Intelligent embroidery path optimization and execution method and system for net cap embroidery

Abstract

The invention relates to an intelligent embroidering path optimization and execution method and system for net cap embroidering in the technical field of intelligent manufacturing, and the method comprises the steps: obtaining a supporting point coordinate from a high-curvature region, employing finite element analysis to simulate a material deformation behavior, obtaining a deformation prediction matrix, and determining the tension distribution of a flexible structure; a supporting path plan is generated and used for deployment of the fixing device; real-time compensation parameters are generated in combination with the curved surface geometric model, material displacement is monitored in the embroidering process through a feedback control mechanism, and a displacement compensation vector is obtained and used for real-time correction of the stitch direction; according to the contact stability after angle deviation data processing, a final embroidery execution sequence is generated, a synchronous driving algorithm is adopted to coordinate the movement of the fixing device and the embroidery machine, and continuous stitch output is obtained to determine that the pattern precision meets the requirement; and obtaining production log data from continuous stitch output, evaluating an overall efficiency index by adopting a data analysis method, and obtaining an optimized feedback cycle for processing and adjusting a next curved surface structure.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent embroidery path optimization and execution method and system for mesh cap embroidery. Background Technology

[0002] Embroidery, as an important component of the textile and apparel industry, is widely used in clothing, home furnishings, and personalized custom products. Its exquisite craftsmanship and production efficiency directly impact a product's market competitiveness. With consumers increasingly demanding personalized and three-dimensional designs, high-precision embroidery, particularly on flexible products with three-dimensional curved structures such as mesh caps, has become a key direction for industry development.

[0003] Traditional embroidery techniques are primarily designed for flat fabrics. However, when faced with the complex geometry of three-dimensional curved surfaces and soft materials, existing technologies struggle to meet the demands for efficient and precise production, revealing significant technological bottlenecks and necessitating innovative breakthroughs to drive industry upgrades. While traditional embroidery equipment performs well on flat fabrics, fixing and positioning flexible three-dimensional structures like mesh caps becomes a primary challenge. The soft and elastic material of mesh caps makes it difficult to stably fix them on the embroidery machine's working plane using traditional clamps or manual assistance. Insecure fixing can easily lead to material slippage or deformation during embroidery, affecting the accuracy of the pattern.

[0004] Furthermore, the curved surface of the mesh cap makes it difficult to directly adapt two-dimensional embroidery patterns. Traditional equipment lacks the ability to perceive three-dimensional contours and cannot dynamically adjust the stitch path according to the curved surface shape. This makes problems such as pattern stretching, wrinkling, and even thread breakage prone to occur during the embroidery process, severely restricting production efficiency and product quality. Against this backdrop, core technical challenges gradually emerge, with the primary issue being how to achieve precise positioning and dynamic support for three-dimensional curved surface structures. The three-dimensional shapes of mesh caps vary, and their size and curvature differ depending on the style. Traditional static fixtures cannot adapt to this diversity or provide uniform tension support. If the three-dimensional contour of the cap cannot be accurately identified and stable local support cannot be provided, the embroidery needle will struggle to maintain the optimal contact angle with the material during high-speed movement, leading to stitch deviation or uneven material stress. Due to the positioning and support issues, another key technical challenge arises: the dynamic adaptation and compensation of embroidery stitches. The curved surface characteristics of the mesh cap require the stitch path to be adjusted in real time according to the actual surface shape, but existing embroidery machines are typically designed based on two-dimensional planes and lack the ability to map two-dimensional patterns onto three-dimensional curved surfaces. This mapping requires not only precise geometric calculations, but also adjustments to stitch density and direction based on the dynamic deformation of the material during the embroidery process. For example, when embroidering the curved area on the side of a net cap, if the stitches are not adjusted according to the curvature, the pattern may be too dense or sparse in some areas, causing visual disharmony or material damage.

[0005] Therefore, achieving precise positioning and dynamic support of three-dimensional curved structures during the embroidery process, while simultaneously completing real-time stitch adaptation and compensation, has become a key issue in promoting automated embroidery of flexible three-dimensional products such as mesh caps. This issue directly affects production efficiency, pattern accuracy, and product consistency, and urgently requires technological innovation to break through existing bottlenecks and achieve efficient and flexible production methods. Summary of the Invention

[0006] This invention provides an intelligent embroidery path optimization and execution method for mesh cap embroidery, mainly including the following steps:

[0007] Point cloud data of the three-dimensional curved surface structure of the mesh cap is acquired using a scanning device. A three-dimensional reconstruction algorithm is used to process the point cloud data to obtain a surface geometric model, which includes curvature distribution information to determine the three-dimensional contour features of the mesh cap. Local curvature values ​​are extracted from the surface geometric model, and the model is divided into multiple regions using a meshing method. The curvature vector of each region is obtained to determine support requirements. If the curvature vector exceeds a preset threshold, it is marked as a high-curvature region for subsequent dynamic adjustment. Support point coordinates are obtained from the high-curvature region, and finite element analysis is used to simulate material deformation behavior, obtaining a deformation prediction matrix to determine the tension distribution of the flexible structure, thereby generating a support path plan for the deployment of the fixing device. Based on the tension distribution output by the support path plan, a path optimization algorithm is used to adjust the embroidery stitch sequence, obtaining a fitted stitch density map to determine surface adaptability. If the deviation value in the density map is lower than a threshold... The process involves confirming the matching of the stitch path with the curved surface; obtaining the adapted stitch density map; generating real-time compensation parameters by combining the curved surface geometric model; using a feedback control mechanism to monitor material displacement during embroidery and obtaining a displacement compensation vector for immediate correction of the stitch direction; extracting the corrected stitch direction from the displacement compensation vector; capturing the contact angle between the embroidery needle and the material using a machine vision system; obtaining angle deviation data to determine contact stability; activating compensation parameters to update the stitch path if the angle deviation exceeds a threshold; generating the final embroidery execution sequence based on the contact stability after processing the angle deviation data; using a synchronous drive algorithm to coordinate the movement of the fixing device and the embroidery machine to obtain continuous stitch output and confirm that the pattern accuracy meets the requirements; obtaining production log data from the continuous stitch output; using data analysis methods to evaluate the overall efficiency index; and obtaining an optimized feedback loop for the next processing and adjustment of the curved surface structure.

[0008] This invention provides an intelligent embroidery path optimization and execution system for mesh cap embroidery, mainly comprising: a point cloud processing module, used to acquire point cloud data of the three-dimensional curved surface structure of the mesh cap through a scanning device, and to process the point cloud data using a three-dimensional reconstruction algorithm to obtain a curved surface geometric model, wherein the model contains curvature distribution information to determine the three-dimensional contour features of the mesh cap; a region segmentation module, used to extract local curvature values ​​based on the curved surface geometric model, and to segment the model into multiple regions using a meshing method, obtaining the curvature vector of each region to determine support requirements, and marking high curvature regions if the curvature vector exceeds a preset threshold for subsequent dynamic adjustment; a finite element analysis module, used to obtain the coordinates of support points from high curvature regions, and to simulate material deformation behavior using finite element analysis to obtain a deformation prediction matrix to determine the tension distribution of the flexible structure, thereby generating a support path plan for the deployment of the fixing device; and a stitch adaptation module, used to output the tension distribution for the support path plan, and to adjust the embroidery stitch sequence using a path optimization algorithm to obtain an adapted stitch density map to determine the curved surface. The system includes several modules: Adaptability: If the deviation value in the density map is below a threshold, the stitch path is confirmed to match the surface; Real-time compensation module: This module acquires the adapted stitch density map, generates real-time compensation parameters based on the surface geometry model, and uses a feedback control mechanism to monitor material displacement during embroidery, obtaining a displacement compensation vector for immediate correction of the stitch direction; Visual monitoring module: This module extracts the corrected stitch direction from the displacement compensation vector, uses a machine vision system to capture the contact angle between the embroidery needle and the material, obtains angle deviation data to judge contact stability, and activates compensation parameters to update the stitch path if the angle deviation exceeds a threshold; Synchronous drive module: This module generates the final embroidery execution sequence based on the contact stability after processing the angle deviation data, uses a synchronous drive algorithm to coordinate the movement of the fixing device and the embroidery machine, and obtains continuous stitch output to ensure the pattern accuracy meets requirements; Data analysis module: This module acquires production log data from the continuous stitch output, uses data analysis methods to evaluate overall efficiency indicators, and obtains an optimized feedback loop for the next processing and adjustment of the surface structure. The technical solution provided by this embodiment of the invention can include the following beneficial effects:

[0009] This invention discloses an intelligent embroidery path optimization and execution method for mesh cap embroidery. Addressing the problems of poor surface adaptability, insufficient stitch stability, and low production efficiency in complex curved surface mesh cap embroidery, it constructs a complete solution by integrating point cloud data processing, finite element analysis, path optimization, and real-time feedback control. First, this invention acquires point cloud data of the mesh cap using a scanning device. A 3D reconstruction algorithm is then used to generate a curved surface geometric model with curvature distribution. Local curvature values ​​are extracted and meshed, and high-curvature areas are marked to determine support requirements. Based on finite element analysis, material deformation is predicted, generating support path planning and tension distribution. The stitch sequence is adjusted using a path optimization algorithm to ensure the stitch density map adapts to the curved surface. Real-time compensation parameters are dynamically corrected by monitoring material displacement and contact angle through feedback control and machine vision, ultimately generating a high-precision embroidery execution sequence. This invention improves overall efficiency by optimizing the feedback loop through data analysis, achieving high adaptability, stability, and efficiency in complex curved surface embroidery, providing technical support for intelligent manufacturing. Attached Figure Description

[0010] Figure 1 This is a flowchart of the intelligent embroidery path optimization and execution method for the mesh cap embroidery of the present invention.

