Surface mounting machine, component supply unit and inspection method

Abstract

To suppress the occurrence of such a problem that a mounting head interferes with a tray or causes erroneous adsorption when an electronic component is taken out from the tray, by detecting abnormality of a surface state of the tray or arrangement abnormality of the electronic component housed in the tray.SOLUTION: A TOF camera 100 irradiates a tray 95 on a pallet 90 pulled out from a storage 81 with light, and acquires a three-dimensional image indicating a surface height of the tray 95 pulled out from the storage 81 and the surface height of a tray component housed in the tray 95 based on reflected light. An inspection device detects at least one of abnormality in a surface state of the tray 95 pulled out from the storage 81 and arrangement abnormality of the tray component with respect to the tray 95 based on the three-dimensional image acquired by the TOF camera 100.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a tray feeder that uses trays to supply electronic components. [Background technology]

[0002] A surface mounter mounts electronic components supplied by a component supplying device onto a board using a mounting head. One such component supplying device is a tray-type component supplying device (tray feeder) that supplies electronic components using a tray. Patent Document 1 is an example of a publication that discloses technology related to a tray-type component supplying device.

[0003] Patent Document 1 discloses that "a tray discharge system in which an empty tray on a pallet pulled out from a tray-type component supply device is sucked up from above by a suction pad and discharged, the system being configured to include a camera which images the tray from above, and a control device which processes the image of the tray captured by the camera to recognize a portion of the tray that can be sucked up by the suction pad and measures the position of the capable portion, and which controls the operation of sucking up the capable portion of the tray with the suction pad and discharging it based on measurement data of the measured position of the capable portion." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2018-107205 A Summary of the Invention [Problem to be solved by the invention]

[0005] If the tray is warped or deformed, the mounting head may interfere with the tray when removing electronic components from the tray.Also, if there is an abnormality in the placement of electronic components on the tray, a pickup error may occur.

[0006] The present invention was developed based on the above-mentioned circumstances, and its objective is to prevent the mounting head from interfering with the tray or causing pickup errors when removing electronic components from the tray by detecting abnormalities in the surface condition of the tray or abnormalities in the arrangement of electronic components contained in the tray. [Means for solving the problem]

[0007] (1) A surface mounting machine including a tray feeder, a head unit that mounts electronic components supplied from the tray feeder onto a substrate, a TOF camera, and an inspection device, wherein the tray feeder includes a storage unit that stores trays containing electronic components, and a pallet that can be pulled out from the storage unit and on which the trays are placed. The TOF camera shines light onto the tray on the pallet pulled out of the storage cabinet, and based on the reflected light, obtains a three-dimensional image showing the surface height of the tray pulled out of the storage cabinet and the surface heights of the electronic components contained in the tray. The inspection device detects at least one of an abnormality in the surface condition of the tray pulled out of the storage facility and an abnormality in the placement of electronic components on the tray based on the three-dimensional image acquired by the TOF camera.

[0008] A TOF camera is a camera that measures the surface height of an object based on the travel distance of light, and TOF is an abbreviation for "Time of flight." A three-dimensional image of the surface height is an image (depth image) that visualizes the distribution of the surface height of an object.

[0009] (2) In the surface mounter described in (1), the inspection device may detect an abnormality in the surface condition of the tray by recognizing a displacement location of the surface height of the tray based on the three-dimensional image. The displacement of the surface height of the tray may be detected by comparing it with the surface height of the tray in a normal state or with the surface height of the surroundings.

[0010] (3) In the surface mounter described in (1), the inspection device may detect an abnormal arrangement of the electronic components on the tray by recognizing a displacement location of the surface height of the electronic components accommodated in the tray based on the three-dimensional image. The displacement of the surface height of the electronic components can be detected, for example, by comparing with the surface height of the electronic components under normal conditions.

[0011] (4) In the surface mounter described in (1), the inspection device includes a model generation unit and a judgment unit, and the model generation unit may be a supervised learning model using a three-dimensional image showing a normal surface height of the tray as teacher data, and may generate a first learning model by machine learning to estimate an abnormality in the surface condition of the tray using the three-dimensional image showing the surface height of the tray as an input, and the judgment unit may judge the abnormality in the surface condition of the tray from the first learning model. In this configuration, by inspecting the surface condition of the tray using a learning model by machine learning, it is expected that the abnormality in the surface condition of the tray can be accurately judged even in cases where it is difficult to judge only the threshold value, such as when the surface condition changes gradually.

[0012] (5) In the surface mount machine described in (1), the inspection device includes a model generation unit and a judgment unit, and the model generation unit is a supervised learning model using a three-dimensional image showing the normal arrangement of the electronic components on the tray as training data, and generates a second learning model by machine learning using the three-dimensional image showing the surface height of the electronic components contained in the tray as input, which estimates an abnormal arrangement of the electronic components on the tray, and the judgment unit may judge the abnormal arrangement of the electronic components on the tray from the second learning model.

[0013] (6) In the surface mounter described in (1), the inspection device may obtain, by the TOF camera, a three-dimensional image indicating the surface height of the tray drawn out from the storage and the surface heights of the electronic components contained in the tray after the tray is drawn out from the storage and before the head unit takes out the electronic components from the tray, and detect at least one of an abnormality in the surface condition of the tray and an abnormal arrangement of the electronic components contained in the tray based on the obtained three-dimensional image. With this configuration, it is possible to prevent the mounting head from interfering with the tray or causing a pickup error when taking out the first electronic component from the tray drawn out from the storage.

