, 24D, first and second drying units 24E, 24F, first and second conveying units 24G, 24H, etc.) of the finishing unit 24, a plurality of input devices 248A that detect data (detection values) necessary for controlling each sub-unit, and a control device 249 that controls the operation of the AC devices 247 and output devices 248B based on the detection values ​​of each input device 248A.

[0069] The control unit 25 includes a control unit 250, a communication unit 251, an input unit 252, an output unit 253, and a storage unit 254. The control unit 25 is configured, for example, by a general-purpose or dedicated computer (see FIG. 8 described later).

[0070] The communication unit 251 is connected to the network 7 and functions as a communication interface for transmitting and receiving various data. The input unit 252 accepts various input operations, and the output unit 253 functions as a user interface by outputting various information via a display screen, signal tower lighting, and buzzer sound.

[0071] The storage unit 254 stores various programs (such as an operating system (OS), application programs, and a web browser) and data (such as apparatus setting information 255 and substrate recipe information 256) used in the operation of the substrate processing apparatus 2. The apparatus setting information 255 and the substrate recipe information 256 are data that can be edited by the user via a display screen.

[0072] The control unit 250 acquires detection values ​​of a plurality of input devices 218A, 228A, 238A, 248A (hereinafter referred to as an "input device group") via a plurality of control devices 219, 229, 239, 249 (hereinafter referred to as a "control device group"), and performs a series of substrate processing by operating a plurality of AC devices 217, 227, 237, 247 (hereinafter referred to as an "AC device group") and a plurality of output devices 218B, 228B, 238B, 248B (hereinafter referred to as an "output device group") in cooperation with each other. In addition, when the control unit 250 receives interlock information 111 from the safety support device 5, it performs interlock control based on the interlock information 111.

[0073] (Hardware configuration of each device) FIG. 8 is a hardware configuration diagram showing an example of the computer 900. Each of the control unit 25, the database device 3, the machine learning device 4, the safety support device 5, and the user terminal device 6 is configured by a general-purpose or dedicated computer 900.

[0074] 8, the computer 900 includes, as its main components, a bus 910, a processor 912, a memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication I / F (interface) unit 922, an external device I / F unit 924, an I / O (input / output) device I / F unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the application of the computer 900.

[0075] The processor 912 is composed of one or more arithmetic processing devices (such as a central processing unit (CPU), a micro-processing unit (MPU), a digital signal processor (DSP), a graphics processing unit (GPU), or a neural processing unit (NPU)) and operates as a control unit that controls the entire computer 900. The memory 914 stores various data and programs 930, and is composed of, for example, a volatile memory (DRAM, SRAM, etc.) that functions as a main memory, a non-volatile memory (ROM), a flash memory, etc.

[0076] The input device 916 is, for example, a keyboard, a mouse, a numeric keypad, an electronic pen, etc., and functions as an input unit. The output device 917 is, for example, a sound (audio) output device, a vibration device, etc., and functions as an output unit. The display device 918 is, for example, a liquid crystal display, an organic EL display, electronic paper, a projector, etc., and functions as an output unit. The input device 916 and the display device 918 may be integrally configured, such as a touch panel display. The storage device 920 is, for example, a HDD, an SSD, etc., and functions as a storage unit. The storage device 920 stores various data necessary for the execution of the operating system and the program 930.

[0077] The communication I / F unit 922 is connected to a network 940 (which may be the same as the network 7 in FIG. 1) such as the Internet or an intranet by wire or wirelessly, and functions as a communication unit that transmits and receives data to and from other computers according to a predetermined communication standard. The external device I / F unit 924 is connected to an external device 950 such as a camera, a printer, a scanner, a reader / writer by wire or wirelessly, and functions as a communication unit that transmits and receives data to and from the external device 950 according to a predetermined communication standard. The I / O device I / F unit 926 is connected to an I / O device 960 such as various sensors and actuators, and functions as a communication unit that transmits and receives various signals and data, such as detection signals from sensors and control signals to actuators, between the I / O device 960. The media input / output unit 928 is composed of, for example, a drive device such as a DVD drive or a CD drive, a memory card slot, and a USB connector, and reads and writes data from and to a medium (non-temporary storage medium) 970 such as a DVD, a CD, a memory card, or a USB memory.

[0078] In the computer 900 having the above configuration, the processor 912 loads the program 930 stored in the storage device 920 into the memory 914, executes the program, and controls each unit of the computer 900 via the bus 910. The program 930 may be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded in the medium 970 in an installable file format or an executable file format, and provided to the computer 900 via the media input / output unit 928. The program 930 may be provided to the computer 900 by downloading it via the network 940 via the communication I / F unit 922. Furthermore, the computer 900 may implement various functions, which are realized by the processor 912 executing the program 930, in, for example, an FPGA (Field Programmable Gate Array (FPGA)). It may also be realized by hardware such as a Field-Programmable Gate Array (FGDA) or an Application Specific Integrated Circuit (ASIC).

