A pu lip edge shaping collaborative control method

Abstract

The present application relates to automobile sunroof glass PU lip shaping control technical field, especially in kind of PU lip shaping collaborative control method, including: the clamping posture image, PU lip contour image, fluorescence image are collected;According to the actual clamping angle of clamping posture image, calculate the angle deviation and generate candidate angle adjustment amount;According to the current profile offset of PU lip contour image, calculate the profile deformation normalization amount;According to the local wrinkle normalization feature and fluorescence intensity continuity index extracted from fluorescence image, the fluorescence deformation score is obtained by combining the profile deformation normalization amount;The clamping force correction amount is calculated, and the candidate clamping force is obtained;The candidate control parameter set is composed of candidate angle adjustment amount, candidate clamping force, candidate driving speed and candidate feed step, and the feasible candidate control parameter set is screened out, and then the target control parameter group is selected, and the unified control instruction is generated.The present application improves the shaping quality and shaping efficiency of PU lip.

Description

Technical Field

[0001] This invention relates to the field of PU lip shaping control technology for automotive sunroof glass, and more particularly to a collaborative control method for PU lip shaping. Background Technology

[0002] The PU lip of automotive sunroof glass needs to achieve a stable inward-curving state before assembly or curing. PU material is viscoelastic; the same angle or clamping force can produce different instantaneous deformations and unloading rebounds depending on the material batch, curing time, and environmental conditions. Existing methods often rely on mechanical settings or manual adjustments to determine shaping parameters, resulting in discrete adjustments that are difficult to continuously correct within the effective control period of the current object.

[0003] Some shaping processes employ conventional visual inspection, clamping force control, or multi-axis synchronous control, but each loop typically determines positioning, overpressure, or timing separately. Conventional images are insensitive to localized wrinkles and occlusions on narrow lips; adjusting angles or clamping forces individually may increase clamping marks or springback while improving inward retraction. Existing methods also lack a processing sequence that converts contour deformation, fluorescence continuity, springback risk, and synchronization status into the same set of candidate control parameters, potentially causing update results from different loops to cancel each other out.

[0004] Therefore, it is necessary to establish a continuous method chain based on the same shaping reference relationship, so that angle deviation, fluorescence deformation score, clamping force correction amount, synchronization ratio and rebound risk have clear sources, downstream consumers and abnormal branches, and complete common constraints and unified arbitration before issuance. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in related technologies. To this end, this invention provides a synergistic control method for PU lip shaping, which solves the problem in the prior art that it is difficult to simultaneously consider contour deformation and fluorescence continuity during PU lip shaping, thereby improving the shaping quality and efficiency of PU lips.

[0006] This invention provides a PU lip contouring and coordinated control method, comprising the following steps: The image acquisition module acquires clamping posture images, PU lip contour images, and fluorescence images; The actual clamping angle is obtained from the clamping posture image. The angle deviation between the actual clamping angle and the target shaping angle is calculated. Candidate angle adjustment amounts are generated based on the angle deviation. The current PU lip contour offset is obtained from the PU lip contour image. The normalized amount of contour deformation is obtained by comparing the current PU lip contour offset with the reference contour offset of the qualified sample of the same model. Local wrinkle normalization features and fluorescence intensity continuity index are extracted from the fluorescence image, and a fluorescence deformation score is obtained by combining the contour deformation normalization amount. The controller acquires the target deformation score, the score deviation gain calibrated by the clamping response and safety range of the same model sample, the springback risk gain calibrated by the historical springback defects and manual re-inspection results, and the springback risk output by the springback risk prediction model. The controller calculates the clamping force correction amount based on the target deformation score, fluorescent deformation score, score deviation gain, springback risk gain and springback risk, and then obtains the candidate clamping force based on the clamping force correction amount. The controller determines the current synchronization ratio and limits the candidate drive speed and candidate feed step size based on the current synchronization ratio. A set of candidate control parameters is formed by candidate angle adjustment amount, candidate clamping force, candidate drive speed and candidate feed step size, and then a set of feasible candidate control parameters is obtained by screening. The controller selects a target control parameter group that meets the set conditions from the set of feasible candidate control parameters, and then forms a unified control command. The controller executes the PU lip shaping operation according to the generated unified control command.

[0007] A further improvement of the PU lip contouring collaborative control method of the present invention is that it further includes the following steps: Once the object to be shaped enters the control cycle, the controller performs object identification and carrier availability checks, reads the object model, PU lip specifications, target inward angle, target deformation score, target deformation range, clamping force safety range, synchronous control boundary, each control cycle, image confidence threshold, temperature rise threshold, communication delay upper limit, and pre-shaping contour reference. Each control cycle together defines the effective control time window. The image confidence threshold, temperature rise threshold, and communication delay upper limit constitute the threshold system of the object to be shaped, and then form a shaping reference relationship based on the same object identifier and timestamp. Determine if the plastic surgery reference relationship is valid. If the result is no, reread the parameters or recalibrate. If the result is still no after rereading the parameters or recalibrating, enter conservative mode, maintain the previous stable parameter group and stop issuing new instructions, and output a re-check mark. If the judgment result is yes, the image acquisition module acquires the clamping posture image, the PU lip contour image, and the fluorescence image.