[0014] (7) In the surface mounter described in (1), the inspection device may acquire, by the TOF camera, a three-dimensional image showing the surface height of the tray and the surface heights of the electronic components contained in the tray during a period from when an electronic component is removed from the tray until the next electronic component is removed from the tray, and detect at least one of an abnormality in the surface condition of the tray and an abnormal arrangement of the electronic components contained in the tray based on the acquired three-dimensional image. With this configuration, it is possible to prevent the mounting head from interfering with the tray or causing a pickup error when removing the second or subsequent electronic components from the tray pulled out from the storage.

[0015] (8) In the surface mount machine described in (2), the inspection device may recognize a displacement location of the surface height of the tray by comparing a 3D image showing the surface height of the tray pulled out of the storage with a 3D image showing the normal surface height of the tray. By using the normal 3D image as a comparison target, it is possible to accurately recognize a change location of the surface height of the tray.

[0016] (9) In the surface mounter described in (1), the TOF camera may be attached to the head unit and may be movable together with the head unit over the tray pulled out of the storage unit. Even if the field of view of the TOF camera is narrow compared to the size of the tray, a 3D image of the entire tray can be obtained by moving over the tray and taking images multiple times.

[0017] (10) In the surface mounting machine described in (1), the TOF camera may be fixed immovably to an upper portion of the storage cabinet.

[0018] (11) The surface mounter according to (1) may include a display unit, and the inspection device may display at least one of an abnormality in the surface condition of the tray and an abnormality in the arrangement of electronic components on the display unit. Since the content of the error (abnormality) can be understood from the display, it becomes possible to take measures according to the error.

[0019] (12) In the surface mounter described in (1), the display unit may display a location of an abnormal placement of electronic components. From the display, it is easy to know in which part of the tray the abnormal placement has occurred.

[0020] (13) A component supply unit, comprising a tray feeder and a TOF camera, the tray feeder comprising a storage unit for storing a tray containing electronic components, and a pallet that can be pulled out of the storage unit and on which the tray is placed, the TOF camera being immovably fixed on the upper part of the storage unit, irradiating light onto the tray on the pallet pulled out of the storage unit, and acquiring a three-dimensional image showing the surface height of the tray and the surface height of the electronic components accommodated in the tray based on the reflected light. With this configuration, a three-dimensional image showing the surface height of the tray and the surface height of the electronic components accommodated in the tray can be acquired. By analyzing the acquired three-dimensional image with a controller of a surface mounter or the like, it is possible to detect an abnormality in the surface condition of the tray and an abnormal arrangement of the electronic components accommodated in the tray. Effect of the Invention

[0021] According to the present invention, it is possible to detect abnormalities in the surface condition of a tray and abnormal arrangement of electronic components contained in the tray from a 3D image of the tray acquired by a TOF camera. Compared to a stereo camera that acquires distance information from the parallax of two cameras with different shooting angles, a TOF camera is small and has a high degree of freedom in placement, which allows for flexible design. [Brief description of the drawings]

[0022] [Figure 1] Plan view of surface mounter [Diagram 2] Front view of X-beam and head unit [Diagram 3] Block diagram showing the electrical configuration of a surface mounter [Figure 4] Perspective view of the tray feeder [Diagram 5] Tray Feeder Side View [Figure 6] Tray top view [Figure 7] Cross-section of the tray [Figure 8] 3D image of the tray [Figure 9] 3D image of the tray [Figure 10] Cross-section of the tray [Figure 11] 3D image of the tray [Figure 12] 3D image of the tray [Figure 13] 3D image of the tray [Figure 14] Flowchart of tray status confirmation process [Figure 15] Implementation control flow chart [Figure 16] Retry operation flowchart [Figure 17] Block diagram of inspection equipment using machine learning [Figure 18] Diagram showing the learning model [Figure 19] Block diagram showing the electrical configuration of a surface mounter [Figure 20] Front view of X-beam and head unit [Figure 21] Cross-section of the tray [Figure 22] Surface mounter block diagram DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] <Embodiment 1> 1. Overall configuration of surface mounter 1A As shown in FIG. 1, the surface mounter 1A includes a base 11, a transport conveyor 20, a driving device 30, a head unit 60, a tape feeder 70, a tray feeder 80, and a TOF camera 100.

[0024] In the following description, the longitudinal direction of base 11 (left-right direction in Figure 1) will be referred to as the X-direction, the depth direction of base 11 (up-down direction in Figure 1) will be referred to as the Y-direction, and the up-down direction in Figure 2 (the ascending and descending direction of mounting head 63) will be referred to as the Z-direction.

[0025] The transfer conveyor 20 is disposed in the center of the base 11. The transfer conveyor 20 includes a pair of transfer belts 21 that are driven to circulate in the X direction, and transfers the printed circuit board P on the transfer belts 21 in the X direction.

[0026] In this embodiment, the left side shown in Fig. 1 is the entrance, and the printed circuit board P is carried into the machine from the left side shown in Fig. 1 via the transport conveyor 20. The carried-in printed circuit board P is carried by the transport conveyor 20 to a work position in the center of the base.

