Fitting sound detection device

Abstract

A chimeric sound detection device is provided with: a microphone (21); a holding member (1) for holding a fitting member provided at an end of the wire harness; a sensor (25) that detects the wire harness in which the fitting member is held by the holding member (1); and a controller (26) for controlling the execution of a determination process for determining whether or not a fitting sound based on the fitting member is included in the sounds collected by the microphone (21) on the basis of sound data based on the sounds collected by the microphone (21), the controller (26) executing the determination process while the sensor (25) detects the wire harness, and controlling the execution of the determination process when the sensor (25) detects the wire harness. The determination processing is not executed during a period in which the sensor (25) does not detect the wire harness.

Description

Technical Field

[0001] This disclosure relates to a device for detecting interlocking sounds, which detects the interlocking sounds produced when interlocking components are fitted together. Background Technology

[0002] In the manufacture of products that include components that require electrical operation, such as automobiles, communication equipment, and medical equipment, wire harnesses are used to connect these components to each other. A wire harness is defined as a cable that transmits power or electrical signals, with a fitting member, such as a connector, attached to the end of the cable for connecting the cable to other components.

[0003] In the manufacturing process of products using wire harnesses, operators need to perform a fitting operation, in which fitting members at the ends of the wire harness fit together with other fitting members. If improper fitting results in the components not being properly electrically connected to each other, it may lead to manufacturing defects in the final product.

[0004] Therefore, Japanese Patent Application Publication Nos. 2023-112729 and 2022-108326 disclose a method for determining the quality of fitting operations by detecting the "fitting sound" produced when fitting members are properly engaged. In these two patents, a watch worn by the operator collects sound data during the fitting operation to detect the fitting sound. The fitting sound is a "predetermined sound" produced when the fitting members are correctly engaged, also known as a "click." For example, the fitting members are equipped with a locking mechanism that holds them in the engaged state when they are correctly fitted together. The fitting sound is produced when the locking mechanism correctly locks the two fitting members together. Summary of the Invention

[0005] To minimize manufacturing defects in products that require interlocking operations during the manufacturing process, it is necessary to improve the operability of interlocking operations and to detect interlocking sounds with high precision.

[0006] This disclosure provides a technique for improving the workability of interlacing operations and detecting interlacing sounds with high accuracy.

[0007] The present invention provides a mating tone detection device comprising: a microphone; a holding member for holding a mating member disposed at an end of a wire harness; a sensor for detecting the wire harness held by the holding member; and a controller for controlling the execution of a determination process, the determination process determining whether the sound collected by the microphone contains a mating tone based on the mating member, the controller performing the determination process during the period when the sensor detects the wire harness, and not performing the determination process during the period when the sensor does not detect the wire harness.

[0008] Other features of the invention will become apparent from the following description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of a chirp detection system that includes a chirp detection device.

[0010] Figure 2 This is a schematic perspective view of an example of a sound detection device.

[0011] Figure 3 This is a diagram illustrating an example of the state in which the wire harness is held by the interlocking tone detection device.

[0012] Figure 4 This is an example diagram showing a schematic appearance of the head of the sound detection device.

[0013] Figure 5 This is an example of the functional configuration of a chiasmus detection device. Detailed Implementation

[0014] The embodiments will now be described in detail with reference to the accompanying drawings. These embodiments do not limit the invention as claimed, and not all combinations of the features described in the embodiments are necessary for the invention. Two or more features described in the embodiments can be combined arbitrarily. The same reference numerals are used for the same or similar elements, and redundant explanations are omitted.

[0015] First Embodiment Figure 1 An example of a sound detection system including a sound detection device 100 is shown. The sound detection device 100 is used during sound detection operations in a factory. The sound detection device 100 can be configured to wirelessly access a local area network (LAN) 200 installed in the factory. A manufacturing system 600 that manages the manufacturing processes in the factory is connected to the LAN 200. The LAN 200 is connected to the Internet 300, and a learning server 400 and a database 500 are connected to the Internet 300.

