Visual alignment system and method of swing arm type chip mounter

Abstract

The invention discloses a visual alignment method and system of a swing arm type chip mounter, the swing arm type chip mounter comprises a carrying table, a suction mechanism and a visual system, and the visual system comprises a first image capturing assembly and a second image capturing assembly; the method comprises the following steps: S1, adsorbing a chip A at a first position through an adsorption mechanism, and adsorbing and positioning a chip B through a carrying table; s2, driving the suction mechanism to rotate to a second position, acquiring a first image of the chip B through the first image acquisition assembly, and acquiring a second image of the chip A through the second image acquisition assembly; s3, projecting the first image and the second image to the same coordinate system, comparing the alignment difference, positioning an error component according to the alignment difference, and calculating and adjusting the compensation amount; s4, driving a corresponding error component to perform error compensation according to the adjustment compensation amount, and repeating the steps S2 and S3 until the alignment difference is smaller than a preset threshold value; the chip mounting alignment precision is obviously improved, and high-precision and high-efficiency alignment adjustment of the swing arm type chip mounter is realized.

Description

Technical Field

[0001] This invention relates to the field of chip technology, and in particular to a vision alignment system and method for a swing-arm type chip mounter. Background Technology

[0002] In the fields of microelectronics manufacturing, semiconductor packaging, and high-precision electronic assembly, the swivel-arm placement machine is a key piece of equipment used to achieve high-speed, high-precision bonding between chips and components. Its core function is to pick up, transport, and precisely mount a chip onto another pre-positioned chip or substrate. The bonding accuracy of this process directly determines the electrical performance and reliability of the final product. Especially in chip-level packaging and advanced packaging scenarios, the alignment accuracy requirements have reached the micrometer level. Any tiny deviation can lead to short circuits, open circuits, or performance degradation, resulting in a decrease in product yield.

[0003] Currently, one common existing technical solution for achieving high-precision alignment is to perform high-precision mechanical calibration beforehand. This method relies on the precision machining and assembly of each motion axis of the pick-and-place machine (such as rotary axes and linear guides), and establishes a theoretical motion model through offline, periodic system calibration to attempt to compensate for inherent mechanical errors. However, this type of method has significant drawbacks: its accuracy maintenance is highly dependent on the initial state and long-term stability of the equipment, and it cannot effectively cope with real-time error drift caused by wear, temperature changes, or mechanical stress relaxation during continuous operation. This static or periodic calibration method lacks dynamic feedback and real-time compensation capabilities for each placement action, resulting in a gradual and imperceptible decrease in placement alignment accuracy during actual production, especially after long-term operation. This makes it difficult to continuously meet the requirements of high-yield production, becoming a key technical bottleneck restricting production efficiency and product consistency.

[0004] Therefore, it is necessary to improve the existing chip placement machine to solve the technical problem of inaccurate chip alignment caused by mechanical errors during its operation. Summary of the Invention

[0005] The purpose of this invention is to provide a vision alignment system and method for a swing-arm type chip mounter, thereby solving the above-mentioned technical problems.

[0006] To achieve this objective, the present invention adopts the following technical solution: A vision alignment method for a swing-arm pick-and-place machine, the swing-arm pick-and-place machine including a stage, a pick-up mechanism, and a vision system, the vision system including a first image-capturing component and a second image-capturing component; the method includes the following steps: S1, the suction mechanism adsorbs chip A at the first position, and the stage adsorbs and positions chip B. S2, drive the suction mechanism to rotate to the second position, acquire the first image of chip B through the first image acquisition component, and acquire the second image of chip A through the second image acquisition component; S3, Project the first image and the second image onto the same coordinate system, compare the alignment difference, locate the error component based on the alignment difference and calculate the adjustment compensation amount; S4, drive the corresponding error component to perform error compensation according to the adjusted compensation amount, and repeat steps S2 and S3 until the alignment difference is less than the preset threshold.