[0027] Feeder installation sections 13A to 13D are provided on the base at four locations around the working position of the printed circuit board P. Feeder installation sections 13A, 13C, and 13D are for tape components, and have tape feeders 70 installed thereon.

[0028] The tape feeder 70 is a device that supplies tape components E1 by feeding a component supply tape (not shown) to a supply position. The tape components are, for example, small electronic components such as chip capacitors.

[0029] The feeder installation section 13B is for tray parts, and has a tray feeder 80 installed therein. The tray feeder 80 is a device that supplies the tray parts E2 by drawing out a tray 95 that contains the tray parts E2 from a storage cabinet 81. The tray parts E2 are, for example, medium- to large-sized electronic parts such as ICs.

[0030] As shown in FIGS. 1 and 2, the driving device 30 is composed of a pair of Y beams 41, an X beam 51, a Y-axis ball screw 45, a Y-axis motor 47, an X-axis ball screw 55, and an X-axis motor 57.

[0031] 1, the pair of Y beams 41 are located on both sides in the X direction of the base 11. The pair of Y beams 41 extend parallel to the Y direction, and support the X beam 51 slidably in the Y direction.

[0032] A Y-axis ball screw 45 extending in the Y direction and a Y-axis motor 47 at its shaft end are disposed on the upper surface of the Y beam 41. The Y-axis ball screw 45 is a drive shaft in the Y-axis direction, and a ball nut (not shown) fixed to the lower surface of the X beam 51 is screwed into the Y-axis ball screw 45.

[0033] By driving the Y-axis motor 47 to rotate the Y-axis ball screw 45 about its axis, the X-beam 51 and the head unit 60 can be moved in the Y direction (Y-axis servo mechanism).

[0034] The X beam 51 is elongated in the X direction, and supports the head unit 60 so that it can slide along the X direction. An X-axis ball screw 55 extending in the X direction and an X-axis motor 57 are arranged on the side of the X beam 51. The X-axis ball screw 55 is a drive shaft in the X-axis direction, and a ball nut (not shown) fixed to the head unit 60 is screwed into the X-axis ball screw 55.

[0035] By driving the X-axis motor 57 to rotate the X-axis ball screw 55 about its axis, the head unit 60 can be moved in the X direction relative to the X-beam 51 (X-axis servo mechanism).

[0036] From the above, by controlling the X-axis motor 57 and the Y-axis motor 47 in a composite manner, it is possible to move the head unit 60 to any position on the base 11.

[0037] The head unit 60 has a plurality of mounting heads 63. The mounting heads 63 are arranged in a row in the X direction relative to the head unit 60. Each mounting head 63 can move in the vertical direction by being driven by a Z-axis motor 67 (Z-axis servo mechanism).

[0038] The mounting head 63 is a shaft that is long in the Z direction, and at its tip (lower end) is provided with a suction nozzle 68. By switching the internal pressure of the suction nozzle 68 between negative and positive pressure, the tape component E1 and the tray component E2 can be held or released.

[0039] With this configuration, by operating the X-axis motor 57, the Y-axis motor 47, and the Z-axis motor 67 at a predetermined timing, the tape components E1 supplied by the feeder 70 and the tray components E2 supplied by the tray feeder 80 can be picked up by the mounting head 63 and mounted on the printed circuit board P.

[0040] 1 denotes a component recognition camera. The component recognition camera 17 is fixed on the base 11 with its imaging surface facing upward, and captures an image (bottom surface image) of the tape component E1 or the tray component E2 picked up by the mounting head 63.

[0041] 2. Electrical configuration of surface mounter 1A 3, the surface mounter 1A has a controller 150. The controller 150 includes a CPU 151, which is a calculation unit, and a memory 153.

[0042] The controller 150 is connected to the conveyor 20, the motor control unit 35, the tape feeder 70, the tray feeder 80, the component recognition camera 17, the TOF camera 100, the display unit 160, the operation panel 170, etc. via a bus BS. The motor control unit 35 is connected to the X-axis motor 57, the Y-axis motor 47, and the Z-axis motor 67.

[0043] The controller 150 is a main control unit of the surface mounter 1, and typically controls the transportation of the printed circuit board P and the mounting of electronic components by the head unit 60. As will be described in detail below, the controller 150 also inspects the surface condition of the tray 95 and the arrangement of the tray components E2 based on a three-dimensional image G of the tray 95 acquired by the TOF camera 100. The controller 150 is an example of the "inspection device" of the present invention.

[0044] 3. Tray Feeder 80 and TOF Camera 100 Fig. 4 is a perspective view of tray feeder 80, and Fig. 5 is a side view of tray feeder 80. Tray feeder 80 is a device that supplies tray parts E2 using trays 95, and includes a storage unit 81 having a drawer opening 82 on the front side, a magazine 83, a pallet base 85, and a drawer device 87.

[0045] The magazine 83 has pallets 90 with trays 95 placed thereon arranged in a stacked manner, and is supported so as to be vertically movable relative to the storage cabinet 81. The pallet base 85 is located at the front of the storage cabinet, below the drawer opening 82. The drawer device 87 is a device that draws the pallets 90 from the storage cabinet 81 onto the pallet base 85. The drawer device 87 can also return the drawn pallets 90 to the storage cabinet 81.

