A visually compensated riveting device and method for loosening screws

The visually compensated non-removable screw rivet device uses a vision module to calculate the center coordinates of the screw and stud, and applies pressure in stages and steps to solve the problem of eccentricity between the punch and the load-bearing base, thus achieving uniform riveting force transmission and unbiased meshing connection.

CN123102853APending Publication Date: 2026-11-17CHANGZHOU BORI ELECTRIC POWER AUTOMATION EQUIP +2
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611278746.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, visual positioning is based on the center of the movable head of the screw that cannot be easily removed, which causes the punch to be off-center from the force-bearing base, resulting in thread damage and skewed riveting.

Method used

A vision-compensated screw tightening device is adopted. The vision module acquires the image of the screw, calculates the coordinates of the stud center and the screw center, and the control module controls the moving platform and the pressurizing module in stages to guide the stud center to align with the punch and then align the screw center with the punch. The tightening is completed by applying pressure in stages.

Benefits of technology

It effectively solves the problems of thread damage and riveting misalignment caused by the eccentricity between the punch and the load-bearing base, ensuring that the riveting force is evenly transmitted along the axial direction, forming a non-eccentric meshing connection, and avoiding thread damage and riveting misalignment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123102853A_ABST
    Figure CN123102853A_ABST
Patent Text Reader

Abstract

This invention relates to the field of sheet metal joining technology, and more particularly to a riveting device and method for captive screws based on vision compensation. The device includes a moving platform, a vision module, a pressure module, and a control module. The moving platform carries the panel to be processed and the captive screw, and moves it horizontally. The vision module acquires an image of the captive screw and calculates the coordinates of the stud center and the screw center. The pressure module is located above the moving platform and includes a punch, a drive unit for raising and lowering the punch, and a pressure sensor for detecting downward pressure. The control module is signal-connected to the moving platform, the vision module, the drive unit, and the pressure sensor. This invention effectively solves the technical problem in the prior art where visual positioning is based on the center of the movable head of the captive screw, leading to eccentricity between the punch and the load-bearing substrate, causing thread damage and misaligned riveting.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of sheet metal connection technology, and in particular to a visually compensated non-removable screw expansion device and expansion method. Background Technology

[0002] Sheet metal joining is a general term for processes used to achieve load transfer and structural fixation between thin sheet metal parts, mainly including three categories: mechanical connection, welding, and bonding. In scenarios such as electronic equipment chassis and precision panels, where ease of assembly and disassembly and vibration resistance are highly demanding, detachable threaded connections and permanent mechanical cold connections are the most widely used. Among them, permanent mechanical cold connections (such as press riveting, expansion riveting, and flange riveting) utilize the plastic deformation of metal to achieve interlocking, forming high-strength inlay or interlocking anchor points on pre-punched holes without a heat source, thereby avoiding the impact of welding thermal deformation on the flatness of thin sheets.

[0003] To improve processing accuracy and efficiency, the industry often combines visual recognition with sheet metal joining equipment to automate mechanical cold joining. For example, patent document CN119281946B discloses a high-precision riveting device based on visual compensation. This device includes an X-axis transfer unit, a pick-up and riveting unit, a Y-axis transfer unit, and a hollow stage. The hollow stage is equipped with a rotating platform assembly with a side-pushing unit and a light-shielding unit. The hollow stage integrates a visual inspection unit composed of imaging components, and unidirectional offset detection units are also arranged at the positions of each moving axis. During operation, the workpiece is first loaded onto the feeding platform. The X-axis side push unit, Y-axis side push unit and L-shaped strip work together to complete the initial positioning of the workpiece. The vision inspection unit collects the image of the workpiece contour groove and identifies the cumulative tolerance of the workpiece in the X, Y and angular directions. Then, the X-axis transfer unit, Y-axis transfer unit and rotating platform group complete the corresponding position compensation. After that, the suction and riveting unit with negative pressure adsorption structure presses down to complete the riveting. The riveting pressure is controlled in real time by the external pressure sensor during the pressing process. During the operation of the equipment, the unidirectional offset detection unit continuously monitors the offset of each axis and stops the machine for visual correction when the tolerance is detected.

[0004] The aforementioned visual compensation device can achieve high-precision positioning and pressing when handling conventional one-piece rigid fasteners (such as ordinary press-fit nuts and press-fit studs). However, for captive screws, the screw shank is sleeved inside the stud body, and the two are supported by a spring, resulting in an inherent clearance. The screw shank is prone to wobbling and deflection. Under normal conditions, the visual inspection unit can usually only capture and identify the center of the top screw head and use this as a reference to move the center of the screw head directly under the punch for pressing. However, due to the objective existence of the clearance and tilt, the punch is only aligned with the center of the screw head, while the actual threaded sleeve base bearing the riveting force is in a deflected state. This incorrect positioning reference will cause the punch to eccentrically impact the thread, damaging it, resulting in unilateral stress on the flange, misalignment of the threaded sleeve during riveting, and gaps on the connection surface. Summary of the Invention

[0005] This invention provides a visually compensated riveting device for captive screws to solve the technical problem in the prior art where visual positioning is based on the center of the movable head of the captive screw, which leads to the eccentricity of the punch and the force-bearing base, causing thread damage and skewed riveting. The purpose of this invention is also to provide a visually compensated riveting method for captive screws.

