Method, device, medium and equipment for three-dimensional reconstruction of aircraft radome machining blank

Abstract

The present invention relates to a three-dimensional reconstruction method, device, medium and equipment for aircraft radome processing blanks, including the steps of measuring scene arrangement, planning measurement path, automatic measurement, point cloud preprocessing and surface reconstruction. By adopting the linkage mode of rotary tooling and measuring robot, it can realize the automatic full coverage measurement of the surface of large composite material blanks, solve the problem of unstable quality of surface data of large blank workpieces obtained by traditional manual measurement, and the measurement process adopts a global positioning camera to capture the spatial coordinate information of the marking point. The scanner splices the acquired multi-frame point cloud data according to the spatial position of the marking point during scanning, thus avoiding the arrangement of marking points on the surface of the blank workpiece and ensuring the molding accuracy of the surface of the blank workpiece. When processing point cloud data, the amount of point cloud data is simplified while retaining the surface feature information of the blank through a non-uniform simplification algorithm, thereby improving the speed of subsequent point cloud data processing and the efficiency of model reconstruction.

Description

Technical Field

[0001] The invention relates to the technical field of mechanical reverse engineering, and in particular to a method, device, medium and equipment for three-dimensional reconstruction of a blank processed by an aircraft radome. Background Art

[0002] With the development of my country's industrial technology and military equipment, there are many precision thin-walled parts in the aerospace field, which have complex structures, large surface curvatures, and high molding precision requirements. Radar radome is a composite component with integrated structure / aerodynamic / wave-transmitting / stealth functions, and has the typical characteristics of large and complex aviation composite materials. During the manufacturing process of radar radome, multiple grinding operations are required, and the grinding process needs to strictly control the grinding surface removal accuracy and grinding force, which not only meets the grinding surface accuracy of the solid core area of ​​the composite material, but also avoids the grinding damage of the ultra-thin skin of the honeycomb sandwich structure. On the other hand, the metal functional structure on the surface of the radar radome needs to be actively avoided during the grinding process. However, the traditional grinding method of this large radar cover mostly adopts traditional manual grinding, which has low grinding efficiency and great dust hazards. Therefore, it is necessary to formulate a digital grinding plan for this type of equipment. The three-dimensional model reconstruction of the surface of the blank workpiece is the primary link of the grinding plan. The quality of the three-dimensional reconstruction will directly affect the final molding accuracy of the blank workpiece. The technical problems of three-dimensional reconstruction of large composite blank workpieces need to be solved urgently.

[0003] At present, there is little research on the 3D reconstruction technology of large components such as aircraft radomes. As for the measurement data collection of rough workpieces, manual scanning is basically adopted. This method has a low level of automation and poor data quality stability, which cannot meet the requirements of my country's digital and intelligent transformation from a manufacturing power to a manufacturing power. In addition, when traditional point cloud data processing methods are used for the reconstruction of large-sized rough workpieces, this application in the field of mechanical processing has extremely high requirements for the accuracy of the reconstructed model. The huge amount of data brought about by this will take a lot of time in the subsequent reconstruction processing, affecting production efficiency. Therefore, it is necessary to propose a 3D reconstruction method for rough workpiece models that meets the requirements of large measurement scale, small data quality fluctuation, fast point cloud processing speed, and high reconstruction accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide a method, device, medium and equipment for three-dimensional reconstruction of aircraft radome processing blanks, so as to solve the problems of low automation level, large fluctuation of collected data and slow data processing speed in the prior art for the reconstruction of large composite material blank workpiece models.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A three-dimensional reconstruction method for an aircraft radome machining blank comprises the following steps:

[0007] S1. Arrange the measurement scene, including the arrangement of the rough workpiece, the rotary tooling, the measuring robot, the 3D scanner and the global positioning camera. The rotary tooling is provided with marking points, and the 3D scanner is provided with high reflective points.

