Blank allowance field driving robot laser scribing path generation method and system

Abstract

The present application belongs to the technical field of digital manufacturing, and specifically relates to a kind of blank allowance field driving robot laser scribing path generation method and system.The measured point cloud of complex aerospace air vane wall blank, CAD nominal model and registration result meeting the allowance constraint are obtained;Then, a signed machining allowance field is constructed based on the registration result, and the target path to be scribed is extracted from the allowance field, CAD design boundary or machining boundary;The target path is projected onto the measured surface of the blank, and a geometric scribing path containing path points, normal and tangent is generated;Subsequently, the laser tool coordinate system expected pose is constructed according to the laser incidence angle, working distance and path direction;Finally, the laser tool pose is converted into the robot flange trajectory using the calibration transformation between the laser tool coordinate system and the robot end flange coordinate system, and the robot executable scribing program is output after kinematics and process constraint checking.The present application can realize the closed loop from measurement, analysis to scribing execution.

Description

Technical Field

[0001] This invention belongs to the field of digital manufacturing technology, and in particular relates to a method and system for generating laser scribing paths for a robot driven by the allowance field of a blank. Background Technology

[0002] Complex aerospace rudder panel blanks typically require machining allowance inspection and scribing before subsequent processing. In traditional processes, operators need to manually mark machining boundaries, allowance indicators, or fitting lines on the part surface according to process requirements and measurement results. For complex panel blanks with free-form surfaces, ribs, hole boundaries, and locally obscured areas, manual scribing suffers from problems such as high labor intensity, poor repeatability, difficulty in controlling positioning errors, and difficulty in quality traceability.

[0003] With the development of industrial robots, structured light 3D measurement, and laser processing technologies, it is now possible to automatically acquire the geometric target to be scribed by using a robot equipped with a 3D measuring instrument to obtain the point cloud of the blank, and then registering and performing allowance analysis between the point cloud and the CAD nominal model. In particular, after adopting allowance-constrained CAD-point cloud registration, the allowance analysis results can better meet the requirements of no negative allowance, blank encompassing the CAD nominal model, datum area fit, and uniform allowance on the machining surface.

[0004] However, the geometric targets obtained from margin analysis cannot be directly used as robot execution commands. When a robot actually performs a marking task, the controller typically requires continuous pose trajectories, joint angle sequences, and laser switch signals in the robot's base coordinate system or end flange coordinate system, rather than the marking target points in the CAD coordinate system. For non-contact laser marking tools, there is also a tool coordinate system calibration relationship between the laser application point and the robot's end flange. If a complete coordinate chain transformation is lacking, the final marking position will produce systematic errors.

[0005] Existing 3D inspection software can typically output CAD edges, allowance contour lines, or point cloud deviation maps. However, these outputs are mostly located in the CAD coordinate system, workpiece coordinate system, or the internal coordinate system of the inspection software, making it difficult to directly convert them into an executable end-effector trajectory for the robot. On the other hand, while robot offline programming software can generate trajectories based on CAD curves, it usually does not fully utilize the measured point cloud and machining allowance field, making it difficult to ensure that the scribing position truly falls on the blank surface and reflects the actual allowance distribution.

[0006] Therefore, there is an urgent need for a method that can automatically convert the geometric scribing target obtained from the margin analysis into a robot-executable laser scribing trajectory to solve the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for generating laser scribing paths for a robot driven by the allowance field of a blank. By combining laser tool coordinate calibration, robot kinematic constraints and laser process constraints, an automated closed loop is formed from point cloud / CAD allowance analysis to robot laser scribing execution.

[0008] To achieve the above objectives, the present invention provides a method for generating laser scribing paths for a robot driven by the allowance field of a blank, comprising the following steps: S1. Obtain the measured point cloud of the blank, the corresponding CAD nominal model, and the registration result between the measured point cloud and the CAD nominal model that satisfies the machining allowance constraint; S2. Based on the registration result, calculate the signed machining allowance of the measured point cloud relative to the CAD nominal model, and construct a machining allowance field associated with the blank surface or the CAD nominal model; S3. Using the machining allowance field as the process criterion, extract the target path to be drawn based on one or more of the following data: CAD design boundary, machining boundary, trimming boundary, hole boundary, rib boundary, target allowance threshold, allowance contour line, and manually specified features. S4. Project the target path to be scribed onto the measured point cloud of the blank and the reconstructed surface of the blank, and preprocess the projected path to obtain a geometric scribing path containing the coordinates of the path points, the path tangent vector and the surface normal vector. S5. Based on the path point coordinates, path tangent vector, surface normal vector, and preset laser incident angle and working distance, construct the desired pose of the laser tool coordinate system at each path point; S6. Obtain the transformation relationship between the calibrated laser tool coordinate system and the robot end flange coordinate system, and combine the transformation relationship between the workpiece coordinate system, CAD coordinate system and robot base coordinate system to convert the desired pose of the laser tool coordinate system into the robot end flange trajectory. S7. Perform kinematic reachability, singularity, collision risk, attitude continuity, velocity continuity and laser process parameters verification on the robot end flange trajectory, and adjust the path points, tool attitude or movement speed according to the verification results to obtain the final robot end flange trajectory.

