Turnover alignment control method and system of single / double-sided turnover plate machine

Abstract

The application relates to the technical field of plate turning alignment control, and provides a turning alignment control method and system of single / double-sided plate turning machines. The method comprises the following steps: performing structure analysis on a turning target, identifying alignment key points and structure features; establishing a pre / post-turning coordinate transformation relationship, calculating a post-turning theoretical position based on a current pose; predicting a turning deviation based on the theoretical position, analyzing compensation parameters and controlling compensation when the deviation exists; obtaining timing process alignment parameters, using the timing process alignment parameters for secondary alignment of the turning target, confirming completion and connecting the next process. The application solves the technical problems that, in the single / double-sided turning of traditional plate turning machines, the structure difference or inaccurate target pose recognition leads to low alignment accuracy and the turning deviation cannot be effectively compensated, and achieves the technical effects of improving the turning alignment accuracy, realizing automatic process connection and the like through secondary alignment and process connection processing.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of plate turning alignment control, in particular to a turning alignment control method and system for single / double-sided plate turning machines. BACKGROUND

[0002] As the core equipment for processes such as double-sided welding of printed circuit boards (PCB), double-sided coating of glass panels and positive / negative processing of precision metal parts, the turning alignment accuracy of single / double-sided plate turning machines directly affects product yield and production line efficiency. Traditional plate turning machines generally use mechanical positioning pins and fixed program control to achieve turning, but when faced with new workpieces with ultra-thin and high-density development, three major technical bottlenecks are exposed: first, the complexity of workpiece structure leads to the failure of traditional mechanical positioning points, and changes in working conditions such as edge warping and thermal deformation cause cumulative errors in the mapping of the fixed coordinate system; second, dynamic deviations caused by inertia impact and transmission clearance during the turning process cannot be covered by static parameter compensation models, and errors are amplified step by step when multiple processes are connected; third, the existing system lacks predictive coordination of positioning needs in subsequent processes, resulting in systematic deviations in the coordinate system matching of turning alignment and downstream equipment. Therefore, developing an intelligent turning alignment control system that integrates feature recognition, pose prediction, dynamic compensation and process coordination is of great significance for improving the automation level of the production line and ensuring product quality consistency. SUMMARY

[0003] The application provides a turning alignment control method and system for single / double-sided plate turning machines, aiming to solve the technical problem of low alignment accuracy and ineffective compensation of turning deviation caused by structural differences or inaccurate target pose recognition when traditional plate turning machines are single / double-sided turned.

[0004] The first aspect of the application provides a turning alignment control method for single / double-sided plate turning machines, which includes: performing structural analysis on the turning target, identifying alignment key points and structural features; establishing a pre / post-turning coordinate transformation relationship, calculating the theoretical position after turning based on the current identified pose, taking the alignment key points and structural features as targets; predicting the turning deviation based on the theoretical position after turning, and when the prediction result has deviation, performing turning parameter compensation analysis according to the prediction deviation, and performing turning deviation compensation control according to the compensation parameters; obtaining timing process alignment parameters, using the timing process alignment parameters to perform secondary alignment on the target after turning, determining the alignment target completion, and entering the next process connection processing.

[0005] Another aspect of the present application provides a turnover alignment control system of a single / double-sided turnover plate machine, which comprises: a structure analysis module that analyzes the structure of a turnover target and identifies alignment key points and structural features; a theoretical position calculation module that establishes a coordinate transformation relationship before / after turnover, calculates a theoretical position after turnover based on a current identified pose and the alignment key points and structural features; a deviation compensation control module that predicts a turnover deviation based on the theoretical position after turnover, performs turnover parameter compensation analysis according to a predicted deviation amount when the prediction result has a deviation, and performs turnover deviation compensation control according to compensation parameters; and a secondary alignment module that obtains timing process alignment parameters, performs secondary alignment of the target after turnover using the timing process alignment parameters, and determines that the alignment target is completed to enter a next process connection process.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0007] The turnover alignment control method of the single / double-sided turnover plate machine first analyzes the structure of a turnover target, identifies key alignment points and structural features, then establishes a transformation relationship between the coordinates before and after turnover, calculates an ideal position after turnover based on a current target pose, predicts possible deviations in the turnover process based on this, and adjusts the turnover parameters according to the prediction result to correct the deviations, finally obtains alignment parameters in a timing process, and uses these parameters to perform secondary accurate alignment of the target after turnover to ensure that the target position after turnover meets the requirements, finally completes alignment and prepares for the smooth connection of subsequent processes.