[0008] A further improvement of the PU lip shaping collaborative control method of the present invention lies in that the calculation formula for the angle deviation is as follows: , in, The angle deviation is the angle deviation in the k-th control cycle. The target shaping angle for the k-th control cycle. This represents the actual clamping angle during the k-th control cycle. Based on the angle deviation and the previous effective angle adjustment amount Generate reference candidate angle adjustment amount : , The reference candidate angle adjustment amount is constrained by an angle limit to obtain the amplitude-limited reference candidate angle adjustment amount. : , Generate a set of candidate angle adjustment values ​​based on the baseline candidate angle adjustment value after amplitude limiting. : in, To adjust the gain for angle adjustment, Candidate step size for angle, The upper limit of angle adjustment, This is the lower limit for angle adjustment. The number of candidate expansions, n is an integer ranging from -N to N; When the pose marker is occluded, the pose fitting confidence is lower than the image confidence threshold, or the target angle version is mismatched, the previous effective angle adjustment amount is maintained, and a re-inspection marker is output.

[0009] A further improvement of the PU lip shaping collaborative control method of the present invention lies in that the calculation formula for the normalized amount of contour deformation is as follows: in, Let be the normalized amount of profile deformation in the k-th control cycle. This represents the current contour offset in the k-th control cycle. This is the reference contour offset of the qualified sample of the same model in the kth control cycle.

[0010] A further improvement of the PU lip contouring collaborative control method of the present invention is that it further includes: The fluorescent lines in the PU lip are identified, and the local wrinkle normalization characteristics are obtained based on the local bending, discontinuity and brightness concentration of the fluorescent lines. The fluorescence intensity continuity index is obtained based on the proportion of effective continuous segments in the length direction. The formula for calculating the fluorescence deformation score is: in, The fluorescence deformation score for the k-th control period. Let be the normalized amount of profile deformation in the k-th control cycle. This represents the local wrinkle normalization feature for the k-th control cycle. The fluorescence intensity continuity index for the k-th control period. For contour deformation weights, For wrinkle feature weights, As fluorescence continuity weight, , and All values ​​range from 0 to 1. .

[0011] A further improvement of the PU lip shaping collaborative control method of the present invention lies in that the calculation formula for the clamping force correction is as follows: in, This is the clamping force correction amount for the k-th control cycle. For scoring bias gain, Score the target deformation. For fluorescence deformation scoring, To mitigate the risk of rebound, To mitigate the risk of rebound, The value ranges from 0 to 1.

[0012] A further improvement of the PU lip shaping collaborative control method of the present invention is that the step of generating the candidate clamping force includes: The controller obtains the clamping force reference value for the kth control cycle. and the correction direction coefficient of the kth control cycle , Calculate the initial candidate clamping force for the k-th control cycle. : Among them, the wrinkling condition is that the local wrinkle normalization feature reaches a preset wrinkle threshold, and the overpressure condition is that the fluorescence intensity continuity index meets a preset overpressure criterion; when the fluorescence deformation score is lower than the target range and the wrinkle condition and overpressure condition are not triggered, When the wrinkle condition or the overpressure condition is triggered, When the fluorescence image is missing, overexposed, underexposed, or has insufficient effective continuous segments, ; The baseline candidate clamping force for the k-th control cycle is obtained by limiting the original candidate clamping force within a safe range. : , A set of candidate clamping forces for the k-th control cycle is generated based on the baseline candidate clamping force after the amplitude limiting. : in, This is the lower limit of the safe clamping force range. This is the upper limit of the safe range for clamping force. The number of candidate expansions is n, which is an integer ranging from -N to N.

[0013] A further improvement of the PU lip contouring collaborative control method of the present invention lies in that the calculation formula for the synchronization ratio is: in, Let be the candidate synchronization ratio for the i-th workstation in the k-th control cycle. Let be the synchronization ratio of the main shaping motion and feed execution action of the i-th station relative to the recipe reference cycle time. The effective synchronization ratio of the i-th workstation is the previous one. Let be the normalized value of the load change at the i-th workstation. Let be the normalized value of the temperature rise at the i-th workstation. The operational health status of the i-th workstation; This is the load penalty coefficient. This is the temperature rise penalty coefficient. This is a health penalty coefficient. Let be the lower limit of the synchronization ratio for the i-th workstation. This represents the upper limit of the synchronization ratio for the i-th workstation; The load change normalized value is obtained by normalizing the effective signal of the execution current, pressure feedback or motion resistance; the temperature rise normalized value is obtained by normalizing the temperature measurement value relative to the allowable temperature rise range of the execution carrier; the operational health is obtained by normalizing the response delay, historical fault records and current image confidence according to the calibrated rules.