[0008] S2. Partition the surface of the rough workpiece design model according to the plan, and plan the measurement path of the partitioned surface in a circular order to achieve full coverage of the measurement path of the rough workpiece surface;

[0009] S3, running the rotary tooling, using the 3D scanner to automatically scan and measure the area to be processed of the rough workpiece, and obtaining the 3D point cloud data of the rough workpiece; turning on the global positioning camera to monitor the marking points on the rotary tooling and the high reflective points on the 3D scanner, obtaining the spatial position relationship between the 3D scanner and the rough workpiece in real time, splicing the obtained 3D point cloud data according to the spatial position of the marking points, and generating edge feature points;

[0010] S4, filtering and simplifying the point cloud data of the rough workpiece;

[0011] S5. Based on the edge feature points and three-dimensional point cloud data, curve and surface fitting is performed to reconstruct the three-dimensional model of the blank workpiece.

[0012] Furthermore, in step S1:

[0013] The rough workpiece is a large composite panel formed by coating glass fiber reinforced material on the surface of a complex curved mold and curing it;

[0014] The rotary tooling consists of a rotary table and a positioning fixture. The rotary table is used to provide rotary motion and cooperate with the measuring robot to complete the measurement of all areas to be processed. Marking points are arranged on the lower edge of the rotary table. The marking points are used as corresponding points for registration and splicing of multi-frame scanning data during scanning.

[0015] The positioning clamp is used to fix the blank workpiece, and fix different blank workpieces at the specified position according to the installation reference, so that the blank workpiece can be clamped on the rotary fixture in a hoisting manner;

[0016] The measuring robot is a six-axis robot installed next to the rotary table. It is used to clamp the 3D scanner to scan and collect the rough workpiece according to the partition planning path, and the rotating axis of the rotary table is used as the seventh axis of the measuring robot to cooperate with the measuring robot for linkage.

[0017] The three-dimensional scanner is a scanning instrument with high reflective points, which is clamped at the execution end of the measuring robot and is used to obtain the spatial coordinate information of the blank workpiece using the laser triangulation principle. The three-dimensional scanner is marked with N high reflective marking points, and N ≥ 32;

[0018] The global positioning camera is hoisted on the side of the rotary tooling, and the acquisition direction of the global positioning camera is facing the blank workpiece. It is used to capture the spatial coordinate information of the marking points, monitor the marking points on the rotary worktable and the high-reflective points on the 3D scanner, obtain the spatial position relationship between the 3D scanner and the blank workpiece in real time, and complete the stitching of multiple frame point clouds during the scanning process.

[0019] Furthermore, in step S1, the spatial positioning method of the blank workpiece includes: using a crane to place the combination of the blank workpiece and the complex curved mold on the table top of the rotary table, adjusting the mold position by means of a positioning block, fixing the mold on the rotary table by means of a positioning clamp, and different molds have different clamping positions, and using radial T-slots reserved on the table top of the rotary table and the mounting surface of the blank workpiece, clamping the blank workpiece in the T-slot, and positioning it by means of reference blocks in the X and Y directions, so that the blank workpiece coincides with the rotation center of the rotary table, and then fixing the blank workpiece on the table top of the rotary table by means of T-screws and a pressure plate, thereby completing the positioning of the blank workpiece.

[0020] Further, in step S2, the path planning measurement method includes: when performing measurement trajectory planning, the surface measurement area needs to be divided, and sector measurement is adopted according to the characteristics of the specific blank workpiece. After the current sector measurement is completed, the blank workpiece is rotated to perform the next sector measurement until the entire blank workpiece is measured.

[0021] Further, in step S3, the scanning measurement method includes:

[0022] Before scanning, a bottom plane is set so that only data points above the bottom plane are retained when the measured data is displayed in real time. The laser line array projected onto the scanned sample is obtained through two sets of cameras set on the 3D scanner. The line array deforms with the shape of the measured object, and the linear 3D point cloud data projected by the laser line array can be obtained by calculation.

[0023] At the same time, the global positioning camera tracks the high-reflective points on the 3D scanner in real time, and obtains the relative information of the scanner's spatial position in real time. After the rotary table rotates, the relative position of the blank workpiece will shift, and the automatic alignment and splicing of the two sets of scanned parts can be achieved through the marking points.