[0009] Furthermore, in step S1, the registration result that satisfies the machining allowance constraint is obtained through the allowance constraint registration algorithm, or it is obtained by using the automatic registration result output by the detection software as the initial value, and then performing signed machining allowance calculation, machining allowance constraint verification and pose adjustment; the constraint conditions of the allowance constraint registration algorithm include one or more of the following: no negative allowance constraint, blank encompassing CAD nominal model constraint, reference area fitting constraint and machining surface allowance uniformity constraint.

[0010] Furthermore, the margin constraint registration algorithm includes the following steps: S101. The measured point cloud data is preprocessed to obtain a rough point cloud containing three-dimensional coordinates and normal vectors; S102. Perform surface discretization and feature recognition on the CAD nominal model in sequence to obtain the CAD discrete model and the feature surface grouping corresponding to the surface entity; S103. Based on the design information of the CAD nominal model and / or the feature surface grouping, determine at least one machined surface region, at least one reference constraint region, and non-machined region, and set corresponding allowance constraint parameters for each machined surface region; S104. Obtain the initial rigid body transformation matrix between the blank point cloud and the CAD nominal model based on one of the following methods: feature coarse registration, principal direction registration, automatic registration results from software, or corresponding features selected manually. S105. Based on the initial rigid body transformation matrix, calculate the signed machining allowance of the blank point cloud in each machining surface region relative to the CAD nominal model. The sign of the signed machining allowance is used to characterize whether the blank point cloud is located outside the CAD nominal model. S106. Establish a registration optimization model that simultaneously includes machining allowance constraints, datum constraints, and allowance uniformity constraints. The machining allowance constraints include a negative allowance constraint and / or a constraint that the blank completely encompasses the CAD nominal model, calculated from the signed machining allowance and allowance constraint parameters. The datum constraint is used to limit the positional error of the datum surface, datum hole, datum axis, or machined area relative to the CAD nominal model. The allowance uniformity constraint is used to limit the fluctuation of the machining allowance within the machined surface area, calculated based on the allowance constraint parameters. S107. Using the initial rigid body transformation matrix as the initial value, solve the registration optimization model to obtain the final rigid body transformation matrix that satisfies the machining allowance constraint, the reference constraint, and the allowance uniformity constraint.

[0011] Furthermore, in step S3, the extraction method of the target path to be drawn includes: extracting the fill contour lines corresponding to a preset fill threshold in the processing fill field; Alternatively, design edges, trimming boundaries, hole boundaries, rib boundaries, or machining area boundaries can be extracted from the CAD nominal model to form candidate paths in the CAD coordinate system, and the candidate paths can be converted to the blank coordinate system or the workpiece coordinate system according to the registration results. Alternatively, based on the topological relationship between the negative margin region, the low margin region, and the target processing boundary, a margin indicator line can be generated to indicate the range to be removed in subsequent processing. The target path to be drawn is represented in one or more forms, such as a point sequence, a curve segment, a polyline segment, a spline curve, or a CAD curve entity.

[0012] Furthermore, in step S4, the preprocessing includes sorting, smoothing, breakpoint repair, and resampling; The sorting includes: sorting discrete path points based on proximity distance, curve parameters, CAD boundary topology, or path direction; The smoothing includes: smoothing the path using moving least squares, spline fitting, B-spline smoothing, circular arc fitting, or polyline simplification methods; The breakpoint repair includes: identifying breakpoints based on the maximum breakpoint distance, normal change angle, curvature change, or path topology, and interpolating, bridging, or segmenting the breakpoints for output. The resampling includes resampling the path according to preset point spacing, curvature adaptive step size, or robot speed planning requirements.