[0006] To solve the above problems, the visual compensation-based non-removable screw rivet tightening device provided by the present invention adopts the following technical solution:

[0007] A vision-compensated riveting device for non-detachable screws includes a moving platform, a vision module, a pressure module, and a control module. The moving platform carries the panel to be processed and the non-detachable screw and moves in a horizontal plane. The vision module acquires an image of the non-detachable screw and calculates the coordinates of the stud center and the screw center. The pressure module is located above the moving platform and includes a punch, a drive unit for raising and lowering the punch, and a pressure sensor for detecting the downward pressure. The control module is signal-connected to the moving platform, the vision module, the drive unit, and the pressure sensor to control the moving platform to align the stud center with the punch, and to control the punch to press down to a first pressure to guide the non-detachable screw. Subsequently, the control module controls the moving platform to align the screw center with the punch, and to control the punch to continue pressing down to a second pressure to complete the riveting.

[0008] The beneficial effects of the vision-compensated non-removable screw rivet expansion device provided by this invention are: By setting up a vision module to acquire images of the lockout screw, the coordinates of the stud center and the screw center are calculated based on the image information. Instead of using the center of the screw head in a floating state as the default position representation of the stud base, the coordinates of the stud center and the screw center are extracted separately as positioning bases. The inherent eccentricity caused by the movement gap can be identified during the measurement process.

[0009] Based on this, the control module performs step-by-step control of the moving platform and the pressurizing module according to the coordinates of the stud center and the screw center. First, the moving platform is controlled to move the captive screw until its stud center coordinates are aligned with the punch. Then, the punch is controlled to press down to reach the first pressure, so that the lower end of the punch enters the inner hole of the stud, correcting the skewed stud base to be coaxial with the punch, and transferring the positioning reference from the movable screw head to the stud base that is actually under force. Next, with the stud already corrected, the moving platform is controlled to move the captive screw until its screw center coordinates are aligned with the punch. Then, the punch is controlled to continue pressing down to reach the second pressure to complete the riveting. At this time, the axis of the punch coincides with the axis of the stud base, and the riveting force is evenly transmitted axially to the flange. The flange opens symmetrically and embeds into the plate, forming a non-eccentric engagement connection, avoiding the technical problems of thread damage and riveting misalignment caused by the eccentricity of the punch and the force-bearing base.

[0010] In summary, this invention extracts the center coordinates of the stud and the screw of the captive screw using a vision module. Combined with the step-by-step alignment and graded pressure application logic executed by the control module, the positioning reference for riveting is corrected from the wobbling screw head to the stressed stud body. This effectively solves the technical problem in the prior art where visual positioning is based on the center of the movable head of the captive screw, leading to misalignment between the punch and the stressed body, causing thread damage and skewed riveting.

[0011] Furthermore, the lower end face of the punch is provided with a clearance hole for avoiding the screw head of a non-detachable screw.

[0012] Beneficial Effects: When the captive screw is in its free state, the screw head inside protrudes from the top of the stud under the action of the spring, and is in an oblique posture due to the existence of the play. If the lower end face of the punch is flat, when pressing down to perform the guiding action, the end face of the punch will directly contact the protruding and oblique screw head, which not only hinders the end of the punch from smoothly entering the inner hole of the stud to complete the guiding action, but also transmits the downward pressure to the threaded part of the screw head, causing the thread to be crushed or damaged. This application provides a clearance hole on the lower end face of the punch, so that when the punch is pressed down, the protruding screw head can be accommodated in the clearance hole, and the end face of the punch can pass over the screw head and act directly on the end face of the stud, thereby accurately transmitting the guiding force to the stud base, which not only protects the threads of the screw head from damage, but also ensures the effective execution of the guiding action, providing a correct posture basis for subsequent riveting.

[0013] Furthermore, the punch includes a forming section and a guiding section arranged sequentially from top to bottom; the forming section has a first conical surface for pressing the stud of the lockout screw under the second pressure, causing the stud to bulge outward; the guiding section has a second conical surface for extending into the inner hole of the lockout screw under the first pressure to guide the stud's posture; the cone angle of the first conical surface is greater than the cone angle of the second conical surface, and the cone angle of the first conical surface is less than or equal to the cone angle of the countersunk hole on the panel to be processed.

[0014] Beneficial effects: By setting a top-down forming section and a guiding section, and ensuring that the cone angle of the first cone surface of the forming section is greater than that of the second cone surface of the guiding section, when the punch presses down for guiding, the guiding section located at the lower end of the punch first contacts the stud that cannot be loosened. The second cone surface extends into the inner hole of the stud, and the relatively gentle slope of the cone surface contour guides and corrects the stud that is offset, correcting the stud's eccentric posture in advance and avoiding unilateral force on the stud during the subsequent forming and extrusion stage. After the guiding is completed, when riveting is performed, the punch continues to press down, and the first cone surface with a larger cone angle gradually fits against the outer wall of the stud, forming a uniform radial extrusion on the stud, causing the stud end to expand and fold outward evenly. The larger cone angle of the forming section results in a flatter extrusion contact surface, ensuring uniform force on the stud folding. After folding, it can fully fit with the countersunk hole of the panel, reducing problems such as local warping and excessive gaps in the fit. In addition, the cone angle of the first cone surface is less than or equal to the cone angle of the countersunk hole, so that the punch contacts and compacts the root of the flange first during the pressing process, thereby improving the flatness of the flange and the countersunk hole slope.

[0015] Furthermore, the mobile platform includes a linear drive module, a stage, and a clamping component; the clamping component is used to press the panel to be processed onto the stage to retract the spring-loaded screws to a preset reference surface.