[0008] The above description is only a summary of the technical solutions of the present application, in order to more clearly understand the technical means of the present application, the specific embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0010] Figure 1 A flowchart of a turnover alignment control method of a single / double-sided turnover plate machine in an embodiment.

[0011] Figure 2 A turnover alignment control system architecture diagram of a single / double-sided turnover plate machine in an embodiment.

[0012] Explanation of reference numerals in the attached diagram: 11 Structural analysis module, 12 Theoretical position calculation module, 13 Deviation compensation control module, 14 Secondary alignment module. Detailed Implementation

[0013] This application provides a flipping alignment control method and system for single and double-sided flipping machines, which solves the technical problem that traditional flipping machines have low alignment accuracy and cannot effectively compensate for flipping deviations due to structural differences or inaccurate target pose recognition when flipping single and double sides.

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0015] It should be noted that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0016] Example 1, as Figure 1 As shown, this application provides a flipping and alignment control method for a single- or double-sided flipping machine, the method comprising:

[0017] Structural analysis is performed on the flipped target to identify key alignment points and structural features.

[0018] In this embodiment, when performing structural analysis on the flipped target, the structural information of the target object is first collected, including its edge contour and calibration features (such as positioning holes and axes of symmetry), which play a role in positioning and guidance during the subsequent flipping process. Then, by analyzing these structural features, the influence evaluation value of each point and the stability of the flipped structure are calculated, providing support for the alignment accuracy during the flipping process. Afterwards, the calculation results are filtered using a set threshold to obtain points that meet the threshold requirements. These points are then used as alignment key points, and the structural features corresponding to these points serve as the structural features of the alignment key points, providing a basis for subsequent deviation prediction and ensuring the accuracy of the target position during the flipping process.

[0019] Furthermore, this application provides structural analysis of the flipped target to identify alignment key points and structural features, including:

[0020] Collecting structure information of the overturning target; identifying edge contour features in the structure information, and calibrating coordinates of the identified features, wherein the calibration-identified features include positioning holes, symmetry axes, identification marks, edge intersection points, and color adjacency; determining the alignment key points and structure features based on the edge contour features and the calibration-identified feature coordinates.

[0021] Preferably, a sensor or an imaging device (such as a laser scanner, a stereo vision camera, etc.) is used to collect three-dimensional data and surface features of the overturning target, to obtain structure information of the overturning target, which contains basic information such as the shape, size, and surface features (such as patterns, colors) of the object. Subsequently, an edge detection algorithm such as Canny edge detection, Sobel operator, etc. is used to extract edge contour features from the collected structure information of the target object. Taking Canny edge detection as an example, the gradient value of each data point in the structure information is calculated using the Sobel operator to obtain the gradient amplitude in each direction (horizontal and vertical) of the structure information. After obtaining the gradient amplitude, a non-maximum suppression algorithm is used to suppress non-edge points and retain the maximum value of the edge intensity, so that the edge lines in the image become clearer. Then, by setting high and low thresholds, the obvious edges are further filtered out. The data points higher than the high threshold are considered as strong edges, the data points lower than the low threshold are considered as non-edges, and the data points between the two thresholds are retained as edges if they are connected to the strong edges, so as to obtain the complete object edges to form the edge contour features and reflect the geometric shape of the object. In addition, the coordinates of the calibration-identified features, including the coordinates of the positioning holes, the coordinates of the symmetry axes, the coordinates of the identification marks, the coordinates of the edge intersection points, and the coordinates of the color adjacency, etc. are obtained in the structure information of the object. The coordinates of the positioning holes and the identification marks can be obtained by template matching; the coordinates of the symmetry axes can be estimated by least squares method; the coordinates of the edge intersection points can be obtained by analyzing the edge contour of the object and using an angle point detection algorithm (such as Harris corner detection) to identify the intersection points between edges; and the coordinates of the color adjacency can be obtained by setting a threshold to extract the color region and then obtaining the position coordinates of the color region boundary. Subsequently, based on the identified edge contour features and calibration-identified feature coordinates, the influence on the posture change and the overturning structure stability of the overturning target is analyzed, and according to the analysis result, the alignment key points and structure features (such as shape, size, color, identification pattern, etc.) are determined by threshold screening. These alignment key points and structure features are the positioning basis in the overturning process, which can ensure that the object maintains the correct alignment during the overturning process and accurately adjusts the overturning position.