[0013] Furthermore, step S5 includes: the process of constructing the desired pose of the laser tool coordinate system includes: The coordinates of the path points are used as the positions of the laser action points; The surface normal vector or the direction corrected for the incident angle is taken as the output axis direction of the laser tool coordinate system; The path tangent vector is used as the line, the direction of travel, or the lateral reference direction of the laser tool coordinate system; A laser tool coordinate system rotation matrix satisfying the right-hand coordinate system constraint is constructed by cross product and orthogonalization. The laser's point of action is compensated along the output axis based on the laser's focal length, defocusing amount, or the rangefinder's working distance.

[0014] Furthermore, in step S6, the trajectory of the robot's end flange is calculated according to the following coordinate chain:

[0015] in, This represents the desired transformation from the robot's base coordinate system B to the robot's end effector flange coordinate system E. This represents the transformation from the robot's base coordinate system B to the workpiece coordinate system P. This represents the transformation from the workpiece coordinate system P to the CAD coordinate system C. This represents the transformation from the CAD coordinate system C to the laser tool coordinate system O, indicating the desired pose. This represents the calibration transformation from the robot end-effector coordinate system E to the laser tool coordinate system O.

[0016] Furthermore, in step S7, the verification specifically includes: Solve the joint angles corresponding to each path point based on the robot's inverse kinematics; Determine whether each joint angle exceeds the joint limit, and determine whether the change in joint angle between adjacent path points exceeds a preset threshold; Determine if the robot is approaching a singular configuration or experiencing a sudden change in posture; Determine if there is a risk of collision between the robot body, the end effector laser, the tooling fixture, the turntable, and the blank to be scribed; Determine whether the laser incident angle, working distance, scribing speed, and laser process parameters meet the preset process window.

[0017] Furthermore, it also includes: based on the final robot end flange trajectory, outputting a robot-executable laser scribing program, end flange trajectory, laser switch control signal, and corresponding machining allowance scribing report; the robot-executable laser scribing program includes one or more of the following: robot motion commands, end flange pose sequence, joint angle sequence, linear velocity, acceleration, laser power, laser switch, pulse frequency, defocusing amount, working distance, and path segment identifier.

[0018] The present invention also provides a blank margin field driven robot laser scribing path generation system for performing the blank margin field driven robot laser scribing path generation method described in any of the above claims.

[0019] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages: 1. The method for generating a robot laser scribing path based on the allowance field of a blank provided by this invention can further convert the geometric scribing target obtained from allowance constraint CAD-point cloud registration and allowance field analysis into a robot executable trajectory. This allows the robot to inherit the results of allowance constraint registration and allowance field analysis, and combine laser tool coordinate calibration, robot kinematic constraints, and laser process constraints to form an automated closed loop from point cloud / CAD allowance analysis to robot laser scribing execution. This reduces manual teaching and coordinate transformation errors, and improves the automation level, trajectory consistency, and processing guidance accuracy of laser scribing on complex blanks.

[0020] 2. This invention improves the fit between the scribing target and the actual blank surface by projecting the path to be scribed onto the measured point cloud or reconstructed surface of the blank.

[0021] 3. By introducing constraints on path tangency, surface normal, laser incident angle and working distance, this invention can generate tool postures that meet the requirements of laser scribing process.

[0022] 4. This invention uses the laser tool coordinate system calibration results to realize the conversion of the laser action point trajectory to the robot end flange trajectory, reducing the errors of manual teaching and coordinate transformation; through accessibility, singularity, collision, attitude continuity and velocity continuity checks, the executability and stability of the robot's line drawing trajectory can be improved.

[0023] 5. This invention can combine the scribing results with retesting for error feedback compensation, thereby improving the repeatability and quality traceability of laser scribing on complex aerospace air rudder panel blanks. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall process of the method of the present invention; Figure 2 A schematic diagram of the three-dimensional measurement point cloud of a complex aerospace air rudder panel blank; Figure 3 A schematic diagram of a CAD model of a complex aerospace air rudder panel blank; Figure 4 A schematic diagram of the extraction of the target path to be marked, driven by the machining allowance field; Figure 5 A schematic diagram of the projection, sorting, smoothing, and resampling process for geometrically drawn paths; Figure 6 This diagram illustrates the coordinate chain transformation between the laser tool coordinate system, the robot end flange coordinate system, the robot base coordinate system, the workpiece coordinate system, and the CAD coordinate system. Figure 7 This is a schematic diagram of the actual path points drawn by the robot after coordinate transformation and path point adjustment. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0026] Please see Figure 1-7 This invention provides a method for generating laser scribing paths for a robot driven by the margin field of a complex aerospace air rudder panel blank. The specific process is as follows: Figure 1 As shown, it includes: S1, acquire the measured point cloud of the complex aerospace air rudder panel blank, the corresponding CAD nominal model, and the registration result between the measured point cloud and the CAD nominal model that satisfies the machining allowance constraint.

