Process track automatic generation method and device and electronic equipment

Abstract

The invention provides a process track automatic generation method and device and electronic equipment, and relates to the technical field of robots, and the method comprises the steps: obtaining point cloud data representing the three-dimensional shape of a to-be-processed workpiece; projecting the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set; in the two-dimensional plane, planning and generating a two-dimensional process path according to the contour represented by the two-dimensional projection point set; and mapping the two-dimensional process path to a three-dimensional space, determining a plurality of three-dimensional space points in the point cloud data, calculating a surface normal vector of each three-dimensional space point, and generating a three-dimensional process track containing three-dimensional coordinates and attitude information. Through the combination of the steps, on the premise that the three-dimensional precision of the final track can be guaranteed, the automation level and the calculation efficiency of track generation are remarkably improved, and efficient and high-precision process track generation is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robots, and in particular to a process trajectory automatic generation method and device and electronic equipment. BACKGROUND

[0002] In the fields of aviation, aerospace, automobile manufacturing, shipbuilding and other high-end equipment manufacturing, the automation process requirements such as polishing, polishing, deburring, spraying and non-destructive testing of complex curved workpieces are increasingly urgent. One of the core technologies to achieve automation of these processes is to plan an accurate process trajectory that fits the surface of the workpiece for the robot.

[0003] The existing trajectory planning algorithm system relies on a single data source and is only based on static CAD models or dynamic point cloud data. However, both static CAD modeling and dynamic point cloud data have actual technical problems: for traditional industrial scenarios, using CAD models to generate process trajectories completely relies on idealized models and cannot perceive and adapt to actual manufacturing errors, clamping deviations, and physical state changes such as deformation and wear caused by long-term use of workpieces; real-time planning algorithms based on dynamic point cloud data are highly dependent on the quality of point cloud data, and the process of generating process trajectories using this method is computationally intensive and time-consuming, making it difficult to meet the efficiency requirements of online real-time generation.

[0004] Therefore, the industry urgently needs a technical solution that can balance the convenience of offline programming and the adaptability of online perception, and can uniformly process three-dimensional data from different sources to achieve efficient and high-precision process trajectory generation. SUMMARY

[0005] The present application provides a process trajectory automatic generation method and device, and electronic equipment to solve the defects that the method based on CAD models in the prior art cannot perceive and adapt to the actual manufacturing errors of the workpiece, and the method based on dynamic point cloud data is computationally intensive, time-consuming and inefficient, and to achieve efficient and high-precision process trajectory generation.

[0006] The present application provides a process trajectory automatic generation method, comprising: obtaining point cloud data representing the three-dimensional topography of a workpiece to be processed; projecting the point cloud data from a three-dimensional space to a pre-set two-dimensional plane to generate a two-dimensional projection point set; in the two-dimensional plane, generating a two-dimensional process path according to the contour represented by the two-dimensional projection point set; mapping the two-dimensional process path to the three-dimensional space to determine a plurality of three-dimensional space points in the point cloud data, and calculating the surface normal vector of each three-dimensional space point to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

[0007] According to the method provided by the application, the point cloud data comprises: model point cloud data generated by a discretization algorithm according to a CAD digital model of the workpiece to be processed; or measured point cloud data directly obtained by collecting the surface of the workpiece to be processed by a three-dimensional scanning device.

[0008] According to the method provided by the application, the discretization algorithm specifically comprises: sampling on the surface of the CAD digital model, and ensuring that the distance between any two points in the generated model point cloud data is not less than a preset minimum distance threshold.

[0009] According to the method provided by the application, the point cloud data is projected from the three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set, specifically comprising: determining a plane that minimizes the projection distortion of the point cloud data by analyzing the distribution characteristics of the point cloud data in the three-dimensional space, as the preset two-dimensional plane; orthogonally projecting each three-dimensional space point in the point cloud data onto the two-dimensional plane to generate the two-dimensional projection point set.

[0010] According to the method provided by the application, in the two-dimensional plane, a two-dimensional process path is planned and generated according to the contour represented by the two-dimensional projection point set, comprising: generating a set of reciprocating paths or circumferential paths for traversing the internal region of the contour according to a preset path spacing inside the contour represented by the two-dimensional projection point set, as the two-dimensional process path.

[0011] According to the method provided by the application, the step of calculating the surface normal vector of each three-dimensional space point comprises: for any three-dimensional space point, finding out the adjacent three-dimensional space points in the point cloud data to which the three-dimensional space point belongs and whose distance from the three-dimensional space point is within a predetermined range; determining a local tangent plane by mathematically fitting the adjacent three-dimensional space points; and taking the normal direction of the local tangent plane as the surface normal vector of the three-dimensional space point.

[0012] According to the method provided by the application, the method further comprises: taking the three-dimensional process trajectory generated based on the model point cloud data as an offline trajectory template; obtaining measured point cloud data of the workpiece to be processed; determining a spatial pose transformation relationship that minimizes the distance between the points on the offline trajectory template and the corresponding points in the measured point cloud data by means of iterative optimization; updating the offline trajectory template according to the spatial pose transformation relationship to generate a final execution trajectory adapted to the workpiece to be processed.

[0013] According to the method provided by the application, the method further comprises: taking the three-dimensional process trajectory generated based on the measured point cloud data as a standard teaching trajectory, so as to be directly called when processing a subsequent workpiece to be processed.

[0014] The application further provides a process trajectory automatic generation device, comprising: a point cloud data acquisition module configured to acquire point cloud data representing a three-dimensional topography of a workpiece to be processed; a two-dimensional point set generation module configured to project the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set; a two-dimensional process generation module configured to plan a two-dimensional process path in the two-dimensional plane according to an outline represented by the two-dimensional projection point set; a three-dimensional process generation module configured to map the two-dimensional process path to the three-dimensional space, determine a plurality of three-dimensional space points in the point cloud data, and calculate a surface normal vector of each three-dimensional space point to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

[0015] The application further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the process trajectory automatic generation method according to any one of the above when executing the computer program.

[0016] The application further provides a non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program is executable by a processor to implement the process trajectory automatic generation method according to any one of the above.

[0017] The application further provides a computer program product comprising a computer program, wherein the computer program is executable by a processor to implement the process trajectory automatic generation method according to any one of the above.