[0079] The computer 900 is, for example, a desktop computer or a portable computer, and is an electronic device of any type. The computer 900 may be a client computer, a server computer, or a cloud computer, or may be, for example, an embedded computer called a control panel, a controller (including a microcomputer, a programmable logic controller, and a sequencer). The computer 900 may also be applied to devices other than the devices 2 to 6.

[0080] (Machine Learning Device 4) 9 is a block diagram showing an example of the machine learning device 4. The machine learning device 4 includes a control unit 40, a communication unit 41, a learning data storage unit 42, a trained model storage unit 43, an input unit 44, and an output unit 45.

[0081] The control unit 40 functions as a learning data acquisition unit 400 and a machine learning unit 401. The communication unit 41 is connected to external devices (e.g., the substrate processing apparatus 2, the database device 3, the safety support device 5, the user terminal device 6, a three-dimensional model device (not shown), etc.) via the network 7, and functions as a communication interface for transmitting and receiving various data. The input unit 44 accepts various input operations, and the output unit 45 functions as a user interface by outputting various information via a display screen or voice.

[0082] The learning data acquisition unit 400 acquires learning data 11 consisting of image data 110 as input data and interlock information 111 as output data. The learning data acquisition unit 400 may acquire the learning data 11 in cooperation with an external device connected via the communication unit 41 and the network 7, for example, or may acquire the learning data 11 by accepting an input operation via the input unit 44 and the output unit 45. The learning data 11 is data used as teacher data (training data), verification data, and test data in supervised learning. The interlock information 11 is data used as a correct answer label in supervised learning.

[0083] The learning data storage unit 42 is a database that stores a plurality of sets of learning data 11 acquired by the learning data acquisition unit 400. The specific configuration of the database that constitutes the learning data storage unit 42 may be designed as appropriate.

[0084] The machine learning unit 401 performs machine learning using multiple sets of learning data 11 stored in the learning data storage unit 42. That is, the machine learning unit 401 inputs multiple sets of learning data 11 to the learning model 10, and generates a trained learning model 10 by having the learning model 10 learn the correlation between the image data 110 and the interlock information 111 included in the learning data 11.

[0085] The trained model storage unit 43 is a database that stores the trained learning model 10 (specifically, a group of adjusted weight parameters) generated by the machine learning unit 401. The trained learning model 10 stored in the trained model storage unit 43 is provided to an actual system (e.g., a safety support device 5) via the network 7, a recording medium, or the like. Note that, although the learning data storage unit 42 and the trained model storage unit 43 are shown as separate storage units in FIG. 9, they may be configured as a single storage unit.

[0086] FIG. 10 is a diagram showing an example of the learning model 10 and the learning data 11. The learning data 11 used for the machine learning of 0 is composed of image data 110 and interlock information 111.

[0087] The image data 110 constituting the learning data 11 is captured by the image capturing unit 63 disposed facing the front of the worker U when the worker U wears the user terminal device 6. The image data 110 includes, for example, images captured when the worker U works on the substrate processing apparatus 2 with at least a part of his / her body within the movable range of a movable part, and each part of the substrate processing apparatus 2 is captured under various capturing conditions with different capturing positions, capturing angles, capturing ranges, etc.

[0088] The interlock information 111 constituting the learning data 11 includes at least one of an interlock on / off state and an interlock control range as an interlock state. The interlock on / off state indicates whether movement is permitted or prohibited for the movable part, with permitted movement being a state in which movement of the movable part is permitted and prohibited movement being a state in which movement of the movable part is prohibited. The interlock control range indicates the permitted movement range or prohibited movement range of the movable part relative to the movable range in the axial coordinate system (position and angle) of the movable part, with the permitted movement range being a range in which movement of the movable part is permitted and the prohibited movement range being a range in which movement of the movable part is prohibited.

[0089] Note that interlock information 111 may include only interlock information 111 for a single specific moving part, or may include interlock information 111 for multiple moving parts as shown in Fig. 10. When interlock information 111 includes interlock information 111 for multiple moving parts, it is sufficient that at least one of an interlock on / off state and an interlock control range is included for each moving part.

[0090] For example, using a substrate processing apparatus 2 for actual production or a substrate processing apparatus 2 for testing, the learning data acquisition unit 400 acquires image data 110 captured by the image capturing unit 63 under various capturing conditions by having an examiner wearing the user terminal device 6 assume a position and posture similar to that of an actual worker U performing work on the substrate processing apparatus 2, and also acquires interlock information 111 of the movable parts for the position and posture when the image data 110 was captured. The interlock information 111 may be input by the examiner via the user terminal device 6, or may be input via the input unit 44 and the output unit 45.