[0027] Specifically, such as Figure 2As shown, firstly, an industrial robot equipped with a surface structured light 3D measuring instrument acquires the point cloud of the blank surface. The point cloud data includes point coordinates and normal vectors. The corresponding CAD nominal model is as follows: Figure 3 As shown, the data is provided by IGES, STEP, STL, PLY, or the inspection software project file. The final pose relationship between the point cloud and the CAD nominal model is obtained through the margin constraint CAD-point cloud registration method. This pose relationship can come from the margin constraint registration algorithm, or from the automatic registration results of the inspection software and subsequent algorithm fine-tuning results.

[0028] The margin-constrained CAD-point cloud registration method may specifically include the following steps: S101. Obtain the measured point cloud data of the aerospace panel blank to be registered, and the CAD nominal model corresponding to the geometry of the finished aerospace panel; preprocess the measured point cloud data to obtain the blank point cloud containing three-dimensional coordinates and normal vectors. S102. Perform surface discretization and feature recognition on the CAD nominal model in sequence to obtain the CAD discrete model and the feature surface grouping corresponding to the surface entity; S103. Based on the design information of the CAD nominal model and / or the feature surface grouping, determine at least one machined surface region, at least one reference constraint region, and non-machined region, and set corresponding allowance constraint parameters for each machined surface region; S104. Obtain the initial rigid body transformation matrix between the blank point cloud and the CAD nominal model based on one of the following methods: feature coarse registration, principal direction registration, automatic registration results from software, or corresponding features selected manually. S105. Based on the initial rigid body transformation matrix, calculate the signed machining allowance of the blank point cloud in each machining surface region relative to the CAD nominal model. The sign of the signed machining allowance is used to characterize whether the blank point cloud is located outside the CAD nominal model. S106. Establish a registration optimization model that simultaneously includes machining allowance constraints, datum constraints, and allowance uniformity constraints. The machining allowance constraints include a negative allowance constraint and / or a constraint that the blank completely encompasses the CAD nominal model, calculated from the signed machining allowance and allowance constraint parameters. The datum constraint is used to limit the positional error of the datum surface, datum hole, datum axis, or machined area relative to the CAD nominal model. The allowance uniformity constraint is used to limit the fluctuation of the machining allowance within the machined surface area, calculated based on the allowance constraint parameters. The objective function of the registration optimization model is shown below:

[0029] in, This represents the rigid body transformation matrix to be solved. Indicates the geometric fitting error. This represents the penalty for negative margin or falling below the minimum margin threshold. This indicates the fluctuation in the distribution of machining allowance, and is a constraint term for the uniformity of allowance. Indicates the reference constraint error. This indicates a deviation from the regularization term. to These are the corresponding weighting coefficients.

[0030] The distance from the transformed blank measuring points to the CAD surface is used to construct the robust geometric fitting error, which is then processed by a robust kernel function.

[0031] In the formula, Represents points in a blank point cloud i The corresponding first on the CAD nominal model j Signed processing margin between sampling points For point i Processing area k The minimum margin threshold;

[0032] In the formula, Indicates the first k The set of measuring points within a processing surface area. Indicates the first i Each measurement point is within the rigid body transformation matrix T Signed machining allowance under action, Var Represents variance. Indicates the first k Weight of allowance uniformity corresponding to each processing surface area; The reference constraint term is calculated from one of the following: the distance from the reference surface point to the surface, the distance from the reference hole point to the axis, the included angle of the reference surface normal, or the included angle of the reference hole axis. For relative to the initial rigid body transformation matrix The pose deviation constraint is used to prevent the optimization result from deviating too much from the automatic matching result.

[0033] S107. Using the initial rigid body transformation matrix as the initial value, solve the registration optimization model to obtain the final rigid body transformation matrix that satisfies the machining allowance constraint, the reference constraint, and the allowance uniformity constraint.

[0034] S2, based on the registration result, calculate the signed machining allowance of the measured point cloud relative to the CAD nominal model, and construct a machining allowance field associated with the blank surface or the CAD nominal model.