[0016] According to this embodiment, the incoherent sound detection device 100 uses a learning model to detect incoherent sounds in an incoherent operation. The learning server 400 collects sound data (sound data) from multiple incoherent sound detection devices 100 during the incoherent operation and performs machine learning based on the collected sound data to generate a learning model.

[0017] Since the sounds produced by a chimera can vary depending on the type of chimera, a learning model can be generated for each type of chimera. In other words, a learning model can be associated with one type of chimera. The type of chimera can be distinguished by its shape, material, etc. If the same learning model can be applied to different types of chimeras, then the same learning model can be used. In other words, a learning model can be used to detect chimeras from multiple types of chimeras. For example, if multiple types of chimeras that can be used with the same learning model are treated as a group of components, then a learning model can be generated associated with a group of components.

[0018] Furthermore, since background noise in a factory can vary from factory to factory, a learning model can be generated for each factory (location) using the inlay sound detection device 100. In other words, a learning model can be associated with a location using that learning model. If the same learning model can be applied even if the locations using the learning model are different, then the same learning model can be used. In other words, a learning model can be used across multiple locations. For example, if multiple locations that can use the same learning model are treated as a location group, a learning model can be generated in association with a location group. Furthermore, a learning model can be generated for each combination of the type of inlay component and the factory (location) using the inlay sound detection device 100. Additionally, a learning model can be generated for each combination of component group and location group.

[0019] Database 500 stores one or more learning models generated by learning server 400. Before starting the chirping operation, chirping detection device 100 downloads and stores the learning models to be used in the chirping operation from database 500. Learning server 400 and database 500 can be the same server. Furthermore, database 500 can be connected to LAN 200, and database 500 can be configured to store only the learning models to be used in the factory where LAN 200 is installed.

[0020] Figure 2 A schematic perspective view of an example of a mortise detection device 100 is shown. Figure 3 The state of the engagement member 31 of the wire harness 3 is shown in the diagram where the engagement tone detection device 200 maintains the engagement tone detection device 200. Figure 4 As shown along Figure 2 A chimera detection device 100 for observation in the -X direction. Figures 2-4This is a schematic diagram for understanding an example of the interlacing tone detection device 100, and does not limit the configuration of the interlacing tone detection device 100. Figure 3 In the figure, reference numeral 310 indicates a locking mechanism provided in the fitting member 31, and reference numeral 32 indicates a cable (multiple) of the wire harness 3.

[0021] The mating tone detection device 100 includes a head 1 and a body 2. The head 1 is configured in various types depending on the shape and size of the mating member 31 of the wire harness 3 to be processed, and is detachable from and attached to the body 2. The operator performs the mating operation by attaching the head 1, which corresponds to the mating member 31 of the wire harness 3 to the body 2. Various components are housed within the housing of the body 2.

[0022] The head 1 is a retaining member for holding the fitting member 31 of the wire harness 3, and is provided with a handle member 11 and a handle member 12. When the fitting member 31 of the wire harness 3 is located between the handle member 11 and the handle member 12, the operator holds the fitting member 31 of the wire harness 3 by grasping (gripping) the handle member 11 and the handle member 12 to bring the handle members 11 and 12 close to each other.

[0023] like Figure 3 As shown, the head 1 can be configured such that, instead of holding the entire mating member 31 with the head 1, the head 1 only grasps a predetermined area on the cable 32 side of the mating member 31, and the remaining area of ​​the side of the mating member 31 to be mated with another mating member protrudes from the head 1.

[0024] like Figure 4 As shown, a hole 13 is provided on the side of the head 1 facing the body 2 to transmit light through an optical sensor 25 disposed on the body 2. Figure 5 The emitted light and its reflected light. The optical sensor 25 is configured to optically detect the cable 32 of the wire harness 3 when the operator holds the fitting member 31 of the wire harness 3 with the head 1.

[0025] Handle members 11 and 12 are configured to move or deform in a direction that brings them closer together, allowing the operator to hold the mating member 31. For example, handle members 11 and 12 can be configured to be elastically deformable. Handle members 11 and 12 can be configured such that only one of them is movable or deformable, thus bringing it closer to the other. Since the head 1 is configured so that the operator holds the mating member 31 by gripping it via handle members 11 and 12, the shape of the surface of the head 1 that contacts the mating member 31 does not need to be the same as the shape of the surface of the mating member 31 that contacts the head 1. Therefore, the same head 1 can be used for multiple types of mating members 31.