[0091] In addition, the learning data acquisition unit 400 acquires, as image data 110, virtual space data that is expected to be captured by the image shooting unit 63 when it is assumed that a worker U is present at a specific position in a virtual space in which the three-dimensional shapes of each part of the substrate processing apparatus 2 are reproduced, for example, using a three-dimensional model provided by a three-dimensional model device, and acquires interlock information 111 for the movable part based on whether or not there is a high risk of the movable part colliding with the worker U when it is assumed that the movable part moves within its movable range.

[0092] The learning model 10 employs, for example, a neural network structure, and includes an input layer 100, an intermediate layer 101, and an output layer 102. Synapses (not shown) that connect each neuron are laid between each layer, and each synapse is associated with a weight. A weight parameter group consisting of the weights of each synapse is adjusted by machine learning.

[0093] The input layer 100 has neurons in a number corresponding to each pixel of image data 110 as input data, and the pixel value of each pixel is input to each neuron. The output layer 102 has neurons in a number corresponding to interlock information 111 as output data, and the prediction result (inference result) of the interlock information 111 for the image data 110 is output as output data. is output.

[0094] The number of learning models 10 stored in the trained model storage unit 43 is not limited to one, and multiple learning models with different conditions may be stored, such as machine learning methods, differences in the mechanisms of the substrate processing apparatus 2, differences in the specifications (resolution and type of image sensor) of the image capturing unit 63, types of data included in the interlock information 111, etc. In this case, the learning data storage unit 42 may be configured to store multiple types of learning data having data configurations respectively corresponding to multiple learning models with different conditions.

[0095] (Machine learning methods) FIG. 11 is a flowchart showing an example of a machine learning method performed by the machine learning device 4.

[0096] First, in step S100, the learning data acquisition unit 400 acquires a desired number of learning data 11 as a preliminary preparation for starting machine learning, and stores the acquired learning data 11 in the learning data storage unit 42. The number of learning data 11 to be prepared here may be set in consideration of the inference accuracy required for the learning model 10 to be finally obtained.

[0097] Next, in step S110, the machine learning unit 401 prepares a pre-learning learning model 10 in order to start machine learning. The pre-learning learning model 10 prepared here is configured with a neural network model, and the weights of each synapse are set to initial values.

[0098] Next, in step S120, the machine learning unit 401 acquires, for example, one set of learning data 11 randomly from the multiple sets of learning data 11 stored in the learning data storage unit .

[0099] Next, in step S130, the machine learning unit 401 inputs input data (image data 110) included in one set of learning data 11 to the input layer 100 of the prepared learning model 10 before learning (or during learning). As a result, output data (interlock information 111) is output as an inference result from the output layer 102 of the learning model 10, but the output data is generated by the learning model 10 before learning (or during learning). Therefore, in the state before learning (or during learning), the output data output as the inference result indicates information different from the correct label (interlock information 111) included in the learning data 11.

[0100] Next, in step S140, the machine learning unit 401 performs machine learning by comparing the correct label included in the set of learning data 11 acquired in step S120 with the output data output from the output layer 102 as an inference result in step S130 and performing a process of adjusting the weight of each synapse (backpropagation). In this way, the machine learning unit 401 causes the learning model 10 to learn the correlation between the input data and the output data.

[0101] Next, in step S150, the machine learning unit 401 determines whether a predetermined learning termination condition has been satisfied, for example, based on an evaluation value of an error function based on the correct label included in the learning data 11 and the output data output as the inference result, or the remaining number of unlearned learning data 11 stored in the learning data memory unit 42.

[0102] In step S150, if the machine learning unit 401 determines that the learning end condition is not satisfied and that machine learning is to be continued (No in step S150), the process returns to step S120, and the process of steps S120 to S140 is performed multiple times on the learning model 10 being trained using untrained learning data 11. If the machine learning end condition is satisfied and the machine learning is to be ended (Yes in step S150), the process proceeds to step S160.

[0103] Then, in step S160, the machine learning unit 401 stores the trained learning model 10 (adjusted weight parameter group) generated by adjusting the weights associated with each synapse in the trained model storage unit 43, and ends the series of machine learning methods shown in Fig. 11. In the machine learning method, step S100 corresponds to a learning data storage step, steps S110 to S150 correspond to a machine learning step, and step S160 corresponds to a trained model storage step.

[0104] As described above, the machine learning device 4 and machine learning method of this embodiment can provide a learning model 10 that can predict (infer) interlock information 111 indicating the interlock state of a movable part from image data 110 captured by the image capturing unit 63 that is positioned facing forward of the worker U when the worker U wears the user terminal device 6.

[0105] (Safety support device 5) Fig. 12 is a block diagram showing an example of the safety support device 5. Fig. 13 is a functional explanatory diagram showing an example of the safety support device 5. The safety support device 5 includes a control unit 50, a communication unit 51, and a storage unit 52.