[0022] Further, the application provides a method for determining the alignment key points and structure features based on the edge contour features and the calibration-identified feature coordinates, comprising:

[0023] analyze the edge contour features and the influence of the calibration recognition feature coordinates on the change of the flip target posture, obtain the influence evaluation value of each structure feature; obtain the flip structure stability of the edge contour features and the calibration recognition feature coordinates; use the influence evaluation value and the flip structure stability to perform key point screening according to a preset quantity threshold, and obtain the alignment key points; and determine the alignment key points and the structure features according to the structure features corresponding to the alignment key points determined through screening.

[0024] Optionally, after obtaining the edge contour features and the calibration recognition feature coordinates, the edge contour features and the calibration recognition feature coordinates are processed by computer-aided design (CAD) software and three-dimensional modeling software, the features are converted into elements in a three-dimensional space, a virtual three-dimensional model is formed, the virtual three-dimensional model is simulated according to predefined flip parameters such as a flip axis and a rotation angle by using simulation software, the coordinates of each structure feature (such as a corner point, a positioning hole, a symmetry axis, a recognition mark, an edge intersection, and a color adjacency) involved in a contour and a mark pattern after flip are recorded, the error between a reference flip template and the corresponding structure feature after flip is calculated by using the Euclidean distance, the error is divided by the maximum error allowed, 1 is subtracted from the quotient, and the influence evaluation value of each structure feature is obtained. The greater the error, the smaller the influence evaluation value, which indicates that the structure feature has less contribution to the alignment effect. In addition, the actual rotation angle during flip is recorded, the actual rotation angle is subtracted from the predefined rotation angle, the calculated error is divided by the maximum tolerance angle, and 1 is subtracted from the quotient. The difference is weighted with the influence evaluation value of each structure feature to obtain the flip structure stability. Then, the influence evaluation value of each structure feature and the flip structure stability are used to calculate the comprehensive evaluation value of each structure feature by weighting, the position coordinates corresponding to each structure feature are arranged in descending order according to the comprehensive evaluation value, and the position coordinates are screened as key points by using a preset quantity threshold. The screened position coordinates are the alignment key points, which can provide the best stability and accuracy during flip. Finally, the alignment key points are associated with the corresponding structure features to form complete alignment key point and structure feature data, thereby providing accurate basis for subsequent flip deviation prediction and alignment control.

[0025] Further, the application provides determining the alignment key points and the structure features, and then further comprising:

[0026] The alignment key points and the structure features are confirmed when the verification evaluation result meets a verification threshold, and auxiliary feature screening is performed from the remaining edge contour features and the calibration recognition feature coordinates until the verification threshold is met when the verification evaluation result does not meet the verification threshold.

[0027] Optionally, alignment abnormal case parameters are configured first, including a flip axis used in historical flipping, a rotation angle, and the like, which are recorded when a larger error occurs in historical flipping. Subsequently, the alignment abnormal case parameters are input into simulation software for flip simulation, and position coordinates of each alignment key point and structure features after flipping are recorded. Errors before and after flipping are calculated through Euclidean distance, and the errors are summarized as a verification evaluation result. Then, the errors recorded in the verification evaluation result are compared with corresponding verification thresholds. If all the errors are less than or equal to the verification thresholds, the key points and the structure features are confirmed to be valid, and can provide a basis for alignment control in the flipping process. If there is a case greater than the verification threshold, it indicates that the currently selected features cannot completely ensure sufficient alignment accuracy in the flipping process. At this time, auxiliary feature screening is performed from the remaining edge contour features and the calibration recognition feature coordinates, that is, the largest position coordinate is extracted as an auxiliary feature from the previously sorted features to supplement the current alignment key points and the corresponding structure features. This process is iterated until a feature combination meeting the threshold requirement is found, ensuring that the selected alignment key points and structure features can meet the accuracy requirement and successfully complete the alignment task.

[0028] A pre-flipping / post-flipping coordinate transformation relationship is established, and a theoretical position after flipping is calculated based on the current recognized pose and the alignment key points and the structure features.

[0029] In one embodiment, after the alignment key points and the structure features are determined, a rigid transformation matrix is established according to structure parameters of a flipping mechanism and original coordinates of the workpiece before and after flipping, to quantify the pre-flipping / post-flipping coordinate transformation relationship. Subsequently, based on the current recognized pose (i.e., the position and direction of the target before flipping), original coordinates of the workpiece in a three-dimensional space are determined. For each alignment key point on the workpiece, a difference value is calculated using the position of the alignment key point and the flipping axis, and the calculated difference value is multiplied by the rigid transformation matrix to obtain a theoretical position of the alignment key point after flipping. These calculation results provide a theoretical basis for subsequent deviation prediction, alignment adjustment, and compensation.

