Neural-guided grating-based manufacturing twin accessibility analysis method and system

Abstract

The application provides a manufacturing twin body accessibility analysis method and system based on neural guided rasterization, relates to the field of subtractive manufacturing technology, and comprises the following steps: obtaining a three-dimensional model of a target manufacturing twin body and tool parameters; based on the tool parameters, discretizing the axial direction that can be used in tool machining, constructing a tool direction set; predicting potential accessibility guide information of all tool directions through a neural network to obtain an accessibility guide matrix; based on the accessibility guide matrix, sequentially verifying the accessibility of each tool direction in the tool direction set; and after the verification of all tool directions is completed, summarizing the accessibility states of all grid vertices; the application does not rely on the direct output of the final accessibility result by the neural network, and the potential accessibility of the model region in the tool direction is predicted and guided by the neural network, and the final determination is completed through the accessibility verification based on rasterization, so that the analysis efficiency and the industrial processing safety are taken into account.

Description

Technical Field

[0001] This invention relates to the field of subtractive manufacturing technology, specifically to a method and system for analyzing the accessibility of manufacturing twins based on neural-guided rasterization. Background Technology

[0002] Manufacturing twins, as one of the core technologies of intelligent manufacturing, are dynamic systems that achieve visualization, analysis, and predictability of the machining process by constructing a digital twin mapping between physical and virtual spaces and integrating manufacturing environment, manufacturing resources, and machining process data. In the construction and application of manufacturing twins, the transformation from design twins to manufacturing twins is a crucial step. Tool accessibility analysis is the fundamental prerequisite for determining the manufacturing feasibility of complex part design features during this transformation process. Its analysis results directly determine whether geometric features in the design twin can be transformed into effective machining features in the manufacturing twin, and are also an important basis for subsequent manufacturing twin process planning, toolpath generation, virtual reproduction of the machining process, and error prediction.

[0003] Tool accessibility analysis in manufacturing twins refers to determining, in the virtual space of the manufacturing twin, whether a specified position on the surface of the 3D model of the design twin can be approached and processed by the tool along an allowed direction without geometric collision, by combining manufacturing resources and environmental constraints such as tool geometry, tool axis posture, and machine tool motion constraints. Its core is to achieve precise coupling between design intent and manufacturing feasibility in the manufacturing twin.

[0004] Existing tool accessibility analysis methods primarily rely on precise geometric calculations. These methods typically require spatial discretization of the model and performing collision detection or visibility assessment for each discrete sampling point across multiple candidate tool orientations. Specifically, they necessitate repeatedly calculating the spatial relationship between the tool and the model, and determining whether intersections or occlusions exist. While these methods can theoretically provide rigorous geometric results, their computational process involves complex spatial queries, numerous repetitive geometric operations, and high-dimensional parameter combinations. When the model's sampling points are dense, the overall computation time increases significantly, making it difficult to meet the efficiency requirements of interactive design and rapid process evaluation.

[0005] To reduce computational costs, recent research has attempted to incorporate deep learning methods, using neural networks to directly predict tool reachability distributions, thereby avoiding extensive geometric calculations during runtime. While these methods improve analysis speed to some extent, because neural networks are primarily data-driven, their predictions struggle to rigorously model complex geometric occlusion relationships. This is especially true when small or multiple occlusion structures exist, easily leading to misclassification of actually inaccessible areas as reachable. Such misclassifications can cause tool collisions with the model during actual machining, resulting in equipment damage or machining accidents. Therefore, their direct application in industrial settings is difficult.

[0006] Therefore, under complex 3D models and multi-tool orientation conditions, existing tool reachability analysis methods struggle to simultaneously address both computational efficiency and decision reliability. Summary of the Invention

[0007] To address the aforementioned issues, this invention proposes a method and system for accessibility analysis of manufacturing twins based on neural-guided rasterization. Instead of relying on neural networks to directly output the final accessibility result, this method utilizes neural networks to predict and guide the potential accessibility of the model region in the tool direction, and then completes the final determination through GPU-based rasterization accessibility verification, thereby balancing analysis efficiency and industrial processing safety.