[0014] A further improvement of the PU lip contouring collaborative control method of the present invention lies in the following steps in selecting a feasible candidate control parameter set: in, Let be the set of feasible candidate control parameters for the k-th control cycle. The k-th control cycle contains candidate elements for candidate angle adjustment, candidate clamping force, candidate drive speed, and candidate feed step size. For the set of candidate control parameters, This is a function for determining the boundary conditions of each project item; when the condition passes... The result is 1; For each element in the feasible candidate control parameter set, the controller calculates the target normalized error of the angle, the target normalized error of the clamping force, the normalized error of the synchronization time, and the normalized error of the process switching disturbance, and then... Calculate the ranking score; The synchronization time normalization error is obtained by normalizing the completion time deviation of the candidate master shaping motion and the feed execution action relative to the same synchronization cycle and the formula reference cycle. in, The ranking score for the m-th element in the k-th control cycle; For the angle target normalization error, To normalize the clamping force target error, To normalize the synchronization time error, This refers to the normalized error caused by process switching disturbances. As the angle error weight, As the weighting factor for clamping force error, As the synchronization time error weight, To switch the perturbation error weights, , , and All values ​​range from 0 to 1. ; choose The elements that have the minimum safety margin and meet the requirements are used as the target control parameter group.

[0015] A further improvement of the PU lip contouring collaborative control method of the present invention is that it further includes: Target control parameter group Form a unified control command; in, To unify control commands; For the target angle adjustment amount, For the target clamping force, Driven by the target speed, For the target feed step size, The linear velocity of the main shaping motion and feed actuator along the direction of PU lip shaping action; The angle, clamping, main shaping motion, and feed execution carriers are respectively issued; After the unified control command is issued, the displacement increment and completion timestamp of the main shaping motion and the feed motion form a synchronous feedback. The controller receives the synchronous feedback and enters the shaping reference relationship for the next control cycle.

[0016] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects: (1) By using the collaborative generation mechanism of angle deviation and clamping force correction, the control error of PU lip contour offset is greatly reduced, and the inward angle deviation is strictly controlled. Compared with the traditional manual debugging or single-loop control method, the shaping accuracy is significantly improved, and there is no need to rely on the operator's experience. The shaping consistency of parts in the same batch is at a high level.

[0017] (2) By integrating local wrinkle features and continuity detection of fluorescence images, micro-wrinkle risks and overpressure clamping hazards can be identified in advance. Combined with the dynamic adjustment of clamping force correction direction, the total incidence of three typical defects, wrinkles, clamping marks and springback, is greatly reduced during the shaping process, effectively reducing subsequent manual re-inspection and rework costs.

[0018] (3) Based on the synchronous proportional adaptive adjustment mechanism of load, temperature rise and operating health, it can automatically adapt to the differences in equipment status at different workstations, effectively control the synchronization error between the main shaping motion and the feed action, reduce the cycle deviation when shaping in parallel at multiple workstations, and avoid significant parameter drift during continuous production, thus effectively avoiding the problem of mutual cancellation of adjustment results from different loops.

[0019] (4) Complete abnormal branch processing and conservative mode design can cope with a variety of common production abnormalities such as attitude marker occlusion, image acquisition failure, and communication timeout. Even under abnormal conditions, the shaping process can still be maintained to run continuously, greatly reducing the unplanned downtime of the equipment. At the same time, it supports rapid parameter switching of multiple models of parts, significantly shortening the changeover time.

[0020] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a collaborative control method for PU lip shaping provided by the present invention.

[0023] Figure 2 This is a comparison chart of the convergence of fluorescence deformation scoring in a PU lip shaping collaborative control method provided by this invention and existing technologies. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention but should not be used to limit the scope of this invention.

[0025] The PU lip shaping equipment includes a clamping mechanism, a shaping mechanism, an image acquisition module, a fluorescence excitation module, and a controller. The PU lip shaping equipment has multiple shaping stations, which can perform shaping operations on multiple PU lips simultaneously. The clamping mechanism is used to clamp the PU lips and adjust their orientation, thereby facilitating the shaping mechanism to perform shaping operations on the PU lips.

[0026] The following is combined Figure 1 The present invention describes a PU lip contouring collaborative control method, comprising the following steps: The image acquisition module acquires clamping posture images, PU lip contour images, and fluorescence images; The controller obtains the actual clamping angle based on the clamping posture image, calculates the angle deviation between the actual clamping angle and the target shaping angle, and generates candidate angle adjustment amounts based on the angle deviation. The controller obtains the current PU lip contour offset based on the PU lip contour image, and obtains the contour deformation normalization amount by comparing the current PU lip contour offset with the reference contour offset of the qualified sample of the same model. The fluorescence image is used to extract local wrinkle normalization features and fluorescence intensity continuity index, and combined with the contour deformation normalization amount to obtain the fluorescence deformation score. The controller acquires the target deformation score, the score deviation gain calibrated by the clamping response and safety range of the same model sample, the springback risk gain calibrated by the historical springback defects and manual re-inspection results, and the springback risk output by the springback risk prediction model. The controller calculates the clamping force correction amount based on the target deformation score, fluorescent deformation score, score deviation gain, springback risk gain and springback risk, and then obtains the candidate clamping force based on the clamping force correction amount. The controller determines the current synchronization ratio and limits the candidate drive speed and candidate feed step size based on the current synchronization ratio. A set of candidate control parameters is formed by candidate angle adjustment amount, candidate clamping force, candidate drive speed and candidate feed step size, and then a set of feasible candidate control parameters is obtained by screening. The controller selects a target control parameter group that meets the set conditions from the set of feasible candidate control parameters, and then forms a unified control command. The controller executes the PU lip shaping operation according to the generated unified control command.