[0106] The control unit 50 functions as an image data acquisition unit 500, an interlock information generation unit 501, a movable part information acquisition unit 502, an object information generation unit 503, and an output processing unit 504. The communication unit 51 is connected to external devices (e.g., the substrate processing apparatus 2, the database device 3, the machine learning device 4, and the user terminal device 6) via the network 7, and functions as a communication interface for transmitting and receiving various data. The storage unit 52 stores various programs (such as an operating system and a user terminal program) and data (learning model 10) used in the operation of the safety support device 5.

[0107] The image data acquisition unit 500 acquires image data 110 captured by the image capturing unit 63 disposed facing the front of the worker U when the worker U wears the user terminal device 6. In this embodiment, the image data acquisition unit 500 acquires (receives) the image data 110 from the user terminal device 6 via the communication unit 51 and the network 7.

[0108] The interlock information generation unit 501 generates interlock information 111 based on the image data 110 acquired by the image data acquisition unit 500. In this embodiment, the interlock information generation unit 501 generates the interlock information 111 for the image data 110 by inputting the image data 110 acquired by the image data acquisition unit 500 to a learning model 10 that has undergone machine learning to learn the correlation between the image data 110 and the interlock information 111.

[0109] The storage unit 52 stores a trained learning model 10 used by the interlock information generation unit 501. The number of learning models 10 stored in the storage unit 52 is not limited to one, and a plurality of trained models having different conditions, such as machine learning techniques, differences in the mechanisms of the substrate processing apparatus 2, differences in the specifications (resolution and type of image sensor) of the image capturing unit 63, types of data included in the interlock information 111, etc., may be stored and used selectively or in parallel. The storage unit 52 may be substituted by a storage unit of an external computer (for example, a server-type computer or a cloud-type computer), and in that case, the interlock information generation unit 501 and the interlock information generation unit 502 may access the external computer.

[0110] The movable part information acquisition unit 502 acquires movable part information 113 indicating the position of the movable part existing around the worker U based on the image data 110 acquired by the image data acquisition unit 500. For example, the movable part information acquisition unit 502 determines whether or not a feature point of the movable part is included within the angle of view when the real space is photographed by the image photographing unit 63, and acquires the movable part information 113 based on the position of the feature point when it detects that the feature point of the movable part is included. The feature point may be based on, for example, the outer shape or outer color of the movable part, or on characters or two-dimensional codes on a sticker attached to the movable part. In addition, when design drawing data of each part of the substrate processing apparatus 2 is stored in the storage unit 52, the movable part information acquisition unit 502 may refer to the design drawing data, and acquire spatial position information of the movable part based on the feature point when it detects that the feature point of the movable part in the design drawing data is included within the photographing range when the real space is photographed by the image photographing unit 63.

[0111] Based on the interlock information 111 generated by the interlock information generation unit 501 and the movable part information 113 acquired by the movable part information acquisition unit 502, the object information generation unit 503 generates object information 112 for superimposing the interlock information 111 on the movable parts present around the worker U when the user terminal device 6 is worn by the worker U.

[0112] The output processing unit 504 performs output processing for outputting the interlock information 111 generated by the interlock information generating unit 501 and the object information 112 generated by the object information generating unit 503. For example, the output processing unit 504 transmits the interlock information 111 to the substrate processing apparatus 2 and the user terminal device 6, whereby the substrate processing apparatus 2 performs interlock control based on the interlock information 111, and the user terminal device 6 outputs a display screen and sound based on the interlock information 111. Furthermore, the output processing unit 504 transmits the object information 112 to the user terminal device 6, whereby the user terminal device 6 outputs a display screen based on the object information 112.

[0113] (User terminal device 6) 14 is a block diagram showing an example of the user terminal device 6. The user terminal device 6 includes a control unit 60, a communication unit 61, a storage unit 62, an image capturing unit 63, an input unit 64, an output unit 65, and a group of sensors 66.

[0114] The control unit 60 functions as an image data transmission processing unit 600, an interlock information processing unit 601, and an object information processing unit 602. The communication unit 61 is connected to an external device (e.g., the substrate processing apparatus 2, the database device 3, the machine learning device 4, and the safety support device 5, etc.) via the network 7, and functions as a communication interface for transmitting and receiving various data. The storage unit 62 stores various programs (such as an operating system and a user terminal program) and data used in the operation of the user terminal device 6. The image capturing unit 63 captures real space and generates image data 110. The input unit 64 accepts various input operations, and the output unit 65 functions as a user interface by outputting various information via a display screen or sound. The sensor group 66 detects the acceleration, angular velocity, attitude, etc. of the device itself.

[0115] The image data transmission processing unit 600 transmits image data 110 captured by the image capturing unit 63 at a predetermined capturing period to the safety support device 5 via the communication unit 61 and the network 7 as needed.

[0116] The interlock information processing unit 601 receives interlock information 111 from the safety support device 5 as a response to the image data 110 transmitted to the safety support device 5, and notifies the worker U of the interlock information 111, for example, via a voice or a display screen from the output unit 65. do.