[0016] Beneficial effects: By setting a clamping component on the moving platform, the panel to be processed can be firmly pressed onto the surface of the stage. The panel, with its countersunk end face as a reference, presses down on the studs of the non-loosening screws, forcing the studs, which were originally in a spring-floating state with uncertain axial position and perpendicularity, to retract to the preset reference surface. This pre-clamping process eliminates the random sway and height dispersion of the studs in the free state, ensuring that all studs on the same panel have a consistent initial position and attitude before riveting, thus providing a unified and repeatable reference condition for subsequent visual positioning and punch alignment.

[0017] Furthermore, the linear drive module includes an X-axis linear drive component and a Y-axis linear drive component arranged perpendicularly to each other.

[0018] To solve the above problems, the visual compensation-based method for tightening and riveting non-removable screws provided by this invention adopts the following technical solution: A method for tightening and riveting captive screws based on visual compensation, implemented using a visual compensation-based captive screw tightening and riveting device, includes the following steps: S1: Obtain an image of the screw that cannot be loosened by the rivet, and calculate the coordinates of the stud center and the screw center based on the image; S2: Control the movement of the captive screw to align the center coordinates of the stud with the punch; then, control the punch to press down, and stop pressing down when the pressing pressure reaches the first pressure, so that the punch enters the front end of the captive screw to achieve center alignment; S3: Control the movement of the lockout screw to align the screw's center coordinates with the punch; then, control the punch to continue pressing down to the second pressure, causing the stud of the lockout screw to expand and fit into the countersunk hole of the panel.

[0019] The beneficial effects of the visual compensation-based riveting method for non-removable screws provided by this invention are as follows: First, an image of the non-removable screw to be processed is acquired, and the coordinates of the stud center and the screw center are calculated separately to obtain two sets of independent positioning bases. In the alignment stage, the stud base that bears the riveting deformation is used as the positioning object. The workpiece movement is controlled to align the stud center with the punch. The punch is pressed down to the first pressure and then paused. The center alignment is completed by the punch extending into the front end of the stud. During alignment, the centering is completed with the force-bearing base as the reference, and the radial eccentricity between the punch and the stud base is eliminated in advance to avoid scraping the threads during the pressing down of the punch. At the same time, the skew posture of the stud is corrected. After the alignment is completed, the positioning reference is switched, and the workpiece movement is controlled to align the screw center with the punch. The punch continues to press down until the second pressure is reached to complete the forming process of the stud flanging and cracking and fitting into the countersunk hole of the panel. The entire processing distinguishes the positioning reference of the guiding process from that of the forming process. The guiding process ensures that the punch and the stud base are coaxial, and the forming process matches the position of the screw head. This avoids the eccentricity problem caused by using only the center of the movable head for positioning, thereby reducing thread damage and riveting misalignment.

[0020] In summary, the present invention effectively solves the technical problem in the prior art where visual positioning is based on the center of the movable head of the screw that cannot be loosened, resulting in the punch being eccentric to the force-bearing base, causing thread damage and riveting misalignment.

[0021] Further, in step S1, before acquiring the image of the screw that cannot be loosened by rivet expansion, the stud center reference point, the screw center reference point, and the corresponding stud rivet expansion reference point and screw rivet expansion reference point are first set in the calibration state. The stud rivet expansion reference point is the projection point of the stud center on the horizontal plane when the stud center moves to be aligned with the punch axis in the calibration state, and the screw rivet expansion reference point is the projection point of the screw center on the horizontal plane when the screw center moves to be aligned with the punch axis in the calibration state.

[0022] Beneficial effects: Before formally acquiring images of screws that are to be riveted but cannot be removed, the stud center reference point, screw center reference point, and corresponding stud riveting reference point and screw riveting reference point are pre-set. The stud center reference point and screw center reference point are the center positions of the circles identified in the visual calibration images, which serve as image comparison benchmarks for calculating the positional deviation of the screws to be riveted but cannot be removed. The stud riveting reference point and screw riveting reference point are taken from the horizontal plane projections when the stud center and screw center are aligned with the punch axis under calibration conditions, respectively. This establishes a positional mapping between the visual recognition coordinates and the actual working axis of the punch, avoiding the introduction of system deviations by directly setting motion coordinates manually. During the processing, the positional deviation of the workpiece is calculated using the stud center reference point and the screw center reference point. This deviation is then compensated to the corresponding riveting reference point. The calculated target riveting coordinates can match the punch working axis, reducing the positioning error caused by the coordinate conversion process. This ensures that the stud center is accurately aligned with the punch axis in the guiding process and that the screw center is aligned in the riveting forming process. This reduces the possibility of radial eccentricity between the workpiece and the punch, reduces the probability of thread damage and riveting misalignment, and improves the positioning reliability of batch processing under the sub-reference alignment mode.

[0023] Further, in step S1, after moving the captive screw to be riveted to the screw center reference point, its image is acquired and its screw center coordinates are calculated. The screw center coordinates are subtracted from the coordinates of the screw center reference point to obtain the first position deviation. After moving the captive screw to be riveted to the stud center reference point, its image is acquired and its stud center coordinates are calculated. The stud center coordinates are subtracted from the coordinates of the stud center reference point to obtain the second position deviation.