[0026] exist Figure 3 In the configuration example shown, if the direction in which head 1 aligns with body 2 when head 1 is attached to body 2 is defined as the X direction, then when head 1 is attached to body 2, handle member 11 and handle member 12 are positioned along the Y direction; however, head 1 can also be configured such that handle member 11 and handle member 12 are positioned along the X direction. In other words, head 1 can be positioned from... Figure 2 The object is attached to the main body 2 in the state shown, after rotating 90 degrees around the Z-axis, which serves as the axis of rotation.

[0027] In addition, Figure 3 In the configuration example shown, the direction in which head 1 aligns with body 2 when head 1 is attached to body 2 is defined as the X direction, and the longitudinal direction of the wire harness 3 held by head 1 is approximately the Z direction. This is because in Figure 3 In the configuration example shown, the head 1 is configured to hold the fitting member 31 of the wire harness 3 at one end in the Z direction. However, for example, when the direction in which the head 1 is aligned with the body 2 is defined as the X direction, the head 1 can be configured to hold the fitting member 31 of the wire harness 3 at the end opposite to the body 2 in the X direction. In other words, the head 1 can be configured to hold the fitting member 31 of the wire harness 3 at the end opposite to the body 2 in the X direction. Figure 2 The device is attached to the body 2 in the state shown, after rotating 90 degrees around the Y-axis, which is the axis of rotation. In this case, the cable 32 of the wire harness 3 held by the head 1 extends toward the body 2 (in the X direction), but does not interfere with the body 2 by pulling the cable 32 out between the handle member 11 and the handle member 12.

[0028] Figure 5 An example of the functional configuration of the interlocking tone detection device 100 is shown. Implementation Figure 5 The hardware of the functional blocks is housed within the casing of the main body 2. For example, the controller 26 contains one or more processors and is configured to control the entire chime detection device 100.

[0029] As described above, the optical sensor 25 detects whether the mating tone detection device 100 holds the wire harness 3. For example, the optical sensor 25 emits light through the hole 13 provided in the head 1 and receives its reflected light. When the head 1 does not hold the wire harness 3, the light emitted by the optical sensor 25 passes between the handle member 11 and the handle member 12, so the optical sensor 25 does not receive the reflected light. On the other hand, when the head 1 holds the wire harness 3, the optical sensor 25 receives the reflected light from the cable 32 of the wire harness 3. Therefore, whether the head 1 holds the wire harness 3 can be detected by whether the optical sensor 25 receives the reflected light.

[0030] Furthermore, a time-of-flight (TOF) sensor can be used as an optical sensor 25. A TOF sensor is an optical sensor that measures the distance to an object. When the head 1 holds the cable 3, the approximate distance between the TOF sensor and the cable 32 can be predetermined. Therefore, by using a TOF sensor, even if reflected light is received when the head 1 is not holding the cable 3, it is possible to detect with high precision whether the head 1 is holding the cable 3.

[0031] In this example, when the operator holds the fitting member 31 of the wire harness 3 with the head 1, the optical sensor 25 detects the cable 32 of the wire harness 3, but it can also detect the fitting member 31 of the wire harness 3. In other words, the optical sensor 25 can be configured to detect whether the operator is holding the fitting member 31 of the wire harness 3 with the head 1 by detecting the wire harness 3 with its fitting member 31 held by the head 1.

[0032] Microphone 21 is, for example, a microelectromechanical system (MEMS) microphone that converts sound into an analog sound signal and outputs it to the preprocessing unit 22. Microphone 21 can be located on the side of the head 1 in the main body 2, thereby enabling high-precision detection of interlaced tones. Furthermore, although... Figure 4 Although not shown, a hole may be provided on the side surface of the head 1 facing the body 2 to facilitate the microphone 21 to collect sound.

[0033] The preprocessing unit 22 has an analog-to-digital converter (ADC) that converts the analog sound signal output from the microphone 21 into digital sound data and outputs the digital sound data to the decision unit 23.