[0035] Specifically, let the blank measuring point be... The corresponding point in CAD is CAD external normal or local surface normal is Then there is a marked machining allowance. It can be represented as:

[0036] Based on all measuring points or sampling points in the processing area Construct a machining allowance field. The machining allowance field can be bound to the blank point cloud, reconstructed mesh, CAD sampling points, or CAD surface parameter domain.

[0037] When the signed processing allowance is less than the preset threshold, the corresponding area is marked as an area with insufficient allowance; when the signed processing allowance is within the preset target range, the corresponding area is marked as a candidate area for drawing lines.

[0038] Based on at least one of the following information: surface normal, curvature, boundary curve, hole axis, surface type, and design level, the CAD nominal model is divided into planar feature surfaces, cylindrical feature surfaces, freeform surface feature surfaces, boundary line features, and non-machined areas.

[0039] S3, using the machining allowance field as the process criterion, at least one of the following: CAD design boundary, machining boundary, target allowance threshold, allowance contour lines, and manually specified features, to extract the target path to be marked. The machining allowance field is still used to determine whether the path needs to be marked, whether it is in the safe allowance area, whether it should be offset / truncated / filtered, and to generate an allowance report.

[0040] Specifically, such as Figure 4 As shown, after the machining allowance field is constructed, the target path to be marked is extracted according to the process requirements. Path extraction methods include, but are not limited to, the following three categories: The first type extracts the target path to be scribed based on the machining boundary, trimming boundary, hole boundary, rib boundary or manually specified features in the CAD nominal model, and maps it to the blank coordinate system according to the final registration relationship; The second type is to extract the margin contour lines based on the preset threshold in the machining margin field, such as the curves corresponding to the target machining margin, the minimum safety margin, or the machining removal boundary. The third type generates a margin indicator line based on the spatial relationship between the negative margin area, the low margin area and the design processing boundary, which is used to indicate the range of subsequent processing or repair.

[0041] For the discretized candidate path points, neighborhood connectivity analysis and curve sorting are performed first. Then, breakpoints, noise points, and abnormal branches are identified based on the path point spacing, angle, and curvature. For path segments with small breakpoints, spline interpolation, straight-line bridging, or local reprojection can be used for repair. For path segments with large breakpoints or uncertain topology, segmented output can be used to avoid abnormal robot movements in discontinuous regions.

[0042] S4. Project the target path to be drawn onto the measured point cloud of the blank and the reconstructed surface of the blank, and sort, smooth, repair the breakpoints and resample the projected path to obtain a geometric drawing path containing the coordinates of the path points, the path tangent vector and the surface normal vector.

[0043] Specifically, projecting the target path to be marked involves: 1) Coordinate system one: that is, transforming the path obtained in S3 into the blank measured point cloud or workpiece coordinate system through the final registration result (rigid body transformation matrix); 2) Surface bonding projection: The path points are then projected onto the measured point cloud of the blank, the reconstructed surface of the blank, or the registered CAD nominal model, so that the path actually falls on the scribing surface and obtains geometric information such as normal and tangential directions.

[0044] Specifically, such as Figure 5 As shown, the target path to be marked is projected onto the measured point cloud of the blank or the reconstructed surface. For each path point... Search for the nearest point or locally fitted surface in the neighborhood of the blank point cloud and obtain the projection point. Surface normal and path tangent .

[0045] Path Tangent It can be calculated from the difference between adjacent path points, the derivative of spline curves, or the direction of CAD curve parameters. Surface normal. The normal vector can be obtained from the original point cloud normal, neighborhood principal component analysis, local surface fitting, or the normal vector of the corresponding CAD surface. To improve path stability, the normal and tangential sequences can be made oriented in a consistent manner to avoid 180-degree flips in the pose of adjacent points.

[0046] The path is then resampled according to the requirements of robot motion and laser processing. For straight lines or gently curved lines with small curvature, a larger point spacing can be used; for areas with large curvature or drastic changes in margin, a smaller point spacing is used to ensure that the scribing trajectory fits the curved surface.

[0047] S5. Based on the path point coordinates, the path tangent vector, the surface normal vector, and the preset laser incident angle and working distance, construct the desired pose of the laser tool coordinate system at each path point; Specifically, for each projection point in the geometrically drawn path Constructing the laser tool coordinate system The desired pose. Let the laser output axis direction be... The direction of travel is marked. The horizontal direction is .