[0011] Figure 2 This is a schematic diagram of the intelligent embroidery path optimization and execution method for mesh cap embroidery of the present invention.

[0012] Figure 3 This is another schematic diagram of the intelligent embroidery path optimization and execution method for mesh cap embroidery of the present invention.

[0013] Figure 4 This is a schematic diagram of the framework of the intelligent embroidery path optimization and execution system for mesh cap embroidery of the present invention. Detailed Implementation

[0014] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] like Figure 1-4 The intelligent embroidery path optimization and execution method for mesh cap embroidery in this embodiment may specifically include:

[0016] Step S101: Obtain point cloud data of the three-dimensional curved surface structure of the net cap through a scanning device, and process the point cloud data using a three-dimensional reconstruction algorithm to obtain a surface geometric model, wherein the model contains curvature distribution information to determine the three-dimensional contour features of the net cap.

[0017] Three-dimensional point cloud data of the mesh cap is acquired using a scanning device to obtain a multi-view point cloud set. The multi-view point cloud set is aligned using a point cloud registration method to obtain an initial point cloud set. A surface reconstruction algorithm is used to generate a curved surface geometric model for this initial point cloud set. Curvature distribution information is extracted from the curved surface geometric model to obtain the local curvature changes of the mesh cap. Based on these local curvature changes, a contour edge detection method is fused to obtain the three-dimensional contour features of the mesh cap. Whether the three-dimensional contour features meet a preset threshold is determined to obtain a three-dimensional surface structure description of the mesh cap.

[0018] Specifically, in one implementation, acquiring point cloud data of the three-dimensional curved surface structure of the net cap using a scanning device is the initial step of the entire process.

[0019] Specifically, the scanning device can be a laser scanner or a structured light scanner. These devices scan the surface of the mesh cap by emitting a beam of light or a laser, capturing the reflected signals to generate a large amount of discrete point coordinate data, forming a point cloud. The point cloud data represents the three-dimensional spatial position information of the mesh cap surface.

[0020] For example, on a mesh cap production line, a fixed laser scanner is used to scan the stationary mesh cap from all angles, ensuring coverage of all curved areas. This method is suitable for mass production scenarios, enabling efficient data acquisition. In the design verification phase, a handheld structured light scanner can be used for flexible scanning to adapt to adjustments needed for different mesh cap shapes. This data acquisition lays the foundation for subsequent reconstruction. Further, processing the point cloud data using a 3D reconstruction algorithm to obtain the surface geometry model is a crucial step. 3D reconstruction algorithms typically include point cloud preprocessing, surface fitting, and mesh generation. First, the point cloud data is preprocessed, such as removing noise and outliers. This can be achieved through statistical filtering methods, such as calculating the neighborhood density of each point and removing isolated points. Next, surface fitting is performed, using parametric surface models such as NURBS surfaces to fit the point cloud data, ensuring a smooth and continuous model. In the surface reconstruction of mesh caps, this algorithm can handle complex topologies, such as the mesh-like surface of the mesh cap, avoiding data loss.

[0021] It's important to note that the core of 3D reconstruction algorithms lies in iterative optimization. By minimizing the distance error between points and surfaces, the model is refined to obtain an accurate geometric representation. This approach is suitable for mesh cap quality inspection scenarios, such as verifying whether surfaces meet design specifications after production.

[0022] In one possible implementation, the surface geometry model includes curvature distribution information to better characterize the surface properties of the mesh cap. Curvature distribution information refers to the distribution of curvature values ​​at various points on the surface. Curvature is a geometric measure describing the degree of surface bending, including Gaussian curvature and mean curvature. Gaussian curvature reflects the inherent bending of the surface; positive values ​​indicate a spherical shape, while negative values ​​indicate a saddle-like shape. Mean curvature describes the average degree of bending. During model generation, local neighborhoods are first extracted from the reconstructed surface mesh, and then the curvature of each vertex is calculated.

[0023] Specifically, for a point on a curved surface, its first-order neighborhood points are selected to construct a local coordinate system. The principal curvature directions and values ​​are estimated by fitting a quadratic surface, for example, by fitting a polynomial surface using the least squares method and solving for the curvature tensor. This calculation process ensures the accuracy of the curvature distribution, which is particularly useful in the design optimization of net caps, as the curved surface of a net cap often involves multiple curved areas, such as the arched top and undulating edges. By analyzing the curvature distribution, potential stress concentration points can be identified, thus guiding material selection or structural adjustments. In practical applications, such as when performing curvature analysis on sports net caps, it was found that high curvature areas may cause discomfort, so the design was adjusted to smooth these areas. This method not only improves the model's descriptive power but also provides a data foundation for subsequent feature determination.

[0024] For example, the 3D contour features of the mesh cap are determined based on the curvature distribution information of the model. The 3D contour features include the overall shape parameters of the mesh cap, such as the location of the point of maximum curvature, the length of the contour line, and the surface convexity / concave distribution. These features are extracted by analyzing the curvature distribution.

[0025] For example, extreme points of curvature are identified as key feature points, and these points are then connected to form the contour skeleton. In determining the 3D contour of the mesh cap, the curvature distribution is first thresholded, with high-curvature areas marked as convex features and low-curvature areas marked as flat features. This extraction process can be applied to personalized customization scenarios, such as scanning the mesh cap according to the user's head shape and determining the contour features to ensure a good fit.

[0026] Preferably, in another embodiment, the acquisition of point cloud data can be combined with multi-view scanning to improve accuracy.

[0027] For example, multiple scanning devices can be used to simultaneously acquire net cap data from different angles, and then a registration algorithm can be used to fuse multiple sets of point clouds into a unified coordinate system. This fusion process involves feature point matching and iterative nearest-point algorithms to ensure that the data has no overlap or gaps, and is suitable for complex net cap structures such as those with decorative edges.

[0028] Specifically, machine learning methods can be introduced to assist in surface fitting in the processing of 3D reconstruction algorithms.

[0029] For example, a neural network model can be used to learn the mapping relationship between point clouds and surfaces, and parameters can be optimized using a training dataset to handle noisy data in net cap reconstruction. This approach enhances the robustness of the algorithm. Furthermore, the calculation of curvature distribution information can be extended to dynamic analysis. In net cap usage simulations, the deformed surface is considered, and the variational curvature distribution is calculated to predict durability.

[0030] For example, virtual pressure can be applied to the net cap, and changes in curvature can be observed to determine the reinforcement zone.

[0031] In one embodiment, the determination of the three-dimensional profile features includes the calculation of quantitative indicators, such as the calculation of the average curvature value and curvature variance of the entire net cap, which are used to classify different net cap types.

[0032] For example, in production quality control, if the curvature variance exceeds a threshold, the product is considered defective. This quantification helps in standardized evaluation.

[0033] Understandably, the overall effect of the above process is to achieve a precise digital representation of the mesh cap surface. By using a model that includes curvature information, the three-dimensional contour features can be effectively determined, thereby improving design and inspection efficiency in the manufacturing industry.

[0034] For example, in batch inspection of mesh caps, the above steps are combined to form a closed-loop process from scanning to feature determination, ensuring the consistency of the contour of each mesh cap.

[0035] Step S102: Extract local curvature values ​​based on the surface geometry model, divide the model into multiple regions using a meshing method, obtain the curvature vector of each region to determine support requirements, and mark the region as a high curvature region for subsequent dynamic adjustment if the curvature vector exceeds a preset threshold.