[0024] Furthermore, the implementation process of step S4 includes:

[0025] a. Point cloud denoising: Use statistical filtering method to remove the noise points generated during the scanning process to ensure the quality of subsequent model reconstruction;

[0026] b. Extract feature points. The average distance d from the point cloud to the tangent plane within the local neighborhood is expressed as

[0027]

[0028] Wherein, p is the characteristic point cloud; k is the number of point clouds within the local neighborhood range of point p; p j is the point cloud within the local neighborhood range of point p; j is any value from 1 to k; n p is the average normal vector at point p;

[0029] If d is greater than the threshold, then this point is considered a characteristic point;

[0030] c. Determine the optimal neighborhood, and determine the lower limit r min and the upper limit r max and the change step r △ , let the search radius for the optimal neighborhood be R, and successively let R = r min + r △ , according to the search radius, calculate the corresponding dimensional features and entropy function values for the point clouds within the local neighborhood respectively until R > r max , finally, select the minimum entropy function value through comparison, and use the neighborhood range obtained from the result as the optimal neighborhood;

[0031] d. Determine the weight, and calculate the weight w i of the local optimal neighborhood of non-characteristic points based on the local neighborhood density, and the calculation formula is

[0032]

[0033] Wherein, p i is the characteristic point cloud; k is p i the number of point clouds within the local neighborhood range of the point; p j is the point cloud within the local neighborhood range of point p; j is any value from 1 to k;

[0034] e. Determine the optimal point. After determining the optimal neighborhood radius R, let x be the optimal point within the optimal neighborhood R, and c be the center point of the point cloud of the selected non-characteristic local neighborhood range. Calculate whether the optimal point is within the optimal neighborhood range, that is, whether it meets the condition ‖x - c‖ < R. If x is within the optimal neighborhood range, use the optimal point x to replace the point cloud within the entire neighborhood range to obtain the thinned point cloud data.

[0035] Furthermore, in step S5, the steps of model three-dimensional reconstruction include:

[0036] a. Based on the edge feature points and three-dimensional coordinate data, select non-uniform rational B-spline curves;

[0037] b. Select non-uniform rational B-spline curve functions and perform curve and surface fitting;

[0038] c. Generate a three-dimensional model of the rough part from the fitted curve and surface.

[0039] Based on the above-mentioned three-dimensional reconstruction method for aircraft radome processing blank, the present invention also provides a model building device of the method, including:

[0040] The first processing unit is used to partition the surface of the rough workpiece design model, plan the measurement path of the partitioned surface in a circular order, and complete the full coverage of the measurement path of the rough workpiece surface;

[0041] The second processing unit is used to control the operation of the rotary tooling and control the 3D scanner to automatically scan and measure the area to be processed of the blank workpiece, turn on the global positioning camera, obtain the spatial position relationship between the 3D scanner and the blank workpiece in real time, and splice the obtained multi-frame point cloud data according to the spatial position of the marking point to generate edge feature points;

[0042] A third processing unit is used to filter and simplify the point cloud data of the rough workpiece;

[0043] The fourth processing unit is used to perform curve and surface fitting based on edge feature points and three-dimensional point cloud data to reconstruct a three-dimensional model of the blank workpiece.

[0044] Based on the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks, the present invention also provides a computer-readable storage medium of the method, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks are implemented.

[0045] Based on the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks, the present invention also provides a computer device for the method, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks when executing the computer program.

[0046] The present invention has the following advantages due to the adoption of the above technical solution:

[0047] 1. By adopting the linkage mode of rotary tooling and measuring robot, it is possible to realize automatic full coverage measurement of the surface of large composite blanks, solving the problem of unstable quality of surface data of large blanks obtained by traditional manual measurement;

[0048] 2. The measurement process uses a global positioning camera to capture the spatial coordinate information of the marking points. The scanner will splice the multi-frame point cloud data acquired according to the spatial position of the marking points during scanning, thus avoiding the arrangement of marking points on the surface of the rough workpiece and ensuring the forming accuracy of the rough workpiece surface;

[0049] 3. When processing point cloud data, the non-uniform simplification algorithm is used to retain the surface feature information of the blank, thereby streamlining the amount of point cloud data and improving the speed of subsequent point cloud data processing and model reconstruction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] By reading the detailed description of the preferred embodiment below, various other advantages and benefits will become clear to those of ordinary skill in the art. The accompanying drawings are only used to illustrate the purpose of the preferred embodiment and are not considered to be limitations of the present invention. In the entire drawings, the same reference numerals are used to represent the same components.