[0007] Optionally, the suction mechanism includes a rotating shaft connected to a drive source; One end of the rotating shaft is fixedly connected to a rotating support arm, and the end of the rotating support arm away from the rotating shaft is provided with a core-picking component, which is used to pick up and fix chip A by vacuum adsorption.

[0008] Optionally, the suction mechanism for adsorbing chip A at the first position specifically includes: The rotating arm is in the first horizontal position with its core-taking surface facing upwards; Chip A is transferred to the core-taking component of the suction mechanism by an external transfer mechanism, and chip A is fixed by the core-taking component through vacuum adsorption.

[0009] Optionally, in step S2, driving the suction mechanism to rotate to the second position specifically involves: The rotating arm rotates 90° around the rotating axis, so that the core extractor and chip A are in a vertical position; the first image-capturing component is positioned above the stage to acquire a first image, and the second image-capturing component is positioned above the vertically positioned core extractor to acquire a second image.

[0010] Optionally, projecting the first image and the second image onto the same coordinate system specifically includes: Extract the alignment feature points of chip B in the first image and the alignment feature points of chip A in the second image; The coordinates of the alignment feature point of chip A are projected into the same coordinate system as the alignment feature point of chip B, based on the known angular relationship of the absorption mechanism rotating from the first position to the second position.

[0011] Optionally, the specific process for comparing the alignment differences is as follows: The alignment feature points of chip A, after being projected onto the same coordinate system, are matched and compared with the alignment feature points of chip B. Calculate the positional deviations in the X and Y directions between the alignment feature points of chip A and chip B, as well as the rotational angle deviations between them; generate the alignment difference data based on the positional deviations and rotational angle deviations.

[0012] Optionally, the alignment difference positioning error component includes: The determination is based on the type of positional deviation and rotation angle deviation; If the alignment difference is mainly manifested as translational deviation in the X or Y direction, then the error component is determined to be the stage or rotation axis related to linear motion. If the alignment difference is mainly manifested as the rotation angle deviation, then the error determination component is the rotating support arm or core extractor related to the rotation angle.

[0013] Optionally, the specific process for calculating the adjustment compensation amount includes: Based on the identified error component type and its corresponding deviation data, calculations are performed using operation rules that match the motion characteristics of the component. If the error component is determined to be a stage or rotation axis related to linear motion, its adjustment compensation amount is a linear compensation displacement, which is calculated as follows: the X-direction position deviation is used as the X-axis compensation displacement of the stage, and the Y-direction position deviation is used as the Y-axis compensation displacement of the stage; or, the component of the position deviation in the radial direction of the rotation axis is used as the radial compensation displacement of the rotation axis. If the error component is determined to be a rotating arm or core extractor related to the rotation angle, its adjustment compensation amount is the angle compensation angle. The calculation method is: directly use the rotation angle deviation as the angle compensation value of the rotating arm or core extractor to correct its rotation angle around the rotation axis or its own deflection angle. The calculation rules also incorporate the known rotation angle of the suction mechanism from the first position to the second position, which is used to convert the pixel deviation in the image coordinate system into the actual physical displacement or angle in the coordinate system.

[0014] Optionally, step S4 specifically includes: S41, based on the calculated adjustment compensation amount, drive the corresponding error component to perform compensation movement, the compensation movement including driving the platform or rotating shaft to perform linear displacement compensation, or driving the rotating support arm or core extractor to perform angular deflection compensation. S42, After the compensation motion is completed, repeat steps S2 and S3 to obtain new alignment differences; S43, determine whether the new alignment difference is less than the preset threshold; If so, then end the alignment process; If not, the new alignment difference is used as input to relocate the error component and calculate the new adjustment compensation amount, then the process returns to step S41 for iterative compensation.