[0030] Further, the application provides a method for establishing a pre-flipping / post-flipping coordinate transformation relationship, including:

[0031] The structure parameters of the turnover plate mechanism are acquired, including the spatial position, rotation direction and rotation angle of the turnover shaft; the original point coordinates of the workpiece before turnover and the original point coordinates of the workpiece after turnover are identified; and a rigid body transformation matrix in three-dimensional space is established according to the offset of the original point coordinates of the workpiece before turnover and the original point coordinates of the workpiece after turnover and the spatial position, rotation direction and rotation angle of the turnover shaft, so as to map the coordinate transformation relationship before / after turnover.

[0032] Preferably, the structure parameters of the turnover plate mechanism are acquired, including the spatial position, rotation direction and rotation angle of the turnover shaft, which describe the specific position of the turnover shaft in three-dimensional space and the rotation direction and rotation angle of the shaft during turnover. These parameters can be collected through mechanical design drawings, so that the motion trajectory of the shaft and the turnover behavior of the object during turnover are known. Then, the original point coordinates of the workpiece before turnover and the original point coordinates of the workpiece after turnover are identified from the historical turnover log. The original point coordinates of the workpiece before turnover are the reference position of the object before turnover, which is usually the geometric center of the object or other specified reference point. After the object is turned over, the original point coordinates of the workpiece after turnover are the new position of the object after turnover. After the turnover transformation, the original point position of the object may change, so the coordinate position after turnover needs to be acquired. Then, a rigid body transformation matrix in three-dimensional space is established according to the offset of the original point coordinates before and after turnover and in combination with the spatial position, rotation direction and rotation angle of the turnover shaft. This rigid body transformation matrix is based on the principle of rigid body motion and describes how the object before turnover is converted to the coordinate system after turnover through rotation and displacement. Specifically, the offset reflects the change of the original point position before and after turnover, while the position, rotation direction and rotation angle of the turnover shaft determine how the object moves in three-dimensional space during rotation. By combining the rotation direction and rotation angle, a rotation matrix is constructed using the Rodrigues formula, and a translation transformation matrix is constructed using the offset of the original point coordinates. By combining the translation transformation matrix and the rotation matrix, the required rigid body transformation matrix is constructed, which can convert the coordinate system before turnover to the coordinate system after turnover. Finally, the rigid body transformation matrix is used to map the coordinate transformation relationship before / after turnover, so that the object can accurately move along the predetermined trajectory during turnover, providing accurate data support for subsequent control and alignment.

[0033] Based on the theoretical position after turnover, turnover deviation prediction is performed, and when the prediction result has deviation, turnover parameter compensation analysis is performed according to the prediction deviation, and turnover deviation compensation control is performed according to the compensation parameters.

[0034] In one embodiment, the deviation during the flipping process is evaluated by comparing the theoretically calculated position before and after flipping with the predicted flipping position based on actual monitoring of the flipping trajectory. When there is a deviation between the two, the flipping parameter compensation analysis is performed according to the predicted deviation amount. The core of this step is to analyze the size and direction of the deviation and determine which flipping parameters (such as angular velocity, acceleration, etc.) need to be adjusted to reduce or eliminate the deviation. In this way, the compensation parameters can be obtained to make the target object as close as possible to the predetermined theoretical position. Then, the compensation parameters are applied to the flipping control system to adjust the control strategy in real time during the flipping process, and by dynamically adjusting the parameters involved in the rotation, the object can be kept in the correct position during the flipping process and the deviation can be minimized, thereby achieving more accurate flipping alignment. This process can be performed in real time during the flipping process, so that the final position of the object is more consistent with the expected target position.

[0035] Further, the present application provides flipping deviation prediction based on the theoretical position after flipping, including:

[0036] According to the alignment key points and structural features, the flipping tracking monitoring is performed, the real-time position and attitude change information during the flipping process is collected, and the actual monitoring flipping trajectory is constructed; the flipping timing transformation prediction is performed according to the actual monitoring flipping trajectory, and the predicted flipping position at the end of the flipping is obtained; and the predicted deviation amount is obtained according to the predicted flipping position and the theoretical position after flipping.