[0046] The tray feeder 80 can supply tray parts E2 by raising and lowering the magazine 83 to move the pallet 90 carrying the tray 95 to the height of the pull-out opening 82, and then moving the pallet 90 carrying the tray 95 from the storage unit 81 onto the pallet base 85.

[0047] 4 and 5, the TOF camera 100 is disposed (immovably fixed) in the upper part of the storage 81 with its imaging surface facing downward. The TOF camera 100 includes a light projecting unit 101 and a light receiving unit 103. The light projecting unit 101 emits light L toward the tray 95, and the light receiving unit 103 receives the reflected light L. The light L is, for example, infrared light (infrared light).

[0048] The TOF camera 100 is a camera that measures the height of an object based on the travel distance of light L, and captures a three-dimensional image G of the tray 95 pulled out of the storage unit 81. The field of view of the TOF camera 100 is wider than the tray 95, and a three-dimensional image G of the entire tray can be captured in one shot.

[0049] The three-dimensional image G of the tray 95 acquired by the TOF camera 100 is an image (depth image) that visualizes the distribution of the surface height of the tray 95 and the surface height of the tray part E2, and includes information on the surface height of the tray 95 and information on the surface height of the tray part E2.

[0050] Fig. 6 is a plan view of the tray 95, and Fig. 7 is a cross-sectional view of the tray 95. As shown in Fig. 6, the tray 95 is generally rectangular with its long side in the X direction and its short side in the Y direction. The tray 95 has a plurality of component storage sections 96 arranged in a matrix. In the example of Fig. 6, the component storage sections 96 have 6 columns in the X direction and 4 rows in the Y direction, for a total of 24 rooms, and the tray 95 can store 24 tray parts E2.

[0051] FIG. 6 shows a state in which the left two rows of the tray 95 contain tray parts E2, and the right four rows contain no parts.

[0052] As shown in FIG. 7, the tray 95 has a recessed component receiving portion 96, and a bottom surface S3 of the component receiving portion 96 is lower than an upper surface S1 of the tray in the height direction (Z direction).

[0053] Figures 8 and 9 are three-dimensional images of the tray 95 captured by the TOF camera 100. Figure 8 is a three-dimensional image G1 when no components are stored, and Figure 9 is a three-dimensional image G2 when a component is stored. The three-dimensional images G1 and G2 are images that represent the distribution of surface heights of the tray 95 and the distribution of surface heights of tray components E2 stored in the tray 95 by changing the display color (gradation).

[0054] As shown in Figure 7, with the tray bottom surface S0 as the reference, the surface height increases in the order of the component accommodating section bottom surface S3, the component top surface S2, and the tray top surface S1. Therefore, in the images of Figures 8 and 9, the display color changes from lighter to lighter in the order of the component accommodating section bottom surface S3, the component top surface S2, and the tray top surface S1.

[0055] Fig. 10 is a cross-sectional view of tray 95 whose upper surface has been curved, and Fig. 11 is a three-dimensional image G3 of tray 95 whose upper surface has been curved. In three-dimensional image G3 (Fig. 11) of tray 95 whose upper surface has been curved, the curved portion (part A) is displayed in a different color than in the three-dimensional image of normal tray 95 (Fig. 9).

[0056] 4. Check the status of Tray 95 If an abnormality such as distortion or deformation occurs on the surface of the tray 95, there is a possibility that the mounting head 63 will interfere with the tray 95 when removing the tray component E2 from the tray 95.

[0057] A three-dimensional image G of the tray 95 pulled out of the storage cabinet 81 is acquired by the TOF camera 100, and the surface condition of the tray 95 is confirmed, whereby abnormalities in the surface height of the tray 95 can be inspected.

[0058] Furthermore, whether or not there is an abnormality in the surface height can be determined by recognizing the displacement points in the surface height of the tray 95 pulled out of the storage cabinet 81 based on the three-dimensional image G of the tray 95.

[0059] For example, by comparing a three-dimensional image acquired by the TOF camera 100 (three-dimensional image G3 shown in FIG. 11) with a three-dimensional image of a tray at a normal surface height (three-dimensional image G2 shown in FIG. 9), it is possible to recognize the location of a displacement in surface height. In this embodiment, when there is a height change of a threshold or more compared to the normal image, it is recognized as a location of a displacement in surface height.

[0060] Furthermore, if a three-dimensional image of the tray 95 at the normal surface height (three-dimensional image G2 shown in Figure 9) is captured in advance by the TOF camera 100 and the data is stored in the memory 153, it can be read out at the time of judgment and used as a comparison target.

[0061] If an abnormality in the surface height of the tray 95 is detected, the supply of the tray part E2 is stopped and the suction operation of the head unit 60 for the tray part E2 is stopped, thereby preventing the mounting head 63 from interfering with the deformed tray 95.

[0062] At this time, an error message may be displayed on the display unit 160, and a three-dimensional image G3 of the tray 95 that caused the error may also be displayed. In addition, it is preferable to specify the deformed part by, for example, encircling the deformed part in the three-dimensional image G3 of the tray that caused the error, as shown in FIG.

[0063] By identifying the deformed portion, the operator can easily recognize which portion of the tray 95 is deformed (abnormal) from the displayed three-dimensional image G3 of the tray 95.

[0064] An operator who recognizes the abnormality in the tray 95 can take measures such as replacing the tray 95 with a normal one.

[0065] 5. Check the placement of tray part E2 Furthermore, the mounting head 63 moves to above the tray component E2 based on the position information of the tray component E2 on the tray, and picks up the tray component E2.