[0117] The object information processing unit 602 receives object information 112 from the safety support device 5 in response to the image data 110 transmitted to the safety support device 5, and based on the object information 112, notifies the worker U of the interlock information 111 by superimposing the interlock information 111 on the movable parts present around the worker U using the output unit 65.

[0118] (Safety support method) 15 is a flowchart showing an example of a safety support method by the substrate processing apparatus 2, the safety support device 5, and the user terminal device 6. In the following, an example of the operation of the substrate processing apparatus 2, the safety support device 5, and the user terminal device 6 will be described when an operator U wearing the user terminal device 6 opens a cover attached to the first polishing unit 22A of the substrate processing apparatus 2 and performs an adjustment operation for adjusting the second transfer position TP2 for transferring the wafer W between the top ring 221 in the first polishing unit 22A and the first linear transporter 230A as an operation for the substrate processing apparatus 2. The flowchart shown in FIG. 15 is executed repeatedly every time an image capturing period by the image capturing unit 63 elapses when the user terminal device 6 receives, for example, an input operation from the operator U instructing the start of the adjustment operation.

[0119] First, in step S200, the image capturing unit 63 of the user terminal device 6 captures an image of the real space in front of the worker U performing the adjustment work, and generates image data 110. Then, the image data transmission processing unit 600 transmits the image data 110 captured by the image capturing unit 63 to the safety support device 5.

[0120] Next, in step S210, the image data acquisition unit 500 of the safety support device 5 receives the image data 110 transmitted in step S200.

[0121] Next, in step S220, the interlock information generation unit 501 uses the image data 110 acquired in step S210 as input data, and generates interlock information 111 for the image data 110 based on output data output by inputting the image data 110 into the learning model 10. Then, in step S221, the output processing unit 504 transmits the interlock information 111 generated in step S220 to the substrate processing apparatus 2 as output processing for outputting the interlock information 111.

[0122] Next, in step S222, upon receiving the interlock information 111 transmitted in step S221, the control unit 25 of the substrate processing apparatus 2 performs interlock control based on the interlock information 111. Note that in step S221, the output processing unit 504 may also transmit the interlock information 111 to the user terminal device 6, and in that case, the interlock information processing unit 601 of the user terminal device 6 may notify the worker U of the interlock information 111 by voice or via a display screen.

[0123] Next, in step S230, the movable part information acquisition unit 502 acquires movable part information 113 indicating the positions of movable parts present around the worker U, based on the image data 110 acquired in step S210.

[0124] Next, in step S231, the object information generating unit 503 generates object information 112 for superimposing the interlock information 111 on the movable parts present around the worker U, based on the interlock information 111 generated in step S220 and the movable part information 113 acquired in step S230. Then, in step S232, Then, the output processing unit 504 transmits the object information 112 generated in step S231 to the user terminal device 6 as an output process for outputting the object information 112.

[0125] Next, in step S233, when the object information processing unit 602 of the user terminal device 6 receives the object information 112 transmitted in step S232, it displays a virtual object indicating interlock information 111 for the movable part in real space on the output unit 65 of the user terminal device 6 based on the object information 112.

[0126] Fig. 16 is a diagram showing a first example in which interlock information 111A is superimposed on a movable part in real space. Fig. 17 is a diagram showing a second example in which interlock information 111B is superimposed on a movable part in real space. In the following, for the sake of simplicity, it is assumed that the angle of view of the image data 110 and the field of view of the worker U match, but the angle of view of the image data 110 and the field of view of the worker U do not have to match, and either one may be wider.

[0127] FIG. 16 illustrates a case in which interlock information 111A generated based on image data 110 captured by the image capturing unit 63 is displayed as virtual objects 120A to 122A when a worker U is standing in front of the cover of the first polishing unit 22A and is within reach of the top ring 221 of the first polishing unit 22A.

[0128] The virtual objects 120A and 121A indicate that, as the interlock information 111A for the top ring 221 of the first polishing unit 22A captured in the image data 110, the interlock of the vertical movement mechanism 221d is on (movement prohibited) and the entire range is set as the prohibited movement range, and the interlock of the swing movement mechanism 221e is on (movement prohibited) and the range from 0 to 60 is set as the prohibited movement range. The virtual object 122A indicates that, as the interlock information 111A for the first linear transporter 230A captured in the image data 110, the interlock of the first linear transporter 230A is off (movement permitted) and the entire range is set as the permitted movement range.

[0129] 16, in the substrate processing apparatus 2, in accordance with the display of the virtual objects 120A to 122A, interlock control is performed in step S222 by the interlock information 111A generated based on the image data 110. Specifically, as the interlock control for the top ring 221 of the first polishing unit 22A captured in the image data 110, the movement of the vertical movement mechanism 221d is restricted over the entire range, and the movement of the swing movement mechanism 221e is restricted over the range of 0 to 60, similar to the virtual objects 120A to 121A. Also, as the interlock control for the first linear transporter 230A captured in the image data 110, the movement of the first linear transporter 230A is permitted over the entire range, similar to the virtual object 122A. In other words, when the worker U is standing in front of the cover of the first polishing section 22A and can reach the top ring 221 of the first polishing section 22A, there is no risk of a collision between the worker U and the first linear transporter 230A, and therefore movement of the first linear transporter 230A is permitted.