[0008] According to some embodiments, the present invention adopts the following technical solution:

[0009] A method for accessibility analysis of fabricated twins based on neural-guided rasterization includes:

[0010] Obtain the 3D model of the target manufacturing twin and the tool parameters;

[0011] Based on the tool parameters, the axial directions that the tool may use during the machining process are discretized to construct a set of tool directions;

[0012] The potential reachability guidance information for all tool directions is predicted by a neural network to obtain an reachability guidance matrix. The reachability guidance matrix records the reachability prediction values ​​of all mesh vertices of the 3D model under each tool direction.

[0013] Based on the reachability guidance matrix, the principle of prioritizing the verification of tool directions that are most likely to confirm a large number of reachable vertices is adopted. The reachability of each tool direction in the tool direction set is verified in turn. Based on the verification results, the reachability status of vertices in the set of mesh vertices that have not yet been confirmed to be reachable is updated.

[0014] After all tool orientations have been verified, the reachability status of all mesh vertices is summarized and mapped back to the original 3D model surface.

[0015] The reachability verification adopts a geometric reachability verification method based on rasterization rendering. It strictly judges whether a collision occurs between the tool and the model through depth map and pooling operation, thereby verifying whether the mesh vertices are reachable.

[0016] According to some embodiments, the present invention adopts the following technical solution:

[0017] A neural-guided rasterization-based manufacturing twin accessibility analysis system includes:

[0018] The acquisition module is configured to acquire the 3D model of the target manufacturing twin and tool parameters;

[0019] The construction module is configured to: discretize the axial directions that the tool may use during the machining process based on the tool parameters, and construct a set of tool directions;

[0020] The prediction module is configured to: predict potential reachability guidance information for all tool directions through a neural network to obtain an reachability guidance matrix, wherein the reachability guidance matrix records the reachability prediction values ​​of all mesh vertices of the 3D model under each tool direction;

[0021] The verification module is configured to: based on the reachability guidance matrix, adopt the principle of prioritizing the verification of tool directions that are most likely to confirm a large number of reachable vertices, sequentially verify the reachability of each tool direction in the tool direction set, and update the reachability status of vertices in the set of mesh vertices that have not yet been confirmed as reachable based on the verification results;

[0022] The summary module is configured to: after all tool orientation verifications are completed, summarize the reachability status of all mesh vertices and map it back to the original 3D model surface;

[0023] The reachability verification adopts a geometric reachability verification method based on rasterization rendering. It strictly judges whether a collision occurs between the tool and the model through depth map and pooling operation, thereby verifying whether the mesh vertices are reachable.

[0024] According to some embodiments, the present invention adopts the following technical solution:

[0025] A computer program product includes a computer program that, when executed by a processor, implements the neural-guided rasterization-based manufacturing twin accessibility analysis method.

[0026] According to some embodiments, the present invention adopts the following technical solution:

[0027] A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned neural-guided rasterization-based manufacturing twin reachability analysis method.

[0028] According to some embodiments, the present invention adopts the following technical solution:

[0029] An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the aforementioned method for analyzing the accessibility of a manufacturing twin based on neural-guided rasterization.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] (1) The manufacturing twin accessibility analysis method proposed in this invention, which combines neural guidance with rasterized geometric verification, significantly improves the overall efficiency of accessibility analysis for complex 3D models while ensuring the strict reliability of the judgment results. By introducing the prediction results of potential accessibility under different tool directions using neural networks, these predictions are used only as guiding information for the geometric verification order and do not directly participate in the final judgment. This allows a large number of truly accessible vertices to be preferentially confirmed under the tool directions that are ranked higher. Once a vertex is confirmed to be accessible in any direction, an early stopping mechanism is triggered and it is removed from the subsequent verification process, thereby effectively avoiding the repeated high-cost geometric calculations for the same vertex under multi-directional and multi-scale conditions. Therefore, even if the number of sampling points on the model surface increases significantly, the computational overhead of this invention will not increase proportionally with the increase in sampling density, and it can maintain high operating efficiency in complex models and high-precision analysis scenarios.

[0032] (2) This invention maintains strict conservatism throughout the entire analysis process, ensuring high accuracy while avoiding misclassifying unreachable areas as reachable. Specifically, the neural network prediction results are used only as prior guidance information, and all final reachability determinations are based on a geometric verification process using rasterized rendering. This geometric verification process conservatively detects potential tool-model collisions based on depth map minimum pooling and tool geometry parameters, and compensates for rasterization discretization errors through boundary expansion and other measures, thereby ensuring that no collision scenarios are missed, exhibiting high engineering safety and industrial applicability.