[0066] If the tray component E2 is not positioned on the tray 95 according to the position information, for example, the mounting head 63 may perform a pickup operation on a location where the tray component E2 is not present, which may cause a pickup error.

[0067] FIG. 12 is a three-dimensional image G4 of the tray 95 with the tray part E2 normally arranged, and FIG. 13 is a three-dimensional image G5 of the tray 95 when a part of the tray part E2 is missing.

[0068] The three-dimensional image G of the tray 95 acquired by the TOF camera 100 includes information on the surface height of the tray 95 as well as information on the surface height of the tray part E2 stored in the part storage section 96, as shown in Figures 12 and 13.

[0069] Therefore, based on the information on the surface height of the tray part E2 included in the three-dimensional image G of the tray 95, it is possible to detect an abnormal arrangement of the tray part E2 with respect to the tray 95.

[0070] Whether or not there is a placement abnormality can be determined by recognizing the displacement location of the surface height of the tray part E2 based on the three-dimensional image G of the tray 95.

[0071] For example, by comparing the 3D image to be inspected (3D image G5 shown in FIG. 13) with a 3D image showing the surface height of a properly placed tray part E2 (3D image G4 shown in FIG. 12), it is possible to recognize the location of displacement of the surface height of the tray part E2. In this embodiment, when there is a height change of a threshold or more compared to the normal image, it is recognized as a location of displacement of the surface height of the tray part E2.

[0072] A three-dimensional image (three-dimensional image G4 shown in FIG. 12) showing the surface height of a properly placed tray part E2 can be captured in advance by the TOF camera 100 and the data stored in the memory 153, so that it can be read out at the time of judgment and used as a comparison target.

[0073] When there is an abnormality in the arrangement of the tray component E2, the head unit 60 stops the suction operation of the tray component E2, thereby preventing the mounting head 63 from making an error in suction of the tray component E2.

[0074] At this time, a three-dimensional image G5 of the tray 95 causing the error may be displayed together with an error message on the display unit 160. In addition, it is preferable to specify the location of the positioning abnormality by, for example, encircling the location of the positioning abnormality with a circle, as shown in FIG.

[0075] By identifying the location of the placement error, the operator can easily recognize from the displayed three-dimensional image G5 of the tray 95 in which part of the tray 95 the placement error of the tray part E2 has occurred.

[0076] Furthermore, the position information of the tray components E2 may be manually input to the surface mounter 1A via the operation panel 170. By comparing the manually input data of the position information of the tray components E2 with the inspection results of the component arrangement based on the three-dimensional image G, it is possible to perform an error check on the manually input information (the position information of the tray components E2).

[0077] 6. Control Flow 14 is a flowchart of the tray status confirmation process. The tray status confirmation process is made up of three steps, S10 to S30.

[0078] The tray condition confirmation process is a process for checking whether the surface condition of a tray 95 pulled out of the storage facility 81 is good or bad before the suction operation of tray part E2 is performed (before the first tray part is sucked up), and is performed when the first tray 95 is pulled out of the storage facility 81, or after the tray part E2 is supplied, when the used tray 95 is collected and a new tray 95 is pulled out of the storage facility 81.

[0079] When the tray 95 is pulled out from the storage 81, the controller 150 captures an image of the tray 95 using the TOF camera 100 and obtains a three-dimensional image G of the tray 95 in S10.

[0080] The three-dimensional image G of the tray 95 corresponds to the "three-dimensional image showing the surface height of the tray and the surface heights of the electronic components accommodated in the tray" of the present invention.

[0081] Thereafter, the process proceeds to S20, where the controller 150 acquires information on the surface height of each part of the tray from the three-dimensional image G of the tray 95, and determines whether or not there is an abnormality in the surface height of the tray 95 pulled out of the storage 81 that may cause interference with the mounting head 63. At this time, in addition to determining whether there is an abnormality in the surface height of the tray 95, an abnormality in the arrangement of the tray part E2 may also be determined.

[0082] If an abnormality in the surface height exists (S20: YES), the process proceeds to S30. When the process proceeds to S30, the controller 150 displays an error message on the display unit 160 and stops the operation of the surface mounter 1A including the head unit 60.

[0083] If no abnormality in the surface height is present (S20: NO), the process proceeds to mounting control shown in Fig. 15. The mounting control is made up of seven steps from S100 to S160. When the mounting control starts, the process proceeds to S100.

[0084] When the process proceeds to S100, the controller 150 moves the head unit 60 above the tray that has been pulled out from the storage unit 81.

[0085] When the head unit 60 moves to above the tray, the process proceeds to S110, and the controller 150 causes the mounting head 63 to pick up the tray component E2 and remove it from the tray 95.

[0086] Thereafter, the process proceeds to S120, where the controller 150 determines whether or not the mounting head 63 has successfully picked up the tray component E2.

[0087] If the component suction has been successful (S120: YES), the process proceeds to S130. In S130, the controller 150 moves the head unit 60 to above the printed circuit board P, and mounts the tape component E1 and tray component E2 that have been sucked and held on the printed circuit board P.

[0088] Furthermore, in parallel with the component mounting in S130, S140 is executed, and the controller 150 captures an image of the tray 95 after the components have been removed using the TOF camera 100, and obtains a three-dimensional image G of the tray 95.