[0130] FIG. 17 illustrates a case in which interlock information 111B generated based on image data 110 captured by the image capturing unit 63 is displayed as virtual objects 120B to 122B when a worker U leans his or her upper body inside the first polishing unit 22A and is in a position where he or she can reach the first linear transporter 230A.

[0131] The interlock information 111B for the top ring 221 of the first polishing unit 22A captured in the image data 110 is set to ON (movement prohibited) for the vertical movement mechanism unit 221d, and the entire range of the virtual objects 120B and 121B is set to the movement prohibited range. This indicates that the interlock of the swing movement mechanism 221e is on (movement prohibited) and the range from 0 to 60 has been set as the prohibited movement range. The virtual object 122B indicates, as the interlock information 111B for the first linear transporter 230A captured in the image data 110, that the interlock of the first linear transporter 230A is on (movement prohibited) and the entire range has been set as the prohibited movement range.

[0132] 17, in the substrate processing apparatus 2, in accordance with the display of the virtual objects 120B to 122B, interlock control is performed in step S222 by the interlock information 111B generated based on the image data 110. Specifically, as the interlock control for the top ring 221 of the first polishing unit 22A captured in the image data 110, the movement of the vertical movement mechanism 221d is restricted over the entire range, and the movement of the swing movement mechanism 221e is restricted over the range of 0 to 60, similar to the virtual objects 120B to 121B. Also, as the interlock control for the first linear transporter 230A captured in the image data 110, the movement of the first linear transporter 230A is prohibited over the entire range, similar to the virtual object 122B. In other words, if the worker U leans his / her upper body inside the first polishing section 22A and is able to reach the first linear transporter 230A, there is a risk of a collision between the worker U and the first linear transporter 230A, so movement of the first linear transporter 230A is prohibited.

[0133] 16 and 17, the interlock information 111A, 111B includes the interlock status for the vertical movement mechanism 221d and the swing movement mechanism 221e of the top ring 221 and the first linear transporter 230A as the movable parts, but may further include the interlock status for other movable parts. The other movable parts may be, for example, movable parts captured in the image data 110, that is, movable parts present within the angle of view of the image capturing unit 63 of the user terminal device 6 worn by the worker U, or movable parts estimated to be present around the worker U.

[0134] As described above, the interlock control is performed by the substrate processing apparatus 2, and the worker U can grasp the interlock state of the movable parts by visually checking the virtual objects 120A-122A and 120B-122B displayed on the output unit 65. Then, as the position and orientation of the body of the worker U are changed and the photographing conditions of the image data 110 are changed, the process of generating the interlock information 111 and the object information 112 is repeatedly performed. This allows the worker U to safely perform the work on the substrate processing apparatus 2. In the above safety support method, step S210 corresponds to an image data acquisition step, step S220 corresponds to an interlock information generation step, step S230 corresponds to a movable part information acquisition step, and step S231 corresponds to an object information generation step.

[0135] According to the safety support device 5 and the safety support method of the present embodiment, when a worker U wearing a user terminal device 6 performs work on the substrate processing apparatus 2 with at least a part of his / her body within the movable range of a movable part, interlock information 111 of the movable part is generated based on image data 110 captured by an image capturing unit 63 arranged facing the front of the worker U, so that the interlock of the movable part can be controlled according to the positional relationship between the worker U and the movable part. Therefore, the work efficiency of the worker U can be improved while appropriately ensuring the safety of the worker U. Moreover, according to the safety support device 5 and the safety support method of the present embodiment, the interlock information 111 is superimposed and displayed on the movable part in the real space, so that the worker U can grasp the interlock state of the movable part.

[0136] (Other embodiments) The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and scope of the present invention. It is contained in the idea.

[0137] In the above embodiment, the database device 3, the machine learning device 4, the safety support device 5, and the user terminal device 6 are described as being configured as separate devices, but the four devices may be configured as a single device, or any two or three of the four devices may be configured as a single device. In addition, at least one of the machine learning device 4 and the safety support device 5 may be incorporated in the control unit 25 of the substrate processing apparatus 2 or the user terminal device 6. For example, the learning model 10 may be stored in the storage unit 62 of the user terminal device 6, and the control unit 60 of the user terminal device 6 may function as the image data acquisition unit 500, the interlock information generation unit 501, the movable part information acquisition unit 502, and the object information generation unit 503.

[0138] In the above embodiment, the substrate processing apparatus 2 has been described as including each of the units 21 to 24. However, the substrate processing apparatus 2 may be a processing apparatus that performs at least one of the substrate processing steps, for example, a polishing process, a cleaning process, and a drying process, and the polishing process may be a physical mechanical polishing process instead of a chemical mechanical polishing process.

