Suction nozzle design apparatus, design method, and design program

Abstract

To calculate a design parameter of a suction nozzle and to manufacture it by a three-dimensional lamination molding device.SOLUTION: The suction nozzle design device 30 includes an imaging device 40 and a calculation device 50. The imaging device 40 captures an image of a component to be sucked. The arithmetic device 50 obtains the three-dimensional shape of the upper surface of the component including the suction surface of the suction nozzle based on the component image captured by the imaging device 40, and calculates the design parameter of the suction nozzle based on the obtained three-dimensional shape of the upper surface of the component.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a design device, a design method, and a program for designing a suction nozzle used in a mounting machine. [Background technology]

[0002] A surface mounter is equipped with a mounting head that mounts components on a board. The mounting head has a suction nozzle at its tip, and is configured to suck up components using negative pressure. Patent Document 1 listed below discloses a device that supports nozzle creation by acquiring component parameters based on information queried from the component mounter and CAD (Computer Aided Design) data. [Patent Document 1] Japanese Patent Application Publication No. 2017-69332 DISCLOSURE OF THE INVENTION [Problem to be solved by the invention]

[0003] However, when mounting new components, it is not always possible to prepare component information and CAD data. In such cases, for example, when mounting components with specially shaped pickup surfaces, it can be difficult to achieve both pickup with existing nozzles and the creation of simple nozzles. An object of the present invention is to provide a design device, a design method, and a program for a suction nozzle that are capable of designing a suction nozzle even when component information such as CAD data is not available. [Means for solving the problem]

[0004] (1) A design device for a suction nozzle, comprising an imaging device and a computing device, wherein the imaging device photographs a component to be picked up, the computing device calculates a three-dimensional shape of the top surface of the component, including the surface to be picked up by the suction nozzle, based on the component image photographed by the imaging device, and determines design parameters for the suction nozzle based on the calculated three-dimensional shape. In the design device for a suction nozzle described in (1), any configuration other than the above is optional and may be used.

[0005] According to (1), it is possible to automatically calculate the design parameters of the suction nozzle, and it is possible to easily manufacture the suction nozzle using a three-dimensional additive manufacturing device, such as a 3D printer.

[0006] (2) In the device for designing a suction nozzle according to (1), the design parameters may include a tip shape of the suction nozzle. The tip shape of the suction nozzle may be such that it does not interfere with the top surface of the component. According to (2), it is possible to avoid poor suction due to interference and unnecessary contact between the suction nozzle and the component, which may scratch the component.

[0007] (3) In the device for designing a suction nozzle according to (1) or (2), the suction nozzle may have a nozzle hole opening at its tip. The design parameters may include the position of the nozzle hole. According to (3), the position of the nozzle hole can be automatically determined according to the part shape.

[0008] (4) In the device for designing a suction nozzle according to any one of (1) to (3), the computing device may calculate the curvature of the upper surface of the component and set the suction position of the suction nozzle in an area where the curvature is equal to or less than a predetermined value. According to (4), the opening position of the suction nozzle is set in an area where the curvature is equal to or less than a predetermined value, so that the component can be stably picked up.

[0009] (5) In the device for designing a suction nozzle according to any one of (1) to (4), the suction nozzle may have a nozzle hole opening at its tip. The design parameters may include the opening size of the nozzle hole. According to (5), the suction force of the suction nozzle is correlated with the nozzle opening area, so it is possible to ensure suction force, and an improvement in the component pickup rate is expected.

[0010] (6) In the device for designing a suction nozzle according to any one of (1) to (5), the suction nozzle has a nozzle hole opening at its tip, and the computing device may calculate, for each of a plurality of regions obtained by dividing the upper surface of the component, at least one of a first score representing the shapeability of the nozzle hole and a second score representing the suction performance when the region is set as a suction position, and select the region to be set as a suction position based on the calculated score. According to (6), the suction position of the suction nozzle can be determined from the viewpoint of the shapeability and suction performance of the suction nozzle.

[0011] (7) In the device for designing a suction nozzle according to (6), the first score may be calculated based on a structural difference between the components in each of the regions and a resolution of a three-dimensional additive manufacturing device.

[0012] (8) In the device for designing a suction nozzle according to (6) or (7), the second score may be calculated based on a suction force generated when a suction position is set in the region.