[0051] In the attached picture:

[0052] Figure 1 It is a flow chart of a method for three-dimensional reconstruction of an aircraft radome machining blank provided by one embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of the measurement scene arrangement of a method for three-dimensional reconstruction of an aircraft radome processing blank provided by this embodiment of the present invention;

[0054] Figure 3 It is a schematic diagram of the profile partition measurement path planning of a three-dimensional reconstruction method for a blank part of an aircraft radome machining provided by this embodiment of the present invention;

[0055] Figure 4 It is a flow chart of point cloud data preprocessing of a method for three-dimensional reconstruction of an aircraft radome processing blank provided by this embodiment of the present invention.

[0056] The symbols in the accompanying drawings are as follows:

[0057] 1-Blank workpiece; 2-Rotary tooling; 3-Measuring robot; 4-3D scanner; 5-Global positioning camera. DETAILED DESCRIPTION

[0058] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0059] Due to the low level of automation in the reconstruction of large composite blank workpiece models, large fluctuations in collected data, and slow data processing speed in the prior art, the present invention can achieve automatic full coverage measurement of the surface of large composite blanks through the steps of measurement scene arrangement-measurement path planning-automatic measurement-point cloud preprocessing-surface reconstruction, and adopts the linkage mode of rotary tooling and measurement robot, so as to solve the problem of unstable quality of surface data of large blank workpieces obtained by traditional manual measurement, and adopts a global positioning camera to capture the spatial coordinate information of the marking points during the measurement process. The scanner splices the acquired multi-frame point cloud data according to the spatial position of the marking points during scanning, thus avoiding the arrangement of marking points on the surface of the blank workpiece, ensuring the molding accuracy of the surface of the blank workpiece, and at the same time, when processing the point cloud data, the non-uniform simplification algorithm is used to retain the surface feature information of the blank, thereby streamlining the amount of point cloud data, and improving the speed of subsequent point cloud data processing and model reconstruction efficiency.

[0060] The scheme of the present invention is described in detail below through examples.

[0061] Example

[0062] like Figure 1 As shown, a method for three-dimensional reconstruction of an aircraft radome machining blank provided by an embodiment of the present invention comprises the following steps:

[0063] S1. Measurement scene layout:

[0064] Arrange the measurement scene, including the arrangement of the blank workpiece 1, the rotary tool 2, the measuring robot 3, the 3D scanner 4 and the global positioning camera 5. The rotary tool 2 is arranged with a marking point 21 (reference Figure 2 ).

[0065] In this step, the blank workpiece 1 is a large composite material panel formed by coating glass fiber reinforced material on the surface of a complex curved mold and curing. The blank workpiece height is ≥ 2m, which is a large composite material blank workpiece. The blank workpiece materials involved include but are not limited to glass fiber composite materials, ceramics, polycarbonate, polyolefin fiber braiding and other materials used in the radome preparation solution.

[0066] The rotary tooling 2 is composed of a rotary table and a positioning fixture. The rotary table is used to provide rotary motion and cooperate with the measuring robot 3 to complete the measurement of all areas to be processed. Marking points 21 are arranged on the lower edge of the rotary table. The marking points 21 are used as corresponding points for the registration and splicing of multiple frames of scanning data during scanning. By not arranging the marking points 21 on the surface of the blank workpiece 1 during measurement, the marking points 21 are arranged on the mold below the blank workpiece, thereby avoiding the arrangement of the marking points 21 on the surface of the blank workpiece, which affects the reconstruction accuracy of the blank workpiece.

[0067] The positioning clamp is used to fix the blank workpiece, and fix different blank workpieces at designated positions according to the installation reference, so that the blank workpiece 1 is clamped on the rotary fixture 2 in a hoisting manner. Among them, the spatial positioning method of the blank workpiece 1 includes: using a crane to place the combination of the blank workpiece 1 and the complex curved mold on the table of the rotary table, adjusting the mold position by the positioning block, and using the positioning clamp to fix the mold on the rotary table, and different molds have different clamping positions, and using the rotary table table and the blank workpiece 1 mounting surface to reserve radial T-slots, which facilitates the installation of the blank workpiece 1 while ensuring that the table can adapt to the clamping of blank workpieces 1 of different shapes and sizes. The blank workpiece 1 is clamped in the T-slot, and positioned by the reference blocks in the X and Y directions, so that the blank workpiece 1 coincides with the rotation center of the rotary table, and then the blank workpiece 1 is fixed on the table of the rotary table using T-screws and pressure plates to complete the positioning of the blank workpiece.