[0048] When the laser is required to be incident along the surface normal, it can be made When it is required that the laser maintains a fixed incident angle relative to the surface normal. At that time, it can be done from the surface normal. and auxiliary directions Corrected within the plane formed:

[0049] in, Can be tangentially along the path With surface normal The cross product direction is determined. To ensure the laser marking direction aligns with the path direction, the path tangent is projected onto a plane perpendicular to the path direction. In the plane:

[0050] Then obtain through cross product This yields the rotation matrix of the laser tool coordinate system relative to the robot base coordinate system or the workpiece coordinate system. Here, normalize is a vector normalization operation used to scale a vector to a unit magnitude.

[0051] The position of the laser tool coordinate system can be determined based on the laser's point of action. Compensation is performed based on the working distance. When the origin of the laser tool coordinate system is defined at the laser's point of action, its position is... When the origin of the laser tool coordinate system is defined inside the laser or before or after the focal point, the coordinates are determined according to the focal length, defocusing distance, or rangefinder readings along... Directional offset compensation is performed.

[0052] S6. Obtain the transformation relationship between the calibrated laser tool coordinate system and the robot end flange coordinate system, and combine the transformation relationship between the workpiece coordinate system, CAD coordinate system and robot base coordinate system to convert the desired pose of the laser tool coordinate system into the robot end flange trajectory.

[0053] The robot's base coordinate system is established at the base position, which is the robot's 0-axis position, while the robot's end effector flange coordinate system is established at the last joint axis, at the end effector flange position connected to the actual working tool (in this application, the laser). Industrial robots typically have 4-6 joint axes. Generally, the relationship between the robot's base coordinate system and the robot's end effector flange coordinate system can be directly obtained through the positive kinematic parameters set by the robot manufacturer.

[0054] Specifically, such as Figure 6 As shown, when a robot performs line marking, the controller typically requires not geometric points in the CAD coordinate system, but rather the pose of the robot's end flange or the sequence of joint angles. Therefore, a complete coordinate chain is needed for transformation.

[0055] set up Representing the robot's base coordinate system B To the j Laser tool coordinate system at each path point The expected transformation, This indicates the calibrated robot end-effector coordinate system. E To the laser tool coordinate system O The transformation of the robot's end flange will result in the desired pose. for:

[0056] When the geometric line path is first in the CAD coordinate system C or workpiece coordinate system P When generated in the middle, it can be further written as:

[0057] in, This represents the transformation from the robot's base coordinate system B to the workpiece coordinate system P. This represents the transformation from the workpiece coordinate system P to the CAD coordinate system C. Indicates the distance from CAD coordinate system C to the first... j The desired transformation of the laser tool coordinate system O at each path point.

[0058] The coordinates can be obtained through laser tool coordinate system calibration. Calibration methods can include non-contact laser ranging constraint calibration, visual calibration, laser tracker calibration, contact-assisted calibration, or tool coordinate system calibration built into the robot controller. For non-contact laser marking tools, a tool coordinate system calibration method that can simultaneously determine the position of the laser's point of application and the direction of the laser's output axis is preferred.

[0059] S7, the robot end flange trajectory is checked for robot kinematic reachability, singularity, collision risk, posture continuity, velocity continuity and laser process parameters, and the path points, tool posture or movement speed are adjusted according to the check results.

[0060] Specifically, such as Figure 7 As shown, after obtaining the initial trajectory of the robot's end flange, the inverse kinematics of the robot are solved for each path point to obtain one or more candidate joint configurations. Based on joint constraints, configuration continuity, singularity indices, and collision detection results, the optimal configuration sequence is selected, and the robot's executable laser scribing program, end flange trajectory, laser switch control signal, and corresponding machining allowance scribing report are output.

[0061] For unreachable points, the following methods can be used: adjust the laser incident angle or the rotation angle around the optical axis; execute the continuous path in segments; change the posture of the robot's external axis, turntable, or tooling; locally offset the path points within the range that does not affect the scribing quality; mark the unreachable area as a manual review area.

[0062] For regions with abrupt attitude changes, quaternion spherical linear interpolation, attitude splines, or rotation vector smoothing methods can be used. For regions with discontinuous velocity, trapezoidal velocity planning, S-curve velocity planning, or curvature-based adaptive velocity planning can be employed.

[0063] Specifically, the above steps acquire and output the final executable laser scribing program and end flange trajectory for the robot. For laser switch control, a pre-on distance can be set before the path start point, and a delayed off distance can be set after the path end point to ensure the quality of the line segment's beginning and end. Combining the above data and steps, a corresponding machining allowance scribing report is output. This invention can be used for automated measurement, allowance analysis, and robotic laser scribing of machining allowances for complex aerodynamic control panels, complex castings, forgings, formed parts, and other blank parts with free-form surfaces and local machining allowance requirements.