[0089] Then, the process proceeds to S150, where the controller 150 acquires information on the surface height of each part of the tray from the three-dimensional image G of the tray 95 acquired in S140, and determines whether there is any abnormality in the surface height of the tray 95 after the components have been removed that may cause interference with the mounting head 63.

[0090] The reason for checking whether the surface height of the tray 95 after the component is removed is that the mounting head 63 may interfere with the tray 95 when removing the component, causing the tray 95 to deform or the adsorbed component E2 to drop onto the tray.

[0091] If there is no abnormality in the surface height (S150: NO), the process returns to S100 and executes a process of moving the head unit 60 onto the tray 95 to pick up the next tray component E2 after waiting for the mounting of the components E1 and E2 taken out from the tape feeder 70 and the tray feeder 80 to be completed.

[0092] On the other hand, if there is an abnormality in the surface height of the tray 95 (S150: YES), the process proceeds to S160. When the process proceeds to S160, the controller 150 displays an error message on the display unit 160 and stops the operation of the surface mounter 1A including the head unit 60.

[0093] Moreover, if the mounting head 63 fails to pick up a component (S120: NO), a retry operation is executed. The retry operation is made up of five steps from S200 to S240, as shown in FIG.

[0094] When the retry operation starts, the process proceeds to S200. When the process proceeds to S200, the controller 150 raises the mounting head 63 to a safe height and moves it away from above the tray at a low speed. Then, the process proceeds to S210.

[0095] In S210, the controller 150 captures an image of the tray 95 using the TOF camera 100, and obtains a three-dimensional image G of the tray 95.

[0096] Then, the process proceeds to S220, where the controller 150 acquires information on the surface height of each part of the tray from the three-dimensional image G of the tray 95 acquired by the TOF camera 100, and determines whether there is an abnormality in the surface height of the tray 95 that may cause interference with the mounting head 63.

[0097] If an abnormality in the surface height exists (S220: YES), the process proceeds to S230. When the process proceeds to S230, the controller 150 displays an error message on the display unit 160 and stops the operation of the surface mounter 1A including the head unit 60.

[0098] On the other hand, if there is no abnormality in the surface height (S220: NO), the process proceeds to S240, where a retry of component pickup is performed.

[0099] 7.Effects According to the present invention, it is possible to detect an abnormality in the surface condition of the tray 95 or an abnormal arrangement of the tray component E2 from the three-dimensional image G of the tray 95 captured by the TOF camera 100. When an abnormality in the surface condition of the tray 95 or an abnormal arrangement of the tray component E2 is detected, a warning or the like is issued, thereby making it possible to prevent the mounting head 63 from interfering with the surface of the tray 95 or causing a pickup error when removing the tray component E2 from the tray 95.

[0100] Furthermore, the TOF camera 100 is smaller and has a higher degree of freedom in terms of layout compared to a stereo camera that obtains distance information based on the parallax between two cameras with different shooting angles, allowing for flexible design.

[0101] <Embodiment 2> The second embodiment differs from the first embodiment in that the surface condition of the tray 95 is inspected for abnormalities using a learning model M1 based on machine learning.

[0102] 17 is a block diagram of a determination device 200 that uses a learning model M1 based on machine learning. The determination device 200 is an example of the “inspection device” of the present invention. The determination device 200 is configured with an information processing unit 210 and a storage unit 220.

[0103] The information processing unit 210 is composed of a learning model generation unit 213 and a judgment execution unit 215. The storage unit 220 is composed of an image storage unit 221, a learned model storage unit 223, and a judgment result storage unit 225. The learning model generation unit 213 corresponds to the "model generation unit" of the present invention, and the judgment execution unit 215 corresponds to the "judgment unit" of the present invention.

[0104] The image storage unit 221 stores the three-dimensional image G of the tray 95 acquired by the TOF camera 100. The learning model generation unit 213 generates a learning model M1 for determining whether the tray 95 has an abnormality in its surface height.

[0105] The learning model M1 is a supervised learning model that uses a three-dimensional image showing the normal surface height of the tray 95 (for example, the three-dimensional image G1 shown in Figure 8) as training data, and is a learning model that estimates abnormalities in the surface height of the tray 95 using the three-dimensional image G showing the surface height of the tray 95 as input.

[0106] The learning model M1 generated by the learning model generation unit 213 is stored in the learned model storage unit 223.

[0107] After the learning model M1 is generated, when three-dimensional images G2, G3 of the tray 95 are captured by the TOF camera 100, the judgment execution unit 215 uses the captured three-dimensional images G2, G3 as input data and judges the surface condition of the tray 95 based on the learning model M1 stored in the learned model memory unit 223.

[0108] The result of the determination is stored in the determination result storage unit 225 and is displayed on the display unit 160.

[0109] The learning model M1 is constructed by a neural network model including an input layer, a hidden layer (intermediate layer), and an output layer, as shown in FIG.

[0110] The input layer is made up of, for example, a number of neurons corresponding to the number of pixels in the three-dimensional image G of the tray 95. The altitude value (surface height data) x of each pixel in the three-dimensional image G of the tray 95 is input to each neuron in the input layer.

[0111] The hidden layer is made up of a plurality of neurons connected to each neuron in the input layer. The hidden layer is a layer that extracts features from the three-dimensional image G of the tray 95, and in this example, there are two hidden layers.