[0027] Furthermore, before acquiring each image, the following steps are also included: Once the object to be shaped enters the control cycle, the controller performs object identification and carrier availability checks, reads the object model, PU lip specifications, target inward angle, target deformation score, target deformation range, clamping force safety range, synchronous control boundary, each control cycle, image confidence threshold, temperature rise threshold, communication delay upper limit, and pre-shaping contour baseline. Each control cycle together defines the effective control time window. The image confidence threshold, temperature rise threshold, and communication delay upper limit constitute the threshold system of the object to be shaped, and then form a shaping reference relationship based on the same object identifier and timestamp. Determine if the plastic surgery reference relationship is valid. If the result is no, reread the parameters or recalibrate. If the result is still no after rereading the parameters or recalibrating, enter conservative mode, maintain the previous stable parameter group and stop issuing new instructions, and output a re-check mark. If the judgment result is yes, the image acquisition module acquires the clamping posture image, the PU lip contour image, and the fluorescence image.

[0028] Preferably, after each image acquisition is completed, the controller binds the acquired image with the corresponding timestamp, stores it in the local temporary cache, and uploads it to the process data archiving unit, waiting for subsequent feature extraction and calculation. If data integrity verification fails or communication latency exceeds the preset communication latency limit during storage or uploading, a re-acquisition command is triggered to restart the image acquisition process. If two consecutive re-acquisitions fail the verification, a fault alarm is triggered, the current control cycle is stopped, and a fault marker is output, waiting for maintenance personnel to investigate and handle the issue.

[0029] Furthermore, the formula for calculating the angular deviation is as follows: , in, The angle deviation is the angle deviation in the k-th control cycle. The target shaping angle for the k-th control cycle. This represents the actual clamping angle during the k-th control cycle. Based on the angle deviation and the previous effective angle adjustment amount Generate reference candidate angle adjustment amount : , The reference candidate angle adjustment amount is obtained by applying angle limit constraints to the reference candidate angle adjustment amount. : , Generate a set of candidate angle adjustment values ​​based on the baseline candidate angle adjustment value after amplitude limiting. : in, To adjust the gain for angle adjustment, Candidate step size for angle, The upper limit of angle adjustment, This is the lower limit for angle adjustment. The number of candidate expansions, n is an integer ranging from -N to N; When the pose marker is occluded, the pose fitting confidence is lower than the image confidence threshold, or the target angle version is mismatched, the previous effective angle adjustment amount is maintained, and a re-inspection marker is output.

[0030] Ideally, based on the calculated angle deviation between the actual clamping angle and the target shaping angle, the deviation of the current clamping posture from the preset target can be sensed in real time, providing accurate quantitative basis for subsequent angle adjustment and avoiding shaping errors caused by blind adjustment. Combined with the processing method of generating a baseline candidate adjustment amount from the previous effective adjustment amount, the continuity and smoothness of the control process are ensured, avoiding instantaneous impacts on the PU lip caused by large changes in adjustment amount, and reducing uneven deformation problems caused by parameter jumps. The angle limit constraint mechanism can strictly limit the adjustment amount within a safe and reasonable range, effectively avoiding excessive lip deformation or equipment stroke exceeding limits caused by over-adjustment, ensuring the safety and reliability of the shaping process. The design of generating multiple sets of candidate adjustment amounts around the baseline adjustment amount provides sufficient feasible solution space for subsequent multi-parameter collaborative screening, facilitating the selection of the optimal adjustment scheme under the premise of considering multi-dimensional constraints such as contour deformation, fluorescence continuity, and clamping force adaptation, thereby improving the overall control effect. The processing logic that maintains the previous effective adjustment value and outputs a re-inspection mark in abnormal scenarios can maintain the stable operation of the shaping process in special cases such as attitude acquisition failure and parameter mismatch, avoiding the interruption of the entire production process due to a single abnormality. At the same time, the re-inspection mark prompts maintenance personnel to troubleshoot problems in a timely manner, taking into account both production continuity and quality controllability.

[0031] Furthermore, the formula for calculating the normalized amount of profile deformation is: in, Let be the normalized amount of profile deformation in the k-th control cycle. This represents the current contour offset in the k-th control cycle. This is the reference contour offset of the qualified sample of the same model in the kth control cycle.

[0032] In a specific implementation case, as shown in Table 1, the current contour offset is divided by the reference contour offset of the qualified sample of the same model to obtain the contour deformation normalization amount, so that the contour offsets under different objects and imaging scales can be compared.

[0033] Table 1

[0034] Furthermore, it also includes: identifying fluorescent lines in the PU lip, obtaining local wrinkle normalization features based on the local bending, discontinuity and brightness concentration of the fluorescent lines, and obtaining a fluorescence intensity continuity index based on the proportion of effective continuous segments in the length direction. The formula for calculating the fluorescence deformation score is: in, The fluorescence deformation score for the k-th control period. Let be the normalized amount of profile deformation in the k-th control cycle. This represents the local wrinkle normalization feature for the k-th control cycle. The fluorescence intensity continuity index for the k-th control period. For contour deformation weights, For wrinkle feature weights, As fluorescence continuity weight, , and All values ​​range from 0 to 1. .