[0034] The determination unit 23 stores the learning model obtained from the database 500 and, by using sound data from the preprocessing unit 22 as input to the learning model, determines whether the sound represented by the sound data contains a spliced ​​tone generated by the splicing member 31. The determination unit 23 notifies the controller 26 of the determination result regarding whether the sound represented by the sound data contains a spliced ​​tone generated by the splicing member 31, i.e., the detection result regarding whether a spliced ​​tone was detected. The function of the determination unit 23 can be implemented by the processor of the controller 26.

[0035] Based on the determination result from the determination unit 23, the controller 26 presents the detection result of whether a chiming sound was detected to the operator via a user interface (UI) 27. As an example, the UI 27 has light-emitting diodes (LEDs) that illuminate in at least two colors. When a chiming sound is detected, the controller 26 controls the LED to illuminate in the first color, and when no chiming sound is detected, it controls the LED to illuminate in a second color different from the first color. Alternatively, the UI 27 has LEDs that emit light in a single color, for example. The controller 26 turns off the LEDs until a chiming sound is detected, and when a chiming sound is detected, it turns the LEDs on until a predetermined condition is met. For example, the predetermined condition can be met by the operator's confirmation input via the UI 27. Alternatively, the predetermined condition can be met when the optical sensor 25 no longer detects the wire harness 3. Furthermore, the predetermined condition can be met after a predetermined time period has elapsed after the optical sensor 25 no longer detects the wire harness 3. Additionally, the UI 27 may have multiple LEDs. The controller 26 changes the lighting / turning off method of the multiple LEDs according to whether a chiming sound is detected. UI 27 may have a display, and the display may be configured to show whether a chord has been detected. Furthermore, UI 27 may have one or more operation buttons for switching the power supply of the chord detection device 100 on / off, and for switching the operating mode of the chord detection device 100, which will be described later.

[0036] The communication unit 24 has the function of wirelessly accessing the LAN 200 and communicating with devices on the LAN 200 and the Internet 300 (such as the manufacturing system 600, the learning server 400 and the database 500).

[0037] The sound blending detection device 100 can be configured to operate in either "operation mode" or "learning mode". The operator can set the operating mode of the sound blending detection device 100 via UI 27. "Operation mode" is the mode used when performing blending operations, and "learning mode" is the mode used when sending sound data to the learning server 400 for machine learning.

[0038] The following describes the operation of the interlocking tone detection device 100 during the interlocking operation. The interlocking tone detection device 100 is set to an "operation mode" for the interlocking operation. The learning model corresponding to the target wiring harness 3 to be interlocked for the interlocking operation is stored in advance in the decision unit 23. Since the interlocking operation does not require the communication unit 24, the controller 26 can set the operating state of the communication unit 24 to a "power-saving state" instead of a "normal state". The normal state is when the communication unit 24 can operate normally, that is, the communication unit 24 can communicate. The power-saving state is when the power consumption is lower than that of the normal state. For example, the state in which the power supply to the communication unit 24 is disconnected is a form of power-saving state. The state in which power is continued to be supplied to a part of the circuit of the communication unit 24 and the power supply to the remaining circuit of the communication unit 24 is cut off to quickly switch to the normal state is also a form of power-saving state.

[0039] Similar to the communication unit 24, the microphone 21 and the pre-processing unit 22 are defined as having a normal state and a power-saving state as their operating states. In the normal state, the microphone 21 collects sound and outputs an audio signal. In the power-saving state, the microphone 21 does not collect sound and therefore does not output an audio signal. In the normal state, when an audio signal is input, the pre-processing unit 22 outputs audio data. In the power-saving state, even if an audio signal is input, the pre-processor 22 does not output audio data.

[0040] Furthermore, similar to communication unit 24, determination unit 23 is defined with a normal state and a power-saving state. In the normal state, when audio data is input, determination unit 23 uses a learning model to determine whether a chirp is detected. On the other hand, in the power-saving state, determination unit 23 does not perform a determination using the learning model.