[0036] Local curvature values ​​are obtained from the surface geometry model. The model is divided into multiple regions using a meshing method, resulting in a curvature vector for each region. The curvature vector is then used to calculate the Gaussian curvature by multiplying the principal curvatures within the region. The Gaussian curvature is defined as the product of the two principal curvatures at a point on the surface. Based on the curvature vector, it is determined whether it exceeds a preset threshold. If it does, it is marked as a high curvature region. For the high curvature regions, a surface texture mapping is integrated. This surface texture mapping is generated by superimposing texture coordinates onto the region's curvature distribution to obtain texture adjustment parameters. Based on these texture adjustment parameters, the support configuration is dynamically adjusted to obtain the adjusted support configuration.

[0037] Specifically, in one implementation, extracting local curvature values ​​using a surface geometry model first requires understanding the basic structure of that model. This model is typically represented as a three-dimensional object surface, composed of points, edges, and faces, used to simulate the object's shape. When extracting local curvature values, a differential geometry method can be used to calculate the principal curvature at each point on the model surface.

[0038] Specifically, for a given point on a curved surface, its curvature in two orthogonal directions is calculated, namely the maximum curvature and the minimum curvature, thus obtaining the average curvature or Gaussian curvature of that point. These curvature values ​​reflect the degree of surface curvature and, for example, are used in the field of 3D printing to assess whether the surface needs additional support to prevent deformation. Further, a meshing method is used to divide the model into multiple regions. This method converts the curved surface geometry into a triangular or quadrilateral mesh, dividing the surface into small regions through uniform or adaptive partitioning. For example...

[0039] In one possible implementation, a uniform mesh is used. The model surface is projected onto a two-dimensional plane, cut into rectangular regions at fixed intervals, and then mapped back to three-dimensional space. Each region obtained in this way contains several surface points, which facilitates subsequent analysis.

[0040] It should be noted that the purpose of mesh generation is to localize the processing, avoid the complexity of global calculations, and optimize computational efficiency during the 3D printing preparation stage. Based on the above mesh generation, the curvature vector of each region is obtained to determine the support requirements. The curvature vector can be understood as a multi-dimensional vector, where each component corresponds to information such as the average curvature value and principal curvature direction of points within the region.

[0041] Specifically, for a given region, the curvature values ​​of all points within it are aggregated to calculate a vector such as (mean curvature, maximum curvature deviation, curvature gradient). These components are obtained through weighted averaging or statistical methods. In 3D printing, this vector is used to quantify the geometric stability of a region. For example, regions with high curvature may be prone to collapse during printing, thus requiring a determination to add support structures. This method ensures the accuracy of model processing. If the curvature vector exceeds a preset threshold, it is marked as a high-curvature region for subsequent dynamic adjustments. The preset threshold can be set based on experience or simulation tests; for example, setting the average curvature threshold to 0.1 per millimeter, and marking the region when the vector magnitude exceeds this value.

[0042] Specifically, during the judgment process, the Euclidean norm or specific components of the vector are calculated and compared with a threshold. If the value exceeds the threshold, it is marked as a high curvature region. This marking supports subsequent steps, such as dynamically adjusting printing parameters or adding support columns.

[0043] In one embodiment, for a spherical surface model, the high curvature region is concentrated at the poles. By marking and adjusting the support density, the printing success rate can be improved.

[0044] For example, in another implementation, when extracting curvature for complex surfaces such as automotive parts models, normal vectors can be used to assist in the calculation, ensuring that the vectors accurately reflect the bending requirements.

[0045] Preferably, the threshold can be dynamically adjusted, with different values ​​set according to the material type, such as plastic or metal, thus adapting to various scenarios within the same 3D printing field. This method enables refined model processing and improves overall manufacturing quality.

[0046] Step S103: Obtain the coordinates of the support points from the high curvature region, use finite element analysis to simulate the material deformation behavior, obtain the deformation prediction matrix to determine the tension distribution of the flexible structure, and thus generate a support path plan for the deployment of the fixing device.

[0047] Finite element analysis is used to mesh the flexible structure, obtain material deformation behavior data, and generate a deformation prediction matrix. Based on the deformation prediction matrix, tension distribution data of the flexible structure is obtained. For this tension distribution data, a support path plan is generated to determine the deployment location of the fixing device. Based on the support path plan, sensor feedback data is obtained to acquire real-time tension distribution information. If the real-time tension distribution information exceeds a preset threshold, the coordinates of the support points are adjusted to obtain an optimized support path plan. Using the optimized support path plan, real-time deformation data is obtained to determine the final deployment scheme of the fixing device, thus generating the support path plan for the flexible structure.

[0048] Specifically, in one implementation, the process of obtaining the coordinates of the support points from the high curvature region first requires analyzing the geometric model of the flexible structure. The flexible structure can be, for example, a flexible catheter used in medical devices, whose surface often exhibits irregular curvature. To identify the high curvature region, the curvature value can be calculated.

[0049] Specifically, curvature values ​​represent a measure of the degree of surface bending. In areas of high curvature, the curvature values ​​are larger, and these areas are prone to deformation. Coordinate points are extracted from the point cloud by scanning surface data of a flexible structure, for example, by using a 3D scanner to acquire point cloud data, and then applying curvature calculation algorithms.

[0050] For example, in the application of flexible conduits, high curvature regions may be located at bends, and obtaining the coordinates of these support points helps improve the accuracy of subsequent simulations. This method ensures that the selection of support points is based on actual geometric features, avoiding errors caused by random selection. Furthermore, simulating material deformation behavior using finite element analysis is a core step. Finite element analysis is a numerical simulation method used to predict the response of materials under external loads. In this embodiment, the geometric model of the flexible structure is first discretized into a finite number of element meshes, for example, using tetrahedral elements to mesh the flexible conduit. Material properties, such as elastic modulus and Poisson's ratio, are defined within each element, and then boundary conditions and loads are applied. The simulation process includes solving equilibrium equations and calculating the displacement and stress at each node. Through this simulation, the deformed structural morphology can be obtained.

[0051] It's important to note that the challenge of finite element analysis lies in setting the mesh quality and boundary conditions. If the mesh is too coarse, it may lead to insufficient simulation accuracy. In flexible conduit scenarios, such as simulating its bending deformation upon insertion into the human body, this analysis helps reveal potential stress concentration points, thus providing a data foundation for tension distribution. This step relies on commercial software or custom programs, but the key is the appropriate selection of parameters to ensure reliable simulation results. Based on the simulation results, a deformation prediction matrix is ​​obtained to determine the tension distribution of the flexible structure. The deformation prediction matrix is ​​a mathematical representation used to quantify the degree of deformation and tension changes.

[0052] Specifically, this matrix can be constructed using the displacement vector and stress tensor output from finite element analysis. First, displacement data for each support point is extracted from the simulation to form a displacement matrix, and then the tension value is calculated using the equations of mechanics of materials. The tension distribution represents the distribution of tensile forces within the flexible structure, which is often higher in regions of high curvature.

[0053] For example, in flexible conduits, the tension distribution can be predicted by matrix operations to determine the axial force variation, where the matrix elements correspond to tension coefficients in different regions. This matrix generation process involves linear algebraic operations, such as matrix multiplication, to integrate deformation data.

[0054] Preferably, a correction factor can be introduced to adjust the matrix accuracy, for example, by fine-tuning element values ​​based on actual test data. The innovation of this process lies in transforming simulated data into a quantifiable predictive tool, which helps identify areas of uneven tension, thereby optimizing structural design. In the field of medical devices, this method can improve the stability of flexible structures and reduce the risk of breakage during use.

[0055] In one possible implementation, the generation of support path planning for the deployment of the fixing device depends on the aforementioned tension distribution. Support path planning refers to planning the placement path of the fixing device based on tension data, such as the order and position of deploying support clamps on a flexible conduit. By analyzing the deformation prediction matrix, areas with high tension are identified as priority fixing points, and then a path optimization algorithm is used to generate the path connecting these points. Path planning can employ a variant of the A* algorithm to ensure that the path minimizes interference and covers critical areas. The output of this planning is a coordinate sequence that guides the automatic deployment of the fixing device. Furthermore, in various implementation scenarios of flexible structures, parameters can be adjusted to adapt to different conditions.

[0056] For example, in scenarios where flexible catheters are used in endoscopes, the coordinates of support points in high-curvature areas can be obtained by combining real-time imaging data to improve accuracy. Finite element analysis can simulate deformation behavior at different temperatures, obtaining a more comprehensive prediction matrix, thereby generating a more adaptable support path. This diversity ensures the versatility of the technical solution.