[0024] Beneficial Effects: By calculating the differences between the screw center coordinates and the screw center reference point coordinates, and the differences between the stud center coordinates and the stud center reference point coordinates, two independent positional deviations were obtained. In actual operation, the return-to-origin operation of the moving platform is affected by the sensor's sensing speed, causing a slight shift in the platform's reset position each time it starts. Simply relying on absolute value positioning commands cannot eliminate the impact of this deviation on riveting accuracy. By sequentially moving the screw to be riveted to the preset screw center reference point and stud center reference point during the calibration stage, taking photos, and calculating their respective coordinate differences, the systematic positioning deviation introduced by the moving platform's return-to-origin operation can be quantified into the respective reference systems of the screw and stud. The independent acquisition of the two deviation values ​​allows subsequent steps to use the deviation values ​​measured at their respective reference points for correction when compensating for the positioning of the stud and screw riveting reference points, without interference. This avoids the secondary errors that might be introduced by compensating for two reference points simultaneously with a single deviation value, thus achieving a unified and accurate transfer of positioning references between the vision system and the motion control system.

[0025] Further, in step S2, the coordinates of the stud riveting reference point are added to the second position deviation to obtain the stud target riveting coordinates. The screw to be riveted is moved to the stud target riveting coordinates to determine that the stud center coordinates are aligned with the punch. The coordinates of the screw riveting reference point are added to the first position deviation to obtain the screw target riveting coordinates. The screw to be riveted is moved to the screw target riveting coordinates to determine that the screw center coordinates are aligned with the punch.

[0026] Beneficial effects: Because the vision module and the punch are not spatially aligned, the vision module cannot directly observe the position of the punch's axis. Therefore, it is impossible to confirm whether the center of the stud or screw is aligned with the punch through real-time visual inspection. This application solves this problem by vector superimposing the coordinates of the rivet reference point stored during the calibration stage with the real-time measured position deviation. The calculated target rivet coordinates are then used as the positioning command for the motion control system. This transforms the alignment, which cannot be directly observed, into a definite coordinate value that can be executed by controlling the movement of the mobile platform. Thus, alignment judgment is indirectly achieved through open-loop positioning, enabling precise positioning without requiring the vision module and punch to be aligned.

[0027] Furthermore, the first pressure is less than the second pressure.

[0028] Beneficial effects: By setting the first pressure to be lower than the second pressure, a clear gradient relationship is formed in the loading conditions of the punch during the two pressing stages. The first pressure only needs to meet the pushing resistance required for the punch's guiding section to extend into the stud's inner hole. At this pressure level, the punch enters the stud's front end with a lower axial load, completing the correction and alignment of the stud's posture, without causing plastic deformation or end damage to the stud before alignment due to excessive pressure. After alignment is completed, a higher second pressure is used for flanging and riveting, causing the stud material to expand outwards uniformly and fit into the countersunk hole of the panel when the punch axis and the stud axis are nearly aligned. This effectively avoids the problems of insufficient alignment or uncontrolled flanging caused by using a single pressure to complete alignment and forming simultaneously, ensuring that each process is completed sequentially under its appropriate pressure conditions. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the visual compensation-based non-removable screw riveting device provided by the present invention. Figure 2 for Figure 1 The left view; Figure 3 for Figure 2 A magnified view of a portion of point A in the middle; Figure 4 for Figure 1 Top view; Figure 5 This is a schematic diagram of the structure of the punch provided by the present invention; Figure 6 for Figure 5 A magnified view of a portion of point A in the middle; Figure 7 Visual acquisition image of the stud profile; Figure 8 This is a visual image of the screw's outline. Figure 9 The flowchart shows the visual compensation-based riveting method for loose screws provided by this invention.

[0030] Explanation of reference numerals in the attached figures: 1. Mounting platform; 2. Moving platform; 21. X-axis linear drive assembly; 22. Y-axis linear drive assembly; 23. Stage; 24. Clamping component; 25. Clamping drive component; 26. Bearing plate; 3. Vision camera; 4. Pressurization module; 41. Punch; 411. Forming section; 412. Guide section; 413. Clearance hole; 42. Drive component; 43. Pressure sensor; 44. Rigid base; 5. Unremovable screw; 6. Panel. Detailed Implementation

[0031] The principles and spirit of the present invention will be explained in detail below with reference to several representative embodiments.

[0032] An embodiment of the visual compensation-based non-removable screw rivet device provided by the present invention: like Figures 1 to 5 As shown, the vision-compensated anti-detachment screw 5 riveting device includes a moving platform 2, a vision module, a pressure module 4, and a control module. The moving platform 2, vision module, and pressure module 4 are all connected to the control module via signals. The control module coordinates the timing of the actions of each module to complete the entire process of calibration and riveting of the anti-detachment screw 5.

[0033] like Figures 1 to 4 As shown, in this embodiment, the mobile platform 2 includes a linear drive module, a stage 23, a support plate 26, a clamping drive component 25, and a clamping component 24.

[0034] like Figure 1 and Figure 4 As shown, in this embodiment, the linear drive module includes an X-axis linear drive assembly 21 and a Y-axis linear drive assembly 22 that are perpendicular to each other. The X-axis linear drive assembly 21 is fixedly mounted on the mounting platform 1, and the Y-axis linear drive assembly 22 is assembled at the output end of the X-axis linear drive assembly 21. The stage 23 is fixedly mounted at the output end of the Y-axis linear drive assembly 22. The X-axis linear drive assembly 21 and the Y-axis linear drive assembly 22 move together, which can drive the stage 23 to complete two-dimensional translation in the horizontal plane.