[0041] In this embodiment, the controller 26 controls the interlacing tone detection device 100 such that the interlacing tone determination process is performed only when the optical sensor 25 detects the wire harness 3, and not during other time periods. For this purpose, the controller 26 sets both the microphone 21 and the pre-processing unit 22 to a power-saving state when the optical sensor 25 does not detect the wire harness 3. In other words, when the optical sensor 25 does not detect the wire harness 3, the controller 26 controls the interlacing tone detection device 100 not to collect sound. Therefore, no sound data is input to the determination unit 23, and thus, no determination process is performed. Instead of setting both the microphone 21 and the pre-processing unit 22 to a power-saving state when the optical sensor 25 does not detect the wire harness 3, only one of the microphone 21 and the pre-processing unit 22 can be set to a power-saving state, while the other is set to a normal state, thereby preventing sound collection.

[0042] When the operation state of the determination unit 23 is configurable, the determination unit 23 can also be set to a power-saving state when the optical sensor 25 does not detect the wire harness 3.

[0043] When the optical sensor 25 detects the wire harness 3, the controller 26 switches the function block that was in power-saving mode when the optical sensor 25 did not detect the wire harness 2 to normal mode. In other words, when the optical sensor 25 detects the wire harness 3, the controller 26 sets the microphone 21, the pre-processing unit 22, and the decision unit 23 to normal mode.

[0044] Therefore, during a determination period from the time the interlacing tone detection device 100 holds the wire harness 3 until the wire harness 3 is no longer held, the microphone 21 and the preprocessing unit 22 collect sound and output sound data. Then, the determination unit 23 determines whether the sound represented by the sound data input from the preprocessing unit 22 contains an interlacing tone and notifies the controller 26 of the determination result. In other words, the determination unit 23 notifies the controller 26 of the detection result of whether an interlacing tone was detected during a determination period.

[0045] The controller 26 presents the determination result of the determination unit 23 to the operator via the UI 27. A determination result indicating that a mixed tone was detected indicates that the mixing operation is good, and a determination result indicating that a mixed tone was not detected indicates that the mixing operation is not good.

[0046] If no chirp is detected during the decision-making process, the controller 26 can be configured to notify the manufacturing system 600 of the error via the communication unit 24. The manufacturing system 600 can then initiate necessary procedures, such as stopping the production line, in response to the notified error. In this case, the controller 26 sets the communication unit 24 to a normal state at least during the execution of the decision-making process.

[0047] The operation of the mating tone detection device 100 when set to the learning model will now be described. The only difference from the operating mode is that the communication unit 24 transmits sound data to the learning server 400 via LAN 200 and Internet 300. Therefore, in the learning model, the controller 26 sets the communication unit 24 to the normal state when the mating tone detection device 100 holds the wiring harness 3. When performing mating operations, in addition to the judgment result of the learning model, the operator individually confirms whether the mating operation is good, and the result confirmed by the operator is transmitted to the learning server 400 via UI 27. For example, the operator can check whether the mating components are locked together by the locking mechanism to check whether the mating operation has been performed correctly. The result confirmed by the operator becomes the correct answer label for the sound data transmitted immediately afterward. In the learning model, the controller 26 can set the judgment unit 23 to a power-saving state because the judgment result of the judgment unit 23 is not needed in the learning model.

[0048] According to this embodiment, the mating tone detection device 100 has a head 1 for the operator to hold the mating member 31 of the wire harness 3. When the operator holds the mating member 31 directly with his / her hand, the size, shape, and position of the mating member 31 in the product may prevent the operator from holding the mating member 31 stably, making it difficult to perform the mating operation correctly. On the other hand, in the mating tone detection device 100 according to this embodiment, the operator holds the mating member 31 via the head 1, thereby enabling the operator to hold the mating member 31 stably, thus improving the workability of the mating operation compared to when the operator holds the mating member 31 directly with his / her hand to perform the mating operation.

[0049] Furthermore, in this embodiment, the microphone 21 in the main body 2 of the fitting tone detection device 100 collects sound during the fitting operation. By using the fitting tone detection device 100 according to this embodiment, sound can be collected closer to the fitting member 31 than in conventional watch-type devices, thus, the accuracy of determining the fitting tone by the learning model can be higher.