[0057] Specifically, the construction process of the deformation prediction matrix needs a detailed explanation of its principles. The rows of the matrix correspond to support points, and the columns correspond to deformation dimensions, such as displacements in the x, y, and z directions. Displacement data obtained through finite element simulation fills the matrix, and then Hooke's law is applied to calculate tension, i.e., stress equals the elastic modulus multiplied by strain. Strain is derived from the displacement gradient, thus presenting the tension distribution in matrix form. In flexible ducts, for example, the matrix shows that the tension values ​​in curved sections are higher than in straight sections, guiding path planning to prioritize fixing high-tension areas. The detail of this process lies in the iterative solution; if the initial simulation has large deviations, the analysis can be run multiple times to refine the matrix. This explanation clarifies the logic of converting the matrix from simulation to tension.

[0058] Understandably, the deployment of fixed equipment is automated through the generated support paths.

[0059] In one embodiment, after path planning is output, the robotic arm places fixing devices, such as grippers, to secure the high-curvature region of the flexible conduit according to the coordinate sequence. This deployment reduces manual intervention and improves efficiency.

[0060] Preferably, in another implementation scenario, given the determined tension distribution of the flexible structure, dynamic load simulation can be incorporated. The finite element analysis is extended to the time domain to predict the deformation matrix that changes over time, thereby generating a dynamic support path. This extension enhances the applicability of the solution under variable load conditions, such as the fixation of a flexible duct during movement.

[0061] For example, in practice, the number of support point coordinates obtained from high-curvature regions can be adjusted according to structural complexity. If the catheter is long, more points are selected for comprehensive coverage, and then the finite element simulation is performed segment by segment, generating local matrices segment by segment, which are finally integrated into an overall deformation prediction matrix. This segmented method reduces computational complexity and ensures the accuracy of tension distribution. Furthermore, this technical solution improves the durability of flexible structures. Through detailed simulation and planning, the fixation device deployment is more precise, reducing failures caused by deformation. In the field of medical devices, this translates to higher safety and reliability.

[0062] Step S104: For the tension distribution output by the support path planning, the path optimization algorithm is used to adjust the embroidery stitch sequence to obtain the adapted stitch density map to judge the surface adaptability. If the deviation value in the density map is lower than the threshold, it is confirmed that the stitch path matches the surface.

[0063] An initial embroidery stitch sequence is obtained from the tension distribution output. This initial embroidery stitch sequence contains multiple stitch points and their tension values. A genetic algorithm is used to optimize the initial embroidery stitch sequence, resulting in an optimized stitch sequence. This optimized stitch sequence generates a preliminary fit density map by calculating the tension differences between sequences. Based on the preliminary fit density map, the deviation value distribution required for surface adaptation is obtained. This deviation value distribution is calculated by comparing the differences between the stitch points and surface coordinates in the preliminary fit density map. For the portion of the deviation value distribution below a preset threshold, the stitch path adjustment direction is determined. Based on the stitch path adjustment direction, tension equilibrium points are obtained from the optimized stitch sequence. These tension equilibrium points are calculated using the tension values ​​at the equilibrium points. A path optimization algorithm is used to generate an extended surface tension model. This extended surface tension model takes the tension equilibrium points as input and outputs an extended tension distribution, resulting in an enhanced fit stitch density. If the overall deviation value distribution in the enhanced fit stitch density is below a preset threshold, the surface matching is considered complete. For the remaining deviation value distribution, the optimized stitch sequence is adjusted to obtain the final stitch path.

[0064] Specifically, in one implementation, the process of output tension distribution for support path planning is first described. Support path planning refers to planning initial stitch paths for curved surfaces such as fabric or leather in embroidery design. These paths need to take into account the curvature of the surface and the support structure to ensure that the stitches do not deform due to uneven tension.

[0065] Specifically, the tension distribution is calculated by simulating the tensile force of the thread during the embroidery process.

[0066] For example, the curved surface model is divided into multiple grid regions, and a virtual mechanical model is applied to the path points in each region to calculate the tension value distribution map. This distribution map shows the tension intensity at each point on the path; if the tension in a certain region is too high, it may cause the stitches to loosen or break. This step provides basic data for subsequent optimization, ensuring the stability of the embroidery pattern on the curved surface. Further, a path optimization algorithm is used to adjust the embroidery stitch sequence. The path optimization algorithm is an iterative algorithm used to modify the initial stitch sequence to adapt to the characteristics of the curved surface.

[0067] For example, the algorithm can work based on the gradient descent method, first taking an initial path and tension distribution as input, then calculating the degree of matching between the curvature of the path and the surface, and if there is a mismatch, adjusting the position of the stitch points, for example, shifting the stitch points in the sequence towards a low-tension region.

[0068] It's important to note that this adjustment process is repeated multiple times until the tension distribution becomes uniform. In embroidery, this helps handle complex curved surfaces such as cuffs or necklines, ensuring the stitch sequence doesn't distort due to the curvature, thus improving embroidery quality. Obtaining the adapted stitch density map is a crucial subsequent step. The stitch density map is generated by statistically optimizing the distribution density of stitch points in the sequence.

[0069] For example, projecting a curved surface onto a two-dimensional plane and calculating the number of stitches per unit area creates a density heatmap. This map visually displays density deviations; if some areas have excessively high density, it may indicate stitch accumulation, leading to uneven embroidery. One possible implementation is to use interpolation to fill the density map, ensuring coverage of the entire curved surface and providing complete data for fit assessment. When assessing surface fit, if the deviation value in the density map is below a threshold, the stitch path is considered to match the surface. The deviation value refers to the difference between the actual density and the ideal uniform density in the density map.

[0070] For example, quantification can be achieved by calculating the standard deviation; if the standard deviation is less than a preset threshold such as 0.05, it is considered a match.

[0071] Preferably, in embroidery production scenarios, this judgment can be integrated into a computer-aided design system to automatically output matching results. If the deviation value exceeds a threshold, the process returns to the path optimization step for readjustment. This mechanism ensures accurate adaptation of the stitch path and reduces the need for manual correction. In another embodiment, the application of different embroidery materials is considered.

[0072] For example, for curved silk surfaces, support path planning needs to consider the material's elasticity coefficient, incorporating an elastic model into the tension distribution calculation to simulate the tension changes of the thread on a soft surface. The path optimization algorithm can employ a variant of a genetic algorithm, simulating multiple generations of path sequence evolution to select the sequence with the most uniform tension distribution to generate a density map. When determining suitability, the threshold can be adjusted according to the material type; for example, a threshold of 0.03 for silk can be set to ensure a fine match. This embodiment demonstrates the flexibility of the technical solution in the embroidery field, adapting to various curved fabric surfaces. Further extending this approach, when handling large embroidery patterns, such as the curved surface adaptation of decorative tapestries, the initial support path planning can use 3D scanning technology to obtain a surface model, and then output the tension distribution. The optimization algorithm for adjusting the stitch sequence can be combined with a machine learning model, learning from historical embroidery data to predict the optimal path. The resulting density map displays deviation areas through visualization tools; if the deviation is below the threshold, the system confirms a match; otherwise, iterative optimization continues. This method can improve efficiency in practical applications, such as reducing the number of trial embroidery attempts and achieving more accurate transfer of curved patterns.

[0073] Understandably, the effectiveness of this technical solution lies in ensuring a seamless match between embroidery stitches and curved surfaces through systematic path adjustment and density judgment. In one scenario, for highly curved surfaces such as shoe upper embroidery, deviation value judgment can effectively prevent stitch deformation and provide reliable fit confirmation.

[0074] Step S105: Obtain the adapted stitch density map, generate real-time compensation parameters by combining the curved surface geometric model, and use a feedback control mechanism to monitor material displacement during the embroidery process to obtain a displacement compensation vector for real-time correction of the stitch direction.

[0075] A curved geometric model is obtained from the surface of the embroidery material, and initial curvature data is extracted from the curved geometric model. Based on a preset density threshold, the initial curvature data is compared in density distribution to obtain a fitted stitch density map. A vector calculation method based on density distribution is used to generate compensation parameters from the fitted stitch density map. If the compensation parameters exceed a preset threshold, material displacement data is captured in real time by a sensor, and the feedback control mechanism is adjusted. The material displacement data obtained by the feedback control mechanism is vector-converted to obtain a displacement compensation vector. Based on the displacement compensation vector, the surface texture deformation of the embroidery material is monitored, and the degree of texture deformation is determined. The embroidery path is adjusted according to the degree of texture deformation to obtain a corrected stitch path.