[0037] Preferably, according to the flip tracking monitoring of the position key points and the structural features, the position and attitude change information of the workpiece during the flipping process is collected in real time using a sensor or an imaging device. These information includes the spatial position of the workpiece position key points and the orientation (such as the rotation angle) of the workpiece. Through the continuous monitoring of these data, the real-time trajectory during the flipping process can be generated, that is, the actual monitoring flipping trajectory, which reflects the position and attitude change of the object at each time during the flipping process. Then, the actual monitoring flipping trajectory at the time to be predicted is input into the flipping prediction model for flipping timing transformation prediction. The flipping prediction model can be constructed based on the long short-term memory network (LSTM). The spatial position of the workpiece position key points and the orientation of the workpiece at each time point in the historical flipping trajectory are arranged into an array according to the timing relationship, which is input into the LSTM as training data. Through the steps of forward propagation, loss calculation, backward propagation and parameter optimization, iterative training is performed until the maximum iteration number is reached or the loss function converges. After receiving the actual monitoring flipping trajectory, the flipping prediction model will predict the next time according to the current time, the current spatial position and the current orientation. This prediction process will continue until the predicted flipping position at the end of the flipping is obtained, including the predicted spatial position and the orientation. Then, the predicted flipping position is compared with the theoretical position after the flipping, and the prediction deviation is obtained by difference calculation. This prediction deviation reflects the error that may occur during the flipping process, which provides a basis for subsequent deviation compensation and control adjustment.

[0038] Further, the application provides flipping parameter compensation analysis according to the prediction deviation and flipping deviation compensation control according to the compensation parameters, including:

[0039] Based on the physical material characteristics of the workpiece, the influence relationship between the flipping parameters and the flipping position is established. The timing relationship chain of the flipping transformation position trajectory is established, the actual monitoring flipping trajectory is time-aligned and compensated using the influence relationship, the compensation parameters of each node in the flipping transformation position trajectory are obtained by taking the prediction deviation as the compensation target, and the flipping trajectory deviation compensation control is performed using the compensation parameters of each node in the flipping transformation position trajectory.

[0040] Optionally, first, based on the physical material characteristics of the workpiece, the influence relationship between the flipping parameters and the flipping position is established. The physical properties of the workpiece, such as mass, density, rigidity, and friction coefficient, will affect its motion performance during the flipping process. By studying the influence of the material of the workpiece on the flipping process, the influence of the flipping parameters (such as angular velocity and angular acceleration) on the position of the workpiece can be determined, and a mathematical model can be established to describe the influence relationship between the flipping parameters and the flipping position, for example, angle = initial angular velocity x time + angular acceleration x time square. For harder or heavier workpieces, higher angular velocity or angular acceleration may be required to complete the flipping, while softer or lighter workpieces may have lower requirements for the parameters of the flipping process. Subsequently, the position and attitude change data of the workpiece are arranged in chronological order to establish the time sequence relationship chain of the flipping transformation position trajectory, and the actual monitoring flipping trajectory is then time-aligned and compensated using the above influence relationship. In this process, the time and corresponding flipping position in the actual monitoring flipping trajectory are added to the predicted deviation, and the added result is input into the mathematical model to generate the theoretical flipping parameters of each node in the flipping transformation position trajectory. The theoretical flipping parameters are then differenced with the actual flipping parameters to determine the compensation parameters required for each trajectory node, including the adjustment amount of angular velocity and angular acceleration, to reduce the deviation. Finally, the compensation parameters of each node in the flipping transformation position trajectory are used to adjust the operating parameters of the flipping control system, such as angular velocity and angular acceleration, to ensure that the flipping trajectory of each stage is closer to the ideal trajectory, thereby reducing the deviation and accurately flipping the workpiece to the predetermined position.

[0041] The time sequence process alignment parameters are obtained, and the secondary alignment of the flipped target is performed using the time sequence process alignment parameters to determine that the alignment target is completed to enter the next process connection processing.

[0042] In one embodiment, the time sequence and process requirements related to the flipping target are extracted from the production or operation system, which typically includes the alignment requirements of the workpiece in different processes, the time interval between processes, the specific operation target of each process, etc. The time sequence process alignment parameters are key data to ensure the smooth connection and completion of each process during the flipping process. Subsequently, the secondary alignment of the flipped target is performed using the time sequence process alignment parameters. The position of the flipped target may be affected by the deviation during the flipping process or external factors, therefore, secondary alignment is needed to ensure the accuracy of the target position. When the secondary alignment is completed and the position and attitude of the target object meet the process requirements, it is confirmed that the alignment target is completed. At this time, the target object is ready to enter the next process, such as assembly, machining or detection, etc. This connection processing ensures smooth transition of the workpiece between processes, avoiding the impact of inaccurate alignment on subsequent work.

[0043] Further, the present application provides a method for determining that the alignment target is completed, comprising:

[0044] Based on the flip platform alignment verification of the alignment key points and the structure features, a platform alignment verification result is determined; according to the timing process alignment parameters and the platform alignment verification result, a process parameter target alignment is performed, when the alignment is matched, a secondary alignment is completed, and an alignment target completion signal is generated.