[0033] (3) Since the accessibility determination core of this invention is based on a general geometric rendering and parametric tool model, it does not depend on a specific model structure or fixed scale features, and has good adaptability to 3D models of different complexity and tools of different sizes, shapes and parameter configurations. Even when the model scale changes or the tool parameters are adjusted, this invention can still maintain stable and reliable analysis performance, has good versatility and engineering robustness, and is suitable for promotion and use in practical application scenarios such as CNC subtractive manufacturing. Attached Figure Description

[0034] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0035] Figure 1 This is a flowchart of the method in Example 1.

[0036] Figure 2 Example diagram of a three-dimensional model of a twin created for the target of Example 1.

[0037] Figure 3 This is an example diagram of tool parameters for Example 1.

[0038] Figure 4 This is an example diagram of the tool direction in Example 1.

[0039] Figure 5 This is a diagram of the neural network architecture based on O-CNN in Example 1.

[0040] Figure 6 This is an example diagram of the tool orientation to be verified in Example 1.

[0041] Figure 7 This is the depth map for Example 1.

[0042] Figure 8 This is a diagram showing the center depth of the tool in Example 1.

[0043] Figure 9 The image shows the pooling depth results for Example 1, where (a) and (b) represent the small radius window and the large radius window, respectively.

[0044] Figure 10 This is a diagram showing the reachability determination results for a certain tool direction in Example 1.

[0045] Figure 11 This is the final three-dimensional result of the accessibility analysis in Example 1. Detailed Implementation

[0046] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0047] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0048] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0049] Example 1

[0050] One embodiment of the present invention provides a method for accessibility analysis of manufacturing twins based on neural-guided rasterization, primarily targeting subtractive manufacturing processes. It analyzes user-input 3D models and, while ensuring strict conservatism in the judgment results, efficiently determines the accessibility distribution of various positions on the model surface under multi-tool orientation conditions. Figure 1 As shown, the specific steps are as follows:

[0051] Step S1: Obtain the 3D model of the target manufacturing twin and the tool parameters.

[0052] The three-dimensional model uses a discretized data structure to approximate and store the shape of continuous three-dimensional objects, such as triangular meshes, point clouds, or SDF (symbolic distance function) representations.

[0053] In this embodiment, as follows Figure 2 Taking the shown 3D model as an example, a triangular mesh is used to represent the 3D model of the target manufacturing twin. Its data format can be STL or OBJ. A triangular mesh is a 3D model data format where vertices determine positions and triangles cover the surface. Each mesh vertex has two values: coordinates and a normal vector. A ball end mill is used, such as... Figure 3 As shown, the tool parameters include the tool ball end radius, tool top radius, total tool length, and tool holder length. The input triangular mesh model must satisfy two key conditions: watertightness and manifold properties.

[0054] 1) The model surface should be free of holes and self-intersections to ensure the integrity of the geometric structure and meet the requirements of subtractive manufacturing process;

[0055] 2) The model must have manifold topological properties, that is, the adjacency relationship of all triangular facets must conform to the standard topological connection rules, excluding abnormal geometric shapes such as boundary points and non-closed regions.

[0056] The above constraints ensure effective compatibility between the model and the manufacturing process, and provide a reliable geometric basis for subsequent accessibility analysis.

[0057] Step S2: Based on the tool parameters, discretize the axial directions that the tool may use during the machining process to construct a set of tool directions.

[0058] Furthermore, the set of tool directions includes:

[0059] Since the tool is machining from above the workpiece, the tool direction is uniformly sampled on the unit upper hemisphere with radius 1, center at the origin and z≥0 to construct a finite set of candidate tool directions. The axial directions that the tool may take during machining are discretized to simulate the entry direction that the tool may take during multi-axis CNC machining.

[0060] Specifically, to facilitate later projection, the tool direction is imagined as a virtual camera, with the camera's shooting angle consistent with the tool direction, such as... Figure 4 As shown by the blue virtual camera in the model, each direction represents a fixed posture of the tool axis in the model coordinate system, which is used to simulate the possible entry direction of the tool during multi-axis CNC machining. By constructing a discrete set of tool directions, the continuous direction space can be transformed into a finite verification problem, providing a clear search space for subsequent reachability analysis, while avoiding the uncontrollable computational problems caused by the continuity of directions.