[0035] Ideally, the normalized contour deformation amount uses qualified samples as a reference benchmark, eliminating the impact of differences in PU lip dimensions between different models on deformation assessment. This allows for the application of the same evaluation logic across models, eliminating the need to separately calibrate judgment thresholds for different specifications of parts, significantly reducing the parameter configuration workload for process engineers. Incorporating the normalized contour deformation amount into the calculation dimension of fluorescence deformation scoring enables coupled assessment of macroscopic contour deformation and microscopic surface defects, avoiding the omission of local hidden quality problems when judging with a single index. The fluorescence deformation scoring mechanism, which integrates local wrinkle normalization characteristics and fluorescence intensity continuity indicators, can simultaneously characterize the wrinkle risk, surface uniformity, and overall deformation state of the PU lip. Compared to traditional single contour detection or pressure feedback, it provides a more comprehensive assessment dimension for the PU lip shaping quality, enabling early identification of hidden quality hazards that are difficult to detect using traditional methods. The multi-weighted adjustable scoring calculation method allows for flexible adjustment of evaluation emphasis according to the quality focus of different parts, adapting to diverse production quality requirements and further improving the method's versatility and adaptability.

[0036] Furthermore, the formula for calculating the clamping force correction is as follows: in, This is the clamping force correction amount for the k-th control cycle. For scoring bias gain, Score the target deformation. For fluorescence deformation scoring, To mitigate the risk of rebound, To mitigate the risk of rebound, The value ranges from 0 to 1.

[0037] Specifically, the steps for generating the candidate clamping force include: The controller obtains the clamping force reference value for the kth control cycle. and the correction direction coefficient of the kth control cycle , Calculate the initial candidate clamping force for the k-th control cycle. : Among them, the wrinkling condition is that the local wrinkle normalization feature reaches a preset wrinkle threshold, and the overpressure condition is that the fluorescence intensity continuity index meets a preset overpressure criterion; when the fluorescence deformation score is lower than the target range and the wrinkle condition and overpressure condition are not triggered, When the wrinkle condition or the overpressure condition is triggered, When the fluorescence image is missing, overexposed, underexposed, or has insufficient effective continuous segments, ; The baseline candidate clamping force for the k-th control cycle is obtained by limiting the original candidate clamping force within a safe range. : , A set of candidate clamping forces for the k-th control cycle is generated based on the baseline candidate clamping force after the amplitude limiting. : in, This is the lower limit of the safe clamping force range. This is the upper limit of the safe range for clamping force. The number of candidate expansions is n, which is an integer ranging from -N to N.

[0038] Preferably, the scene-specific correction direction coefficient design based on fluorescence deformation scoring, defect triggering state, and image validity can dynamically adapt to the clamping force adjustment requirements under different shaping conditions, ensuring deformation convergence efficiency while avoiding quality problems caused by incorrect force adjustment direction. The safety range limiting mechanism strictly constrains the clamping force within a preset reasonable range, avoiding PU lip crushing due to force overload or incomplete shaping due to insufficient force. The processing method of generating a multi-level candidate set around the benchmark candidate clamping force provides a feasible solution space for subsequent multi-parameter collaborative screening, balancing control accuracy and condition adaptability, and effectively improving the matching degree between clamping force parameters, angle adjustment, and feed actions. The conservative strategy of maintaining the benchmark force value in abnormal scenarios avoids control logic chaos caused by invalid images or abnormal features, ensuring continuous and stable operation of the shaping process.

[0039] Furthermore, the formula for calculating the synchronization ratio is: in, Let be the candidate synchronization ratio for the i-th workstation in the k-th control cycle. Let be the synchronization ratio of the main shaping motion and feed execution action of the i-th station relative to the recipe reference cycle time. The effective synchronization ratio of the i-th workstation is the previous one. Let be the normalized value of the load change at the i-th workstation. Let be the normalized value of the temperature rise at the i-th workstation. The operational health status of the i-th workstation; This is the load penalty coefficient. This is the temperature rise penalty coefficient. This is a health penalty coefficient. Let be the lower limit of the synchronization ratio for the i-th workstation. This represents the upper limit of the synchronization ratio for the i-th workstation; The load change normalized value is obtained by normalizing the effective signal of the execution current, pressure feedback or motion resistance; the temperature rise normalized value is obtained by normalizing the temperature measurement value relative to the allowable temperature rise range of the execution carrier; the operational health is obtained by normalizing the response delay, historical fault records and current image confidence according to the calibrated rules.

[0040] Ideally, a synchronous proportional dynamic calculation mechanism based on load, temperature rise, and operational health can sense changes in the operating status of each shaping station in real time, automatically adjusting the matching rhythm of the main shaping motion and the feed motion. This adapts to differences in execution efficiency under different working conditions without manual intervention, effectively avoiding asynchrony issues caused by equipment status fluctuations. Incorporating the previous effective synchronization ratio into the baseline calculation ensures the continuity and smoothness of the rhythm adjustment, avoiding motion shocks or rhythm disturbances caused by sudden proportional changes, and improving the stability of multi-station collaborative operation. The upper and lower limit constraint mechanism of the synchronization ratio can control the rhythm within a reasonable and safe range, effectively avoiding under-shaping or over-shaping problems caused by abnormal ratios, and ensuring consistent shaping quality across all stations.