[0045] Preferably, after the compensation is completed, the alignment key points and the structure features are subjected to flip platform alignment verification. This process is performed by comparing the position of the workpiece on the flip platform with the preset ideal alignment key points and structure features. The sensor or imaging device on the flip platform monitors the position and posture of the target object in real time and compares it with the theoretical position. If the position of the target object on the flip platform meets the preset alignment requirements, the verification is successful; otherwise, the verification fails. The verification result reflects the positioning accuracy of the object during the flipping process. Subsequently, according to the timing process alignment parameters and the platform alignment verification result, a process parameter target alignment is performed. Specifically, the verification result is checked to determine whether a secondary alignment is needed. If the verification result shows that the verification is passed, it indicates that the flipping target position has reached the process requirements. At this time, the secondary alignment is skipped, and an alignment target completion signal is generated. This signal is used to indicate that the alignment target has been successfully completed, and the next process can be smoothly entered. On the contrary, if the verification result shows that the verification fails, the fine adjustment of the XY platform is performed through the fine adjustment cylinder to ensure that the process requirements are met.

[0046] Further, the present application provides that, according to the timing process alignment parameters and the platform alignment verification result, a process parameter target alignment is performed, and then the following steps are further included:

[0047] When the alignment is not matched, the fine adjustment of the XY platform is performed through the fine adjustment cylinder according to the alignment deviation to achieve the alignment target of the timing process alignment parameters.

[0048] Optionally, when the verification result shows that the verification fails, it means that there is an error between the position or posture of the workpiece and the theoretical position, at this time, the alignment deviation is determined by calculating the position difference of the workpiece on the X axis and the Y axis, which can be a position deviation or a posture deviation, depending on the positioning requirements of the workpiece on the platform. Subsequently, according to the alignment deviation, accurate adjustment is performed through the fine adjustment electric cylinder of the XY platform, which is a two-dimensional platform that can be adjusted by a small displacement along the X axis and the Y axis. The fine adjustment electric cylinder will adjust the alignment according to the size of the deviation, by controlling the extension or rotation of the electric cylinder, to ensure that each key point and structural feature of the target object can align with the target position in the timing process alignment parameter, until the alignment error is reduced to an acceptable range, meeting the alignment target requirements in the timing process alignment parameter. After fine adjustment, the position and posture of the workpiece will be checked again to ensure that it fully meets the alignment requirements, if there is still deviation, the adjustment will continue until the accurate alignment is achieved.

[0049] In summary, the embodiments of the present application have at least the following technical effects:

[0050] The embodiments of the present application first analyze the structure of the target to be flipped, identify the alignment key points and structural features; then, the pre-flipping / post-flipping coordinate transformation relationship is established, based on the current identified pose, the alignment key points and structural features are taken as the target, the theoretical position after flipping is calculated; then, based on the theoretical position after flipping, the flipping deviation is predicted, when the prediction result has deviation, the flipping parameter compensation analysis is performed according to the prediction deviation, the flipping deviation compensation control is performed according to the compensation parameter; finally, the timing process alignment parameter is obtained, the secondary alignment of the target after flipping is performed using the timing process alignment parameter, the alignment target is determined to be completed, to enter the next process linking processing. These technical effects collectively solve the technical problems of low alignment accuracy and ineffective compensation of flipping deviation caused by structural differences or inaccurate target pose recognition in traditional plate flipping machines during single / double-sided flipping, achieving the technical effects of improving the flipping alignment accuracy and realizing automatic process linking through secondary alignment and process linking processing.

[0051] Embodiment two, based on the same inventive concept as the flipping alignment control method of the single / double-sided plate flipping machine in the foregoing embodiments, such as Figure 2As shown, the present application provides a turnover alignment control system of a single / double-sided turnover plate machine, which comprises: a structure analysis module 11: performing structure analysis on a turnover target, identifying alignment key points and structure features; a theoretical position calculation module 12: establishing a pre / post-turnover coordinate transformation relationship, calculating a post-turnover theoretical position based on a current identified pose, taking the alignment key points and structure features as targets; a deviation compensation control module 13: performing turnover deviation prediction based on the post-turnover theoretical position, performing turnover parameter compensation analysis according to the predicted deviation amount when the prediction result has a deviation, and performing turnover deviation compensation control according to the compensation parameters; a secondary alignment module 14: obtaining timing process alignment parameters, performing secondary alignment on the post-turnover target using the timing process alignment parameters, determining an alignment target completion, and entering a next process connection processing.