[0015] The present invention also provides a vision alignment system for a swing-arm type placement machine, which adopts the vision alignment method of the swing-arm type placement machine as described above. The vision alignment system further includes a processing unit, which is electrically connected to the pick-up mechanism, the vision system, and the stage, and is configured to: control the vision system to acquire images, project the first image and the second image onto the same coordinate system for comparison to obtain alignment differences, locate error components based on the alignment differences and calculate and adjust the compensation amount, and control the corresponding error components to perform compensation movements until the alignment difference is less than a preset threshold.

[0016] Compared with existing technologies, the present invention has the following advantages: First, chip A is adsorbed at a first position by a suction mechanism, and chip B is adsorbed and positioned by a stage; then, the suction mechanism is driven to rotate to a second position, a first image of chip B is acquired using a first imaging component, and a second image of chip A is acquired using a second imaging component; next, the first and second images are projected onto the same coordinate system to compare the alignment difference, and the error component is located based on the alignment difference and the adjustment compensation amount is calculated; according to the adjustment compensation amount, the corresponding error component is driven to perform error compensation, and the image acquisition, comparison, and compensation steps are repeated until the alignment difference is less than a preset threshold; this solution combines visual alignment with iterative compensation. After acquiring images of chip A and chip B, the images are projected onto the same coordinate system for difference comparison, thereby accurately locating the error component and calculating the compensation amount. The alignment error is gradually reduced through iterative adjustment until the preset accuracy requirement is met; this method significantly improves the chip placement alignment accuracy and realizes high-precision and high-efficiency alignment adjustment of the swing arm type chip mounter. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] The structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0019] Figure 1 This is a schematic diagram of the overall system layout of the swing arm type chip mounter in this embodiment 1; Figure 2 This is a schematic diagram of the pick-up mechanism of the swing arm type chip mounter in the first embodiment when it is in the first position. Detailed Implementation

[0020] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] In the description of this invention, it should be understood that the terms "upper," "lower," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. It should be noted that when a component is considered to be "connected" to another component, it can be directly connected to the other component or there may be a component positioned centrally in the connection.

[0022] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1: Combination Figures 1 to 2 As shown, this embodiment of the invention provides a visual alignment method for a swing-arm pick-and-place machine. The swing-arm pick-and-place machine includes a stage 10, a pick-and-place mechanism, and a vision system. The vision system includes a first image-capturing component 20 and a second image-capturing component 30. The first image-capturing component 20 is positioned directly above the stage and is used to capture and acquire images of the chip B on the upper surface of the stage. The second image-capturing component 30 is positioned opposite the pick-and-place mechanism and is used to acquire images of the pick-and-place mechanism at a second position (as shown in the attached diagram). Figure 1 The image of chip A (when it is in the vertical position).

[0024] Specifically, the visual alignment method includes the following steps: S1, chip A is adsorbed at the first position by the suction mechanism, and chip B is adsorbed and positioned by the stage 10; Combination Figure 2 As shown, the suction mechanism first moves to a preset first position (usually a horizontal position), and the chip-taking component 60 at its end picks up and fixes the chip A to be mounted by vacuum adsorption or other means. At the same time, the stage 10 uses the suction holes provided below it to vacuum adsorb and precisely position the chip B that has been placed, so that it is stably placed in the alignment position.

[0025] S2, drive the suction mechanism to rotate to the second position, acquire the first image of chip B through the first image acquisition component 20, and acquire the second image of chip A through the second image acquisition component 30; After the chip is fixed, the suction mechanism is driven to rotate around its rotation axis 40 to a preset second position (usually a vertical position). When the mechanism is in position, the vision system is synchronously triggered to acquire images: the first image acquisition component 20 faces the upper surface of the stage 10 and acquires a clear image of the positioned chip B, i.e., the first image; the second image acquisition component 30 faces the end of the suction mechanism in the second position and acquires a clear image of the adsorbed chip A, i.e., the second image.