[0061] Step S3: Predict the potential reachability guidance information for all tool directions using a neural network to obtain the reachability guidance matrix. The reachability guidance matrix records the reachability prediction values ​​of all mesh vertices of the 3D model under each tool direction.

[0062] Furthermore, the prediction of potential reachability guidance information for all tool directions via neural networks includes:

[0063] The 3D model is represented as an octree structure. The triangular mesh model of the octree structure, along with the tool parameters, is input into the trained neural network for inference. The binary potential reachability prediction value of each mesh vertex under each tool direction is output, and an reachability guidance matrix is ​​constructed. ,in, The number of grid vertices, The number of tool directions in the tool direction set, matrix elements Represents vertices In the direction of the tool It has potential accessibility. This indicates that the prediction is potentially unreachable.

[0064] Specifically, the neural network employs, as follows: Figure 5 The Unet architecture based on O-CNN shown is used to predict the potential reachability distribution of each vertex in the model under different tool orientations. Binary (0 / 1) reachability prediction is performed on all mesh vertices under each candidate tool orientation. Figure 5 The red areas on the 3D model represent the directions with the highest probability, thus constructing a vertex-direction accessibility guidance matrix. This matrix serves only as guidance information in the subsequent geometric accessibility verification process and is not directly used as the final basis for tool accessibility determination. Its role is to provide prior reference for the subsequent verification stage, thereby reducing unnecessary geometric calculations.

[0065] This invention randomly selected 11,000 triangular mesh models from the public dataset ABC (A Big CAD Model Dataset For Geometric Deep Learning). Based on traditional geometric methods, it calculated the reachability of sampling points in each unit hemisphere sampling direction and constructed the dataset. After scale normalization, 8,800 models were used to train the neural network, 2,000 models constituted the test set, and the remaining 200 models were used for quantitative evaluation of computational efficiency and accuracy beyond the map library results. The training process employed a cross-entropy loss function.

[0066] Step S4: Based on the reachability guidance matrix, the principle of prioritizing the verification of tool directions that are most likely to confirm a large number of reachable vertices is adopted. The reachability of each tool direction in the tool direction set is verified in turn. Based on the verification results, the reachability status of vertices in the set of mesh vertices that have not yet been confirmed to be reachable is updated.

[0067] The reachability verification adopts a geometric reachability verification method based on rasterization rendering. It strictly judges whether a collision occurs between the tool and the model through depth map and pooling operation, thereby verifying whether the mesh vertices are reachable.

[0068] Specifically, a multi-round reachability verification is used to confirm the feasibility of all mesh vertices in the mesh vertex set. Once a vertex is confirmed to be reachable in any tool direction, it is removed from the set of mesh vertices that have not yet been confirmed to be reachable.

[0069] In each round of reachability verification, the tool orientation set is sorted based on the neural network prediction results to optimize the calculation order of subsequent geometric reachability verifications. Specifically, a coverage-based greedy sorting strategy is adopted. In each round of reachability verification, the tool orientation that covers the largest number of currently unconfirmed reachable vertices is selected as the tool orientation to be verified in the current round. The tool direction selected in the wheel satisfy:

[0070]

[0071] in, This represents the set of mesh vertices that have not yet been confirmed as reachable. This represents the set of verified tool directions. The sorting method described above allows for priority verification of tool directions more likely to confirm a large number of reachable vertices in the subsequent rasterization-based geometric verification stage. This increases the probability of triggering the early stopping mechanism and effectively reduces overall computational overhead while ensuring the conservatism of the final decision.

[0072] Regarding the first Tool direction selected by the wheel ,like Figure 6 As shown, taking the downward tool direction as an example, geometric reachability verification based on rasterization rendering is performed on currently unconfirmed reachable mesh vertices. This verification method utilizes the axisymmetric properties of the tool geometry to transform the tool-model collision detection problem in 3D space into a 2D parallel computation process based on depth maps. This significantly reduces computational costs while ensuring strict conservatism in the judgment results. The reachability verification of the wheel is specifically as follows:

[0073] Step S4-1: Set the current tool orientation to be verified. Orientation normalization is performed, and the normalized tool orientation is set as the camera viewing direction. The rotation matrix from the world coordinate system to the camera coordinate system is calculated through spatial transformation. This rotational transformation uniformly transforms the model's vertices and their normal vectors to the camera coordinate system.