[0052] Further, the structure analysis module 11 is further configured to perform the following method:

[0053] Collecting structure information of a turnover target; identifying edge contour features and calibration identification feature coordinates in the structure information, wherein the calibration identification features include positioning holes, symmetry axes, identification marks, edge intersection points, and color adjacency; determining the alignment key points and structure features according to the edge contour features and calibration identification feature coordinates.

[0054] Further, the structure analysis module 11 is further configured to perform the following method:

[0055] Analyzing the influence relationship of the edge contour features and calibration identification feature coordinates on the posture change of the turnover target, obtaining an influence evaluation value of each structure feature; obtaining turnover structure stability of the edge contour features and calibration identification feature coordinates; using the influence evaluation value and the turnover structure stability to perform key point screening according to a preset quantity threshold, to obtain the alignment key points; determining the alignment key points and structure features according to the structure features corresponding to the alignment key points determined by the screening.

[0056] Further, the structure analysis module 11 is further configured to perform the following method:

[0057] Configuring alignment exception case parameters, performing alignment identification verification based on the alignment key points and structure features, obtaining a verification evaluation result; when the verification evaluation result meets a verification threshold, confirming the alignment key points and structure features; when the verification threshold is not met, performing auxiliary feature screening from the remaining edge contour features and calibration identification feature coordinates, until the verification threshold is met.

[0058] Further, the theoretical position calculation module 12 is further configured to perform the following method:

[0059] Obtain the structure parameters of the turnover plate mechanism, including the spatial position, rotation direction and rotation angle of the turnover shaft; identify the original coordinates of the workpiece before turnover and the original coordinates of the workpiece after turnover; according to the offset of the original coordinates of the workpiece before turnover and the original coordinates of the workpiece after turnover, and the spatial position, rotation direction and rotation angle of the turnover shaft, establish a rigid body transformation matrix in three-dimensional space to map the coordinate transformation relationship before / after turnover.

[0060] Further, the deviation compensation control module 13 is also used to execute the following method:

[0061] According to the turnover tracking monitoring of the alignment key points and structural features, real-time position and attitude change information in the turnover process is collected to construct an actual monitoring turnover trajectory; according to the actual monitoring turnover trajectory, turnover timing transformation prediction is performed to obtain a predicted turnover position at the end of turnover; according to the predicted turnover position and the theoretical position after turnover, a predicted deviation amount is obtained.

[0062] Further, the deviation compensation control module 13 is also used to execute the following method:

[0063] Based on the physical material characteristics of the workpiece, an influence relationship between turnover parameters and turnover positions is established; a timing relationship chain of the turnover transformation position trajectory is established, the actual monitoring turnover trajectory is time-aligned and compensated using the influence relationship, the predicted deviation amount is taken as the compensation target, and compensation parameters of each node in the turnover transformation position trajectory are obtained; the compensation parameters of each node in the turnover transformation position trajectory are used to perform turnover trajectory deviation compensation control.

[0064] Further, the secondary alignment module 14 is also used to execute the following method:

[0065] Based on the turnover platform alignment verification of the alignment key points and structural features, a platform alignment verification result is determined; according to the timing process alignment parameters and the platform alignment verification result, process parameter target alignment is performed, when the alignment matches, secondary alignment is completed, and an alignment target completion signal is generated.

[0066] Further, the secondary alignment module 14 is also used to execute the following method:

[0067] When the alignment does not match, according to the alignment deviation amount, fine adjustment of the XY platform is performed through the fine adjustment cylinder to achieve the alignment target of the timing process alignment parameters.

[0068] It should be noted that the above-mentioned embodiment sequence of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the above describes a specific embodiment of the present application. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0069] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0070] The present application is only an exemplary description of the present application, and is considered to cover any and all modifications, changes, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various modifications and changes to the present application without departing from the scope of the present application. Thus, if these modifications and changes of the present application belong to the scope of the present application and its equivalents, the present application intends to include these modifications and changes.