[0018] The process trajectory automatic generation method and device provided by the application can greatly improve the calculation efficiency and robustness by acquiring point cloud data representing a three-dimensional topography of a workpiece to be processed, projecting the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set, and planning a two-dimensional process path in the two-dimensional plane. Finally, the two-dimensional process path is mapped to the three-dimensional space to determine a plurality of three-dimensional space points in the point cloud data and calculate a surface normal vector of each three-dimensional space point, so that the finally generated trajectory points can be fitted in the original three-dimensional space. Through the combination of the above steps, the application can significantly improve the automation level and calculation efficiency of trajectory generation while ensuring the three-dimensional precision of the final trajectory, and achieve efficient and high-precision process trajectory generation. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to make the technical solutions in the present application or the prior art clearer, the accompanying drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative work based on the accompanying drawings are within the protection scope of the present application.

[0020] Figure 1 is a flowchart of the process trajectory automatic generation method provided by the present application.

[0021] Figure 2 is a schematic diagram of converting a CAD digital model into model point cloud data provided by the present application.

[0022] Figure 3 is an effect schematic diagram of the alpha-shapes algorithm provided by the present application.

[0023] Figure 4 is a decision flowchart of the sampling points provided by the present application.

[0024] Figure 5 is a method schematic diagram of updating the trajectory template through visual guidance provided by the present application.

[0025] Figure 6 is a structural schematic diagram of the process trajectory automatic generation device provided by the present application.

[0026] Figure 7 is a structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0027] In order to make the technical solutions in the present application or the prior art clearer, the accompanying drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative work based on the accompanying drawings are within the protection scope of the present application.

[0028] For the two types of methods mentioned in the prior art: The first type is an offline programming method based on a CAD digital model. This method imports a three-dimensional CAD model of a workpiece in a virtual environment, plans a trajectory on the ideal model through CAM (computer aided manufacturing) software or a dedicated offline programming software, then generates a robot program and downloads it to a robot controller for execution. The advantage of this method is that programming can be completed without occupying production time, and the generated trajectory is theoretically accurate. However, its disadvantage is that it is seriously divorced from the actual working condition. Due to manufacturing tolerances, workpiece clamping errors, thermal deformation and other factors, there is always a deviation between the actual workpiece surface profile and the theoretical CAD model. This results in the ideal trajectory generated offline not being able to truly fit the actual workpiece, which may cause uneven machining quality (such as missed grinding or excessive grinding), or even cause the robot end tool to collide with the workpiece, damaging the equipment.

[0029] The second type is an online generation method based on three-dimensional sensors such as vision sensors. This method uses a line laser profiler, a structured light camera or other three-dimensional sensors to scan the surface of the actual workpiece in real time, obtains three-dimensional point cloud data, and then plans and generates a trajectory directly on the point cloud data. This method can perceive the real appearance of the workpiece, thereby generating a trajectory that can adapt to the actual deviation and ensure the machining quality. However, its disadvantages are also obvious: first, it is highly dependent on physical samples and cannot be pre-programmed in the product design stage or when there is no physical object; second, the process of on-site scanning, data processing and trajectory calculation is time-consuming, which reduces the production rhythm and efficiency. In addition, the point cloud data collected by the sensor may have noise, holes and data missing, which poses a great challenge to stable and reliable trajectory generation.

[0030] In addition, the trajectories generated by existing methods are mostly simple straight line or circular arc interpolation, lacking advanced waveform interpolation functions that are coupled with specific processes and affect the machining quality, and the process adaptability is poor, requiring experienced engineers to make a lot of tedious manual adjustments.

[0031] In order to solve the problems existing in the prior art, the embodiment of the present application provides a process trajectory automatic generation method, the core of which is to complete the core path planning step in an efficient and stable two-dimensional plane, and then map it back to the original three-dimensional space, thereby combining the efficiency of path planning in a two-dimensional plane and the three-dimensional process trajectory accuracy after mapping the two-dimensional plane to a three-dimensional space. This method can not only process theoretical CAD models, but also directly process measured point cloud data, and has high flexibility and industrial applicability.

[0032] In the specific implementation process, the method provided by the embodiment of the present application can be deployed in a robot automation system, which includes: Electronic device: used for data processing and algorithm operation, such as industrial computer, server or embedded computer, which includes processor and memory, and the memory stores computer program for executing the method of the present application; Robot system: including robot body, robot controller and end effector (such as polishing head, spray gun or detection probe); And three-dimensional scanning device configured as needed, such as line laser scanner or structured light camera; Auxiliary connection components: including data transmission module (ensuring high-speed data transmission between devices), tool fixture (used for fixing workpieces to be processed to ensure processing stability) and the like. The above devices are connected through wired or wireless communication mode.

[0033] Before introducing the technical scheme of the embodiment of the present application, the terms of the embodiment of the present application are illustratively explained.

[0034] Point cloud data: a set composed of a large number of three-dimensional coordinate points, used for fully characterizing the three-dimensional topographic features of the workpiece to be processed, and is the basic data source for trajectory generation.

[0035] Model point cloud data: point cloud data generated based on the CAD digital model of the workpiece to be processed through a discretization algorithm (such as Monte Carlo sampling), which belongs to theoretical three-dimensional topographic data.

[0036] Measured point cloud data: point cloud data reflecting the real physical state of the workpiece obtained by directly collecting the surface of the workpiece to be processed through a three-dimensional scanning device (such as a line laser scanner or a structured light camera).

[0037] Two-dimensional projection point set: a two-dimensional coordinate point set formed after orthogonal projection of three-dimensional point cloud data from three-dimensional space to a preset two-dimensional plane (a plane with minimized projection distortion).

[0038] Two-dimensional process path: a trajectory generated according to a preset path interval within the contour of the workpiece represented by the two-dimensional projection point set, including reciprocating path or circumferential path, used for simplifying the complexity of three-dimensional trajectory planning.

[0039] Surface normal vector: a vector perpendicular to the local tangent plane to which the three-dimensional space point belongs, used for representing the surface posture of the point, and is the core parameter for generating three-dimensional process trajectory posture information.