[0013] (9) In the device for designing a suction nozzle according to any one of (6) to (8), the calculation device may combine adjacent regions and recalculate the score if the score does not satisfy a standard. According to (9), combining regions can be expected to have the effect of increasing suction force, enabling flexible parameter design of the suction nozzle.

[0014] (10) In the device for designing a suction nozzle according to any one of (1), (2), (3), or (5), the arithmetic unit may extract a plurality of feature points from images of the upper surface of the component photographed from a plurality of directions by the imaging device using a feature point matching method to generate point cloud data. An area where the density of the point cloud is equal to or less than a threshold may be determined as the suction position of the suction nozzle. According to (10), score calculations and the like are not required, thereby reducing the computational load on the arithmetic unit.

[0015] This technology can be applied to suction nozzle design methods and design programs. [Effects of the Invention]

[0016] According to the present invention, it is possible to automatically calculate the design parameters of the suction nozzle, and the suction nozzle can be easily manufactured using a three-dimensional additive manufacturing device, such as a 3D printer. [Brief explanation of the drawings]

[0017] [Figure 1] FIG. 1 is a block diagram of a design device for a suction nozzle according to an embodiment of the present invention. [Figure 2] FIG. 2 is a flowchart showing the process of the suction nozzle. [Figure 3] FIG. 3 is a diagram showing how an RGB camera is used as the imaging device in the first embodiment, and input images for the SfM algorithm are acquired by photographing a component from multiple directions. [Figure 4] FIG. 4 is a diagram for explaining imaging setup conditions and acquired depth images using a TOF camera for imaging in the first embodiment. [Figure 5] FIG. 5 is a diagram for explaining how a curve is approximated for a pickup target area from a plurality of points acquired as the three-dimensional shape of the component pickup surface. [Figure 6] FIG. 6 is a cross-sectional view of a component and a suction nozzle, showing that the flat portion is the suction position. [Figure 7]FIG. 7 is a cross-sectional view of a component and a suction nozzle, showing that the protrusion is set as the suction position. [Figure 8] FIG. 8 is a diagram showing a cross section of a part with a complex shape. [Figure 9] FIG. 9 is a flowchart showing how the top surface of a component is divided into small regions, a score is calculated for each small region, and a pickup position is determined. [Figure 10] FIG. 10 is a diagram for explaining division of the upper surface of a component into a plurality of regions. [Figure 11] FIG. 11 is a diagram for explaining the structural differences between regions on the upper surface of the component. [Figure 12A] FIG. 12A is a diagram showing the relationship between structural difference and score. [Figure 12B] FIG. 12B is a diagram showing the relationship between structural difference and score. [Figure 13A] FIG. 13A shows the scores calculated for each region. [Figure 13B] FIG. 13B shows the scores calculated for each region. [Figure 14] FIG. 14 shows a part with a complex shape and the resulting pickup positions. [Figure 15] FIG. 15 is a diagram showing scores for each region in which adsorption is difficult with only a single connected region, but becomes possible by setting multiple adsorption regions. [Figure 16] FIG. 16 is a cross-sectional view of a part that has a mixture of structured and less structured areas. [Figure 17] FIG. 17 is a diagram showing that the obtained point cloud data of the part includes both sparse and dense regions. DETAILED DESCRIPTION OF THE INVENTION

[0018] <Embodiment 1> 1. Design equipment and design flow for suction nozzle 20 1 is a block diagram of a design device 30 for a suction nozzle 20. The design device 30 for a suction nozzle 20 is composed of a camera 40 and a computing device 50. The camera 40 is an example of the "imaging device" of the present invention.

[0019] The camera 40 is used to photograph parts, and can be an RGB camera, a stereo camera, an RGBD camera, a TOF camera, etc. Also, an optical system set system such as an inspection machine may be used.

[0020] The calculation device 50 is, for example, a computer equipped with a CPU and an image memory, and includes a 3D data acquisition unit 51 and a parameter design unit 55.

[0021] The 3D data acquisition unit 51 acquires data on the three-dimensional shape of the part 10 from an image of the part 10 captured by the camera 40. The three-dimensional shape data is a collection of point cloud data, but may also be data in other forms such as a mesh. The shape of the suction surface may also be formulated as an approximation curve such as a quadratic equation using the least squares method or the like.