[0112] The output layer is composed of two neurons connected to each neuron in the hidden layer, and outputs the probability of the surface condition of the tray 95 being good or bad.

[0113] The estimation accuracy of the learning model M can be improved by optimizing the connection weights w(n), v(n), and q(n) of each layer based on training data.

[0114] By determining the surface condition of the tray 95 using the machine learning learning model M1, it is expected that abnormalities in the surface condition of the tray 95 can be accurately determined even in cases where it is difficult to determine the threshold value alone, such as when there is a gradual change in the surface condition.

[0115] In the above example, an abnormality in the surface condition of the tray 95 was determined using a learning model M1 based on machine learning, but an abnormal placement of the tray part E2 relative to the tray 95 may also be determined using a learning model M2 based on machine learning.

[0116] The learning model M2 is a supervised learning model that uses a three-dimensional image showing the normal positioning of the tray part E2 (e.g., the three-dimensional image G4 shown in FIG. 12) as training data, and is a learning model that estimates abnormal positioning of the tray part E2 relative to the tray 95 using a three-dimensional image showing the surface height of the tray part E2 contained in the tray 9595 (e.g., the three-dimensional image G5 shown in FIG. 13) as input.

[0117] The learning model M2 can be constructed using a neural network or the like, similar to the learning model M1.

[0118] 19 is a block diagram of a surface mounter 1B. The surface mounter 1B disclosed in the second embodiment is obtained by adding a determination device 200 to the surface mounter 1A of the first embodiment. In this example, the determination device 200 is installed separately from the controller 150, but a configuration in which some or all of the functions of the determination device 200 are executed by the controller 150 is also possible. <Other embodiments> The present invention is not limited to the embodiments described above and illustrated in the drawings, and for example, the following embodiments are also included within the technical scope of the present invention.

[0119] (1) In the first embodiment, the TOF camera 100 is attached to the front of the storage 81. As shown in Fig. 20, the TOF camera 100 may be attached to the head unit 60. The TOF camera 100 may be movable together with the head unit 60 above the tray 95 pulled out from the storage 81. Even with a TOF camera 100 having a narrow field of view, a three-dimensional image G of the entire tray can be acquired by moving above the tray and taking images multiple times.

[0120] (2) In the first embodiment, the surface height of the tray 95 and the positional abnormality of the tray part E2 are inspected based on the three-dimensional image G of the tray 95 acquired by the TOF camera 100. It is also possible to inspect only one of the surface height of the tray 95 and the positional abnormality of the tray part E2.

[0121] (3) In embodiment 1, the displacement points of the surface height of the tray 95 were recognized by comparing the three-dimensional image of the tray 95 acquired by the TOF camera 100 (three-dimensional image G3 shown in FIG. 11) with the three-dimensional image of the tray 95 at the normal surface height (three-dimensional image G2 shown in FIG. 9). The object of comparison when recognizing the displaced portion is not limited to the 3D image G of the normal surface height, but may be, for example, shape data of the tray 95. The shape data of the tray 95 is data D1 of the thickness of the tray bottom 97 and data D2 of the height of the tray side wall 98, etc., shown in FIG.

[0122] The same applies when recognizing a displaced portion of the surface height of the tray part E2, and the comparison target is not limited to the 3D image G in the normal arrangement, but may be, for example, data on the surface height of the tray part E2 when accommodated in the tray (D1+D3 shown in FIG. 21). D1 is data on the thickness of the tray bottom 97, and D3 is data on the height of the tray part E2.

[0123] (4) In the first embodiment, the TOF camera 100 is installed on the tray feeder 80. When the TOF camera 100 is installed on the tray feeder 80, the tray feeder 80 and the TOF camera 100 can be handled as one unit. The surface mounter 1C shown in FIG. 22 has a component supply unit U including the tray feeder 80, the TOF camera 100, and a feeder controller 250. The feeder controller 250 is connected to the bus BS of the surface mounter 1C via the I / O 175, and controls the tray feeder 80 in response to a command from the controller 150. Specifically, the tray component E2 is supplied by drawing out the pallet 90 from the storage 81. The feeder controller 250 is a so-called control board, and may be built into the storage of the tray feeder 80 or may be externally attached to the tray feeder 80. The TOF camera 100 is arranged with the imaging surface facing downward in the upper part of the storage 81, as in the surface mounter 1A. The TOF camera 100 is connected to the bus BS of the surface mounter 1C, and in response to a command from the controller 150, acquires a three-dimensional image of the tray 95 and the tray components E2 accommodated in the tray 95. Data of the acquired three-dimensional image is output to the controller 150 via the bus BS, and the controller 150 of the surface mounter IC inspects for abnormalities in the surface height of the tray 95 and abnormal arrangement of the tray components E2.

[0124] Furthermore, the inspection of the surface height of the tray 95 and the arrangement of the tray components E2 may be performed by the controller 150 of the surface mounter 1C, or may be performed by the feeder controller 250 of the component supply unit U. That is, the 3D image data may be sent from the TOF camera 100 to the feeder controller 250 of the component supply unit U, and the feeder controller 250 may inspect the surface height of the tray 95 and the like. In this case, the controller 150 of the surface mounter 1C may be configured to be notified of only the judgment result via the I / O 175.