[0076] Specifically, in one implementation, a adapted stitch density map is first obtained. This process involves converting the initial embroidery design pattern into a density distribution representation suitable for a specific material surface.

[0077] Specifically, a stitch density map refers to the distribution density of embroidery stitches per unit area, which is generated by scanning the surface of the embroidery material and applying image processing technology.

[0078] For example, optical sensors can be used to capture material texture, and then density values ​​can be adjusted according to the outline of the design pattern to ensure uniform density distribution in curved areas. This adaptation helps avoid pattern deformation caused by material stretching during embroidery. Through this step, the system can provide basic data for subsequent compensation, ensuring the stability of embroidery quality. Furthermore, real-time compensation parameters are generated by combining a surface geometry model. The surface geometry model is a three-dimensional mathematical representation of the embroidery material surface, typically constructed based on parametric surfaces such as Bézier surfaces. In the field of embroidery, this model is used to simulate the bending and wrinkling of fabric. The specific process includes first measuring the surface point cloud data of the material, then fitting the surface equation, and calculating the curvature and normal vector.

[0079] For example.

[0080] In one possible implementation, the system acquires the material position on the embroidery machine in real time and uses this data to generate compensation parameters, such as scaling factors and rotation angles. These parameters are used to adjust the stitch path to match the geometry of the curved surface.

[0081] It's important to note that the key to this integration process lies in the dynamic updating of parameters. For example, when the material undergoes slight deformation, the model recalculates the compensation value, thereby achieving precise alignment of the embroidery pattern. This technology can effectively handle different fabric types in embroidery production, such as adapting to the curved surfaces of silk or cotton materials, ensuring that the pattern is not distorted due to geometric differences.

[0082] Preferably, a feedback control mechanism is employed to monitor material displacement during the embroidery process. The feedback control mechanism is a closed-loop system comprising sensors, a controller, and actuators. In the embroidery scenario, this mechanism monitors material position changes in real time using displacement sensors mounted on the embroidery machine.

[0083] Specifically, sensors such as laser rangefinders or cameras capture the minute movements of the material beneath the needle, creating a displacement data stream. The controller then analyzes this data based on preset thresholds; if the displacement exceeds an allowable range, an adjustment signal is triggered. This process operates on the principle of negative feedback, generating control commands by comparing the deviation between the actual displacement and the expected position.

[0084] For example, when embroidering lace patterns, if the material slips due to tension, the mechanism will immediately detect and record the displacement vector, thereby maintaining the continuity of the embroidery.

[0085] In one embodiment, a displacement compensation vector is obtained for real-time correction of the stitch direction. The displacement compensation vector is a vector representing the direction and magnitude of material movement, calculated from displacement data acquired through a feedback mechanism. Specifically, the process involves first extracting the displacement components along the x, y, and z axes from the monitoring data, and then applying vector addition to synthesize the compensation vector.

[0086] For example.

[0087] Understandably, during the embroidery process, if the material shifts 2 millimeters to the right, the system generates a reverse vector to correct the stitch direction. This correction is achieved by adjusting the stepper motor of the embroidery machine, ensuring the needle enters the material along the correct path. In practical applications, this vector compensation is particularly useful for embroidering complex patterns, such as embroidering curved surfaces at garment cuffs, reducing scrap rates and improving production efficiency. Through this instant correction, the embroidered pattern maintains the integrity and aesthetics of the design.

[0088] For example, in another implementation, the above process can be applied to an industrial embroidery machine production line.

[0089] Specifically, after acquiring the density map, parameters are generated using the model. Then, a feedback mechanism monitors displacement and calculates a compensation vector. This integrated approach demonstrates the technology's versatility; for example, when processing patterns of different densities, the system can flexibly adjust parameters to adapt to various embroidery tasks. Furthermore...

[0090] It should be noted that the construction of the surface geometry model can include multi-layer mesh representation to improve the accuracy of the compensation parameters. In the field of embroidery, this model helps to handle dynamic deformations, such as the natural stretching of fabric during the embroidery process.

[0091] In one possible implementation, the feedback control mechanism can be integrated with a visual recognition module to further improve the accuracy of displacement monitoring.

[0092] For example, a camera can be used to capture the edges of materials in real time to help calculate the displacement vector.

[0093] Preferably, the application of the displacement compensation vector includes a threshold filtering step to avoid noise interference with the correction process. This step ensures stable correction of the embroidery direction.

[0094] For example, in the production of embroidered pillowcases, the entire process, from density map adaptation to vector correction, can achieve highly efficient automated operation.

[0095] Understandably, these implementation methods are all limited to the field of embroidery, demonstrating various application forms of the technical solutions in the same business scenario.

[0096] Step S106: Extract the corrected stitch direction from the displacement compensation vector, use a machine vision system to capture the contact angle between the embroidery needle and the material, obtain angle deviation data to judge contact stability, and activate compensation parameters to update the stitch path if the angle deviation exceeds the threshold.

[0097] The contact angle between the embroidery needle and the material is captured by a machine vision system, and the difference between the contact angle and a preset standard angle is obtained to obtain angle deviation data. The contact stability is determined by comparing the angle deviation data with a preset threshold. If the angle deviation data exceeds the preset threshold, compensation parameters are activated, and the stitch path is updated to obtain an adjusted stitch path. The material contact analysis result is obtained by comparing the contact point coordinates of the adjusted stitch path with the stitch path before adjustment. The pattern boundary offset value is obtained by comparing the material contact analysis result with the boundary overlap of the ideal pattern, and the boundary stability is determined. An offset compensation matrix is ​​generated for the pattern boundary offset value, and the boundary offset value is vector-weighted and corrected using the offset compensation matrix to obtain pattern integrity data. Based on the pattern integrity data, the embroidery execution sequence is updated to obtain an optimized embroidery execution sequence.

[0098] Specifically, in one implementation, the process of extracting the corrected stitch direction from the displacement compensation vector first involves acquiring displacement data of the embroidery machine during operation.

[0099] Specifically, the displacement compensation vector is calculated using the offset generated when the embroidery needle moves across the material surface.

[0100] For example, when an embroidery needle shifts position due to material tension or mechanical vibration, the system records the displacement components along the x and y axes, forming a two-dimensional vector. By analyzing the direction of this vector, the corrected stitch direction can be extracted, which helps ensure that the stitch path matches the design pattern. On an embroidery production line, this extraction can be performed in real time to adapt to the embroidery needs of different materials. Capturing the contact angle between the embroidery needle and the material using a machine vision system is a crucial step. Machine vision systems typically include a high-resolution camera mounted on the embroidery machine and an image processing unit.

[0101] For example, the camera captures an image of the embroidery needle falling at a frequency of 30 frames per second, and then an edge detection algorithm identifies the contact point between the needle tip and the material surface. The contact angle is obtained by calculating the angle between the needle axis and the material plane.

[0102] For example, if the needle axis is perpendicular to the material, the angle is 0 degrees; if there is tilt, the specific deviation value is calculated. This capture method can be applied to various pattern embroidery scenarios in the embroidery field, such as floral patterns on textiles or decorative lines on clothing, ensuring the accuracy of angle data. After obtaining the angle deviation data, it is necessary to determine the contact stability. The angle deviation data is obtained from the difference between the captured contact angle and the preset standard angle.

[0103] Specifically, the standard angle is usually set to the value when the embroidery needle enters the machine ideally vertically, such as 90 degrees. The system compares the actual angle with the standard value. If the absolute value of the difference is less than a preset threshold (such as 5 degrees), the contact is considered stable; otherwise, it is considered unstable. This judgment process is based on real-time data processing and can respond quickly even when the embroidery machine is running at high speed.

[0104] In one possible implementation, for embroidery on heavy materials such as leather, the threshold can be adjusted to a more lenient range to accommodate natural deviations caused by the material's characteristics, thereby improving overall embroidery efficiency. Furthermore, if the angular deviation exceeds the threshold, compensation parameters are activated to update the stitch path. These compensation parameters include adjustment coefficients for the stitch path, such as scaling factors for the displacement vector or directional rotation angles.

[0105] For example, when the deviation exceeds a threshold, the system calculates a compensation vector based on the deviation data and superimposes it onto the original stitch path to update the path. This update can dynamically correct the error during the embroidery process, ensuring that the final pattern is distortion-free.

[0106] In one embodiment, for embroidery scenarios with multiple layers of materials, activating compensation can reduce the scrap rate caused by stitch deviation and provide a more stable output effect.