[0026] S3, Project the first image and the second image onto the same coordinate system, compare the alignment difference, locate the error component based on the alignment difference and calculate the adjustment compensation amount; First, the first and second images are projected onto the same coordinate system based on the known spatial geometric relationship (e.g., a 90° rotation) of the extraction mechanism as it rotates from the first position to the second position, making the two images comparable. Image processing techniques are used to extract the chip alignment features (e.g., edges, corners, or specific markings) from the two images, and these features are compared to calculate the positional and rotational angle deviations, i.e., the alignment differences. Then, based on the type of alignment difference (e.g., pure translation, pure rotation, or a combination of deviations) and its mapping relationship with the kinematic models of each moving component (stage 10, rotating shaft 40, rotating support arm 50, and core extractor 60), the main error components causing the deviation are located, and the required adjustment compensation amount (e.g., linear displacement or rotational angle value) for that component is calculated.

[0027] S4. Drive the corresponding error component to perform error compensation according to the adjusted compensation amount, and repeat steps S2 and S3 until the alignment difference is less than the preset threshold.

[0028] Based on the calculated adjustment compensation amount, the corresponding error components (such as stage 10, rotating shaft 40, or rotating support arm 50) are driven to perform fine-tuning movements to compensate for the identified deviations. After this compensation is completed, the system repeats steps S2 and S3: that is, the suction mechanism is driven to rotate to the second position again, the images of the two chips are acquired again, and the new alignment difference is calculated. The new alignment difference is compared with the preset threshold. If the accuracy requirement is not met, the error positioning, compensation calculation and execution are performed again based on the new difference data, forming a closed-loop control cycle of "acquisition-analysis-compensation-verification".

[0029] The iterative process continues until the alignment difference converges and is less than a preset threshold. At this point, the system determines that the visual alignment is complete and the two chips have reached a relative position state that allows for high-precision bonding.

[0030] The working principle of this invention is as follows: First, chip A is adsorbed at a first position by the suction mechanism, and chip B is adsorbed and positioned by the stage 10; then, the suction mechanism is driven to rotate to a second position, and a first image of chip B is acquired using the first image acquisition component 20, while a second image of chip A is acquired using the second image acquisition component 30; next, the first and second images are projected onto the same coordinate system to compare the alignment difference, and the error component is located based on the alignment difference and the adjustment compensation amount is calculated; according to the adjustment compensation amount, the corresponding error component is driven to perform error compensation, and the image acquisition, comparison, and compensation steps are repeated until the alignment difference is less than a preset threshold; this solution combines visual alignment with iterative compensation. After acquiring images of chip A and chip B, the images are projected onto the same coordinate system for difference comparison, thereby accurately locating the error component and calculating the compensation amount. The alignment error is gradually reduced through iterative adjustment until the preset accuracy requirement is met; this method significantly improves the chip placement alignment accuracy and realizes high-precision and high-efficiency alignment adjustment of the swing arm type chip mounter.

[0031] In this embodiment, the suction mechanism includes a rotating shaft 40 connected to a drive source; a rotating support arm 50 is fixedly connected to one end of the rotating shaft 40, and a core-picking component 60 is provided at the end of the rotating support arm 50 away from the rotating shaft 40. The core-picking component 60 is used to pick up and fix the chip A by vacuum adsorption.

[0032] It should be noted that the rotating shaft 40, directly connected to an external drive source (such as a servo motor), serves as the rotational center of the entire mechanism. A rotating support arm 50 is fixedly connected to one end of the rotating shaft 40, acting as a rigid component for load-bearing and transmission. A core-picking component 60 is located at the free end of the rotating support arm 50, away from the rotating shaft 40. This core-picking component 60 is a key terminal for performing the chip-grabbing function. It is connected to a negative pressure system via a built-in vacuum channel, enabling reliable vacuum adsorption to pick up and securely fix chip A. This structural design constitutes the functional design for chip picking, spatial transfer, and angular oscillation.