[0040] Three-dimensional process trajectory: a complete trajectory containing the coordinates of each three-dimensional space point and the surface normal vector (posture information) formed after mapping the two-dimensional process path back to the three-dimensional space, which can be directly used to guide the robot to perform processing operation.

[0041] Three-dimensional space point: the basic unit constituting the point cloud data or the three-dimensional process trajectory, each point containing explicit three-dimensional coordinate information (X, Y, Z) for positioning the workpiece surface position.

[0042] Local tangent plane: the plane obtained by fitting the adjacent three-dimensional space points of a certain three-dimensional space point, which is tangent to the workpiece surface at the point and is the basis for calculating the surface normal vector.

[0043] Offline trajectory template: a three-dimensional process trajectory generated based on model point cloud data, which can be completed in advance during non-production period and used for subsequent batch processing for quick calling and adaptation.

[0044] Spatial pose transformation: a spatial transformation relationship (including translation, rotation, etc.) determined by iterative optimization, used to correct the deviation of the offline trajectory template and the measured point cloud data, so that the trajectory adapts to the actual pose of the workpiece.

[0045] Final execution trajectory: the trajectory obtained by compensating the offline trajectory template through spatial pose transformation, which adapts to the actual physical state of the workpiece to be processed, and is the basis for the robot to finally execute the processing operation.

[0046] The following will be described Figures 1-5 The process trajectory automatic generation method provided by the embodiment of the application.

[0047] Figure 1 is one of the process trajectory automatic generation method provided by the embodiment of the application, and the method comprises the following steps: Step 101, acquiring point cloud data representing the three-dimensional appearance of a workpiece to be processed.

[0048] This step is the starting stage of the method of the application, and its function is to provide initial digital geometric input for all subsequent calculation steps. In this step, acquisition means providing original data for representing the three-dimensional geometric shape of the workpiece through one or more technical means. Point cloud data, as a standard three-dimensional data structure, refers to a collection of a series of discrete data points in a shared coordinate system, which can effectively describe the outer surface contour of an object. Each data point contains its coordinate information in three-dimensional space.

[0049] The purpose of this step is to establish a unified data processing starting point. In a complex industrial environment, the three-dimensional appearance information of the workpiece may exist in various forms, such as a CAD digital model as a design drawing or a physical entity. This step converts the appearance information from different sources into standardized point cloud data, so that the subsequent projection, planning, mapping and other steps can be based on a unified data format, thereby simplifying the complexity of the algorithm and improving the universality of the method.

[0050] Step 102, project the point cloud data from the three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set.

[0051] In step 102, projection is a mathematical mapping operation that transforms each three-dimensional space point in the obtained three-dimensional point cloud data into a corresponding two-dimensional coordinate point on a certain two-dimensional plane.

[0052] In addition, the preset two-dimensional plane refers to a reference plane defined in three-dimensional space, which serves as the target carrier for this projection operation. The selection and positioning of this plane aim to maximize the preservation of the overall topological shape and contour features of the workpiece represented by the original point cloud, and avoid severe geometric distortion caused by improper projection angles.

[0053] The purpose of step 102 is to simplify the complex and computationally intensive three-dimensional space path planning problem into a mature and efficient two-dimensional plane path planning problem. By transforming massive three-dimensional point cloud data into a two-dimensional point set with simpler structure, the subsequent path planning algorithm can run in a lower-dimensional space, significantly reducing the computational complexity and execution time of the algorithm, and improving the robustness of path generation.

[0054] The final output of this step is a two-dimensional projection point set, which is a collection of multiple two-dimensional coordinate points. This collection intuitively outlines the boundary contour of the workpiece to be processed on the two-dimensional plane, providing a clear geometric basis for subsequent steps to plan specific process paths within this contour.

[0055] Step 103, in the two-dimensional plane, generate a two-dimensional process path according to the contour represented by the two-dimensional projection point set.

[0056] In this step, planning generation is an algorithm-driven process aimed at systematically and organizedly creating a path that effectively covers the target area, rather than simply connecting points. The key to this planning process lies in the contour represented by the two-dimensional projection point set. This means that the two-dimensional projection point set generated in the previous step is used as a geometric constraint boundary. The planning algorithm will ensure that the path it generates is strictly located within this contour, thereby ensuring that all subsequent machining actions of the robot occur within the effective area of the workpiece, avoiding empty travel or moving outside the workpiece.

[0057] The two-dimensional process path obtained in this step is a path defined by a sequence of ordered two-dimensional coordinate points or mathematical curves. It is essentially an abstract motion pattern, such as a reciprocating scanning pattern or a spiral contraction pattern from the outside to the inside. The two-dimensional path itself does not contain three-dimensional height and attitude information, but it completely encodes the lateral movement logic and coverage strategy that the robot tool should follow.

[0058] By this step, the two-dimensional process path output by the embodiment of the present application will serve as a clear index to provide guidance for the next step of recovering the three-dimensional morphology of the track from the original three-dimensional point cloud.

[0059] Step 104, mapping the two-dimensional process path to the three-dimensional space, determining a plurality of three-dimensional space points in the point cloud data, and calculating the surface normal vector of each three-dimensional space point to generate a three-dimensional process track containing three-dimensional coordinates and attitude information.

[0060] The main purpose of this step is to combine the two-dimensional process path with the original high-precision three-dimensional point cloud data to generate a three-dimensional process track containing three-dimensional coordinates and attitude information.

[0061] This step includes two main processes: The first process is "mapping the two-dimensional process path to the three-dimensional space to determine a plurality of three-dimensional space points". Here, the mapping is an operation of indexing the two-dimensional process path and querying and positioning in the original point cloud data. Specifically, for each point on the two-dimensional process path, the method of the embodiment will find the three-dimensional space point corresponding to it in the projection relationship in the obtained point cloud data. By repeating this operation, a series of ordered three-dimensional space points corresponding to the two-dimensional process path can be determined. This process ensures that the final generated track can be accurately attached to the original workpiece surface in position.