[0107] It should be noted that the integration of the above processes can be achieved through the control software of the embroidery machine, forming a closed-loop control from displacement vector extraction to path update. In practical applications, this technical solution is suitable for both household embroidery machines and industrial embroidery equipment, demonstrating its versatility in the embroidery field. For example, in fine embroidery on silk or large-scale production on cotton, stitch quality can be optimized by adjusting thresholds and compensation parameters.

[0108] Preferably, in another embodiment, the machine vision system can be combined with infrared illumination to improve capture accuracy in low-light environments, further enhancing the reliability of angular deviation judgment. This optional feature does not change the core process but can be extended to embroidery scenarios with continuous nighttime production.

[0109] Understandably, through these steps, the system can effectively handle the dynamic interaction between the embroidery needle and the material, ensuring contact stability and stitch accuracy. In various scenarios within the embroidery field, such as designs with complex patterns, this solution can significantly reduce the impact of deviations and provide reliable technical support.

[0110] Step S107: Based on the contact stability after processing the angle deviation data, the final embroidery execution sequence is generated. A synchronous drive algorithm is used to coordinate the movement of the fixing device and the embroidery machine to obtain continuous stitch output and ensure that the pattern accuracy meets the requirements.

[0111] Angle deviation data is acquired by measuring the relative angle between the fixing device and the embroidery machine needle position. Based on the difference between the angle deviation data and a preset standard angle, a contact stability index is calculated. It is then determined whether the contact stability index reaches a preset threshold, resulting in a stable contact sequence. Based on the stable contact sequence, the synchronization drive algorithm generates an initial embroidery execution sequence by matching the timing of the fixing device position and the embroidery machine needle position. To address fabric tension variations, the initial embroidery execution sequence is adjusted by monitoring the deviation between tension sensor data and the path of the initial embroidery execution sequence, resulting in a continuous stitch output path. If the continuous stitch output path meets the pattern accuracy requirements, trajectory calibration verification is performed by comparing the matching degree between the continuous stitch output path points and the preset pattern outline, yielding the final pattern accuracy index.

[0112] Specifically, in one implementation, the angle deviation data is first processed to assess contact stability.

[0113] Specifically, the angular deviation data is derived from the relative position monitoring of the embroidery machine's fixing device and the embroidery needle, with real-time deviation values ​​collected by sensors. These deviation values ​​reflect the contact state between the fabric and the needle during the embroidery process. If the deviation exceeds a threshold, it may lead to unstable stitches. The processing includes filtering algorithms to remove noise, followed by calculating the average deviation and variance to quantify stability.

[0114] For example, during the embroidery pattern design stage, this data is input into a stability model to generate a stability score, which is used for subsequent sequence adjustments. Furthermore, based on the processed contact stability, the final embroidery execution sequence is generated.

[0115] It should be noted that the execution sequence consists of a series of stitch instructions, including the needle position, speed, and direction. The innovation lies in incorporating a stability score into the sequence generation algorithm, ensuring that the sequence adapts to different fabric types.

[0116] In one possible implementation, for cotton fabrics, the sequence would include buffering instructions to compensate for low stability, while for silk fabrics, it would be optimized into a high-precision short stitch sequence. In this way, driven by stability data, the sequence can be dynamically adjusted to avoid pattern distortion.

[0117] Preferably, a synchronous drive algorithm is used to coordinate the movement of the fixing device and the embroidery machine. The core of this algorithm is a real-time synchronization mechanism to ensure that the movement of the fixing device is consistent with the stitch movement of the embroidery machine. The specific process includes establishing a motion model, in which the drive motor of the fixing device receives the speed signal of the embroidery machine and adjusts its position through feedback loops.

[0118] For example, when handling angular deviations, if an increase in deviation is detected, the algorithm sends a synchronization pulse to slow down the speed of the fixing device to match the embroidery machine, thereby maintaining contact stability. The key to this coordination mechanism is timestamp synchronization, which avoids stitch misalignment caused by delays. In this way, the system achieves precise motion matching, supporting the embroidery of complex patterns.

[0119] In one embodiment, a continuous stitch output is obtained to determine if the pattern accuracy meets requirements. The continuous stitch output is achieved stitch by stitch on the embroidery machine by executing a generated sequence.

[0120] Specifically, the output process monitors the error between the actual and expected positions of each stitch. If the error is less than a preset value, such as 0.1 mm, the accuracy is considered satisfactory. For example, when embroidering floral patterns, the sequence ensures continuous stitches, forming smooth curves, while stability processing prevents thread breaks caused by fabric shifting. The technical effect of this output mechanism is to improve the overall accuracy of the pattern, making it suitable for large-scale embroidery production.

[0121] Understandably, in another implementation, for multi-layer fabric embroidery scenarios, the angular deviation data processing will additionally consider interlayer friction, incorporating friction compensation parameters when generating the sequence. The synchronous drive algorithm is extended to multi-axis coordination, with the fixing device and embroidery machine moving synchronously on the X and Y axes. The final stitch output is verified for accuracy through visual feedback, ensuring that the pattern remains consistent across different fabric thicknesses. Furthermore, the assessment of contact stability can be optimized using historical data.

[0122] For example, in continuous embroidery tasks, the system accumulates deviation data to form a learning model, which is used to predict potential instabilities and thus adjust the execution sequence in advance. This approach enhances the system's adaptability.

[0123] In one embodiment, the synchronous drive algorithm is specifically implemented using proportional-integral-derivative (PID) control to precisely coordinate the movement speed. This control shortens the response time of the fixing device, ensuring the continuity of the stitch output.

[0124] For example, in the context of embroidery machine maintenance, testing contact stability before generating a sequence can detect device wear early and improve overall efficiency.

[0125] It should be noted that these implementation methods are all limited to the field of embroidery, supporting the universality of the technical solutions in the claims.

[0126] Step S108: Obtain production log data from continuous stitch output, use data analysis methods to evaluate overall efficiency indicators, and obtain an optimized feedback loop for the next processing and adjustment of the curved structure.

[0127] Production log data is acquired from continuous stitch output, and data analysis methods are used to evaluate overall efficiency indicators, yielding preliminary optimization feedback. Based on this preliminary optimization feedback, adjustment parameters for surface structure processing are determined through log data fusion, and continuous stitch monitoring results are obtained. Based on these continuous stitch monitoring results, it is determined whether the adjustment parameters exceed a preset threshold, resulting in feedback iteration requirements. Based on these feedback iteration requirements, the surface structure processing parameters are adjusted, and adjusted continuous stitch monitoring results are obtained. Using these adjusted continuous stitch monitoring results, overall efficiency indicators are re-evaluated, yielding optimized efficiency feedback. Based on this optimized efficiency feedback, the surface structure processing parameters are updated, and the final production adjustment plan is determined.

[0128] Specifically, in one implementation, production log data can be obtained from continuous stitch output using sensors deployed on the sewing equipment. These sensors monitor the stitch formation process in real time, capturing data on the depth, speed, and position of each stitch, for example, when sewing curved surfaces such as the curved surfaces of textiles.

[0129] Specifically, the sensors include optical sensors and force feedback sensors. The former tracks the visual continuity of stitches, while the latter measures the pressure distribution of the stitches on the material. This data generates a production log, including timestamps, stitch counts, and anomaly markers, forming a complete data stream for subsequent analysis. This approach ensures data continuity and reliability, and is suitable for various sewing scenarios in the manufacturing of curved textile structures, such as the joining of curved surfaces in garments or tents.

[0130] For example, stitch output refers to the continuous stitch marks produced by a sewing machine on curved materials, which carry key information about the production process.

[0131] It should be noted that the process of acquiring production log data involves a data acquisition module, which is integrated into the equipment control system and records stitch parameters in real time.

[0132] For example, in scenarios involving composite material surfaces, log data might include stitch density and tension values, which are directly output from sensors and stored in a local database. This acquisition method not only captures real-time data but also handles noise interference, using filtering algorithms to initially clean the data and ensure log accuracy. In practical applications, this helps track production bottlenecks, such as surface deformation caused by uneven stitching. When using data analysis methods to evaluate overall efficiency metrics, the collected log data can be processed using statistical models.

[0133] Specifically, efficiency metrics include production speed, defect rate, and resource utilization, which are evaluated based on stitch data in the logs.

[0134] For example, production speed can be obtained by dividing the number of stitches by the time interval, while the defect rate is quantified by identifying the proportion of abnormal stitches.