[0074]

[0075] in, Represents the set of vertices of a 3D model. This represents the vertex coordinates of the model in the world coordinate system. Represents vertices Transform the vertex coordinates to the camera coordinate system. This represents the vertex normal vector of the model in the world coordinate system. Normal vector Transform the vertex normals to the camera coordinate system. Based on the bounding box of the rotated model, adaptively set the orthogonal projection camera parameters so that the entire model falls completely within the viewport of a fixed resolution.

[0076] Step S4-2: In the current tool direction The following method uses GPU-based rasterization rendering to perform orthogonal projection on the rotated 3D model, generating the corresponding depth map. .

[0077] A depth map is a two-dimensional image used to record three-dimensional spatial information. It looks like a photograph, but the colors in the photograph do not record the brightness or texture of objects, but rather how far away objects in the scene are from the observer.

[0078] In this embodiment, a GPU-based rasterization rendering method is used. The virtual camera is aligned with the rotated 3D model, and an orthographic projection matrix is ​​used to project the 3D model onto a 2D plane in parallel. This preserves the precise dimensions of the object and eliminates the perspective effect of "objects appearing larger when closer and smaller when farther away," resulting in the following: Figure 7 In the depth map visualization shown, lighter colors typically represent farther distances, while darker colors represent closer distances.

[0079] Each pixel in the depth map records the depth value of the most visible surface at the foremost point in the current tool direction. Pixels not covered by any geometry are assigned an infinite depth. To ensure the conservatism of the analysis results, a local minimum expansion operation is performed on the model boundary region after the depth map is generated to compensate for the discrete errors that may be caused by the limited resolution.

[0080] Step S4-3: After obtaining the depth map, first filter out the non-negative triangle IDs from the pixel-triangle mapping table of the rasterized output to obtain the set of visible triangles in the current view. Then, flatten the vertex IDs of all visible triangles and take their unique values ​​to obtain the initial set of visible vertices. And project it onto the screen space.

[0081] For each visible vertex Offset the radius of the tool cutting area along the direction of its unit normal vector after rotation. Construct the corresponding tool center position:

[0082]

[0083] Then the tool center Projected onto screen space, its pixel coordinates and depth values ​​are obtained, used as follows: Figure 8 The tool center depth diagram shown is used to represent this. Figure 8 Each point in the process is a visible vertex that forms the center of the tool.

[0084] Step S4-4: Using the projected center position of the tool as the center, in the depth map... A circular detection window corresponding to the tool geometry is constructed on top.

[0085] Specifically, such as Figure 9As shown, based on the structural characteristics of the cutting tool, which consists of cylindrical segments of different radii, the following settings are respectively configured: Figure 9 The small radius window shown in (a) He Ru Figure 9 The large radius window shown in (b) And perform minimum pooling on the depth map within the corresponding window, expressed by the formula:

[0086]

[0087]

[0088] in, This indicates a circle with the center of the tool's projection as the center and a radius of... Circular pixel area Represents pixels, This represents the minimum depth value within a small-radius pooling window. This represents the minimum depth value within the large radius pooling window.

[0089] Although this embodiment uses a tool composed of two cylindrical segments as an example, the problem of tools composed of non-cylindrical segments can be solved by modifying the pooling window. The determination of tools with more segments can be solved by slightly modifying the determination conditions and adding a pooling window. The solution is based on the depth difference determination.

[0090] Step S4-5: By comparing the difference between the tool center depth and the minimum depth value within each circular window, determine whether the tool has a geometric collision with the model in the current direction, and obtain... Figure 10 The reachability determination results shown are as follows: Figure 10 Medium red indicates reachable, gray indicates unreachable.

[0091] Specifically, the depth difference is calculated:

[0092]

[0093]

[0094] in, This represents the depth value at the center of the tool. If the height exceeds the corresponding cylindrical segment height, the vertex is determined to be unreachable in the current tool direction; otherwise, the vertex is determined to be reachable in that direction.

[0095] Complete the current wheel tool direction After the rasterization-based reachability verification, the set of vertices that are still not confirmed to be reachable is updated, and early stopping is performed to avoid unnecessary subsequent geometric calculations.