[0033] In this embodiment, the suction mechanism specifically includes the following components for adsorbing chip A at the first position: The rotating arm 50 is in the first horizontal position with its core-taking surface facing upwards; The chip A is transferred to the core-taking component 60 of the suction mechanism by an external transfer mechanism, and the core-taking component 60 fixes the chip A by vacuum adsorption.

[0034] This achieves accurate transfer and initial fixation of chip A from the external feeding system to the working head of this pick-and-place machine.

[0035] In this embodiment, step S2, driving the suction mechanism to rotate to the second position specifically involves: Rotate the support arm 50 90° around the rotation axis 40 so that the core extractor 60 and the chip A are in a vertical position; the first image acquisition component 20 is positioned above the stage 10 to acquire the first image, and the second image acquisition component 30 is positioned above the vertical core extractor 60 to acquire the second image.

[0036] It should be noted that the specific process is as follows: After the adsorption of chip A is completed, the rotating arm 50, driven by the drive source, rotates 90° around the rotating axis 40, moving from a horizontal state to a vertical state. This position is the second position for alignment imaging. In this state, the two imaging components work synchronously: the first imaging component 20, because its fixed installation position is directly above the stage 10, can vertically downward capture the first image of chip B on the stage 10; at the same time, the second imaging component 30 is specifically set in an orientation corresponding to the vertical position of the adsorption mechanism, so that its lens is directly facing the side of the vertically positioned core-taking component 60 and chip A, thereby capturing the second image of chip A. This design ensures that the two cameras can simultaneously capture the feature surface image most conducive to alignment analysis after a specific change in the chip's spatial position.

[0037] In this embodiment, specifically, projecting the first image and the second image onto the same coordinate system includes: Extract the alignment feature points of chip B in the first image and the alignment feature points of chip A in the second image; The coordinates of the alignment feature points of chip A are projected onto the same coordinate system as the alignment feature points of chip B, based on the known angular relationship of the suction mechanism rotating from the first position to the second position.

[0038] Several feature points (such as corners or the center of specific marks) of chip B are extracted from the first image, and corresponding feature points of chip A are extracted from the second image. Then, the key lies in utilizing a known and precise mechanical motion relationship—namely, the 90° angular transformation of the extraction mechanism from the first position (horizontal) to the second position (vertical)—to spatially transform the coordinates of the feature points of chip A (projection calculation). Through this step, the feature points of chip A, originally captured from different viewpoints and positions, are converted into the same two-dimensional coordinate system as the feature points of chip B.

[0039] In this embodiment, the specific process of comparing alignment differences is further explained as follows: The alignment feature points of chip A, after being projected onto the same coordinate system, are matched and compared with the alignment feature points of chip B. Calculate the positional deviations of the alignment feature points of chip A and chip B in the X and Y directions, as well as the rotational angle deviations between them; generate alignment difference data based on the positional and rotational angle deviations.

[0040] It's important to note that the quantitative comparison to generate specific deviation data involves two steps: First, feature point matching is performed, meaning that the projected feature points of chip A are correlated one-to-one with the feature points of chip B. Then, based on these matched point pairs, precise deviation calculations are performed. These calculations are primarily conducted in three degrees of freedom: the positional deviation in the X direction (horizontal translation), the positional deviation in the Y direction (vertical translation), and the rotational angle deviation (the deflection angle of chip A relative to chip B). Integrating these calculated translational and rotational amounts constitutes a complete, quantified alignment difference dataset. This data objectively describes the specific state and extent of misalignment between the two chips in their current state.

[0041] In this embodiment, the alignment difference positioning error component further includes: The judgment is based on the type of positional deviation and rotation angle deviation; If the alignment difference is mainly manifested as translational deviation in the X or Y direction, then the error component is determined to be the stage 10 or the rotation axis 40, which are related to linear motion. If the alignment difference is mainly manifested as a deviation in rotation angle, then the error determination component is the rotating support arm 50 or the core extractor 60, which are related to the rotation angle.