[0068] The measuring robot 3 is a six-axis robot installed next to the rotary table, which is used to clamp the 3D scanner 4 to scan and collect the rough workpiece according to the partition planning path, and the rotary table axis is used as the seventh axis of the measuring robot 3 to cooperate with the measuring robot 3. In this way, when scanning, the rotary table axis is used as the seventh axis of the industrial robot to cooperate with the robot, which can realize the automatic scanning of the surface of large rough workpieces.

[0069] The three-dimensional scanner 4 is a scanning instrument with high reflective points, which is clamped at the execution end of the measuring robot 3 and is used to obtain the spatial coordinate information of the blank workpiece using the principle of laser triangulation. The three-dimensional scanner 4 is marked with N high reflective marking points, and N ≥ 32, so that no matter what position the scanner is in during scanning and measurement, the global positioning camera 5 can obtain the relative spatial position of the scanner.

[0070] The global positioning camera 5 is hoisted beside the rotary tooling, and the acquisition direction of the global positioning camera 5 is toward the blank workpiece, which is used to capture the spatial coordinate information of the marking point 21, and monitor the marking point 21 on the rotary worktable and the high-reflective points on the three-dimensional scanner 4, and obtain the spatial position relationship between the three-dimensional scanner 4 and the blank workpiece in real time, and complete the splicing of multiple frame point clouds during the scanning process.

[0071] S2. Plan the measurement path:

[0072] According to the design model surface of the blank workpiece, the measurement path is planned for the partitioned surface in circular order to achieve full coverage of the measurement path of the blank workpiece surface (reference Figure 3 ).

[0073] In this step, the path planning measurement method includes: when performing measurement trajectory planning, the surface measurement area needs to be divided, and sector measurement is adopted according to the characteristics of the specific blank workpiece. After the current sector measurement is completed, the blank workpiece is rotated to perform the next sector measurement until the entire blank workpiece is measured.

[0074] Due to the large size and complex surface shape of special-shaped composite materials, the surface measurement area needs to be divided when planning the measurement trajectory. After the entire blank workpiece is measured once, it is rotated a certain angle and processed again. This method can avoid the accumulation of point cloud data caused by the same splicing position of the tool each time, and reduce the amount of point cloud processing calculations.

[0075] S3. Automated measurement:

[0076] The rotary tooling 2 is operated, and the 3D scanner 4 automatically scans and measures the area to be processed of the rough workpiece 1 to obtain the 3D point cloud data of the rough workpiece 1; the global positioning camera 5 is turned on to monitor the marking point 21 on the rotary tooling 2 and the high reflective point on the 3D scanner 4, and the spatial position relationship between the 3D scanner 4 and the rough workpiece 1 is obtained in real time, and the obtained 3D point cloud data is spliced ​​according to the spatial position of the marking point 21 to generate edge feature points;

[0077] In this step, in order to reduce the stitching calculation pressure of the measurement software, the bottom plane is set before scanning, so that when the measurement data is displayed in real time and dynamically, only the data points above the bottom plane are retained, and then the global positioning camera 5 and the scanner system are turned on, and the control program of the rotary table and the measurement robot 3 is run. The measurement robot 3 drives the 3D scanner at the execution end to scan and measure the area to be processed in cooperation with the rotary table.

[0078] During the measurement process, two groups of cameras arranged on the three-dimensional scanner 4 respectively obtain the laser line array projected onto the scanned sample. The line array deforms with the shape of the measured object, and the linear three-dimensional point cloud data collection projected by the laser line array can be obtained by calculation;

[0079] At the same time, the global positioning camera 5 tracks the high-reflective points on the three-dimensional scanner 4 in real time, and obtains the relative information of the scanner's spatial position in real time. After the rotary table rotates, the relative position of the blank workpiece will shift, and the automatic alignment and splicing function of the two sets of scanned parts can be realized through the marking point 21.