[0096] Early stop determination refers to utilizing the monotonicity of tool reachability determination. Once a vertex is strictly confirmed to be reachable in any tool direction, there is no need to continue verifying the vertex in other directions. All vertices that have been confirmed to be reachable will be marked and removed from the geometric verification process of subsequent tool directions, while vertices that have not yet been confirmed to be reachable will continue to participate in the verification process of subsequent directions.

[0097] Step S5: After all tool orientations have been verified, summarize the reachability status of all mesh vertices and map it back to the original 3D model surface.

[0098] After all tool orientation verifications are completed, the vertex-level reachability determination results are summarized, and the final reachability labels are mapped back to the original 3D model surface. For example... Figure 11 As shown, based on the final determination state of the vertices, the model surface is divided into reachable and inaccessible regions, and further output in the form of color marking, region segmentation or data files for subsequent CNC machining process evaluation, tool path planning or manufacturing feasibility analysis.

[0099] Example 2

[0100] One embodiment of the present invention provides a manufacturing twin accessibility analysis system based on neural-guided rasterization, comprising:

[0101] The acquisition module is configured to acquire the 3D model of the target manufacturing twin and tool parameters;

[0102] The construction module is configured to: discretize the axial directions that the tool may use during the machining process based on the tool parameters, and construct a set of tool directions;

[0103] The prediction module is configured to: predict potential reachability guidance information for all tool directions through a neural network to obtain an reachability guidance matrix, wherein the reachability guidance matrix records the reachability prediction values ​​of all mesh vertices of the 3D model under each tool direction;

[0104] The verification module is configured to: based on the reachability guidance matrix, adopt the principle of prioritizing the verification of tool directions that are most likely to confirm a large number of reachable vertices, sequentially verify the reachability of each tool direction in the tool direction set, and update the reachability status of vertices in the set of mesh vertices that have not yet been confirmed as reachable based on the verification results;

[0105] The summary module is configured to: after all tool orientation verifications are completed, summarize the reachability status of all mesh vertices and map it back to the original 3D model surface;

[0106] The reachability verification adopts a geometric reachability verification method based on rasterization rendering. It strictly judges whether a collision occurs between the tool and the model through depth map and pooling operation, thereby verifying whether the mesh vertices are reachable.

[0107] Example 3

[0108] One embodiment of the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned neural-guided rasterization-based manufacturing twin accessibility analysis method.

[0109] Example 4

[0110] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned method for analyzing the accessibility of manufacturing twins based on neural-guided rasterization.

[0111] Example 5

[0112] One embodiment of the present invention provides an electronic device, including: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the aforementioned method for analyzing the accessibility of a manufacturing twin based on neural-guided rasterization.

[0113] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0114] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0115] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for accessibility analysis of fabricated twins based on neural-guided rasterization, characterized in that, include: Obtain the 3D model of the target manufacturing twin and the tool parameters; Based on the tool parameters, the axial directions that the tool may use during the machining process are discretized to construct a set of tool directions; The potential reachability guidance information for all tool directions is predicted by a neural network to obtain an reachability guidance matrix. The reachability guidance matrix records the reachability prediction values ​​of all mesh vertices of the 3D model under each tool direction. Based on the reachability guidance matrix, the principle of prioritizing the verification of tool directions that are most likely to confirm a large number of reachable vertices is adopted. The reachability of each tool direction in the tool direction set is verified in turn. Based on the verification results, the reachability status of vertices in the set of mesh vertices that have not yet been confirmed to be reachable is updated. After all tool orientations have been verified, the reachability status of all mesh vertices is summarized and mapped back to the original 3D model surface. The principle of prioritizing the verification of tool directions most likely to confirm a large number of reachable vertices is to adopt a greedy sorting strategy based on coverage. In each round of verification, the tool direction that can cover the largest number of currently unconfirmed reachable vertices is selected from the set of remaining unverified tool directions as the current verification object. The reachability verification adopts a geometric reachability verification method based on rasterization rendering. It strictly determines whether a collision occurs between the tool and the model through depth map and pooling operation, thereby verifying whether the mesh vertices are reachable. The geometric reachability verification method based on rasterization rendering specifically includes: The current tool direction is normalized, and the normalized tool direction is set as the camera viewing direction. The world coordinate system is aligned with the camera coordinate system through rotation transformation, and the model vertices and normal vectors are transformed to the camera coordinate system. Generate depth maps under orthographic projection using GPU rasterization rendering; For the visible vertices in the model, offset the radius of the tool cutting area along its normal vector direction to construct the tool center position and project it onto the screen space; On the depth map, a circular detection window corresponding to the tool geometry is constructed with the tool center projection position as the center. The minimum depth within the window is calculated by minimum pooling and compared with the tool center depth to determine whether the vertex is reachable in the current tool direction.