[0041] Furthermore, the selection of feasible candidate control parameters involves the following steps: in, Let be the set of feasible candidate control parameters for the k-th control cycle. The k-th control cycle contains candidate elements for candidate angle adjustment, candidate clamping force, candidate drive speed, and candidate feed step size. For the set of candidate control parameters, This is a function for determining the boundary conditions of each project item; when the condition passes... The result is 1; For each element in the feasible candidate control parameter set, the controller calculates the target normalized error of the angle, the target normalized error of the clamping force, the normalized error of the synchronization time, and the normalized error of the process switching disturbance, and then... Calculate the ranking score; The synchronization time normalization error is obtained by normalizing the completion time deviation of the candidate master shaping motion and the feed execution action relative to the same synchronization cycle and the recipe reference cycle. in, The ranking score for the m-th element in the k-th control cycle; For the angle target normalization error, To normalize the clamping force target error, To normalize the synchronization time error, This refers to the normalized error caused by process switching disturbances. As the angle error weight, As the weighting factor for clamping force error, As the synchronization time error weight, To switch the perturbation error weights, , , and All values ​​range from 0 to 1. ; choose The elements that have the minimum safety margin and meet the requirements are used as the target control parameter group.

[0042] Furthermore, it also includes: target control parameter groups according to Form a unified control command; in, To unify control commands; For the target angle adjustment amount, For the target clamping force, Driven by the target speed, For the target feed step size, The linear velocity of the main shaping motion and feed actuator along the direction of PU lip shaping action; The angle, clamping, main shaping motion, and feed execution carriers are respectively issued; After the unified control command is issued, the displacement increment and completion timestamp of the main shaping motion and the feed motion form a synchronous feedback. The controller receives the synchronous feedback and enters the shaping reference relationship for the next control cycle.

[0043] Ideally, a feasible candidate control parameter set screening mechanism based on item-by-item engineering boundary determination can preemptively exclude parameter combinations that exceed process safety limits or conflict with equipment operation constraints, preventing invalid parameters from entering subsequent evaluation stages from the outset, significantly reducing the computational complexity of parameter screening, and ensuring that all candidate solutions are engineering-feasible. A multi-dimensional normalized error-weighted ranking and scoring design can simultaneously consider multiple control objectives such as angle control accuracy, clamping force adaptability, synchronization matching degree, and process switching smoothness, achieving globally optimal parameter selection under multiple constraints and avoiding the degradation of other control indicators caused by prioritizing a single dimension. Integrating multiple control parameters into a unified control command issuance method ensures that the action triggering time of each execution carrier is highly synchronized with the control objective, avoiding timing deviations caused by dispersed command issuance and improving the consistency of multi-mechanism collaborative actions. A synchronous feedback closed-loop design can incorporate the actual execution effect into the reference of subsequent control cycles in real time, achieving dynamic iterative optimization of control parameters and further improving the stability of shaping quality in continuous production processes.

[0044] The following calculations demonstrate the improvement in PU lip shaping accuracy using an example (the collaborative control method of this invention) and a comparative example (existing technology, where fluorescence deformation scoring is not included in clamping force correction, feasible candidate screening, and unified sorting arbitration; only a fixed clamping force of 155N and linear response are used). Both the example and the comparative example use the same model of PU lip, with an initial contour offset of 0.90mm, a reference contour offset of 3.00mm for qualified samples of the same model, and a target deformation score of 0.52. The fixed input parameters for the shaping equipment are: deformation score weights: contour deformation 0.70, wrinkle characteristics 0.20, fluorescence continuity 0.10; fluorescence characteristic indices: local wrinkle normalization characteristics 0.1, fluorescence intensity continuity index 0.9; springback risk: 0.18; initial clamping force: 155N; clamping force safety range: 120N~260N.

[0045] The examples and comparative examples were configured with a total of 10 effective control cycles, from k=0 to k=9, to simulate the continuous progress of the shaping action. Table 2 records the fluorescence deformation scores of the examples and comparative examples under the same profile deformation normalization amount at different control cycles.

[0046] Table 2

[0047] Figure 2 The comparison curves of fluorescence deformation score with control period k are given. It can be seen that after adopting the method of the present invention, through contour normalization and fluorescence deformation score to enter clamping force correction, feasible candidate screening and unified sorting arbitration, the fluorescence deformation score increases with the increase of control period and remains controllable within the common physical boundary; while in the comparative scheme, the fluorescence deformation score only retains the linear response and the end correction gain is small.

[0048] This invention utilizes closed-loop control to rapidly increase the fluorescence deformation score with increasing period k, reaching 0.507 in the 9th period, extremely close to the target score of 0.52. Existing technologies, lacking closed-loop correction, have minimal end-correction gain, resulting in a score that ultimately remains around 0.305, failing to meet acceptable shaping standards. Experimental data demonstrates that "fluorescence deformation score incorporating clamping force correction and unified arbitration" is a key technical means to solve the problem of insufficient inward retraction in PU lip shaping and improve yield.