Claims

1. A method for controlling the flipping and aligning of a single or double-sided flipping machine, characterized in that, Comprise: Structural analysis of the overturning target, identifying the alignment key points and structural features; Establish the coordinate transformation relationship before / after overturning, based on the current recognized pose, calculate the theoretical position after overturning based on the alignment key points and structural features as the target; Based on the theoretical position after overturning, predict the overturning deviation, when the prediction result has deviation, compensate and analyze the overturning parameters according to the prediction deviation, and control the overturning deviation compensation according to the compensation parameters; Obtain the timing process alignment parameters, use the timing process alignment parameters to perform secondary alignment on the target after overturning, determine the alignment target completion, and enter the next process connection processing; Structural analysis of the overturning target, identifying the alignment key points and structural features, comprising: Collecting the structural information of the overturning target; Identify the edge contour features and calibration identification feature coordinates in the structural information, wherein the calibration identification features include positioning holes, symmetry axes, identification marks, edge intersection points, and color adjacency; Determine the alignment key points and structural features according to the edge contour features and calibration identification feature coordinates; Determine the alignment key points and structural features according to the edge contour features and calibration identification feature coordinates, comprising: Analyze the influence relationship of the edge contour features and calibration identification feature coordinates on the posture change of the overturning target, and obtain the influence evaluation value of each structural feature; Obtain the overturning structural stability of the edge contour features and calibration identification feature coordinates; Use the influence evaluation value and overturning structural stability to screen key points according to a preset quantity threshold to obtain the alignment key points; Determine the alignment key points and structural features according to the structural features corresponding to the alignment key points determined by screening; Determine the alignment key points and structural features, and then further comprise: Configure the alignment exception case parameters, perform alignment identification verification based on the alignment key points and structural features, and obtain the verification evaluation result; When the verification evaluation result meets the verification threshold, the alignment key points and structural features are confirmed; When the verification threshold is not met, auxiliary feature screening is performed from the remaining edge contour features and calibration identification feature coordinates until the verification threshold is met.

2. The method according to claim 1, wherein Determine the alignment target completion, comprising: Performing overturning platform alignment verification based on the alignment key points and structural features, and determining the platform alignment verification result; According to the process parameter target alignment of the timing process alignment parameters and the platform alignment verification result, when the alignment matches, complete the secondary alignment, and generate an alignment target completion signal.

3. The method according to claim 2, wherein According to the process parameter target alignment of the timing process alignment parameters and the platform alignment verification result, further comprising: When the alignment does not match, align the fine adjustment of the XY platform according to the alignment deviation amount through the fine adjustment cylinder to achieve the alignment target of the timing process alignment parameters.

4. The method of claim 1, wherein the method further comprises: Obtain the structural parameters of the overturning mechanism, including the spatial position, rotation direction and rotation angle of the overturning shaft; Identify the origin coordinates of the workpiece before overturning and the origin coordinates of the workpiece after overturning; ​ Based on the origin coordinates of the workpiece before flipping, the offset of the origin coordinates of the workpiece after flipping, and the spatial position, rotation direction, and rotation angle of the flipping axis, a rigid body transformation matrix in three-dimensional space is established to map the coordinate transformation relationship before / after flipping.

5. The method of claim 1, wherein the method further comprises: Based on the theoretical position after the flip, the flip deviation is predicted, including: Based on the alignment key points and structural features, flipping tracking and monitoring are performed, real-time position and attitude change information during the flipping process is collected, and the actual monitoring flipping trajectory is constructed. Based on the actual monitored flipping trajectory, a flipping timing transformation prediction is performed to obtain the predicted flipping position at the end of the flipping. The prediction deviation is obtained based on the predicted flip position and the theoretical position after flipping.

6. The method according to claim 5, wherein Based on the predicted deviation, the flipping parameter compensation is analyzed, and the flipping deviation compensation control is performed according to the compensation parameters, including: Based on the physical material characteristics of the workpiece, the influence relationship between the flipping parameters and the flipping position is established. Establish a time-series relationship chain for the flip-over position trajectory, and use the influence relationship to perform time-series alignment compensation on the actual monitored flip-over trajectory. With the predicted deviation as the compensation target, obtain the compensation parameters of each node in the flip-over position trajectory. The compensation parameters of each node in the flip-transformation position trajectory are used to perform flip trajectory deviation compensation control.

7. A turnover alignment control system for a single or double facer turner, characterised in that, The system is used to execute the flipping and alignment control method of the single- or double-sided flipping machine according to any one of claims 1-6, including: Structural analysis module: Performs structural analysis on the flipped target, identifying alignment key points and structural features; Theoretical position calculation module: Establishes the coordinate transformation relationship before / after flipping, and calculates the theoretical position after flipping based on the current recognized pose, with the alignment key points and structural features as targets; Deviation compensation control module: Based on the theoretical position after flipping, the flipping deviation is predicted. When there is a deviation in the prediction result, the flipping parameter compensation is analyzed according to the predicted deviation amount, and the flipping deviation compensation control is performed according to the compensation parameter. Secondary alignment module: acquires the timing process alignment parameters, uses the timing process alignment parameters to perform secondary alignment of the flipped target, confirms that the alignment target is completed, and proceeds to the next process connection processing.

Images