[0062] The second process is "calculating the surface normal vector of each three-dimensional space point". This process aims to assign attitude information to each three-dimensional position point determined in the previous step. Specifically, for each three-dimensional space point, the method of the embodiment analyzes the distribution of points in its local neighborhood in the original point cloud data, and determines the normal direction of the local surface at the point through mathematical fitting or other means. This normal direction is taken as the surface normal vector of the three-dimensional space point, which is used to guide the attitude of the robot end tool.

[0063] The final output three-dimensional process track is an ordered data sequence. Each element in the sequence contains a three-dimensional coordinate value and a surface normal vector value. The track data can be directly used for subsequent robot motion control.

[0064] The process trajectory automatic generation method provided by the embodiment of the present application simplifies the three-dimensional space path planning problem which is complex and has a large amount of calculation into a mature and efficient two-dimensional plane path planning problem, greatly improves the calculation efficiency and robustness, by obtaining point cloud data representing the three-dimensional topography of the workpiece to be processed, then projecting the point cloud data from the three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set, and planning a two-dimensional process path in the two-dimensional plane, finally mapping the two-dimensional process path to the three-dimensional space to determine a plurality of three-dimensional space points in the point cloud data and calculate the surface normal vector of each three-dimensional space point, so as to ensure that the finally generated trajectory points can be fitted in the original three-dimensional space. Through the combination of the above steps, the present application significantly improves the automation level and calculation efficiency of trajectory generation under the premise of ensuring the three-dimensional accuracy of the final trajectory, and realizes efficient and high-precision process trajectory generation.

[0065] Further, in the embodiment of the present application, the point cloud data acquisition method includes the following two methods: The first method is based on a CAD digital model, and a model point cloud data is generated by a discretization algorithm.

[0066] The path is suitable for scenarios with a theoretical design model of a workpiece. Specifically, when the design department provides a CAD digital model (for example, a file in a standard format such as STEP, IGES, Parasolid, etc.) of the workpiece to be processed, the embodiment performs a discretization algorithm to densely sample the surface of the CAD model to generate model point cloud data.

[0067] In a preferred scheme, to ensure the stability and accuracy of subsequent calculations, the discretization algorithm adopts a Poisson disk sampling (Poisson Disk Sampling) method. Compared with simple random sampling (Monte Carlo sampling), the significant advantage of Poisson disk sampling is that it can generate a very uniform point set. The core of the algorithm is to ensure that the Euclidean distance between any two points in the finally generated model point cloud data is not less than a preset minimum distance threshold. This feature effectively avoids the phenomenon of excessive aggregation or sparseness of point cloud in local areas, thereby providing a high-quality and unbiased data basis for subsequent steps of accurate local surface fitting and surface normal vector calculation. The number of sampling points or the minimum distance threshold can be set according to the size of the workpiece, the complexity of the surface, and the accuracy requirements of the subsequent process.

[0068] The second method is to collect the surface of the workpiece to be processed by a three-dimensional scanning device to obtain measured point cloud data.

[0069] The path is suitable for industrial scenarios without CAD models or requiring reverse teaching based on a physical standard sample. Specifically, the embodiment directly obtains measured point cloud data containing the real topography, size, and potential deformation of a physical workpiece by scanning the surface of the workpiece with a configured three-dimensional scanning device.

[0070] The three-dimensional scanning device can include a line laser profile scanner fixedly installed on a robot workstation or a handheld structured light scanner held by another robot / operator. The scanning process converts the surface of the workpiece into millions or even tens of millions of data points with three-dimensional coordinates (X, Y, Z). In actual applications, due to factors such as environmental lighting and object surface reflection characteristics, the obtained original measured point cloud data can contain noise and outliers. Therefore, after this step, necessary preprocessing of the measured point cloud data is also included, such as using filtering algorithms such as Statistical Outlier Removal or Radius Outlier Removal, to improve the purity and quality of the point cloud data.

[0071] Through any of the above techniques, whether the virtual model is derived from theoretical design or the real workpiece is derived from the physical world, the three-dimensional topography information is successfully converted into a unified point cloud data format, laying a solid foundation for subsequent standardized and automated trajectory generation processes.

[0072] In the embodiment of the application, the discretization algorithm is a key step for converting the CAD digital model into model point cloud data. The algorithm generates model point cloud data by sampling the surface of the CAD digital model. To ensure the quality and uniformity of the point cloud data, the algorithm particularly designs the following key steps: First, the algorithm uniformly samples the surface of the CAD digital model. The sampling process is based on the geometric characteristics of the model surface and uses an adaptive sampling strategy. Specifically, for each local region of the model surface, the sampling density is dynamically adjusted according to its curvature and complexity. In regions with large curvature or complex geometric features, the sampling points are more dense to ensure that the details of these key regions can be fully captured; while in regions with small curvature or relatively flat, the sampling points are relatively sparse to reduce unnecessary computational overhead.

[0073] In the process of generating sampling points, the algorithm introduces a preset minimum distance threshold. This threshold is used to ensure that the distance between any two points in the generated model point cloud data is not less than the threshold. In this way, redundant calculations caused by too dense sampling points can be avoided, while ensuring the uniform distribution of point cloud data. In specific implementation, after generating a new sampling point, the algorithm checks the distance between the point and the generated sampling points. If the distance is less than the preset minimum distance threshold, the point will be discarded and a new sampling point will be generated until the distance requirement is met.

[0074] After completing the initial sampling, the algorithm optimizes the generated model point cloud data. The optimization process includes removing duplicate points, filling holes, and smoothing the point cloud data. Through these optimization steps, the generated model point cloud data not only has uniform distribution characteristics, but also better reflects the geometric features of the CAD digital model, providing a high-quality data basis for subsequent process trajectory generation.

[0075] Through the above steps, the discretization algorithm of the embodiment of the present application can efficiently convert the CAD digital model into model point cloud data, while ensuring the uniformity and quality of the point cloud data. This algorithm is not only suitable for various complex CAD models, but also can flexibly adjust the sampling strategy according to different processing needs, providing reliable data support for the automatic processing of complex curved surface workpieces.

[0076] Taking the storage of the triangular facets of the CAD model as an example, referring to Figure 2 For static model input, first, according to the data format of the imported CAD model, different data formats are standardized to triangular facet storage. The specific steps are as follows: 1) STL file processing. For STL type files, since their structure is already composed of triangular facets, these triangular facets can be directly extracted and stored as a triangular facet set. This step ensures the direct use of data without additional conversion or processing.