[0035] like Figure 1 and Figure 2As shown, in this embodiment, the clamping drive 25 is installed on the stage 23, and the clamping member 24 is disposed above the stage 23 and connected to the output end of the clamping drive 25. Under the action of the clamping drive 25, it can move vertically relative to the stage 23. The stage 23 is also equipped with a support plate 26 for supporting the panel 6 to be processed. When the panel 6 to be processed is placed on the support surface of the support plate 26, the clamping member 24 can move down to clamp and fix the panel 6 to be processed, compress the spring structure inside the non-detachable screw 5, and retract the non-detachable screw 5, which is in a spring floating state, to a preset reference surface, thereby eliminating the influence of the axial floating of the screw on the visual recognition accuracy and the quality of the riveting process.

[0036] In this embodiment, the movement distance is determined by controlling the pulses of the X-axis linear drive component 21 and the Y-axis linear drive component 22.

[0037] In this embodiment, both the X-axis linear drive assembly 21 and the Y-axis linear drive assembly 22 are linear modules; the clamping drive component 25 is a rotary cylinder; and the clamping component 24 is a pressure plate. In other embodiments, the X-axis linear drive assembly 21 and the Y-axis linear drive assembly 22 may also be synchronous belt modules; the clamping drive component 25 may also be a standard cylinder, hydraulic cylinder, or linear motor; and the clamping component 24 may be configured according to the shape of the panel 6 as an annular pressure plate, multi-point distributed independent pressure claws, or contoured pressure blocks, etc., as long as uniform clamping of the panel 6 is achieved.

[0038] like Figure 1 , Figure 2 and Figure 4 As shown, in this embodiment, the vision module is fixedly installed on the mounting platform 1 and located above the right end of the X-axis linear drive assembly 21. The vision module includes a vision camera 3, with the imaging end of the vision camera 3 facing downwards. The vision module can acquire images of the end face of the captive screw 5 placed on the support plate 26, and process and calculate the acquired images to identify and solve the coordinates of the stud center and the screw center of the captive screw 5. At the same time, it can convert the actual millimeter-level coordinates into control values ​​of the corresponding pulse equivalent and transmit them to the control module, providing a data basis for subsequent alignment compensation.

[0039] In this embodiment, the visual camera 3 uses a nine-point calibration method to determine the pixel equivalent between the visual coordinate system and the motion coordinate system, thereby determining the actual physical distance corresponding to each pixel. In other embodiments, a four-point calibration method, a multi-point calibration method, or an automatic calibration method based on a standard calibration board can also be used, as long as an accurate mapping relationship between the visual coordinate system and the motion coordinate system is established.

[0040] In other embodiments, the visual camera 3 can also be mounted on an independent adjustable bracket, and the height and horizontal position of the camera can be finely adjusted by adjusting the bracket to adapt to the shooting needs of different sizes of non-removable screws 5.

[0041] like Figure 1 , Figure 2 and Figure 4 As shown, in this embodiment, the pressurizing module 4 is fixedly installed on the mounting platform 1, located at the right end of the X-axis linear drive assembly 21 and below the vision module. The projections of the pressurizing module 4 and the vision module on the horizontal plane are misaligned. The pressurizing module 4 includes a drive component 42, a pressure sensor 43, a punch 41, and a rigid base 44. The output end of the drive component 42 is arranged vertically. The pressure sensor 43 is assembled between the output end of the drive component 42 and the punch 41. The punch 41 is connected to the lower end of the pressure sensor 43. The drive component 42 can drive the punch 41 to complete the downward pressing and lifting actions in the vertical direction. The pressure sensor 43 can collect the downward pressure value applied by the punch 41 in real time and feed it back to the control module. The rigid base 44 is arranged below the punch 41 and is used to provide rigid support for the bearing plate 26 during riveting.

[0042] In this embodiment, the drive unit 42 is a servo electric cylinder, which can receive pulse commands and complete precise lifting control. In other embodiments, the drive unit 42 can also be a hydraulic cylinder, a linear motor, or a stepper motor with lead screw transmission, as long as it can output controllable downward pressure and displacement according to control commands.

[0043] like Figure 5 and Figure 6 As shown, in this embodiment, the punch 41 includes a forming section 411 and a guiding section 412 arranged sequentially from top to bottom. The forming section 411 is provided with a first conical surface, and the guiding section 412 is provided with a second conical surface. The cone angle of the first conical surface is greater than the cone angle of the second conical surface and less than the cone angle of the countersunk hole on the panel to be processed. A clearance hole 413 is provided on the lower end face of the punch 41. The clearance hole 413 can accommodate the screw head of the non-removable screw 5 during the downward pressing of the punch 41, avoiding damage to the part caused by the punch 41 directly squeezing the screw head. The guiding section 412 is located in the lower region of the punch 41. During the downward movement of the punch 41, the guiding section 412 first contacts the inner hole of the stud, and corrects the stud posture by relying on the guiding effect of the second conical surface. The forming section 411 is located above the guiding section 412. When the punch 41 continues to move downward, the first conical surface contacts the end of the stud, applying radial extrusion force to the stud to cause the stud to fold outward.