[0049] While pursuing high scores, this invention strictly controls all physical parameters within safety boundaries, and all verification results are passed. The verification results are shown in Table 3.

[0050] Table 3

[0051] In summary, the collaborative control method of this invention acquires the target shaping angle, generates multiple sets of candidate adjustment quantities based on the angle deviation, and completes arbitration optimization through contour deformation normalization and fluorescence deformation scoring. This achieves closed-loop precise correction of the PU lip shaping process, effectively solving the problems of insufficient inward retraction and low yield in existing technologies. All control parameters of this invention operate within preset safety boundaries, eliminating the risk of out-of-range operating conditions. Stability and reliability meet mass production requirements, effectively reducing the shaping defect rate and improving the quality stability of PU lip shaping, demonstrating promising prospects for industrial application. Simultaneously, the traceability design of the entire process parameters enables complete data link transmission from image acquisition to command issuance, facilitating subsequent process optimization and quality issue backtracking. The coupled evaluation mechanism of multi-dimensional indicators avoids the degradation of other quality indicators caused by prioritizing a single control objective, achieving a multi-objective balance of shaping accuracy, surface quality, and operational stability. The closed-loop iterative design of control commands and execution feedback can continuously optimize the control logic with the accumulation of production data, and shaping accuracy and defect suppression capabilities can be gradually improved with the mass production process.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A collaborative control method for PU lip contouring, characterized in that, Includes the following steps: The image acquisition module acquires clamping posture images, PU lip contour images, and fluorescence images; The actual clamping angle is obtained from the clamping posture image. The angle deviation between the actual clamping angle and the target shaping angle is calculated. Candidate angle adjustment amounts are generated based on the angle deviation. The current PU lip contour offset is obtained from the PU lip contour image. The normalized amount of contour deformation is obtained by comparing the current PU lip contour offset with the reference contour offset of the qualified sample of the same model. Local wrinkle normalization features and fluorescence intensity continuity index are extracted from the fluorescence image, and a fluorescence deformation score is obtained by combining the contour deformation normalization amount. The controller acquires the target deformation score, the score deviation gain calibrated by the clamping response and safety range of the same model sample, the springback risk gain calibrated by the historical springback defects and manual re-inspection results, and the springback risk output by the springback risk prediction model. The controller calculates the clamping force correction amount based on the target deformation score, fluorescent deformation score, score deviation gain, springback risk gain and springback risk, and then obtains the candidate clamping force based on the clamping force correction amount. The controller determines the current synchronization ratio and limits the candidate drive speed and candidate feed step size based on the current synchronization ratio. A set of candidate control parameters is formed by candidate angle adjustment, candidate clamping force, candidate drive speed and candidate feed step size, and then a set of feasible candidate control parameters is obtained by screening. The controller selects a target control parameter group that meets the set conditions from the set of feasible candidate control parameters, and then forms a unified control command. The controller executes the PU lip shaping operation according to the generated unified control command.

2. The PU lip shaping collaborative control method according to claim 1, characterized in that, It also includes the following steps: Once the object to be shaped enters the control cycle, the controller performs object identification and carrier availability checks, reads the object model, PU lip specifications, target inward angle, target deformation score, target deformation range, clamping force safety range, synchronous control boundary, each control cycle, image confidence threshold, temperature rise threshold, communication delay upper limit, and pre-shaping contour reference. Each control cycle together defines the effective control time window. The image confidence threshold, temperature rise threshold, and communication delay upper limit constitute the threshold system of the object to be shaped, and then form a shaping reference relationship based on the same object identifier and timestamp. Determine if the plastic surgery reference relationship is valid. If the result is no, reread the parameters or recalibrate. If the result is still no after rereading the parameters or recalibrating, enter conservative mode, maintain the previous stable parameter group and stop issuing new instructions, and output a re-check mark. If the judgment result is yes, the image acquisition module acquires the clamping posture image, the PU lip contour image, and the fluorescence image.

3. The PU lip shaping collaborative control method according to claim 2, characterized in that, The formula for calculating angular deviation is: , in, The angle deviation is the angle deviation in the k-th control cycle. The target shaping angle for the k-th control cycle. This represents the actual clamping angle during the k-th control cycle. Based on the angle deviation and the previous effective angle adjustment amount Generate reference candidate angle adjustment amount : , The reference candidate angle adjustment amount is constrained by an angle limit to obtain the amplitude-limited reference candidate angle adjustment amount. : , Generate a set of candidate angle adjustment values ​​based on the baseline candidate angle adjustment value after amplitude limiting. : in, To adjust the gain for angle adjustment, For the candidate step size of the angle, The upper limit of angle adjustment, This is the lower limit for angle adjustment. The number of candidate expansions, where n is an integer ranging from -N to N; When the pose marker is occluded, the pose fitting confidence is lower than the image confidence threshold, or the target angle version is mismatched, the previous effective angle adjustment amount is maintained, and a re-inspection marker is output.

4. The PU lip shaping collaborative control method according to claim 3, characterized in that, The formula for calculating the normalized profile deformation is: in, Let be the normalized amount of profile deformation in the k-th control cycle. This represents the current contour offset in the k-th control cycle. This is the reference contour offset of the qualified sample of the same model in the kth control cycle.