[0077] 2) STEP file processing. For STEP type files, their data structure usually contains parameterized surface description. First, parameterize these parameterized surfaces in a planar unit in the UV space. Then, by sampling control points in the plane, use the Delaunay triangulation algorithm to convert these control points to triangular facets. Finally, map the generated triangular facets back to the three-dimensional surface to complete the normalization of data analysis.

[0078] 3) Generation of triangular facet set. Through the above steps, whether it is an STL file or a STEP file, a unified triangular facet set can be obtained. This set not only contains the geometric information of the workpiece, but also provides a basis for the subsequent discretization algorithm.

[0079] 4) Discretization. After obtaining the triangle mesh set, further discretization is performed on the mesh. By sampling on each mesh and ensuring that the distance between any two points is not less than the preset minimum distance threshold, the model point cloud data is generated. This step ensures the uniform distribution and high quality of the point cloud data, providing accurate data support for subsequent process trajectory generation.

[0080] Taking the discretization of point cloud data by Monte Carlo sampling method as an example, it specifically includes: 1) Calculate the total area of the triangle mesh. First, traverse all the triangle meshes, calculate the area of each mesh, and calculate the total area ΣS i . This step provides the basis for subsequent sampling point allocation.

[0081] 2) Determine the total number of sampling points. According to the preset number of sampling points per unit area n, calculate the total number of sampling points N=nΣS i . This total number N will be used for sampling the entire model.

[0082] 3) Assign sampling points to the triangle mesh. For each triangle mesh A i , according to its area proportion in the total area S i / ΣS i , calculate the number of sampling points it should have n i = n S i / ΣS i .

[0083] 4) Monte Carlo sampling. In each triangle mesh, sampling points are generated by Monte Carlo sampling method. Specifically, for the selected triangle mesh, the sampling point coordinates are randomly generated using the barycentric coordinate system: P=αA+βB+γC where A, B, C are the vertices of the triangle mesh, and α, β, γ are positive numbers satisfying α+β+γ=1. These coefficients are calculated by uniformly distributed random numbers r1, r2: where r1, r2~μ(0,1).

[0084] 5) Generate discrete point cloud. By the above sampling method, the continuous geometric features are transformed into discrete point cloud, which is the input of the next step process trajectory generation. This step ensures the uniform distribution and high quality of the point cloud data, providing accurate data support for subsequent process trajectory generation.

[0085] Through the description of the above embodiments, it is elaborated in detail how to convert the CAD model into a unified triangular facet storage, and a foundation is provided for subsequent discretization algorithms and process trajectory generation. This process not only improves the efficiency of data processing, but also ensures the accuracy and consistency of the data, providing reliable data support for the automatic machining of complex curved surface workpieces.

[0086] In step 102, projecting point cloud data from three-dimensional space to a pre-set two-dimensional plane is a key step to achieve efficient path planning. The specific implementation process is as follows: (1) Determine the projection plane.

[0087] First, analyze the distribution characteristics of the point cloud data in three-dimensional space. By calculating the principal component analysis (PCA) of the point cloud data, the main distribution direction of the point cloud data is determined. PCA can help identify the main geometric features of the point cloud data in three-dimensional space, especially the long axis, short axis and normal direction of the point cloud data. Based on the results of PCA, select a plane that best fits the distribution of the point cloud data, so that the projection distortion of the point cloud data on this plane is minimized. This plane is used as the pre-set two-dimensional plane for subsequent projection operations.

[0088] (2) Orthogonal projection After determining the pre-set two-dimensional plane, each three-dimensional space point in the point cloud data is orthogonally projected onto the two-dimensional plane. Specifically, for each three-dimensional space point P 3D , calculate its orthogonal projection point P 2D on the pre-set two-dimensional plane. The calculation method of orthogonal projection is to project the three-dimensional point along the normal direction of the plane onto the plane by the distance formula from the point to the plane, thereby generating a set of two-dimensional projection points. This process not only preserves the geometric features of the point cloud data, but also significantly reduces the complexity of the data, providing a foundation for subsequent two-dimensional path planning.

[0089] (3) Optimize the projection result After completing the projection, the generated two-dimensional projection point set is optimized. The optimization process includes removing overlapping points or abnormal points that may be introduced during the projection process, and smoothing the projection point set to ensure the quality and consistency of the two-dimensional projection point set. Through these optimization steps, the generated two-dimensional projection point set can more accurately reflect the contour and shape features of the original point cloud data, providing high-quality input data for subsequent two-dimensional process path planning.

[0090] Through the above steps, the embodiments of the present application can efficiently project three-dimensional point cloud data onto a two-dimensional plane while ensuring the accuracy and minimizing distortion of the projection. This method not only simplifies the complexity of path planning, but also provides reliable data support for subsequent two-dimensional process path generation, significantly improving the efficiency and accuracy of the overall process trajectory generation.

[0091] In step 103, the planning of the two-dimensional process path is based on the contour represented by the two-dimensional projection point set, and the purpose is to generate a set of paths that can efficiently traverse the internal area of the contour. The specific implementation process is as follows: (1) Contour extraction and analysis.

[0092] First, the two-dimensional projection point set is subjected to contour extraction. The contour lines are extracted from the point set by computational geometry methods such as the α-shapes algorithm or the convex hull algorithm. These contour lines can clearly define the boundary of the region to be processed. For complex contours, there may be an outer contour and multiple inner holes (i.e., inner contours). By analyzing the geometric characteristics of these contours, the starting point, ending point, and key feature points of the contour are determined, providing a basis for subsequent path planning.

[0093] (2) Path spacing setting.

[0094] According to the process requirements and workpiece characteristics, the path spacing is preset. Path spacing refers to the distance between adjacent two processing paths, which directly affects the processing efficiency and quality. For example, in polishing or spraying processes, the path spacing needs to be adjusted according to the coverage range of the tool and the processing precision requirements. After the path spacing is preset, the number and distribution of paths are calculated according to the geometric shape and size of the contour.

[0095] (3) Path generation method selection.

[0096] When generating two-dimensional process paths inside the contour, reciprocating paths or encircling paths can be selected.