[0044] In this embodiment, the cone angle of the first cone surface is 90°, and the cone angle of the second cone surface is 60°. In other embodiments, the cone angle of the first cone surface can be selected from the range of 80° to 120°, and the cone angle of the second cone surface can be selected from the range of 45° to 75°. The specific angle values ​​can be adapted and adjusted according to the specifications, material, and flange requirements of the screw 5 to be expanded and loosened. It is only necessary to ensure that the cone angle of the first cone surface is greater than the cone angle of the second cone surface and less than the cone angle of the countersunk hole on the panel to be processed.

[0045] It should be noted that the experiment found that if the cone angles of the first and second conical surfaces are the same and both are 90°, then during riveting, the second conical surface is difficult to penetrate into the inner hole of the stud, the flange material is prone to folding inward, and the edge of the countersunk hole cannot form a complete flange, resulting in poor riveting quality. If the cone angles of the first and second conical surfaces are the same and both are 60°, then during riveting, the cone angle of the first conical surface does not match the cone angle of the countersunk hole, the flange cannot be flush with the inclined surface of the countersunk hole, and the head of the non-detachable screw is prone to protrusion or loosening after installation, and the riveting quality also cannot meet the requirements. In this embodiment, the cone angle of the first cone surface is greater than that of the second cone surface, and the cone angle of the first cone surface is less than or equal to the cone angle of the countersunk hole on the panel to be processed. When the punch 41 presses down, the second cone surface can first insert into the stud of the non-removable screw 5 to straighten the stud that is offset, correct the eccentric posture of the stud in advance, and avoid the stud being subjected to force on one side in the subsequent forming and extrusion stage. Subsequently, the first cone surface gradually fits against the outer wall of the stud, forming a uniform radial extrusion on the stud, causing the end of the stud to expand and turn outward evenly.

[0046] In this embodiment, the control module has visual communication, motion control and pressure synchronization processing functions. It can receive pulse coordinate data transmitted by the vision module and pressure data fed back by the pressure sensor 43, and output pulse control commands after completing coordinate compensation calculations to drive each actuator to act according to a preset timing sequence.

[0047] The working principle of the visual compensation-based non-removable screw rivet device provided by this invention is as follows: Before riveting the standard captive screw 5, a calibration operation is performed. The calibration workpiece, loaded with the standard captive screw 5, is placed on the support plate 26. The clamping component 24 clamps the panel 6 of the calibration workpiece, causing the captive screw 5 to retract to a preset reference surface. The control module drives the linear drive module to move the calibration workpiece into the field of view of the vision module. The vision module acquires an image of the captive screw 5 in the calibration state (e.g., ...). Figure 7 and Figure 8As shown in the diagram, the coordinates of the stud center reference point and the screw center reference point are calculated and stored in the control module. The control module continues to drive the linear drive module to move the calibration workpiece, and observes the position of the screw center with the naked eye or through a camera or other vision device, until the screw center in the calibration state moves to a position aligned with the axis of the punch 41. The projection point of the screw center on the horizontal plane at this time is recorded as the screw expansion reference point. Subsequently, the control module drives the linear drive module to move again, so that the stud center in the calibration state moves to a position aligned with the axis of the punch 41. The projection point of the stud center on the horizontal plane at this time is recorded as the stud expansion reference point, completing the storage of all calibration parameters.

[0048] When the center of the stud is located at the stud expansion reference point, the control drive 42 drives the punch 41 to press down at a speed of 3mm / s. The change in pressure value monitored by the pressure sensor 43 is observed. When the pressure value changes slightly, the pressing movement of the drive 42 is paused. It is observed whether the punch 41 contacts the stud of the captive screw 5 and generates compression. If it has contacted the stud of the captive screw 5 and generated compression, the current pressure F1 is recorded. Then, move the captive screw 5 so that its center coincides with the stud expansion reference point. Then, control the drive component 42 to drive the punch 41 to continue pressing down at a speed of 2 mm / s. When the pressure value monitored by the pressure sensor 43 rises in a curve and the pulse distance of the drive component 42 changes from a steady increase to a slow increase, stop the movement of the drive component 42 and observe whether the stud of the captive screw 5 has fully opened. If it has expanded, it means that the pressure F2 at this time is the control target value. That is, at this time, the stroke of the punch 41 is physically limited by the countersunk end face. The axial force applied by the drive component 42 is only applied to the already mated metal contact surface and is no longer converted into plastic deformation of the stud material. In this state, although the F value detected by the pressure sensor 43 can continue to increase under the pressing action of the drive component 42, the pressing position of the punch 41 can no longer advance. Continuing to pressurize will not help to further change the geometry of the stud, but may damage the mold or the thread. At this time, the first pressure (F1) and the second pressure (F2) have also been calibrated.

[0049] During formal processing, the panel 6 to be processed, equipped with the anti-detachable screw 5 to be riveted, is placed at the designated position on the support plate 26. The clamping member 24 moves downward to press the panel 6, causing the anti-detachable screw 5 to retract to the preset reference surface. The control module drives the linear drive module to translate the panel 6, first moving the anti-detachable screw 5 to the position corresponding to the screw center reference point. The vision module acquires an image of the end face of the anti-detachable screw 5 at this position and calculates the real-time screw center coordinates. The control module performs a difference calculation between the real-time screw center coordinates and the coordinates of the screw center reference point to obtain the first position deviation. Subsequently, the linear drive module continues to translate the panel 6, moving the anti-detachable screw 5 to the position corresponding to the stud center reference point. The vision module again acquires an end face image, calculates the real-time stud center coordinates, and the control module performs a difference calculation between the real-time stud center coordinates and the coordinates of the stud center reference point to obtain the second position deviation.