[0022] The parameter design unit 55 determines design parameters of the suction nozzle 20 based on data of the three-dimensional shape of the component 10.

[0023] 2 is a flowchart of the process of the suction nozzle 20. The process (design flow) of the suction nozzle 20 is made up of three steps, S10 to S30.

[0024] S10 is the photographing step of the parts. S20 is a step for acquiring the three-dimensional shape. S30 is a step for determining design parameters.

[0025] Each step will be explained in order below. <S10、S20> (1) RGB camera and SFM technology As a first method, we will explain the case where an RGB camera is used as the imaging device. In this case, the part 10 is photographed from multiple directions, as shown in Figure 3. Then, the 3D data acquisition unit 51 can create a 3D model of the part 10 by using an SFM (Structure from Motion) algorithm.

[0026] SFM is a method for estimating the point cloud of the part 10 by estimating information about the photographing position and orientation of each image.

[0027] (2) Use of TOF (Time Of Flight) camera As a second method, we will explain the case where a TOF camera is used as the imaging device. In this case, as shown in Figure 4, a depth image of the component 10A can be obtained by capturing an image of the component 10A with a TOF camera 40A. The depth image (x, y, d) indicates that it has information on the depth d corresponding to the two-dimensional plane coordinates x, y.

[0028] The size of one pixel P in the horizontal direction (X direction in FIG. 4) of the depth image can be calculated using formula (1).

[0029] P=2×h×Tan(θ / 2)×(1 / n)···(1) h is the camera height [mm], θ is the horizontal angle of view of the camera, and n is the number of pixels in the horizontal direction of the camera.

[0030] In the two example methods described above, the 3D data acquisition unit 51 can calculate point cloud data as 3D data. Furthermore, the 3D data acquisition unit 51 may obtain an approximation curve of the suction surface of the part shape. This method will be described using the TOF camera 40. In this case, the 3D data acquisition unit 51 targets the protrusion 13 in the image of the part 10A captured by the TOF camera 40, and approximates multiple points corresponding to the part surface with a curve, thereby obtaining the shape of the protrusion 13. For example, an approximation curve of the part shape can be obtained using the least squares method, and this can be used as the 3D shape of the part.

[0031] It is also possible to express a 3D shape in terms of curvature. For example, as shown in Figure 5, if protrusion 13 of component 10A has a shape that is line-symmetric about the Z axis, the shape of protrusion 13 can be approximated by the quadratic function F of equation (2), where 2|a| is the curvature. The symbol || indicates the absolute value.

[0032] F=ar 2 +br+c···(2) r=sqrt(x 2 +y 2 )···(3)

[0033] <s30> The parameter setting unit 55 calculates the design parameters of the suction nozzle 20 based on the 3D shape of the component 10 acquired in S20. The design parameters include the shape of the nozzle tip 21 and the position of the nozzle hole 25.

[0034] The parameter setting unit 55 may determine the shape of the nozzle tip 21 to be the same as the shape of the suction surface of the component 10A so that the nozzle tip 21 does not interfere with the component 10 and there is no gap between the nozzle tip 21 and the component 10A.

[0035] 6 and 7, because the component upper surface 11 is used as the suction surface, the shape of the nozzle tip 21 is determined to be the same as the component upper surface 11, in other words, a shape with the concaves and convexes reversed. In this example, a depression 23 with the concaves and convexes reversed is formed in the nozzle tip 21 in correspondence with the protrusion 13 on the component upper surface 11.

[0036] Fig. 6 shows an example in which the flat portion 12 on the component upper surface 11 is the suction position, and Fig. 7 shows an example in which the protrusion 13 on the component upper surface 11 is the suction position. The suction position is the position of the nozzle hole 25. The symbol S shown in Figs. 6 and 7 represents the opening size (opening area) of the nozzle hole 25.

[0037] 7, when setting a pickup position on a curved surface of component upper surface 11, such as protrusion 13, parameter setting unit 55 preferably calculates curvature a of component upper surface 11 and sets the pickup position in a range where curvature a is equal to or less than a predetermined value. By picking up a position in a range where curvature a is small, component 10B can be picked up stably.