[0097] Reciprocating paths are suitable for regular-shaped contours such as rectangles or circles. The path starts from one end of the contour, moves linearly in the preset direction, and then reverses when it reaches the other end, forming a zigzag-like path. This path method can efficiently cover the entire processing area and is suitable for large-area processing tasks.

[0098] Encircling paths are suitable for complex-shaped contours, especially for areas with multiple inner holes. The path starts from the boundary of the contour and gradually shrinks inward along the shape of the contour, forming an encircling path. This path method can better adapt to the geometric characteristics of complex contours, avoiding the intersection and overlap between paths.

[0099] (4) Path generation and optimization.

[0100] According to the selected path generation mode, two-dimensional process paths are generated inside the contour at a preset path interval. For a reciprocating path, a series of parallel line segments are generated by calculating the boundary points and directions of the contour; for a surrounding path, a series of closed paths are generated by gradually shrinking the contour boundary. After generating the path, the path is optimized, including smoothing the path corner, adjusting the path density, and removing possible redundant paths, to ensure the continuity and processing efficiency of the path.

[0101] Through the above steps, the embodiment of the present application can efficiently plan two-dimensional process paths in a two-dimensional plane that adapt to the internal area of the contour. These paths not only can fully cover the processing area, but also can flexibly adjust the path form and interval according to different process requirements, thereby providing high-quality two-dimensional basic paths for subsequent three-dimensional process trajectory generation, and significantly improving the efficiency and accuracy of the overall process trajectory generation.

[0102] In the embodiment of the present application, principal component analysis (PCA) and rigid body transformation are used to transform three-dimensional point cloud data into two-dimensional point cloud data on a main plane, and further through a contour extraction algorithm, the outer contour and inner hole of the planar polygon are found out to plan two-dimensional process paths. The specific steps are as follows: First, according to the input point cloud features (whether generated based on CAD model discretization or directly obtained by a perception system), the principal component direction of the point cloud data is determined by PCA, and then the point cloud data is mapped to the main plane by applying rigid body transformation to generate two-dimensional point cloud data.

[0103] Secondly, the alpha-shapes algorithm is used to extract the contour of the two-dimensional point cloud data to identify the outer contour and inner hole of the planar polygon. The alpha-shapes algorithm can flexibly extract different levels of detail features by adjusting the shape parameter, such as Figure 3 as shown.

[0104] Thirdly, the starting search direction and the horizontal interval h and the vertical interval v are set, and a line scan is performed on the planar point cloud. Whether the sampling point is inside the planar polygon is determined by the ray method. For the points generated inside the inner contour, the bisection method is used to gradually approach the contour boundary point to ensure that the sampling point can accurately cover the contour area.

[0105] Finally, according to the extracted contour information, two-dimensional process paths are planned. Inside the contour represented by the two-dimensional projection point set, a set of reciprocating paths or surrounding paths for traversing the internal area of the contour are generated at a preset path interval as two-dimensional process paths.

[0106] In the embodiment of the present application, in order to accurately extract the contour and plan the process path, it is necessary to determine the relationship between the sampling point and the planar polygon M. Figure 4A decision flowchart for sampling points is provided to handle the relationship between the current sampling point P1 and the next sampling point P2 and the planar polygon M. Specifically, it includes: 1) Acquisition of sampling points. Acquire the current sampling point P1 and the next sampling point P2.

[0107] 2) Relationship judgment. Determine the relative position relationship between the sampling points P1 and P2 and the planar polygon M. This involves four main cases: Case 1: P1 and P2 are both inside the polygon M.

[0108] Case 2: P1 and P2 are both outside the polygon M.

[0109] Case 3: P1 is inside the polygon M, and P2 is outside the polygon M.

[0110] Case 4: P1 is outside the polygon M, and P2 is inside the polygon M.

[0111] 3) Decision processing. According to different cases, take corresponding processing measures: For case 1 (P1, P2 are both inside M), keep both points as they are inside the polygon and have no effect on contour extraction.

[0112] For case 2 (P1, P2 are both outside M), do not keep both points as they are not inside the polygon and have no contribution to contour extraction.

[0113] For case 3 (P1 is inside M, P2 is outside M), keep P1 point and find contour points between P1 and P2, because the transition from P1 to P2 may cross the contour boundary.

[0114] For case 4 (P1 is outside M, P2 is inside M), keep P2 point and find contour points between P2 and P1, also because the transition from P2 to P1 may cross the contour boundary.

[0115] 4) Contour point search. In the case of finding contour points (case 3 and case 4), use appropriate algorithms (such as ray method or dichotomy method) to determine the contour points between P1 and P2. These contour points are points on the boundary of the polygon, which are crucial for accurately defining the polygon contour.

[0116] Further, in order to ensure the accuracy of the process trajectory and the processing quality, the embodiments of the present application adopt an efficient local fitting strategy when calculating the surface normal vector of the three-dimensional space point. The specific implementation process is as follows: Firstly, for each three-dimensional space point, the system retrieves neighboring points in its belonging point cloud data. These neighboring points are determined by setting a predetermined distance range, which takes into account the density of the point cloud and the geometric complexity of the workpiece surface. To improve retrieval efficiency, the system utilizes a spatial indexing structure (such as KD-tree) to quickly locate all neighboring points within the predetermined range from the target point.

[0117] Next, the system mathematically fits these neighboring points to determine the local tangent plane they lie on. The fitting process usually employs the Moving Least Squares (MLS) method, which minimizes the sum of squared distances from neighboring points to the fitted plane, resulting in an optimal fitting plane. This plane accurately reflects the geometric characteristics of the local region, providing a basis for subsequent normal vector calculation.

[0118] Finally, the system obtains the surface normal vector of the three-dimensional space point by calculating the normal direction of the fitted plane. The direction of the normal vector directly reflects the geometric "orientation" of the point on the local surface, which is crucial for the adjustment of the robot tool's pose. To further improve the accuracy and stability of the normal vector, the system also optimizes the calculation results, such as reducing noise impact through smoothing algorithms, ensuring the consistency and continuity of the normal vector within the local region.