[0139] In the above embodiment, the machine learning device 4, the safety support device 5, and the user terminal device 6 have been described as being applied to the substrate processing apparatus 2 that performs substrate processing, but they may be applied to any processing apparatus as long as it is a processing apparatus that performs a predetermined processing by moving a movable part having a predetermined movable range. The processing apparatus may be a closed type apparatus in which a cover or door is provided on the housing 20 and the movable part is arranged in the space inside the cover or door, as with the substrate processing apparatus 2, or an open type apparatus in which the movable part is arranged in a space such as indoors or outdoors without a cover or door.

[0140] In the above embodiment, a case has been described in which a neural network is used as a learning model for implementing machine learning by the machine learning unit 401, but other machine learning models may be used. Examples of other machine learning models include tree types such as decision trees and regression trees, ensemble learning such as bagging and boosting, recurrent neural networks, convolutional neural networks, and neural network types such as LSTM (including deep learning). (including hierarchical clustering, non-hierarchical clustering, k-nearest neighbors, k-means, etc.) Examples of such methods include multivariate analyses such as filtering, principal component analysis, factor analysis, and logistic regression, as well as support vector machines.

[0141] In the above embodiment, the interlock information generating unit 501 of the safety support device 5 uses the learned learning model 10 when generating the interlock information 111 based on the image data 110 acquired by the image data acquiring unit 500. However, other methods may be adopted. For example, a rule base based on image processing may be used as another method. That is, the interlock information generating unit 501 may acquire worker information indicating the position of the worker U by performing image processing on the image data 110 acquired by the image data acquiring unit 500, and generate the interlock information 111 for the image data 110 based on the acquired worker information. The position of the worker U may be, for example, an absolute position with an arbitrary position in the housing 20 of the substrate processing apparatus 2 as the origin, or a relative position from the movable part with the position of the movable part as the reference. Then, the interlock information generating unit 501 may determine whether or not there is a high risk of the movable part colliding with the worker U based on the positional relationship (distance and height) between the worker U and the movable part, and generate the interlock information 111 based on the determination result. At this time, the interlock information generating unit 501 may further take into consideration the position of the movable part at the time of the judgment and generate the interlock information 111. The position of the movable part may be acquired by, for example, the movable part information acquiring unit 502 performing image processing on the image data 110, or may be acquired by the control unit 25 of the substrate processing apparatus 2. Alternatively, the report R may be acquired by receiving the report R from the communication unit 51 via the network 7.

[0142] In the above embodiment, the safety support device 5 operates according to the flowchart shown in Fig. 15, but the order of execution of each step may be changed as appropriate, or some steps may be omitted. For example, steps S230 to S232 may be omitted.

[0143] (Machine learning program and safety support program) The present invention can also be provided in the form of a program (machine learning program) that causes the computer 900 to function as each unit included in the machine learning device 4, or a program (machine learning program) that causes the computer 900 to execute each step included in the machine learning method. The present invention can also be provided in the form of a program (safety support program) that causes the computer 900 to function as each unit included in the safety support device 5 or the user terminal device 6, or a program (safety support program) that causes the computer 900 to execute each step included in the safety support method according to the above embodiment.

[0144] (Inference device, inference method, and inference program) The present invention can be provided not only in the form of the safety support device 5 (safety support method or safety support program) according to the above embodiment, but also in the form of an inference device (inference method or inference program) used to infer interlock information. In this case, the inference device (inference method or inference program) can include a memory and a processor, and the processor executes a series of processes. The series of processes includes image data processing (image data acquisition step) for acquiring image data 110, and inference processing (inference step) for inferring interlock information 111 indicating the interlock state of the movable part based on the image data acquired by the image data acquisition process.

[0145] By providing it in the form of an inference device (inference method or inference program), it can be easily applied to various devices compared to the case where a safety support device is implemented. It can be naturally understood by those skilled in the art that when the inference device (inference method or inference program) infers interlock information, the inference method implemented by the interlock information generation unit may be applied using a trained learning model generated by the machine learning device and machine learning method according to the above embodiment. [Explanation of symbols]

[0146] 1... substrate processing system, 2... substrate processing apparatus, 3... database device, 4...machine learning device, 5...safety support device, 6...user terminal device (worker device), 7...Network, 10...Learning model, 11...Learning data 20...housing, 21...load / unload unit, 22...polishing unit, 23: substrate transport unit, 24: finishing unit, 25: control unit, 40: control unit, 41: communication unit, 42: learning data storage unit, 43... trained model memory unit, 44... input unit, 45... output unit, 50: control unit, 51: communication unit, 52: storage unit, 60: control unit, 61: communication unit, 62: storage unit, 63: image capture unit, 64: input unit, 65: output unit; 66: sensor group; 110: image data, 111, 111A, 111B: interlock information, 112... object information, 113... moving part information, 120A~122A, 120B~122B...Virtual objects 400...learning data acquisition unit, 401...machine learning unit, 500: image data acquisition unit; 501: interlock information generation unit; 502: movable part information acquisition unit; 503: object information generation unit; 504: output processing unit; 600: image data transmission processing unit; 601: interlock information processing unit; 602... object information processing unit, 900...Computer, U...Worker, W...Wafer