[0080] S4. Point cloud preprocessing:

[0081] The statistical filtering method is used to remove the error point cloud measured in step S3, and the point cloud data is simplified based on the non-uniform simplification algorithm (refer to Figure 4 ).

[0082] In this step, point cloud simplification methods include:

[0083] a. Point cloud denoising: The statistical filtering method is used to remove the noise points generated during the scanning process to ensure the quality of subsequent model reconstruction;

[0084] b. Extract feature points: According to the average distance d from the point cloud within the local neighborhood range to the tangent plane, the expression is

[0085]

[0086] In the formula, p is the feature point cloud; k is the number of point clouds within the local neighborhood range of point p; p j is the point cloud within the local neighborhood range of point p; j is any value from 1 to k; n p is the average normal vector at point p;

[0087] If d is greater than the threshold, then this point is considered a feature point;

[0088] c. Determine the optimal neighborhood: Determine the lower limit r min and the upper limit r max of the neighborhood search radius and the change step r △ , let the search radius of the optimal neighborhood be R, and successively let R = r min +r △ , according to the search radius, calculate the corresponding dimensional features and entropy function values for the point clouds within the local neighborhood respectively until R > r max , finally, select the minimum entropy function value through comparison, and use the neighborhood range obtained from the result as the optimal neighborhood;

[0089] d. Determine the weight: Calculate the weight w i of the local optimal neighborhood of non-feature points based on the local neighborhood density, and the calculation formula is

[0090]

[0091] In the formula, p i is the feature point cloud; k is the number of point clouds within the local neighborhood range of point p i ; p j is the point cloud within the local neighborhood range of point p; j is any value from 1 to k;

[0092] e. Determine the optimal point: After determining the optimal neighborhood radius R, let x be the optimal point within the optimal neighborhood R, and c be the center point of the point cloud within the selected non-feature local neighborhood range. Calculate whether the optimal point is within the optimal neighborhood range, that is, whether it meets the condition ‖x - c‖ < R. If x is within the optimal neighborhood range, use the optimal point x to replace the point cloud within the entire neighborhood range to obtain the thinned point cloud data.

[0093] S5. Surface reconstruction:

[0094] Based on edge feature points and 3D point cloud data, curve and surface fitting is performed to reconstruct the 3D model of the blank workpiece.

[0095] In this step, the steps of three-dimensional reconstruction of the model include:

[0096] a. Based on edge feature points and three-dimensional coordinate data, select non-uniform rational B-spline curve;

[0097] b. Select non-uniform rational B-spline curve function to perform curve and surface fitting;

[0098] c. Generate a three-dimensional model of the blank by fitting curves and surfaces.

[0099] It can be seen that the above technical solution can realize the automatic full coverage measurement of the surface of large composite blanks by adopting the linkage between the rotary fixture 2 and the measuring robot 3, and solve the problem of unstable quality of surface data of large blanks obtained by traditional manual measurement. In addition, the measurement process uses the global positioning camera 5 to capture the spatial coordinate information of the marking point. The scanner will splice the acquired multi-frame point cloud data according to the spatial position of the marking point 21 during scanning, thus avoiding the arrangement of the marking point 21 on the surface of the blank workpiece and ensuring the molding accuracy of the surface of the blank workpiece. When processing the point cloud data, the non-uniform simplification algorithm is used to reduce the amount of point cloud data while retaining the surface feature information of the blank, thereby improving the speed of subsequent point cloud data processing and the efficiency of model reconstruction.

[0100] Based on the above-mentioned three-dimensional reconstruction method for aircraft radome processing blank, the present invention also provides a model building device of the method, including:

[0101] The first processing unit is used to partition the surface of the rough workpiece design model, and plan the measurement path of the partitioned surface in a circular order to achieve full coverage of the measurement path of the rough workpiece surface;

[0102] The second processing unit is used to control the operation of the rotary tooling 2, and control the three-dimensional scanner 4 to automatically scan and measure the to-be-processed area of ​​the blank workpiece 1, turn on the global positioning camera 5, obtain the spatial position relationship between the three-dimensional scanner 4 and the blank workpiece 1 in real time, and splice the obtained multi-frame point cloud data according to the spatial position of the marking point 21 to generate edge feature points;

[0103] A third processing unit is used to filter and simplify the point cloud data of the blank workpiece 1;

[0104] The fourth processing unit is used to perform curve and surface fitting based on edge feature points and three-dimensional point cloud data to reconstruct a three-dimensional model of the blank workpiece.