[0050] After completing the deviation calculation, the control module performs coordinate compensation calculation, adding the coordinates of the stud rivet reference point to the second position deviation to obtain the target rivet coordinates of the stud.

[0051] Subsequently, the control module drives the linear drive module to translate according to the target riveting coordinates of the stud, moving the captive screw 5 to the target riveting coordinates. At this point, the center of the stud is aligned with the axis of the punch 41. The control module controls the drive component 42 to move the punch 41 downward. The pressure sensor 43 monitors the downward pressure value in real time. When the downward pressure reaches the first pressure, the drive component 42 stops outputting, and the punch 41 maintains its current downward position. The guide section 412 of the punch 41 extends into the inner hole of the stud front end of the captive screw 5, and corrects the eccentric posture of the stud by relying on the guiding effect of the second conical surface, thus completing the center alignment.

[0052] After the alignment process is completed, the control module adds the coordinates of the screw riveting reference point to the first position deviation to obtain the target riveting coordinates of the screw. The control module drives the linear drive module to move the panel 6 to be processed to the target riveting coordinates of the screw. At this time, the center of the screw is aligned with the axis of the punch 41. The control module controls the drive component 42 to continue to drive the punch 41 downward. When the pressure sensor 43 detects that the downward pressure reaches the second pressure, the drive component 42 stops pressing down. The first conical surface of the forming section 411 of the punch 41 applies a uniform extrusion force to the end of the stud, causing the stud to expand outward and tightly fit into the inner wall of the countersunk hole of the panel 6, completing the riveting process of a single non-removable screw 5. After the processing is completed, the drive component 42 drives the punch 41 to lift and reset, the clamping component 24 releases the panel 6 to be processed, and the linear drive module moves the carrier plate 26 out of the processing station. After changing the workpiece, the above process is repeated to complete the riveting operation of batch workpieces.

[0053] An embodiment of the visual compensation-based riveting method for loose screws provided by this invention: like Figure 9 As shown, the vision-compensated screw tightening method is implemented using a vision-compensated screw tightening device, and includes the following steps: S1: Obtain an image of the screw that cannot be loosened by the rivet, and calculate the coordinates of the stud center and the screw center based on the image; S2: Control the movement of the captive screw to align the center coordinates of the stud with the punch; then, control the punch to press down, and stop pressing down when the pressing pressure reaches the first pressure, so that the punch enters the front end of the captive screw to achieve center alignment; S3: Control the movement of the lockout screw to align the screw's center coordinates with the punch; then, control the punch to continue pressing down to the second pressure, causing the stud of the lockout screw to expand and fit into the countersunk hole of the panel.

[0054] In step S1, before acquiring the image of the screw that cannot be loosened by rivet expansion, the stud center reference point, the screw center reference point, and the corresponding stud rivet expansion reference point and screw rivet expansion reference point are first set in the calibration state. The stud rivet expansion reference point is the projection point of the stud center on the horizontal plane when the stud center moves to be aligned with the punch axis in the calibration state. The screw rivet expansion reference point is the projection point of the screw center on the horizontal plane when the screw center moves to be aligned with the punch axis in the calibration state.

[0055] In this embodiment, the calibration state refers to the preparatory operation process carried out before formal mass production to establish all the reference parameters required for subsequent visual compensation and motion control.

[0056] When acquiring the stud center reference point, screw center reference point, stud expansion reference point, and screw expansion reference point of the lock-up screw under calibration conditions, the calibration workpiece loaded with the standard lock-up screw is placed in the working position, and a clamping force is applied to the panel to retract the lock-up screw to the preset reference surface. The calibration workpiece is then translated to the visual acquisition position, and an image of the lock-up screw under calibration conditions is acquired visually. The coordinates of the stud center reference point and the screw center reference point are calculated and stored.

[0057] Continue translating the calibrated workpiece, determining the position of the screw center by visual inspection or observation, until the screw center in the calibrated state is aligned with the punch axis. Record the projection point of the screw center on the horizontal plane at this point as the screw expansion reference point. Translate the calibrated workpiece again until the stud center in the calibrated state is aligned with the punch axis, and record the projection point of the stud center on the horizontal plane at this point as the stud expansion reference point.

[0058] When the stud center is located at the stud rivet reference point, the punch is controlled to press down at the first speed while monitoring the pressure value. When the pressure value changes, pressing is paused, and it is observed whether the punch has contacted the stud and generated compression. If contact has occurred, the current pressure is recorded as the first pressure. Then, the stud center is aligned with the stud rivet reference point again, and the punch is controlled to continue pressing down at the second speed, where the second speed is less than or equal to the first speed. When the pressure value increases in a curve and the travel distance of the drive mechanism changes from a steady increase to a slow increase, pressing is paused, and it is observed whether the stud is fully opened. If riveted, the current pressure is recorded as the second pressure. This completes the storage of all calibration parameters. Subsequently, the riveting of the captive screw on the panel can be performed based on the calibrated stud center reference point, screw center reference point, stud rivet reference point, and screw rivet reference point.

[0059] In step S1, after moving the captive screw to be riveted to the screw center reference point, its image is acquired and its screw center coordinates are calculated. The screw center coordinates are subtracted from the coordinates of the screw center reference point to obtain the first position deviation. After moving the captive screw to be riveted to the stud center reference point, its image is acquired and its stud center coordinates are calculated. The stud center coordinates are subtracted from the coordinates of the stud center reference point to obtain the second position deviation.