5. The PU lip shaping collaborative control method according to claim 4, characterized in that, Also includes: Fluorescent lines in the PU lip are identified, and the local wrinkle normalization characteristics are obtained based on the local bending, discontinuity and brightness concentration of the fluorescent lines. The fluorescence intensity continuity index is obtained based on the proportion of effective continuous segments in the length direction. The formula for calculating the fluorescence deformation score is: in, The fluorescence deformation score for the k-th control period. Let be the normalized amount of profile deformation in the k-th control cycle. This represents the local wrinkle normalization feature for the k-th control cycle. The fluorescence intensity continuity index for the k-th control period. For contour deformation weights, For wrinkle feature weights, As fluorescence continuity weight, , and All values ​​range from 0 to 1. .

6. The PU lip shaping collaborative control method according to claim 5, characterized in that, The formula for calculating the clamping force correction is: in, This is the clamping force correction amount for the k-th control cycle. For scoring bias gain, Score the target deformation. For fluorescence deformation scoring, To mitigate the risk of rebound, To mitigate the risk of rebound, The value ranges from 0 to 1.

7. The PU lip shaping collaborative control method according to claim 6, characterized in that, The step of generating the candidate clamping force includes: The controller obtains the clamping force reference value for the kth control cycle. and the correction direction coefficient of the kth control cycle , Calculate the initial candidate clamping force for the k-th control cycle. : Among them, the wrinkling condition is that the local wrinkle normalization feature reaches a preset wrinkle threshold, and the overpressure condition is that the fluorescence intensity continuity index meets a preset overpressure criterion; when the fluorescence deformation score is lower than the target range and the wrinkle condition and overpressure condition are not triggered, When the wrinkle condition or the overpressure condition is triggered, When the fluorescence image is missing, overexposed, underexposed, or has insufficient effective continuous segments, ; The baseline candidate clamping force for the k-th control cycle is obtained by limiting the original candidate clamping force within a safe range. : , A set of candidate clamping forces for the k-th control cycle is generated based on the baseline candidate clamping force after the amplitude limiting. : in, This is the lower limit of the safe clamping force range. This is the upper limit of the safe range for clamping force. The number of candidate expansions is n, which is an integer ranging from -N to N.

8. The PU lip shaping collaborative control method according to claim 7, characterized in that, The formula for calculating the synchronization ratio is: in, Let be the candidate synchronization ratio for the i-th workstation in the k-th control cycle. Let be the synchronization ratio of the main shaping motion and feed execution action of the i-th station relative to the recipe reference cycle time. The effective synchronization ratio of the i-th workstation is the previous one. Let be the normalized value of the load change at the i-th workstation. Let be the normalized value of the temperature rise at the i-th workstation. The operational health status of the i-th workstation; This is the load penalty coefficient. This is the temperature rise penalty coefficient. This is a health penalty coefficient. Let be the lower limit of the synchronization ratio for the i-th workstation. This represents the upper limit of the synchronization ratio for the i-th workstation; The load change normalized value is obtained by normalizing the effective signal of the execution current, pressure feedback or motion resistance; the temperature rise normalized value is obtained by normalizing the temperature measurement value relative to the allowable temperature rise range of the execution carrier; the operational health is obtained by normalizing the response delay, historical fault records and current image confidence according to the calibrated rules.

9. The PU lip shaping collaborative control method according to claim 8, characterized in that, The process of selecting a feasible set of candidate control parameters includes the following steps: in, Let be the set of feasible candidate control parameters for the k-th control cycle. The k-th control cycle contains candidate elements for candidate angle adjustment, candidate clamping force, candidate drive speed, and candidate feed step size. For the set of candidate control parameters, This is a function for determining the boundary conditions of each project item; when the condition passes... The result is 1; For each element in the feasible candidate control parameter set, the controller calculates the target normalized error of the angle, the target normalized error of the clamping force, the normalized error of the synchronization time, and the normalized error of the process switching disturbance, and then... Calculate the ranking score; The synchronization time normalization error is obtained by normalizing the completion time deviation of the candidate master shaping motion and the feed execution action relative to the same synchronization cycle and the formula reference cycle. in, The ranking score for the m-th element in the k-th control cycle; For the angle target normalization error, To normalize the clamping force target error, To normalize the synchronization time error, This refers to the normalized error caused by process switching disturbances. As the angle error weight, As the weighting factor for clamping force error, As the synchronization time error weight, To switch the perturbation error weights, , , and All values ​​range from 0 to 1. ; choose The elements that have the minimum safety margin and meet the requirements are used as the target control parameter group.

10. The PU lip shaping collaborative control method according to claim 9, characterized in that, Also includes: Target control parameter group Form a unified control command; in, To unify control commands; For the target angle adjustment amount, For the target clamping force, Driven by the target speed, For the target feed step size, The linear velocity of the main shaping motion and feed actuator along the direction of PU lip shaping action; The angle, clamping, main shaping motion, and feed execution carriers are respectively issued; After the unified control command is issued, the displacement increment and completion timestamp of the main shaping motion and the feed motion form a synchronous feedback. The controller receives the synchronous feedback and enters the shaping reference relationship for the next control cycle.

Images