[0038] After the design parameters are determined in S30, the suction nozzles 20A and 20B can be manufactured based on the parameters using a three-dimensional additive manufacturing device, such as a 3D printer.

[0039] <Embodiment 2> Generally, stable suction is expected by suctioning a part of the component shape that has minimal structural change. For example, in the case of component 10C shown in Figure 8, when two protrusions 13A and 13B exist on the component top surface 11, it is desirable to suction the tops K of protrusions 13A and 13B, as this part has minimal structural change. Structural change is synonymous with shape change.

[0040] In the second embodiment, a method for automatically setting the pickup position of the component 10 is disclosed. Figure 9 is a flowchart that divides the top surface of a component into small regions, calculates the score for each small region, and determines the pickup position. The flowchart consists of three steps: S50, S60, and S70.

[0041] S50 is a step of dividing the upper surface of the component into a plurality of regions G. S60 is a step for outputting the score C for each region. S70 is a step for determining the pickup position based on the score C.

[0042] <s50> Fig. 10 is a diagram for explaining the division of the upper surface of a component into multiple regions. As shown in Fig. 10, the parameter setting unit 55 divides the upper surface of the component 10 into multiple regions G using a lattice-like grid. The size d of one side of the region G can be determined as follows.

[0043] (a) Determine the size d based on the xy resolution of the 3D additive manufacturing device. (b) Use a predetermined size d.

[0044] In the case of (a), the size d of the region G is determined to be d≧r, where r is the xy resolution of the 3D additive manufacturing device. For example, d=r, r×1.2, r×1.5, etc.

[0045] In the case of (b), the size d of the region G is a predetermined value, regardless of the xy resolution r of the 3D additive manufacturing device.

[0046] If a predetermined value is used as the size d of region G, the score of the xy resolution of the 3D additive manufacturing device may also be calculated when calculating the score, and region G may be enlarged if necessary when selecting the region.

[0047] <s60> After dividing the upper surface of the component into a plurality of regions G, the parameter setting unit 55 calculates two scores C1 and C2 for each region G.

[0048] The first score C1 represents the formability of the nozzle hole 25 when the region G is set as the suction position, specifically, the ease of production by a three-dimensional additive manufacturing device.

[0049] Fig. 11 shows the structural difference (synonymous with shape change) U in the Z direction of the part 10 in the region G, and Fig. 12 is an example of a graph showing the relationship between the structural difference U and the first score C1. The first score C1 takes a value between 0 and 1 depending on the structural difference U.

[0050] Specifically, if the structural difference U in the Z direction is larger than the z-direction resolution R of the 3D additive manufacturing device (also called the layer pitch in the z direction), more specifically, if the structure is steep or has fine details, it becomes difficult to manufacture the nozzle hole 25 accurately and stably. In such cases, it is difficult to manufacture the nozzle hole 25 accurately and with high precision, and it is desirable that the first score C1 be a value close to 0, as shown in FIG. 12. On the other hand, if the structural difference U in the Z direction is smaller than the z-direction resolution R of the 3D additive manufacturing device, it is easy to manufacture the nozzle hole 25, and the first score C1 is a value close to 1.

[0051] The first score C1 is not limited to the formability in the Z direction, and may be evaluated using the formability in the X direction or the Y direction (structural difference in the X direction or the Y direction) instead of the Z direction. A composite evaluation of these may also be performed. Furthermore, the object does not necessarily need to be layered parallel to the adsorption surface, and may be layered in the X direction, the Y direction, or another direction if this is advantageous for the object from the viewpoint of accuracy. The first score C1 may be calculated using an index other than the resolution described above, such as the object's accuracy (the object's stability). The object's formability may also include the time required for the object to be formed. For example, with a 3D additive manufacturing device that has a variable layer pitch, setting the layer pitch small makes it possible to create detailed structures, but at the same time, the number of layers to be created also increases, resulting in an increase in production time and a decrease in the number of layers that can be created per unit time.

[0052] The second score C2 is a score that evaluates the pickup performance, and can be defined as the ratio of the generated pickup force Vg to the required pickup force Vm, as shown in equation (4). The generated pickup force Vg is the pickup force that is generated when the area G is set as the pickup position. The required pickup force Vm is the pickup force that is required to hold a component when the area G is set as the pickup position.