[0042] It should be noted that the intelligent judgment logic for diagnosing the source of mechanical errors based on deviation data is built upon the analysis of the correspondence between deviation types and the motion characteristics of mechanical components. Specifically: if the calculated alignment difference data mainly manifests as translational deviations in the X or Y directions, it indicates that the root cause of the problem lies in the inaccurate positioning or drift of the component performing linear motion. Therefore, the erroneous component is determined to be stage 10 (responsible for carrying the planar movement of chip B) or rotating shaft 40 (whose radial runout or axial movement may cause linear errors).

[0043] Conversely, if the alignment difference is mainly manifested as a rotation angle deviation, it indicates that the root cause of the problem lies in the angular deviation of the rotation-related components. Therefore, the error component is identified as the rotating support arm 50 (whose swing angle may be inaccurate) or the core extractor 60 (which may itself have a deflection installation error). This judgment logic enables rapid and accurate tracing from the "image deviation phenomenon" to the "specific deviation component".

[0044] In this embodiment, the specific process of calculating the adjustment compensation amount includes: Based on the identified error component type and its corresponding deviation data, calculations are performed using operational rules that match the component's motion characteristics. The purpose is to ensure that the calculation logic is consistent with the actual motion of the component, thereby guaranteeing the effectiveness and physical feasibility of the compensation command.

[0045] If the error component is determined to be related to the linear motion of the stage 10 or the rotation axis 40, then its adjustment compensation amount is the linear compensation displacement, which is calculated as follows: the position deviation in the X direction is taken as the X-axis compensation displacement of the stage 10, and the position deviation in the Y direction is taken as the Y-axis compensation displacement of the stage 10; or, the component of the position deviation in the radial direction of the rotation axis 40 is taken as the radial compensation displacement of the rotation axis 40. If the error component is determined to be the rotating arm 50 or the core extractor 60 related to the rotation angle, its adjustment compensation amount is the angle compensation angle. The calculation method is: directly use the rotation angle deviation as the angle compensation value of the rotating arm 50 or the core extractor 60 to correct its rotation angle around the rotation axis 40 or its own deflection angle. For linear motion components (stage 10 or rotating shaft 40), the compensation amount is linear displacement. During calculation, the pixel deviations in the X and Y directions obtained from image comparison are directly converted into physical displacement amounts using camera calibration parameters, and used as the compensation command for the corresponding axis of the component. For rotating components (rotating arm 50 or core extractor 60), the compensation amount is rotation angle. The calculation is more direct; typically, the chip rotation angle deviation obtained from image comparison is directly used as the angle command value for driving the component to perform reverse rotation correction.

[0046] The calculation rules also incorporate the known rotation angle of the absorbing mechanism from the first position to the second position, which is used to convert the pixel deviation in the image coordinate system into the actual physical displacement or angle in the coordinate system.

[0047] Since the alignment difference data is initially calculated based on image pixels, while the motion mechanism requires displacement or angle commands from the physical world, coordinate transformation is necessary. In this process, acquiring the known rotation angle (e.g., 90°) of the mechanism from the first position to the second position plays a crucial bridging role. The attitude transformation of chip A in space is defined, enabling the system to accurately decompose and map the two-dimensional pixel deviation on the image plane to the actual physical quantities corresponding to each motion axis (X-axis, Y-axis, rotation axis) in the world coordinate system.

[0048] In this embodiment, step S4 specifically includes: S41, based on the calculated adjustment compensation amount, drive the corresponding error component to perform compensation motion, the compensation motion includes driving the stage 10 or the rotating shaft 40 to perform linear displacement compensation, or driving the rotating support arm 50 or the core extractor 60 to perform angular deflection compensation. Specifically, if the error component is the stage 10 or the rotating shaft 40, the system controls its drive mechanism (such as a linear motor or servo motor) to perform linear displacement compensation, that is, to move a specified compensation distance in the X, Y, or radial directions. If the error component is the rotating arm 50 or the core extractor 60, the system controls its rotation drive source to perform angular deflection compensation, that is, to rotate around the axis by a specified compensation angle.