[0050] Furthermore, the components of the interlocking tone detection device 100 are controlled such that sound is collected when the operator holds the wire harness 3 with their head 1, i.e., when the operator is performing an interlocking operation, and sound is not collected when the operator does not hold the wire harness 3 with their head 1, i.e., when the operator is not performing an interlocking operation. With this configuration, the power consumption of the interlocking tone detection device 100 can be reduced.

[0051] Second Embodiment The focus will then be on describing the differences between the second embodiment and the first embodiment. In the first embodiment, a configuration is described in which a learning model is generated for each location (factory) or group of locations using the inflection sound detection device 100. This is because the background noise of each factory may be different. In this embodiment, noise information representing the background noise of the location (factory) using the inflection sound detection device 100 is stored in a database 500. The noise information may be digital data generated by collecting the background noise of each location (factory) or group of locations using the inflection sound detection device 100.

[0052] The controller 26 obtains noise information of the location where the sound blending detection device 100 is used from the database 500 via the communication unit 24 and stores it in the preprocessing unit 22. The preprocessing unit 22 outputs sound data representing the sound after subtracting the background noise represented by the noise information from the sound represented by the sound signal. Therefore, the determination unit 23 receives the sound data after subtracting the background noise, and thus can detect sound blending with high accuracy.

[0053] Third Embodiment Next, the focus will be on explaining the differences between the third embodiment and the second embodiment. In the second embodiment, noise information is pre-generated and stored in the database 500. In this embodiment, the noise information is generated and updated by the chirp detection device 100.

[0054] The controller 26 initiates a determination process when the optical sensor 25 detects the wire harness 3, and stops the determination process when the optical sensor 24 no longer detects the wire harness 3. In this embodiment, when the controller 26 stops the determination process, a background noise measurement process begins. The period for performing the measurement process can be preset in the controller 26. In this case, the controller 26 terminates the measurement process after a predetermined time period has elapsed since the start of the measurement process. Alternatively, the measurement process can be performed until the optical sensor 25 detects the wire harness 3. In this case, when the determination process ends, the controller 26 starts the measurement process, and the measurement process continues until the start of the next determination process is triggered. When neither the determination process nor the measurement process is performed, the controller 26 sets at least one of the microphone 21 and the pre-processing unit 22 to a power-saving state.

[0055] During the measurement process, the controller 26 sets the microphone 21 and the preprocessing unit 22 to a normal state. The preprocessing unit 22 then determines the background noise based on the sound signal input from the microphone 21 during the measurement process and generates noise information. Alternatively, the preprocessing unit 22 updates the noise information already generated based on the sound signal input from the microphone 21 during the measurement process.

[0056] As mentioned above, measuring background noise during chimera operations can reduce the influence of background noise that may vary depending on the date and time, thus allowing for high-precision determination of chimeras.

[0057] Other embodiments exist Figure 1 In the system configuration shown, learning server 400 and database 500 are connected to the Internet 300, but learning server 400 and database 500 can also be connected to LAN 200. In this case, LAN 200 can be a closed network within the factory that is not connected to the Internet 300.

[0058] In addition, the following configuration can be adopted: wherein a chirping sound detection device is provided separately for chirping operations only, that is, a chirping sound detection device that operates only in the above-mentioned operating mode, and a chirping sound detection device is provided separately for collecting learning data only, that is, a chirping sound detection device that operates only in the above-mentioned learning model.

[0059] The chord detection device 100 described above has a communication unit 24 and acquires learning models, noise information, etc., via a network. However, the chord detection device 100 can be configured to store the learning models, noise information, etc., in the chord detection device 100 via a Universal Serial Bus (USB) storage device or the like, and the communication unit 24 can be omitted. Furthermore, the communication unit 24 can be a wired communication interface such as a USB interface.

[0060] In the above embodiment, whether the head 1 is holding the fitting member 31 of the wire harness 3 is detected by the optical sensor 25. However, it can also be configured to use a different sensor than the optical sensor to detect whether the head 1 is holding the fitting member 31 of the wire harness 3. For example, a marker that changes its location depending on whether the head 1 is holding the fitting member 31 of the wire harness 3 can be installed in the head 1, and a mechanical sensor can be used to detect whether the head 1 is holding the wire harness 3 by the position of the marker. In addition, an ultrasonic sensor with a frequency that does not interfere with the sound used to detect whether the head 1 is holding the fitting member 31 of the wire harness 3 can also be used.