[0053] C2=Vg / Vm (4) Vg = negative pressure × opening area of ​​nozzle hole 25 (5) Vm = M × (1 + L) (6) M is the weight of the part. L is the distance from the center of gravity O of the part to the pickup position (area G) (see Figure 10).

[0054] The second score C2 of each region G can be calculated using the above formula, with the negative pressure being a constant and the opening area of ​​the nozzle hole 25 being equal to the area of ​​the region G.

[0055] As a result of the calculation, if Vg≧Vm, the required suction force is obtained, and C2≧1. If C2≧1, then by setting C2=1, the second score C2 takes a value between 0 and 1, just like the first score C1.

[0056] In this embodiment, the total score C of each region G is calculated by multiplying the first score C1 and the second score C2.

[0057] C = C1 × C2 (7)

[0058] Then, the parameter setting unit 55 compares the total scores C of each region G and determines the region G with the maximum score as the suction position of the suction nozzle 20. In addition, the size (area) of the region G with the maximum score is determined as the opening size (opening area) of the nozzle hole 25.

[0059] 13A shows the calculation result of the total score C. In this example, of the regions G1 to G7, the region G4 has the highest score (C=1), and the region G4 is determined as the suction position (suction range) of the suction nozzle 20. The size of the region G4 is then determined as the opening size of the nozzle hole 25.

[0060] Furthermore, if the maximum score C does not meet a standard (for example, "1"), multiple regions G may be combined. Combining regions G increases the overall score C for the following reasons.

[0061] (A) As a result of the wider opening area, the generated adsorptive force Vg increases. (B) By expanding the area, for example, if a sufficient area can be secured in the xy direction and the modeling accuracy is stable, it becomes possible to manufacture using a three-dimensional additive manufacturing device.

[0062] 13B, for example, the four regions G1 to G4 are combined into one region, and the total score C is recalculated. Then, when the total score C reaches a target value (for example, "1"), the combined regions G1 to G4 are set as the suction position of the suction nozzle 20. In this case, the total size (total area) of the combined regions G1 to G4 is determined as the opening size (opening area) of the nozzle hole 25.

[0063] Figure 14 shows the pickup position of component 10C, and Figure 15 shows the score for each region G. In this example, the score is low at a maximum of 0.2, so the four regions G1 to G4 and the four regions G5 to G8 shown in Figure 15 were combined and the score C was recalculated.

[0064] After joining, the score C of regions G1 to G4 and regions G5 to G8 is 0.8 (= 0.2 × 4). The sum of the scores C of regions G1 to G4 and G5 to G8 is 1.6, which exceeds 1, so two locations, regions G1 to G4 and regions G5 to G8, are set as pickup positions (pickup ranges).

[0065] In the second embodiment, the top surface of the component is divided into multiple regions G, and two scores C1 and C2 are calculated for each region G. The pickup position is then determined based on the product of the two scores C1 and C2. However, the pickup position may be determined by calculating only one of the two scores C1 and C2.

[0066] Furthermore, even when both the scores C1 and C2 are used, the score C may be calculated using a sum or other operation instead of a product.

[0067] <Embodiment 3> In the first embodiment, an RGB camera was used to capture images of the part 10 from multiple directions, and an SFM algorithm was used to create a 3D model of the part 10. In the third embodiment, a feature point matching method is used to create a 3D model of the part 10. The feature point matching method is a method of detecting feature points between two images (in this example, between multiple images captured with an RGB camera) and matching similar feature points.

[0068] The feature point matching method represents the part shape as a collection of feature points (hereafter referred to as a point cloud), and the density of the point cloud varies depending on the area. Specifically, areas with many structures such as protrusions have a large number of feature points, and a dense point cloud with many matching feature points is generated. In contrast, areas with few structures such as flat areas have a small number of feature points, and a sparse point cloud with few matching feature points is generated.

[0069] In the third embodiment, the area with the smallest density of the point cloud is determined as the pickup position. For example, in the case of a part 10D shown in Fig. 16, which has a flat center 15 and protrusions on the outer periphery 16, the feature points M are concentrated on the outer periphery 16 and are almost absent from the center 15, as shown in Fig. 17.