[0049] S42, After the compensation motion is completed, repeat steps S2 and S3 to obtain new alignment differences; After completing one physical compensation, the system does not directly confirm successful alignment. Instead, it re-executes the complete image acquisition and difference calculation process. That is, it drives the suction mechanism to rotate to the second position again, and the first and second image acquisition components 30 re-acquire the current images of chip B and chip A, and repeat the processes of projection, feature comparison, and deviation calculation to obtain new alignment difference data.

[0050] S43, determine whether the new alignment difference is less than a preset threshold; If so, then end the alignment process; If not, the new alignment difference is used as input to relocate the error component and calculate the new adjustment compensation amount, then return to step S41 for iterative compensation.

[0051] The logic is as follows: the newly obtained alignment difference data is compared with a pre-set threshold representing the accuracy standard. This is a clear binary decision: if the alignment difference is less than the threshold, it proves that after compensation, the alignment accuracy of the two chips meets the process requirements, and the system can end the alignment process and proceed to the subsequent bonding process. If the alignment difference is still greater than or equal to the threshold, it means that the accuracy does not meet the standard. The system uses the current new difference data as new input, returns and re-executes S41 and subsequent steps, starting a new round of the "positioning-calculation-compensation-verification" cycle.

[0052] Example 2: The present invention also provides a vision alignment system for a swing-arm type pick-and-place machine, which adopts the vision alignment method of the swing-arm type pick-and-place machine as described in Embodiment 1. The vision alignment system further includes a processing unit, which is electrically connected to the pick-up mechanism, the vision system, and the stage 10, and is configured to: control the vision system to acquire images, project the first image and the second image onto the same coordinate system for comparison to obtain alignment differences, locate the error component based on the alignment differences and calculate the adjustment compensation amount, and control the corresponding error component to perform compensation movement until the alignment difference is less than a preset threshold.

[0053] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A visual alignment method for a swing-arm type chip mounter, characterized in that, The swing-arm type placement machine includes a stage, a pick-and-place mechanism, and a vision system, wherein the vision system includes a first image acquisition component and a second image acquisition component; the method includes the following steps: S1, the suction mechanism adsorbs chip A at the first position, and the stage adsorbs and positions chip B. S2, drive the suction mechanism to rotate to the second position, acquire the first image of chip B through the first image acquisition component, and acquire the second image of chip A through the second image acquisition component; S3, Project the first image and the second image onto the same coordinate system, compare the alignment difference, locate the error component based on the alignment difference and calculate the adjustment compensation amount; S4, drive the corresponding error component to perform error compensation according to the adjusted compensation amount, and repeat steps S2 and S3 until the alignment difference is less than the preset threshold.

2. The visual alignment method for a swing-arm type chip mounter according to claim 1, characterized in that, The suction mechanism includes a rotating shaft connected to a drive source; A rotating support arm is fixedly connected to one end of the rotating shaft, and a core-picking component is provided at the end of the rotating support arm away from the rotating shaft. The core-picking component is used to pick up and fix chip A by vacuum adsorption.

3. The visual alignment method for a swing-arm type chip mounter according to claim 2, characterized in that, The suction mechanism for adsorbing chip A at the first position specifically includes: The rotating arm is in the first horizontal position with its core-taking surface facing upwards; Chip A is transferred to the core-taking component of the suction mechanism by an external transfer mechanism, and chip A is fixed by the core-taking component through vacuum adsorption.

4. The visual alignment method for a swing-arm type chip mounter according to claim 3, characterized in that, In step S2, driving the suction mechanism to rotate to the second position specifically involves: The rotating arm rotates 90° around the rotating axis, so that the core extractor and chip A are in a vertical position; the first image-capturing component is positioned above the stage to acquire a first image, and the second image-capturing component is positioned above the vertically positioned core extractor to acquire a second image.