[0125] (5) In the second embodiment, a neural network having an input layer, two hidden layers, and an output layer is exemplified as the learning models M1 and M2 by machine learning. The learning model M is not limited to the example of the embodiment. If the learning model M is used to inspect anomalies in the surface height of the tray 95 or anomalies in the arrangement of the tray parts E2, it may be a neural network with a single hidden layer. Furthermore, a learning model other than a neural network may be used. [Explanation of symbols]

[0126] 1A~1C: Surface mount machine 20...Transport conveyor 60…Head unit 63...Mounting head 70…Tape feeder 80…Tray feeder 81…Storage 90…Palette 95…Tray 100…TOF camera 101...Light emitting element 103...Photo-receiving element 150...Controller (inspection device) 200...Judgment device (inspection device) 250…Feeder controller

Claims

1. A surface mount machine, A tray feeder; a head unit that mounts electronic components supplied from the tray feeder onto a substrate; A time-of-flight camera; An inspection device, The tray feeder comprises: a storage unit for storing trays containing electronic components; a pallet capable of being withdrawn from the storage unit and carrying the tray; The TOF camera irradiates light onto the tray on the pallet pulled out from the storage unit, and obtains a three-dimensional image showing a surface height of the tray pulled out from the storage unit and a surface height of the electronic components accommodated in the tray based on the reflected light. The inspection device detects at least one of an abnormality in the surface condition of the tray pulled out of the storage facility and an abnormality in the placement of electronic components on the tray based on the three-dimensional image acquired by the TOF camera.

2. 2. The surface mounter according to claim 1, The inspection device includes: The surface mounter detects abnormalities in the surface condition of the tray by recognizing displacement locations of the surface height of the tray based on the three-dimensional image.

3. 2. The surface mounter according to claim 1, The inspection device includes: The surface mounter detects abnormal placement of the electronic components relative to the tray by recognizing displacement locations of the surface height of the electronic components accommodated in the tray based on the three-dimensional image.

4. 2. The surface mounter according to claim 1, The inspection device includes: A model generation unit; A determination unit, the model generation unit is a supervised learning model using a three-dimensional image showing a normal surface height of the tray as training data, and generates, by machine learning, a first learning model that estimates an abnormality in the surface condition of the tray using the three-dimensional image showing the surface height of the tray as an input; The determination unit determines an abnormality in the surface condition of the tray from the first learning model.

5. 2. The surface mounter according to claim 1, The inspection device includes: A model generation unit; A determination unit, The model generation unit a second learning model is generated by machine learning using a three-dimensional image showing a normal arrangement of electronic components on the tray as training data, the second learning model being configured to estimate an abnormal arrangement of the electronic components on the tray using the three-dimensional image showing surface heights of the electronic components accommodated on the tray as an input; The determination unit determines an abnormality in placement of electronic components on the tray from the second learning model.

6. 2. The surface mounter according to claim 1, After the tray is drawn out from the storage cabinet and before the head unit removes the electronic components from the tray, the inspection device acquiring, by the TOF camera, a three-dimensional image showing a surface height of the tray pulled out from the storage unit and a surface height of the electronic components accommodated in the tray; The surface mounter detects at least one of an abnormality in the surface condition of the tray and an abnormality in the arrangement of electronic components accommodated in the tray based on the acquired three-dimensional image.

7. 2. The surface mounter according to claim 1, The inspection device, during a period from when an electronic component is taken out of the tray until when a next electronic component is taken out of the tray, acquiring a three-dimensional image indicating a surface height of the tray and a surface height of the electronic components accommodated in the tray by the TOF camera; The surface mounter detects at least one of an abnormality in the surface condition of the tray and an abnormality in the arrangement of electronic components accommodated in the tray based on the acquired three-dimensional image.

8. 3. The surface mounter according to claim 2, The inspection device recognizes the locations of displacement in the surface height of the tray by comparing a three-dimensional image indicating the surface height of the tray pulled out of the storage facility with a three-dimensional image indicating the normal surface height of the tray.

9. 2. The surface mounter according to claim 1, The TOF camera is attached to the head unit and is movable together with the head unit above the tray pulled out from the storage cabinet.

10. 2. The surface mounter according to claim 1, A surface mount machine, wherein the TOF camera is immovably fixed to the upper part of the storage cabinet.

11. 2. The surface mounter according to claim 1, A display unit is provided, The inspection device displays at least one of an abnormality in the surface condition of the tray and an abnormality in the arrangement of electronic components on the display unit.

12. 12. The surface mounter according to claim 11, The display unit is a surface mounter that displays identified positions of electronic components that are abnormally placed.

13. A parts supply unit, A tray feeder; A time-of-flight camera; The tray feeder comprises: a storage unit for storing trays containing electronic components; a pallet capable of being withdrawn from the storage unit and carrying the tray; The TOF camera is fixed immovably at the top of the storage unit, and shines light onto the tray on a pallet pulled out of the storage unit, and based on the reflected light, obtains a three-dimensional image showing the surface height of the tray and the surface heights of the electronic components contained in the tray.

14. 1. A testing method comprising: A three-dimensional image of the tray pulled out from the storage unit is acquired by a TOF camera; An inspection method for detecting at least one of abnormalities in the surface condition of the tray pulled out of the storage facility and abnormalities in the arrangement of electronic components contained on the tray based on a three-dimensional image of the tray acquired by the TOF camera.

Images