2. The method for accessibility analysis of fabricated twins based on neural-guided rasterization as described in claim 1, characterized in that, The set of tool directions specifically includes: The tool direction is uniformly sampled on the unit upper hemisphere to construct a finite set of candidate tool directions. The axial direction that the tool may take during the machining process is discretized to simulate the entry direction that the tool may take during multi-axis CNC machining.

3. The method for accessibility analysis of fabricated twins based on neural-guided rasterization as described in claim 1, characterized in that, The method of predicting potential reachability guidance information for all tool directions using a neural network specifically includes: The 3D model is encoded into a 3D representation that can be processed by a neural network, and input along with tool parameters into the trained neural network for inference. The binary potential reachability prediction value of each grid vertex under each tool direction is output, and an reachability guidance matrix is ​​constructed. ,in, For grid vertex data, The number of tool directions in the tool direction set, matrix elements Represents vertices In the direction of the tool It has potential accessibility. This indicates that the prediction is potentially unreachable.

4. The method for accessibility analysis of fabricated twins based on neural-guided rasterization as described in claim 1, characterized in that, The principle of prioritizing the verification of tool directions most likely to confirm a large number of reachable vertices is expressed by the formula: in, This represents the set of mesh vertices that have not yet been confirmed as reachable. For tool direction set, This represents the set of verified tool orientations. This represents the set of remaining unverified tool directions.

5. The method for accessibility analysis of fabricated twins based on neural-guided rasterization as described in claim 1, characterized in that, The construction of the circular detection window corresponding to the tool geometry specifically includes: Based on the structural characteristics of the cutting tool, which consists of cylindrical segments of different radii, small radius windows are set respectively. and large radius window These correspond to different geometric segments of the cutting tool; Calculate the difference between the tool center depth and the minimum depth value within each window. and ,like or If the height exceeds the corresponding cylindrical segment height, the vertex is determined to be unreachable in the current tool direction; otherwise, it is determined to be reachable.

6. The method for accessibility analysis of fabricated twins based on neural-guided rasterization as described in claim 1, characterized in that, The process of sequentially verifying the reachability of each tool direction in the tool direction set also includes an early stop determination step: For any mesh vertex that has been confirmed to be reachable, it is immediately marked and removed from the subsequent geometric verification process in the tool direction, and will no longer participate in the subsequent verification, so as to reduce unnecessary geometric calculations by taking advantage of the monotonicity of tool reachability determination.

7. A manufacturing twin accessibility analysis system based on neural-guided rasterization, characterized in that, The method for accessibility analysis of fabricated twins based on neural-guided rasterization as described in any one of claims 1-6 includes: The acquisition module is configured to acquire the 3D model of the target manufacturing twin and tool parameters; The construction module is configured to: discretize the axial directions that the tool may use during the machining process based on the tool parameters, and construct a set of tool directions; The prediction module is configured to: predict potential reachability guidance information for all tool directions through a neural network to obtain an reachability guidance matrix, wherein the reachability guidance matrix records the reachability prediction values ​​of all mesh vertices of the 3D model under each tool direction; The verification module is configured to: based on the reachability guidance matrix, adopt the principle of prioritizing the verification of tool directions that are most likely to confirm a large number of reachable vertices, sequentially verify the reachability of each tool direction in the tool direction set, and update the reachability status of vertices in the set of mesh vertices that have not yet been confirmed as reachable based on the verification results; The summary module is configured to: after all tool orientation verifications are completed, summarize the reachability status of all mesh vertices and map it back to the original 3D model surface; The reachability verification adopts a geometric reachability verification method based on rasterization rendering. It strictly judges whether a collision occurs between the tool and the model through depth map and pooling operation, thereby verifying whether the mesh vertices are reachable.

8. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the neural-guided rasterization-based manufacturing twin accessibility analysis method as described in any one of claims 1-6.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the method for analyzing the accessibility of a manufacturing twin based on neural-guided rasterization as described in any one of claims 1-6.

Images