5. The visual alignment method for a swing-arm type chip mounter according to claim 1, characterized in that, Projecting the first image and the second image onto the same coordinate system specifically includes: Extract the alignment feature points of chip B in the first image and the alignment feature points of chip A in the second image; The coordinates of the alignment feature point of chip A are projected into the same coordinate system as the alignment feature point of chip B, based on the known angular relationship of the absorption mechanism rotating from the first position to the second position.

6. The visual alignment method for a swing-arm type chip mounter according to claim 5, characterized in that, The specific process for comparing the alignment differences is as follows: The alignment feature points of chip A, after being projected onto the same coordinate system, are matched and compared with the alignment feature points of chip B. Calculate the positional deviations in the X and Y directions between the alignment feature points of chip A and chip B, as well as the rotational angle deviations between them; generate the alignment difference data based on the positional deviations and rotational angle deviations.

7. The visual alignment method for a swing-arm type chip mounter according to claim 6, characterized in that, The alignment difference positioning error component includes: The determination is based on the type of positional deviation and rotation angle deviation; If the alignment difference is mainly manifested as translational deviation in the X or Y direction, then the error component is determined to be the stage or rotation axis related to linear motion. If the alignment difference is mainly manifested as the rotation angle deviation, then the error determination component is the rotating support arm or core extractor related to the rotation angle.

8. The visual alignment method for a swing-arm type chip mounter according to claim 7, characterized in that, The specific process for calculating the adjustment compensation amount includes: Based on the identified error component type and its corresponding deviation data, calculations are performed using operation rules that match the motion characteristics of the component. If the error component is determined to be a stage or rotation axis related to linear motion, its adjustment compensation amount is a linear compensation displacement, which is calculated as follows: the X-direction position deviation is used as the X-axis compensation displacement of the stage, and the Y-direction position deviation is used as the Y-axis compensation displacement of the stage; or, the component of the position deviation in the radial direction of the rotation axis is used as the radial compensation displacement of the rotation axis. If the error component is determined to be a rotating arm or core extractor related to the rotation angle, its adjustment compensation amount is the angle compensation angle. The calculation method is: directly use the rotation angle deviation as the angle compensation value of the rotating arm or core extractor to correct its rotation angle around the rotation axis or its own deflection angle. The calculation rules also incorporate the known rotation angle of the suction mechanism from the first position to the second position, which is used to convert the pixel deviation in the image coordinate system into the actual physical displacement or angle in the coordinate system.

9. The visual alignment method for a swing-arm type chip mounter according to claim 1, characterized in that, Step S4 specifically includes: S41, based on the calculated adjustment compensation amount, drive the corresponding error component to perform compensation movement, the compensation movement including driving the platform or rotating shaft to perform linear displacement compensation, or driving the rotating support arm or core extractor to perform angular deflection compensation. S42, After the compensation motion is completed, repeat steps S2 and S3 to obtain new alignment differences; S43, determine whether the new alignment difference is less than the preset threshold; If so, then end the alignment process; If not, the new alignment difference is used as input to relocate the error component and calculate the new adjustment compensation amount, then the process returns to step S41 for iterative compensation.

10. A vision alignment system for a swing-arm type chip mounter, characterized in that, The visual alignment method of the swing-arm type placement machine as described in any one of claims 1 to 7 is further comprising a processing unit, wherein the processing unit is electrically connected to the pick-up mechanism, the vision system and the stage respectively, and is configured to: control the vision system to acquire images, project the first image and the second image onto the same coordinate system for comparison to obtain alignment differences, locate error components according to the alignment differences and calculate and adjust the compensation amount, and control the corresponding error components to perform compensation movements until the alignment differences are less than a preset threshold.

Images