[0135] In one possible implementation, a machine learning model is introduced to analyze this data. This model first extracts features from the logs, such as feature vectors for stitch spacing and force variation, and then uses a regression algorithm to predict efficiency indicators. This analytical process is widely used in textile production lines with curved surfaces, identifying key factors affecting efficiency, such as the impact of material curvature on stitch stability, thus providing data support for optimization. Furthermore, the detailed process of the data analysis method needs to be clearly explained, where the statistical model refers to a probability distribution-based framework used to quantify efficiency.

[0136] For example, assuming the log data forms a time series, the analysis engine will calculate the average stitch speed as an efficiency benchmark. If the speed is below a threshold, it will be marked as an inefficient area.

[0137] Preferably, when evaluating overall efficiency, a multi-dimensional indicator fusion strategy is adopted, which weights and sums speed, defect rate, and energy consumption indicators to form a comprehensive efficiency score. This strategy includes data preprocessing steps, such as normalizing stitch parameters, followed by applying a weighted formula to calculate the score. In the field of curved surface processing, this helps to assess the differences in production efficiency between different surface types, such as spherical or wavy surfaces, ensuring the comprehensiveness of the analysis results. Through this method, the entire chain of business processes from data input to indicator output is fully represented, enabling real-time monitoring of the production line.

[0138] In one embodiment, the optimized feedback loop for the next adjustment of the surface structure is achieved through a feedback module. This module receives efficiency index analysis results and generates adjustment parameters.

[0139] For example, if the analysis shows that the stitch speed is too low, the feedback loop will suggest increasing the sewing machine's speed or adjusting the stitch angle. These adjustment parameters are input into the next production cycle, forming a closed loop.

[0140] Specifically, the feedback loop includes iterative algorithms that optimize parameters based on historical data, such as using gradient descent to gradually adjust variable values. In the processing of textile surface structures, this is suitable for continuous production scenarios, ensuring that each adjustment improves the accuracy and efficiency of surface processing.

[0141] Understandably, the core of optimizing the feedback loop lies in its closed-loop characteristic, meaning that the analysis results directly affect subsequent operations.

[0142] For example, when dealing with complex curved surfaces such as automotive interior materials, feedback may involve modifying the stitch path planning to reduce material waste.

[0143] In one embodiment, the loop is integrated into the control software, automatically applying adjustments to reduce manual intervention. This approach brings technical benefits in business operations, such as improved overall quality consistency of curved structures and prevention of recurring defects. Furthermore, the evaluation of efficiency indicators can be extended to more scenarios.

[0144] In one possible implementation, the analysis method includes clustering algorithms that group stitch data to identify efficiency patterns.

[0145] For example, clustering reveals high stitch density in certain curved surface regions, leading to inefficiency; the feedback loop then adjusts parameters for these regions. This extension demonstrates the versatility of the technical solution, applicable to a variety of processing tasks ranging from simple planes to complex curved surfaces within the same textile manufacturing field.

[0146] For example, the implementation of the feedback loop may also include a simulation verification step, in which the effect of the parameters is tested through a virtual model before the application is adjusted.

[0147] Specifically, the virtual model reconstructs the surface structure based on log data and simulates stitch output to predict optimization results. This step ensures the reliability of feedback and reduces trial-and-error costs in actual production. In another implementation, the entire process, from stitch data acquisition to feedback adjustments, forms an integrated system. This system is deployed in the production workshop, running all modules in real time to ensure seamless integration.

[0148] For example, log data flows directly into the analytics engine, generating metrics and immediately triggering feedback, thus enabling dynamic optimization. In businesses processing curved structures, this system is suitable for batch production scenarios, improving overall throughput.

[0149] It should be noted that this technical solution, through data-driven feedback, can improve the production process without changing the hardware. For example, when processing flexible curved materials, the optimized stitches are more uniform and reduce structural weaknesses.

[0150] This invention provides an intelligent embroidery path optimization and execution system for mesh cap embroidery, mainly comprising: a point cloud processing module, used to acquire point cloud data of the three-dimensional curved surface structure of the mesh cap through a scanning device, and to process the point cloud data using a three-dimensional reconstruction algorithm to obtain a curved surface geometric model, wherein the model contains curvature distribution information to determine the three-dimensional contour features of the mesh cap; a region segmentation module, used to extract local curvature values ​​based on the curved surface geometric model, and to segment the model into multiple regions using a meshing method, obtaining the curvature vector of each region to determine support requirements, and marking high curvature regions if the curvature vector exceeds a preset threshold for subsequent dynamic adjustment; a finite element analysis module, used to obtain the coordinates of support points from high curvature regions, and to simulate material deformation behavior using finite element analysis to obtain a deformation prediction matrix to determine the tension distribution of the flexible structure, thereby generating a support path plan for the deployment of the fixing device; and a stitch adaptation module, used to output the tension distribution for the support path plan, and to adjust the embroidery stitch sequence using a path optimization algorithm to obtain an adapted stitch density map to determine the curved surface. The system includes several modules: Adaptability: If the deviation value in the density map is below a threshold, the stitch path is confirmed to match the surface; Real-time compensation module: This module acquires the adapted stitch density map, generates real-time compensation parameters based on the surface geometry model, and uses a feedback control mechanism to monitor material displacement during embroidery, obtaining a displacement compensation vector for immediate correction of the stitch direction; Visual monitoring module: This module extracts the corrected stitch direction from the displacement compensation vector, uses a machine vision system to capture the contact angle between the embroidery needle and the material, obtains angle deviation data to judge contact stability, and activates compensation parameters to update the stitch path if the angle deviation exceeds a threshold; Synchronous drive module: This module generates the final embroidery execution sequence based on the contact stability after processing the angle deviation data, uses a synchronous drive algorithm to coordinate the movement of the fixing device and the embroidery machine, and obtains continuous stitch output to ensure the pattern accuracy meets requirements; Data analysis module: This module acquires production log data from the continuous stitch output, uses data analysis methods to evaluate overall efficiency indicators, and obtains an optimized feedback loop for the next processing and adjustment of the surface structure. Based on the above embodiments of the present invention, and through the above description, those skilled in the art can make various changes and modifications without departing from the technical concept of this invention. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for intelligent embroidery path optimization and execution in mesh cap embroidery, characterized in that, The method includes the following steps: Step S101: Obtain point cloud data of the three-dimensional curved surface structure of the mesh cap through a scanning device, and process the point cloud data using a three-dimensional reconstruction algorithm to obtain a curved surface geometric model, wherein the model contains curvature distribution information to determine the three-dimensional contour features of the mesh cap; Step S102: Extract local curvature values ​​based on the curved surface geometric model, and divide the model into multiple regions using a meshing method to obtain the curvature vector of each region to determine the support requirements. If the curvature vector exceeds a preset threshold, it is marked as a high curvature region for subsequent dynamic adjustment; Step S103: Obtain the coordinates of the support points from the high curvature region, simulate the material deformation behavior using finite element analysis, obtain the deformation prediction matrix to determine the tension distribution of the flexible structure, and thus generate a support path plan for the deployment of the fixing device; Step S104: For the tension distribution output by the support path plan, use a path optimization algorithm to adjust the embroidery stitch sequence to obtain an adapted stitch density map to determine the surface adaptability. If the deviation value in the density map is lower than the threshold, it is confirmed that the stitch path matches the surface; Step S104 further includes: An initial embroidery stitch sequence is obtained from the tension distribution output, the initial embroidery stitch sequence containing multiple stitch points and their tension values; The initial embroidery stitch sequence is optimized using a genetic algorithm to obtain an optimized stitch sequence. The optimized stitch sequence generates a preliminary fit density map by calculating the tension difference between sequences. Based on the preliminary adaptation density map, the distribution of deviation values ​​required for surface adaptation is obtained. The distribution of deviation values ​​is calculated by comparing the differences between the stitch points and the surface coordinates in the preliminary adaptation density map. For the portion of the deviation value distribution that is below a preset threshold, determine the direction of the stitch path adjustment; Adjust the direction according to the stitch path, obtain the tension equilibrium point from the optimized stitch sequence, and calculate the tension equilibrium point through the tension value of the equilibrium point; An extended surface tension model is generated using a path optimization algorithm. The extended surface tension model takes the tension equilibrium point as input and outputs the tension extension distribution to obtain the enhanced adaptive stitch density. If the overall distribution of deviation values ​​in the enhanced adaptation stitch density is lower than a preset threshold, then the surface matching is confirmed to be complete. For the distribution of remaining deviation values, the optimized stitch sequence is adjusted to obtain the final stitch path; in step S105, the adapted stitch density map is obtained, and real-time compensation parameters are generated by combining the curved surface geometric model. A feedback control mechanism is used to monitor material displacement during the embroidery process to obtain a displacement compensation vector for real-time correction of the stitch direction. Step S106: Extract the corrected stitch direction from the displacement compensation vector, capture the contact angle between the embroidery needle and the material using a machine vision system, obtain angle deviation data to judge contact stability, and activate compensation parameters to update the stitch path if the angle deviation exceeds the threshold; Step S107: Generate the final embroidery execution sequence based on the contact stability after processing the angle deviation data, use a synchronous drive algorithm to coordinate the movement of the fixing device and the embroidery machine, and obtain continuous stitch output to confirm that the pattern accuracy meets the requirements; Step S108: Obtain production log data from the continuous stitch output, use data analysis methods to evaluate the overall efficiency index, and obtain an optimized feedback loop for the next processing and adjustment of the curved structure.

2. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S101 includes: The three-dimensional point cloud data of the net cap is collected by scanning equipment to obtain a multi-view point cloud set; The multi-view point cloud set is aligned using a point cloud registration method to obtain an initial point cloud set; For the initial point cloud set, a surface reconstruction algorithm is used to generate a curved surface geometric model; Curvature distribution information is extracted from the surface geometry model to obtain the local curvature changes of the net cap; Based on the local curvature changes, the contour edge detection method is fused to obtain the three-dimensional contour features of the net cap; Determine whether the three-dimensional contour features meet a preset threshold to obtain a three-dimensional curved surface structure description of the mesh cap.

3. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S102 includes: Local curvature values ​​are obtained from the surface geometry model. The model is divided into multiple regions using a meshing method to obtain the curvature vector of each region. The curvature vector is used to calculate the Gaussian curvature by multiplying the principal curvatures within the region. The Gaussian curvature is defined as the product of the two principal curvatures at a point on the surface. Based on the curvature vector, it is determined whether it exceeds a preset threshold. If it exceeds the preset threshold, it is marked as a high curvature region. For the high curvature region, a surface texture mapping is integrated. The surface texture mapping is generated by superimposing texture coordinates on the region curvature distribution to obtain texture adjustment parameters. Based on the texture adjustment parameters, the support configuration is dynamically adjusted to obtain the adjusted support configuration.

4. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S103 includes: The flexible structure is meshed using finite element analysis to obtain material deformation behavior data and thus a deformation prediction matrix. Based on the deformation prediction matrix, obtain the tension distribution data of the flexible structure; Based on the tension distribution data, a support path plan is generated to determine the deployment location of the fixing device; Based on the support path planning, sensor feedback data is obtained to acquire real-time tension distribution information; If the real-time tension distribution information exceeds a preset threshold, the coordinates of the support points are adjusted to obtain an optimized support path plan. By using the optimized support path planning, real-time deformation data is obtained, the final deployment scheme of the fixed device is determined, and the support path planning of the flexible structure is generated.

5. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S105 includes: Obtain a curved surface geometric model from the surface of the embroidery material, and extract the initial curvature data from the curved surface geometric model; Based on a preset density threshold, the initial curvature data is compared in terms of density distribution to obtain a fitted stitch density map. A vector calculation method based on density distribution is used to generate compensation parameters from the adapted stitch density map; If the compensation parameter exceeds the preset threshold, the material displacement data is captured in real time by the sensor, and the feedback control mechanism is adjusted. The material displacement data obtained by the feedback control mechanism is vector-converted to obtain the displacement compensation vector; Based on the displacement compensation vector, monitor the surface texture deformation of the embroidery material and determine the degree of texture deformation; Adjust the embroidery path according to the degree of texture deformation to obtain a corrected stitch path.

6. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S106 includes: The contact angle between the embroidery needle and the material is captured by a machine vision system, and the difference between the contact angle and the preset standard angle is obtained to obtain angle deviation data. The contact stability is determined by comparing the angle deviation data with a preset threshold. If the angle deviation data exceeds the preset threshold, the compensation parameter is activated, the stitch path is updated, and the adjusted stitch path is obtained. By comparing the differences in contact point coordinates between the adjusted stitch path and the original stitch path, the material contact analysis results are obtained. By comparing the material contact analysis results with the boundary overlap of the ideal pattern, the pattern boundary offset value is obtained, and the boundary stability is determined. For the pattern boundary offset value, an offset compensation matrix is ​​generated, and the boundary offset value is vector-weighted and corrected by the offset compensation matrix to obtain the pattern integrity data; Based on the pattern integrity data, the embroidery execution sequence is updated to obtain the optimized embroidery execution sequence.

7. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S107 includes: Obtain angle deviation data, which is obtained by measuring the relative angle between the fixing device and the needle position of the embroidery machine; Based on the difference between the angle deviation data and the preset standard angle, the contact stability index is calculated, and it is determined whether the contact stability index reaches the preset threshold to obtain a stable contact sequence. Based on the stable contact sequence, the synchronous drive algorithm matches the time points of the fixed device position and the embroidery machine needle position. Generate the initial embroidery execution sequence; In response to changes in fabric tension, the initial embroidery execution sequence is adjusted by monitoring the deviation between the tension sensor data and the initial embroidery execution sequence path to obtain a continuous stitch output path; If the continuous stitch output path meets the pattern accuracy requirements, then the trajectory calibration verification is performed by comparing the matching degree between the continuous stitch output path points and the preset pattern outline to obtain the final pattern accuracy index.

8. The intelligent embroidery path optimization and execution method for mesh cap embroidery according to claim 1, characterized in that, Step S108 includes: Production log data is obtained from continuous stitch output, and data analysis methods are used to evaluate overall efficiency indicators and obtain preliminary optimization feedback. Based on the preliminary optimization feedback, the surface structure processing adjustment parameters were determined by fusion of log data, and the continuous stitch monitoring results were obtained. Based on the continuous stitch monitoring results, it is determined whether the adjustment parameters exceed the preset threshold, and feedback loop iteration requirements are obtained; Based on the feedback loop iteration requirements, adjust the surface structure processing parameters to obtain the adjusted continuous stitch monitoring results; Using the adjusted continuous stitch monitoring results, the overall efficiency index is re-evaluated to obtain optimized efficiency feedback; Based on the optimized efficiency feedback, the surface structure processing parameters are updated to determine the final production adjustment plan.

9. An intelligent embroidery path optimization and execution system for mesh cap embroidery, characterized in that, This system is used to implement the intelligent embroidery path optimization and execution method for net cap embroidery as described in any one of claims 1-8. The system includes: a point cloud processing module, used to acquire point cloud data of the three-dimensional curved surface structure of the net cap through a scanning device, and process the point cloud data using a three-dimensional reconstruction algorithm to obtain a curved surface geometric model, wherein the model contains curvature distribution information to determine the three-dimensional contour features of the net cap; a region segmentation module, used to extract local curvature values ​​based on the curved surface geometric model, and divide the model into multiple regions using a mesh partitioning method, obtaining the curvature vector of each region to determine support requirements; if the curvature vector exceeds a preset threshold, it is marked as a high curvature region for subsequent dynamic adjustment; a finite element analysis module, used to obtain the coordinates of support points from the high curvature region, use finite element analysis to simulate material deformation behavior, obtain a deformation prediction matrix to determine the tension distribution of the flexible structure, thereby generating a support path plan for the deployment of the fixing device; and a stitch adaptation module, used to output the tension distribution for the support path plan, and use a path optimization algorithm to adjust the embroidery stitch sequence to obtain an adapted stitch. The system employs several mechanisms: a stitch density map to assess surface adaptability, confirming a match between the stitch path and the surface if the deviation value in the density map is below a threshold; a real-time compensation module to obtain the adapted stitch density map, generate real-time compensation parameters based on the surface geometry model, and use a feedback control mechanism to monitor material displacement during embroidery, obtaining a displacement compensation vector for immediate correction of the stitch direction; a visual monitoring module to extract the corrected stitch direction from the displacement compensation vector, capturing the contact angle between the embroidery needle and the material using a machine vision system, obtaining angle deviation data to assess contact stability, and activating compensation parameters to update the stitch path if the angle deviation exceeds a threshold; a synchronous drive module to generate the final embroidery execution sequence based on the contact stability after processing the angle deviation data, using a synchronous drive algorithm to coordinate the movement of the fixing device and the embroidery machine, obtaining continuous stitch output to ensure the pattern accuracy meets requirements; and a data analysis module to obtain production log data from the continuous stitch output, using data analysis methods to evaluate overall efficiency indicators, obtaining an optimized feedback loop for the next surface structure processing and adjustment.

Images