[0119] Through the above method, the embodiment of the present application can efficiently and accurately calculate the surface normal vector of each three-dimensional space point, providing accurate pose information for complex surface machining, thereby significantly improving the machining quality and efficiency.

[0120] Further, based on the generated three-dimensional process trajectory, the embodiment of the present application further provides two optimization and application schemes to improve the adaptability and machining efficiency of the trajectory.

[0121] (1) Optimization of offline trajectory template.

[0122] Firstly, the three-dimensional process trajectory generated based on the model point cloud data is used as an offline trajectory template. This template is generated based on the idealized CAD model, which can guarantee the theoretical accuracy and process requirements of the trajectory. The generation process of the offline trajectory template does not depend on specific workpiece instances, so it can be completed in the design stage, saving programming time in actual production.

[0123] Before actual machining, the measured point cloud data of the workpiece to be machined is obtained through a three-dimensional scanning device. These measured data reflect the actual shape and size of the workpiece, including manufacturing errors, clamping deviations, and other actual physical state changes.

[0124] To adapt the offline trajectory template to the actual workpiece, the system determines the spatial pose transformation relationship between the points on the offline trajectory template and the corresponding points in the measured point cloud data by an iterative optimization algorithm. Specifically, the iterative closest point (ICP) algorithm or other optimization methods are used to gradually adjust the spatial position and attitude of the template trajectory until the distance between the template trajectory and the measured point cloud is minimized. This process can effectively compensate for the actual deviation of the workpiece and ensure the accuracy of the trajectory.

[0125] According to the determined spatial pose transformation relationship, the offline trajectory template is updated to generate the final execution trajectory adapted to the workpiece to be processed. The updated trajectory not only retains the process accuracy of the theoretical trajectory, but also adapts to the actual state change of the workpiece, thereby realizing high-precision automated operation in actual processing.

[0126] As shown in Figure 5 The embodiment of the present application provides a method for updating trajectory template through visual guidance to adapt to the pose change of the actual workpiece and improve the flexibility of the automated system. The specific steps are as follows: 1) Generation of trajectory template and feature surface point cloud. The trajectory template P and the feature surface point cloud C are generated using a three-dimensional model. The trajectory template P is a predetermined trajectory generated based on an ideal workpiece model, while the feature surface point cloud C contains key surface features of the workpiece.

[0127] 2) Acquisition of actual point cloud data. The point cloud data Cr of the actual workpiece is acquired through a visual perception system. These data reflect the state of the workpiece in actual clamping and positioning, which may deviate from the theoretical model.

[0128] 3) Point cloud registration. The iterative closest point (ICP) algorithm or other registration techniques are used to register the theoretical point cloud C and the actual point cloud Cr to determine the pose deviation T. The pose deviation T contains the difference between the actual pose of the workpiece and the theoretical pose.

[0129] 4) Update of trajectory template. The trajectory template P is updated according to the pose deviation T. This step generates process point information adapted to the actual pose of the current tooling by compensating the deviation T to the theoretical trajectory template, thereby obtaining the actual trajectory Pr.

[0130] 5) Execution of actual trajectory. The updated trajectory Pr is used to guide the robot for processing operation. Since the trajectory has been adjusted according to the pose of the actual workpiece, it can more accurately fit the workpiece surface, improving the processing quality and efficiency.

[0131] Through the description of the above embodiments, it is shown how to achieve automatic updating of the trajectory template to adapt to the pose changes of the actual workpiece through visual guidance and point cloud registration technology. This process not only improves the flexibility of the automation system, but also ensures the accuracy and consistency of the machining operation.

[0132] (2) Application of standard teaching trajectory.

[0133] In addition to the optimization of the offline trajectory template, the present application also provides a standard teaching trajectory generated based on the measured point cloud data. This trajectory is directly generated from the point cloud data of the actual workpiece, and can truly reflect the actual shape and machining requirements of the workpiece. The generation process of the standard teaching trajectory takes into account the actual physical state of the workpiece, so it has higher adaptability in subsequent machining.

[0134] In subsequent machining of the workpiece to be machined, the standard teaching trajectory can be directly called. This method is particularly suitable for multi-batch, few-variety production scenarios, and can significantly improve machining efficiency and reduce the workload of repeated programming. By directly calling the standard teaching trajectory, the robot can quickly achieve consistent machining results on different workpieces while ensuring the stability of the machining quality.

[0135] Through the above two schemes, the embodiments of the present application not only effectively solve the problem of disconnection between traditional offline programming and the actual state of the workpiece, but also achieve efficient and stable machining in multi-batch production. The optimization of the offline trajectory template ensures accurate matching of the theoretical trajectory and the actual workpiece, while the application of the standard teaching trajectory further improves the flexibility and efficiency of machining. The combination of these two methods provides an efficient, flexible and high-precision solution for the automated machining of complex curved surface workpieces.

[0136] The effects achieved by the method of the embodiments of the present application are as follows: 1) The method of the embodiments of the present application supports static CAD model and dynamic point cloud input, making it flexible to adapt to actual production needs of multi-variety or large quantities. This flexibility allows users to choose the most suitable data input method according to different production scenarios and workpiece characteristics, whether based on idealized CAD models in the design stage or based on real-time scanning point cloud data of actual workpieces.

[0137] 2) By inputting process parameters and performing process point waveform interpolation, the method can quickly adapt to multiple different process requirements for the same workpiece. This feature significantly improves the efficiency and flexibility of the automation system, and also improves the machining quality. The input of process parameters allows the system to automatically adjust the trajectory planning according to different machining goals (such as grinding, polishing, spraying, etc.), achieving optimization of the process.

[0138] 3) The method of the embodiment of the present application supports the combination of process templates and point cloud registration technology, providing accurate processing schemes for freely placed tooling. This combination not only improves the flexibility of the automation system, but also enables the system to adapt to actual physical state changes of the workpiece, such as manufacturing errors, clamping deviations, etc., thereby ensuring the accuracy and consistency of the processing process.

[0139] The process trajectory automatic generation device provided by the embodiment of the present application is described below, and the process trajectory automatic generation device described below can be referred to in correspondence with the process trajectory automatic generation method described above.