Claims

1. A safety support device that supports the safety of a worker who performs a predetermined operation on a processing device that moves a movable part having a predetermined movable range and performs a predetermined process by placing at least a part of the body within the movable range, an image data acquisition unit that acquires image data captured by the image capture unit, which is disposed facing forward of the worker when the worker wears the worker device having the image capture unit; an interlock information generation unit that generates interlock information indicating an interlock state of the movable part based on the image data acquired by the image data acquisition unit, Safety support equipment.

2. The processing device The movable part is provided inside the processing device, The interlock information generation unit generating the interlock information based on the image data acquired by the image data acquisition unit when the worker places at least a part of his / her body inside the processing device; The safety support device according to claim 1 .

3. The interlock information generation unit acquiring worker information indicating the position of the worker by performing image processing on the image data acquired by the image data acquisition unit, and generating the interlock information for the image data based on the acquired worker information. The safety support device according to claim 1 .

4. The interlock information generation unit generating the interlock information for the image data by inputting the image data acquired by the image data acquisition unit into a learning model that has undergone machine learning to learn the correlation between the image data and the interlock information; The safety support device according to claim 1 .

5. a movable part information acquisition unit that acquires movable part information indicating positions of the movable parts present around the worker based on the image data acquired by the image data acquisition unit; an object information generation unit that generates the object information for superimposing the interlock information on the movable parts present around the worker when the worker device is worn by the worker, and that further includes an output unit that is capable of superimposing object information on an object in real space based on the interlock information generated by the interlock information generation unit and the movable part information acquired by the movable part information acquisition unit; The safety support device according to claim 1 .

6. The interlock information includes, as the interlock state, an interlock on / off state indicating whether movement of the movable part is permitted or prohibited; an interlock control range indicating a permitted movement range or a prohibited movement range of the movable part relative to the movable range, The safety support device according to any one of claims 1 to 5.

7. An inference device comprising a memory and a processor, The processor: an image data acquisition process for acquiring image data captured by an image capturing unit disposed facing forward when a worker wears a worker device equipped with an image capturing unit and performs a predetermined task with a processing device that moves a movable part having a predetermined movable range and performs a predetermined process by placing at least a part of the body within the movable range; When the image data is acquired in the image data acquisition process, an inference process is executed to infer interlock information indicating an interlock state of the movable part based on the image data. Reasoning device.

8. a learning data storage unit that stores multiple sets of learning data, each set consisting of image data captured by an image capturing unit that is positioned facing forward of a worker wearing the worker device, the image capturing unit being arranged to face the worker and that performs a predetermined task on a processing device that moves a movable unit having a predetermined range of motion to perform a predetermined process and that places at least a part of the worker's body within the movable range; and interlock information that indicates an interlock state of the movable unit; a machine learning unit that inputs a plurality of sets of the learning data into a learning model to cause the learning model to learn a correlation between the image data and the interlock information; a learned model storage unit that stores the learned model in which the correlation is learned by the machine learning unit, Machine learning device.

9. 1. A safety support method for supporting the safety of a worker who performs a predetermined operation on a processing device that moves a movable part having a predetermined movable range and performs a predetermined process by placing at least a part of the worker's body within the movable range, an image data acquisition step of acquiring image data captured by the image capture unit, which is disposed facing forward of the worker when the worker wears the worker device having the image capture unit; an interlock information generating step of generating interlock information indicating an interlock state of the movable part based on the image data acquired in the image data acquiring step, Safety support methods.

10. An inference method executed by an inference device having a memory and a processor, The processor: an image data acquisition step of acquiring image data captured by an image capturing unit disposed facing forward of a worker wearing a worker device equipped with an image capturing unit, the worker performing a predetermined task with a processing device that moves a movable part having a predetermined movable range and places at least a part of the body within the movable range; When the image data is acquired in the image data acquisition step, an inference step is executed of inferring interlock information indicating an interlock state of the movable part based on the image data. Reasoning method.

11. a learning data storage step of storing, in a learning data storage unit, a plurality of sets of learning data each consisting of image data captured by an image capturing unit disposed facing forward of a processing device that moves a movable part having a predetermined range of motion to perform a predetermined process and that places at least a part of the body within the movable range, the image capturing unit being attached to the processing device; and interlock information indicating an interlock state of the movable part; a machine learning process of inputting a plurality of sets of the learning data into a learning model to allow the learning model to learn a correlation between the image data and the interlock information; and a learned model storage step of storing the learned model, which has learned the correlation through the machine learning step, in a learned model storage unit. Machine learning methods.

Images