[0070] Therefore, in the case of part 10D, the central portion 15 where the density of the point cloud is lowest is set as the pickup position. Note that the feature points M may be the density in an XYZ cube (in three-dimensional space) or the density on the XY plane (in two-dimensional plane) onto which the Z direction of the point cloud is projected.

[0071] When the pickup position is determined based on the density of the point cloud, fitting is not required and processing can be performed using the point cloud data as is, which has the advantage of reducing the amount of calculation.

[0072] <Other embodiments> The present invention is not limited to the embodiments described above and illustrated in the drawings, and the following embodiments, for example, are also included within the technical scope of the present invention.

[0073] (1) As another embodiment, an image of a component may be acquired by an imaging device, the acquired image may be transferred, and the three-dimensional shape of the component's top surface may be calculated by a computing device at the transfer destination, and the design parameters of the suction nozzle may be determined. With this method, for example, a component may be imaged by a camera in a factory, the image may be transferred, and the suction nozzle may be designed and manufactured using a computing device installed at a distance. The image may be transferred via a storage medium such as a USB memory, or via a network. [Explanation of symbols]

[0074] 1 part 30 Design equipment 40 Camera (imaging device) 50 Arithmetic unit 51 3D Data Acquisition Unit 55 Parameter Design Department

Claims

1. A suction nozzle design device, An imaging device; a computing device; the imaging device takes an image of the component to be picked up; The computing device calculates the three-dimensional shape of the top surface of the component, including the surface to be picked up by the suction nozzle, based on the component image captured by the imaging device, and determines design parameters of the suction nozzle based on the calculated three-dimensional shape.

2. 2. The device for designing a suction nozzle according to claim 1, the design parameters include a tip shape of the suction nozzle, The tip shape of the suction nozzle is a shape that does not interfere with the upper surface of the component.

3. 3. The device for designing a suction nozzle according to claim 1, the suction nozzle has a nozzle hole opening at a tip end thereof, The design parameters include the position of the nozzle hole.

4. 3. The device for designing a suction nozzle according to claim 1, The arithmetic unit calculates the curvature of the upper surface of the component, and sets the suction position of the suction nozzle in an area where the curvature is equal to or less than a predetermined value.

5. 3. The device for designing a suction nozzle according to claim 1, the suction nozzle has a nozzle hole opening at a tip end thereof, A device for designing a suction nozzle, wherein the design parameters include an opening size of the nozzle hole.

6. 3. The device for designing a suction nozzle according to claim 1, the suction nozzle has a nozzle hole opening at a tip end thereof, The computing device A suction nozzle design device that calculates, for each of multiple regions obtained by dividing the top surface of the part, at least one of a first score representing the shapeability of the nozzle hole and a second score representing the suction performance when that region is set as a suction position, and selects the region to be set as a suction position based on the calculated score.

7. 7. The device for designing a suction nozzle according to claim 6, A suction nozzle design device, wherein the first score is calculated based on the structural difference of the part for each region and the resolution of a three-dimensional additive manufacturing device.

8. 7. The device for designing a suction nozzle according to claim 6, The second score is calculated based on the suction force generated when a suction position is set in the area.

9. 7. The device for designing a suction nozzle according to claim 6, The arithmetic unit is a device for designing a suction nozzle, and if the score does not satisfy the standard, the arithmetic unit combines adjacent regions and recalculates the score.

10. 3. The device for designing a suction nozzle according to claim 1, the computing device extracts a plurality of feature points from images of the top surface of the component captured from a plurality of directions by the imaging device using a feature point matching method to generate point cloud data; A design device for a suction nozzle that determines a region where the density of the point cloud is equal to or less than a threshold value as a suction position of the suction nozzle.

11. A method for designing a suction nozzle, comprising: taking an image of the component with an imaging device; determining a three-dimensional shape of the upper surface of the component, including the surface to be picked up by the suction nozzle, based on a component image captured by an imaging device; and determining design parameters of the suction nozzle based on the obtained three-dimensional shape.

12. A suction nozzle design program, On the computer, determining a three-dimensional shape of the upper surface of the component, including the surface to be picked up by the suction nozzle, based on the component image captured by the imaging device; and determining design parameters of the suction nozzle based on the obtained three-dimensional shape.

Images