[0061] While the invention has been described with reference to exemplary embodiments, it should be understood that the invention is not limited to the disclosed exemplary embodiments. The scope of the appended claims should be given the broadest interpretation, thereby covering all modifications, equivalents, and functions.

Claims

1. A device for detecting interlocking tones, wherein, The incoherence detection device includes: Microphone (21); A retaining member (1) is used to retain a fitting member disposed at the end of the wire harness; Sensor (25) detects the wire harness held by the retaining member (1) of the fitting member; and The controller (26) controls the execution of a determination process that determines whether the sound collected by the microphone (21) contains a mating tone based on the mating member, according to the sound data collected by the microphone (21). The controller (26) performs the determination process during the period when the sensor (25) detects the wire harness, and does not perform the determination process during the period when the sensor (25) does not detect the wire harness.

2. The device for detecting interlocking tones according to claim 1, wherein, The controller (26) does not collect sound based on the microphone (21) during the period when the sensor (25) does not detect the harness, and therefore does not perform the determination process.

3. The device for detecting interlocking tones according to claim 1, wherein, The retaining member (1) is configured to be detachable relative to the housing that houses the microphone (21), the sensor (25) and the controller (26).

4. The device for detecting interlocking sounds according to claim 3, wherein, The sensor (25) detects the wire harness by the situation that light is emitted through a hole provided in the holding member (1).

5. The device for detecting interlocking tones according to claim 1, wherein, The sound detection device further includes a preprocessing unit (22) which stores noise information representing background noise and outputs sound data representing the sound collected from the microphone (21) after subtracting the background noise from the sound collected from the microphone (21) based on the sound collected from the microphone (21) and the noise information.

6. The device for detecting interlocking tones according to claim 5, wherein, The controller (26) controls the operating state of at least one of the microphone (21) and the preprocessing unit (22) so that the power consumption of at least one of the microphone (21) and the preprocessing unit (22) when the determination process is not performed is less than the power consumption when the determination process is performed.

7. The device for detecting interlocking tones according to claim 5, wherein, In response to the sensor (25) no longer detecting the harness, the controller (26) begins a measurement process to determine the background noise based on the sound collected by the microphone (21) and to generate or update the noise information.

8. The device for detecting interlocking sounds according to claim 7, wherein, The controller (26) controls the operating state of at least one of the microphone (21) and the preprocessing unit (22) so that the power consumption of at least one of the microphone (21) and the preprocessing unit (22) when neither the determination process nor the measurement process is executed is less than the power consumption when the determination process is executed.

9. The device for detecting interlocking tones according to any one of claims 1 to 8, wherein, The incoherent sound detection device also includes a determination mechanism (23) that uses a learning model to determine whether the sound data represented by the sound data from the start of the determination process to the end of the determination process contains the incoherent sound.

10. The device for detecting interlocking sounds according to claim 9, wherein, The chirp detection device also includes a communication mechanism that communicates with a server device storing one or more learning models to obtain the learning models.

11. The device for detecting interlocking sounds according to claim 9, wherein, The learning model is associated with the location where the chiaroscuro detection device is used.

12. The device for detecting interlocking sounds according to claim 9, wherein, The learning model is associated with the type of the interlocking component.

13. The device for detecting interlocking tones according to any one of claims 1 to 8, wherein, The sound detection device also includes a communication mechanism that, when operating in a first mode, transmits the sound data during the period when the sensor (25) detects the wire harness to the network, and when operating in a second mode, does not transmit the sound data during the period when the sensor (25) detects the wire harness to the network.

14. The device for detecting interlocking tones according to any one of claims 1 to 8, wherein, The incoherent sound detection device also includes a communication mechanism, which, during the determination process, if the determination result is obtained that the incoherent sound is not included in the sound collected by the microphone (21), wirelessly notifies a predetermined device of an error.

Images