[0140] The embodiment of the present application provides a process trajectory automatic generation device, referring to Figure 6 , comprising: A point cloud data acquisition module 610 is configured to acquire point cloud data representing a three-dimensional topography of a workpiece to be processed. A two-dimensional point set generation module 620 is configured to project the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set. A two-dimensional process generation module 630 is configured to plan and generate a two-dimensional process path in the two-dimensional plane according to the contour represented by the two-dimensional projection point set. A three-dimensional process generation module 640 is configured to map the two-dimensional process path to the three-dimensional space, determine a plurality of three-dimensional space points in the point cloud data, and calculate a surface normal vector of each three-dimensional space point to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

[0141] Figure 7 An example of a schematic diagram of the physical structure of an electronic device is shown in Figure 7 The electronic device can include a processor 810, a communications interface 820, a memory 830, and a communications bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other through the communications bus 840. The processor 810 can invoke the logical instructions in the memory 830 to execute the process trajectory automatic generation method, which includes: acquiring point cloud data representing a three-dimensional topography of a workpiece to be processed; projecting the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set; planning and generating a two-dimensional process path in the two-dimensional plane according to the contour represented by the two-dimensional projection point set; mapping the two-dimensional process path to the three-dimensional space, determining a plurality of three-dimensional space points in the point cloud data, and calculating a surface normal vector of each three-dimensional space point to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

[0142] Further, the logic instructions in the memory 830 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the part of the prior art that contributes essentially or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0143] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the process trajectory automatic generation method provided by the above-mentioned methods. The method comprises: acquiring point cloud data representing the three-dimensional topography of a workpiece to be processed; projecting the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set; in the two-dimensional plane, a two-dimensional process path is planned and generated according to the contour represented by the two-dimensional projection point set; mapping the two-dimensional process path to the three-dimensional space, determining a plurality of three-dimensional space points in the point cloud data, and calculating the surface normal vector of each three-dimensional space point to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

[0144] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the process trajectory automatic generation method provided by the above-mentioned methods. The method comprises: acquiring point cloud data representing the three-dimensional topography of a workpiece to be processed; projecting the point cloud data from a three-dimensional space to a preset two-dimensional plane to generate a two-dimensional projection point set; in the two-dimensional plane, a two-dimensional process path is planned and generated according to the contour represented by the two-dimensional projection point set; mapping the two-dimensional process path to the three-dimensional space, determining a plurality of three-dimensional space points in the point cloud data, and calculating the surface normal vector of each three-dimensional space point to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

[0145] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0146] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0147] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for automatically generating process trajectories, characterized in that, include: Acquire point cloud data characterizing the three-dimensional morphology of the workpiece to be processed; The point cloud data is projected from three-dimensional space onto a preset two-dimensional plane to generate a two-dimensional projection point set; In the two-dimensional plane, a two-dimensional process path is planned and generated based on the contour represented by the two-dimensional projection point set; The two-dimensional process path is mapped to the three-dimensional space, multiple three-dimensional spatial points are determined in the point cloud data, and the surface normal vector of each three-dimensional spatial point is calculated to generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

2. The method according to claim 1, characterized in that, The point cloud data includes: model point cloud data generated by a discretization algorithm based on the CAD digital model of the workpiece to be processed; or measured point cloud data directly obtained by acquiring the surface of the workpiece to be processed using a 3D scanning device.

3. The method according to claim 2, characterized in that, The discretization algorithm specifically includes: sampling on the surface of the CAD digital model, and ensuring that the distance between any two points in the generated model point cloud data is not less than a preset minimum distance threshold.

4. The method according to claim 1, characterized in that, Projecting the point cloud data from three-dimensional space onto a preset two-dimensional plane to generate a two-dimensional projection point set specifically includes: By analyzing the distribution characteristics of the point cloud data in three-dimensional space, a plane that minimizes its projection distortion is determined as the preset two-dimensional plane; Each three-dimensional spatial point in the point cloud data is orthogonally projected onto the two-dimensional plane to generate the two-dimensional projection point set.

5. The method according to claim 1, characterized in that, In the two-dimensional plane, based on the contour represented by the two-dimensional projection point set, a two-dimensional process path is planned and generated, including: Within the contour represented by the two-dimensional projection point set, a set of reciprocating or circling paths for traversing the internal region of the contour is generated according to a preset path spacing, serving as the two-dimensional process path.

6. The method according to claim 1, characterized in that, The step of calculating the surface normal vector of each of the three-dimensional space points includes: For any given three-dimensional spatial point, find the neighboring three-dimensional spatial points whose distance from the given three-dimensional spatial point is within a predetermined range in the point cloud data to which it belongs; The local tangent plane is determined by mathematically fitting the neighboring three-dimensional spatial points; The normal direction of the local tangent plane is taken as the surface normal vector of the point in the three-dimensional space.

7. The method according to claim 2, characterized in that, The method further includes: The three-dimensional process trajectory generated based on the model point cloud data is used as an offline trajectory template; Obtain the measured point cloud data of the workpiece to be processed; By iteratively optimizing, a spatial pose transformation relationship is determined that minimizes the distance between points on the offline trajectory template and corresponding points in the measured point cloud data. The offline trajectory template is updated according to the spatial pose transformation relationship to generate a final execution trajectory adapted to the workpiece to be processed.

8. The method according to claim 2, characterized in that, The method further includes: The three-dimensional process trajectory generated based on the measured point cloud data will be used as a standard teaching trajectory for direct reference when processing subsequent workpieces.

9. An automatic process trajectory generation device, characterized in that, include: The point cloud data acquisition module is used to acquire point cloud data that characterizes the three-dimensional shape of the workpiece to be processed; The two-dimensional point set generation module is used to project the point cloud data from three-dimensional space onto a preset two-dimensional plane to generate a two-dimensional projection point set; A two-dimensional process generation module is used to plan and generate a two-dimensional process path in the two-dimensional plane based on the contour represented by the two-dimensional projection point set. The three-dimensional process generation module is used to map the two-dimensional process path to the three-dimensional space, determine multiple three-dimensional space points in the point cloud data, calculate the surface normal vector of each three-dimensional space point, and generate a three-dimensional process trajectory containing three-dimensional coordinates and attitude information.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the automatic process trajectory generation method as described in any one of claims 1 to 8.

Images