[0060] In step S2, the coordinates of the stud riveting reference point are added to the second position deviation to obtain the stud target riveting coordinates. The screw to be riveted is moved to the stud target riveting coordinates to determine that the stud center coordinates are aligned with the punch. The coordinates of the screw riveting reference point are added to the first position deviation to obtain the screw target riveting coordinates. The screw to be riveted is moved to the screw target riveting coordinates to determine that the screw center coordinates are aligned with the punch.

[0061] Furthermore, it should be noted that in this embodiment, the first pressure is less than the second pressure.

[0062] Based on the above description in this specification, those skilled in the art will also understand that the following terms, such as "upper," "lower," "front," "back," "left," "right," "inner," and "outer," which indicate orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings of this specification and should not be understood or interpreted as limiting the present invention.

[0063] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. A visually compensated screw tightening device, characterized in that, The system includes a moving platform, a vision module, a pressurizing module, and a control module. The moving platform carries the panel to be processed and the captive screws and moves them horizontally. The vision module acquires images of the captive screws and calculates the coordinates of the stud center and the screw center. The pressurizing module is located above the moving platform and includes a punch, a drive unit for raising and lowering the punch, and a pressure sensor for detecting the downward pressure. The control module is signal-connected to the moving platform, the vision module, the drive unit, and the pressure sensor to control the moving platform to align the stud center coordinates with the punch, and to control the punch to press down to a first pressure to guide the captive screws. Subsequently, the control module controls the moving platform to align the screw center coordinates with the punch, and to control the punch to continue pressing down to a second pressure to complete the riveting process.

2. The visual compensation-based non-removable screw rivet tightening device according to claim 1, characterized in that, The lower end face of the punch is provided with a clearance hole for avoiding the screw head of a screw that cannot be removed.

3. The visually compensated non-removable screw rivet tightening device according to claim 1 or 2, characterized in that, The punch includes a forming section and a guiding section arranged sequentially from top to bottom; the forming section has a first conical surface for pressing the stud of the lockout screw under the second pressure, causing the stud to fold outward; the guiding section has a second conical surface for extending into the inner hole of the lockout screw under the first pressure to guide the stud's posture. The cone angle of the first cone surface is greater than that of the second cone surface, and the cone angle of the first cone surface is less than or equal to the cone angle of the countersunk hole on the panel to be processed.

4. The visually compensated non-removable screw rivet tightening device according to claim 1 or 2, characterized in that, The mobile platform includes a linear drive module, a stage, and a clamping component; the clamping component is used to press the panel to be processed onto the stage to retract the spring-loaded screws to a preset reference surface.

5. The visually compensated non-removable screw rivet tightening device according to claim 4, characterized in that, The linear drive module includes an X-axis linear drive component and a Y-axis linear drive component arranged perpendicularly to each other.

6. A method for tightening and riveting non-removable screws based on visual compensation, characterized in that, The implementation using the vision-compensated non-removable screw rivet device according to any one of claims 1 to 5 includes the following steps: S1: Obtain an image of the screw that cannot be loosened by the rivet, and calculate the coordinates of the stud center and the screw center based on the image; S2: Control the movement of the captive screw to align the center coordinates of the stud with the punch; then, control the punch to press down, and stop pressing down when the pressing pressure reaches the first pressure, so that the punch enters the front end of the captive screw to achieve center alignment; S3: Control the movement of the lockout screw to align the screw's center coordinates with the punch; then, control the punch to continue pressing down to the second pressure, causing the stud of the lockout screw to expand and fit into the countersunk hole of the panel.

7. The method for riveting a non-detachable screw based on visual compensation according to claim 6, characterized in that, In step S1, before acquiring the image of the screw that cannot be loosened by rivet expansion, the stud center reference point, the screw center reference point, and the corresponding stud rivet expansion reference point and screw rivet expansion reference point are first set in the calibration state. The stud rivet expansion reference point is the projection point of the stud center on the horizontal plane when the stud center moves to be aligned with the punch axis in the calibration state. The screw rivet expansion reference point is the projection point of the screw center on the horizontal plane when the screw center moves to be aligned with the punch axis in the calibration state.

8. The method for riveting a non-detachable screw based on visual compensation according to claim 6, characterized in that, In step S1, after moving the captive screw to be riveted to the screw center reference point, its image is acquired and its screw center coordinates are calculated. The screw center coordinates are subtracted from the coordinates of the screw center reference point to obtain the first position deviation. After moving the captive screw to be riveted to the stud center reference point, its image is acquired and its stud center coordinates are calculated. The stud center coordinates are subtracted from the coordinates of the stud center reference point to obtain the second position deviation.

9. The method for riveting a non-detachable screw based on visual compensation according to claim 6, characterized in that, In step S2, the coordinates of the stud riveting reference point are added to the second position deviation to obtain the stud target riveting coordinates. The screw to be riveted is moved to the stud target riveting coordinates to determine that the stud center coordinates are aligned with the punch. The coordinates of the screw riveting reference point are added to the first position deviation to obtain the screw target riveting coordinates. The screw to be riveted is moved to the screw target riveting coordinates to determine that the screw center coordinates are aligned with the punch.

10. The method for riveting a non-detachable screw based on visual compensation according to claim 6, characterized in that, The first pressure is less than the second pressure.

Citation Information

Patent Citations

  • A high-precision riveting equipment based on visual compensation

    CN119281946B