[0064] In summary, this invention can automatically convert the geometric scribing target obtained from machining allowance analysis into the end flange trajectory required for actual robot execution, reducing manual teaching and coordinate transformation errors, and improving the automation level, trajectory consistency, and machining guidance accuracy of laser scribing on complex aerospace air rudder panel blanks. This solves the problems of existing allowance analysis results not being able to directly generate robot end flange trajectories, insufficient fit between the scribing path and the actual blank surface, difficulty in controlling coordinate transformation errors of non-contact laser tools, and lack of kinematic and process constraint verification for robot scribing trajectories.

[0065] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for generating laser scribing paths for a robot driven by the allowance field of a blank, characterized in that, Includes the following steps: S1. Obtain the measured point cloud of the blank, the corresponding CAD nominal model, and the registration result between the measured point cloud and the CAD nominal model that satisfies the machining allowance constraint; S2. Based on the registration result, calculate the signed machining allowance of the measured point cloud relative to the CAD nominal model, and construct a machining allowance field associated with the blank surface or the CAD nominal model; S3. Using the machining allowance field as the process criterion, extract the target path to be drawn based on one or more of the following data: CAD design boundary, machining boundary, trimming boundary, hole boundary, rib boundary, target allowance threshold, allowance contour line, and manually specified features. S4. Project the target path to be scribed onto the measured point cloud of the blank and the reconstructed surface of the blank, and preprocess the projected path to obtain a geometric scribing path containing the coordinates of the path points, the path tangent vector and the surface normal vector. S5. Based on the path point coordinates, path tangent vector, surface normal vector, and preset laser incident angle and working distance, construct the desired pose of the laser tool coordinate system at each path point; S6. Obtain the transformation relationship between the calibrated laser tool coordinate system and the robot end flange coordinate system, and combine the transformation relationship between the workpiece coordinate system, CAD coordinate system and robot base coordinate system to convert the desired pose of the laser tool coordinate system into the robot end flange trajectory. S7. Perform robot kinematics and process constraint verification on the robot end flange trajectory, and adjust the path points, tool posture or movement speed according to the verification results to obtain the final robot end flange trajectory.

2. The method for generating laser scribing paths for a robot driven by the allowance field of a blank according to claim 1, characterized in that, In step S1, the registration result that satisfies the machining allowance constraint is obtained through the allowance constraint registration algorithm, or it is obtained by using the automatic registration result output by the detection software as the initial value, and then performing signed machining allowance calculation, machining allowance constraint verification and pose adjustment; the constraint conditions of the allowance constraint registration algorithm include one or more of the following: no negative allowance constraint, blank encompassing CAD nominal model constraint, reference area fitting constraint and machining surface allowance uniformity constraint.

3. The method for generating laser scribing paths for a robot driven by the allowance field of a blank according to claim 2, characterized in that, The margin-constrained registration algorithm includes the following steps: S101. The measured point cloud data is preprocessed to obtain a rough point cloud containing three-dimensional coordinates and normal vectors; S102. Perform surface discretization and feature recognition on the CAD nominal model in sequence to obtain the CAD discrete model and the feature surface grouping corresponding to the surface entity; S103. Based on the design information of the CAD nominal model and / or the feature surface grouping, determine at least one machined surface region, at least one reference constraint region, and non-machined region, and set corresponding allowance constraint parameters for each machined surface region; S104. Based on one of the following methods: feature coarse registration, principal direction registration, automatic registration results from software, or corresponding features selected manually, obtain the initial rigid body transformation matrix between the blank point cloud and the CAD nominal model. S105. Based on the initial rigid body transformation matrix, calculate the signed machining allowance of the blank point cloud in each machining surface region relative to the CAD nominal model. The sign of the signed machining allowance is used to characterize whether the blank point cloud is located outside the CAD nominal model. S106. Establish a registration optimization model that simultaneously includes machining allowance constraints, datum constraints, and allowance uniformity constraints. The machining allowance constraints include a negative allowance constraint and / or a constraint that the blank completely encompasses the CAD nominal model, calculated from the signed machining allowance and allowance constraint parameters. The datum constraint is used to limit the positional error of the datum surface, datum hole, datum axis, or machined area relative to the CAD nominal model. The allowance uniformity constraint is used to limit the fluctuation of the machining allowance within the machined surface area, calculated based on the allowance constraint parameters. S107. Using the initial rigid body transformation matrix as the initial value, solve the registration optimization model to obtain the final rigid body transformation matrix that satisfies the machining allowance constraint, the reference constraint, and the allowance uniformity constraint.

4. The method for generating laser scribing paths for a robot driven by the allowance field of a blank according to claim 1, characterized in that, In step S3, the extraction method of the target path to be drawn includes: extracting the fill contour lines corresponding to a preset fill threshold in the processing fill field; Alternatively, design edges, trimming boundaries, hole boundaries, rib boundaries, or machining area boundaries can be extracted from the CAD nominal model to form candidate paths in the CAD coordinate system, and the candidate paths can be converted to the blank coordinate system or the workpiece coordinate system according to the registration results. Alternatively, based on the topological relationship between the negative margin region, the low margin region, and the target processing boundary, a margin indicator line can be generated to indicate the range to be removed in subsequent processing. The target path to be drawn is represented in one or more forms, such as a point sequence, a curve segment, a polyline segment, a spline curve, or a CAD curve entity.

5. The method for generating laser scribing paths for a robot driven by the allowance field of a blank according to claim 1, characterized in that, In step S4, the preprocessing includes sorting, smoothing, breakpoint repair, and resampling; The sorting includes: sorting discrete path points based on proximity distance, curve parameters, CAD boundary topology, or path direction; The smoothing includes: smoothing the path using moving least squares, spline fitting, B-spline smoothing, circular arc fitting, or polyline simplification methods; The breakpoint repair includes: identifying breakpoints based on the maximum breakpoint distance, normal change angle, curvature change, or path topology, and interpolating, bridging, or segmenting the breakpoints for output. The resampling includes resampling the path according to preset point spacing, curvature adaptive step size, or robot speed planning requirements.

6. The method for generating laser scribing path for a robot driven by the allowance field of a blank according to claim 1, characterized in that, Step S5 includes: the process of constructing the desired pose of the laser tool coordinate system includes: The coordinates of the path points are used as the positions of the laser action points; The surface normal vector or the direction corrected for the incident angle is taken as the output axis direction of the laser tool coordinate system; The path tangent vector is used as the line, the direction of travel, or the lateral reference direction of the laser tool coordinate system; A laser tool coordinate system rotation matrix satisfying the right-hand coordinate system constraint is constructed by cross product and orthogonalization. The laser's point of action is compensated along the output axis based on the laser's focal length, defocusing amount, or the rangefinder's working distance.

7. The method for generating laser scribing paths for a robot driven by the allowance field of a blank according to claim 1, characterized in that, In step S6, the trajectory of the robot's end flange is calculated according to the following coordinate chain: in, This represents the desired transformation from the robot's base coordinate system B to the robot's end effector flange coordinate system E. This represents the transformation from the robot's base coordinate system B to the workpiece coordinate system P. This represents the transformation from the workpiece coordinate system P to the CAD coordinate system C. This represents the transformation from the CAD coordinate system C to the laser tool coordinate system O, indicating the desired pose. This represents the calibration transformation from the robot end-effector coordinate system E to the laser tool coordinate system O.

8. The method for generating laser scribing path for a robot driven by the allowance field of a blank according to claim 1, characterized in that, In step S7, the kinematic and process constraint verification specifically includes: Solve the joint angles corresponding to each path point based on the robot's inverse kinematics; Determine whether each joint angle exceeds the joint limit, and determine whether the change in joint angle between adjacent path points exceeds a preset threshold; Determine if the robot is approaching a singular configuration or experiencing a sudden change in posture; Determine if there is a risk of collision between the robot body, the end effector laser, the tooling fixture, the turntable, and the blank to be scribed; Determine whether the laser incident angle, working distance, scribing speed, and laser process parameters meet the preset process window.

9. The method for generating laser scribing path for a robot driven by the allowance field of a blank according to any one of claims 1-8, characterized in that, Also includes: Based on the final robot end flange trajectory, the system outputs a robot-executable laser scribing program, end flange trajectory, laser switch control signal, and corresponding machining allowance scribing report. The robot-executable laser scribing program includes one or more of the following: robot motion commands, end flange pose sequence, joint angle sequence, linear velocity, acceleration, laser power, laser switch, pulse frequency, defocusing amount, working distance, and path segment identifier.

10. A system for generating laser scribing paths for a robot driven by a blank's allowance field, characterized in that, The method for generating laser scribing paths for a robot driven by the allowance field of a blank as described in any one of claims 1-9.

Images