[0105] Based on the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks, the present invention also provides a computer-readable storage medium of the method, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks are implemented.

[0106] Based on the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks, the present invention also provides a computer device for the method, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for three-dimensional reconstruction of aircraft radome processing blanks when executing the computer program.

[0107] The present invention is described in terms of flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to specific embodiments. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional reconstruction method for aircraft radome blanks, characterized in that: The steps include: S1. Arrange the measurement scene, including the arrangement of the rough workpiece, the rotary tooling, the measuring robot, the 3D scanner and the global positioning camera. The rotary tooling is provided with marking points, and the 3D scanner is provided with high reflective points. S2. Partition the surface of the rough workpiece design model according to the plan, and plan the measurement path of the partitioned surface in a circular order to achieve full coverage of the measurement path of the rough workpiece surface; S3, running the rotary tooling, using a 3D scanner to automatically scan and measure the area to be processed of the rough workpiece, and obtaining 3D point cloud data of the rough workpiece; The global positioning camera is turned on to monitor the marking points on the rotary tooling and the high-reflective points on the 3D scanner, and the spatial position relationship between the 3D scanner and the blank workpiece is obtained in real time. The obtained 3D point cloud data is spliced ​​according to the spatial position of the marking points to generate edge feature points; S4, filtering and simplifying the point cloud data of the rough workpiece; S5. Based on the edge feature points and the three-dimensional point cloud data, curve and surface fitting is performed to reconstruct the three-dimensional model of the blank workpiece; In step S3, the scanning measurement method includes: Before scanning, a bottom plane is set so that only data points above the bottom plane are retained when the measured data is displayed in real time. The laser line array projected onto the scanned sample is obtained through two sets of cameras set on the 3D scanner. The line array deforms with the shape of the measured object, and the linear 3D point cloud data projected by the laser line array can be obtained by calculation. At the same time, the global positioning camera tracks the high reflective points on the 3D scanner in real time, and obtains the relative information of the scanner's spatial position in real time. After the rotary table rotates, the relative position of the blank workpiece will shift, and the automatic alignment and splicing function of the two sets of scanned parts can be realized through the marking points; The implementation process of step S4 includes: a. Point cloud denoising: Use statistical filtering method to remove the noise points generated during the scanning process to ensure the quality of subsequent model reconstruction; b. Extract feature points based on the average distance from the point cloud to the tangent plane within the local neighborhood The expression is Where p is the feature point cloud; is the number of point clouds within the local neighborhood of point p; j is the point cloud within the local neighborhood of point p; From 1 to Any value of n p is the average normal vector at point p; If d is greater than the threshold, the point is considered a feature point; c. Determine the optimal domain and the lower limit r of the domain search radius min and the upper limit r max And the change step length r △ , let the radius r of the optimal neighborhood search be R, and let R=r min +r △ , according to the search radius, the corresponding dimensional features and entropy function values ​​are calculated for the point cloud in the local neighborhood until R>r max ,Finally, the minimum entropy function value is selected by comparison, and the neighborhood range obtained by its result is taken as the optimal neighborhood; d. Determine the weight and calculate the weight of the local optimal neighborhood of the non-feature point based on the local neighborhood density , the calculation formula is In the formula, p i is the feature point cloud; For p i The number of point clouds within the local neighborhood of the point; p j is the point cloud within the local neighborhood of point p; From 1 to Any value of ; e. Determine the optimal point. After the optimal neighborhood radius R is determined, set is the optimal point in the optimal neighborhood R, For the center point of the selected non-feature local neighborhood point cloud, calculate whether the optimal point is within the optimal neighborhood, that is, whether it meets the conditions ,like Located in the optimal neighborhood, use the optimal point The point cloud within the entire neighborhood is replaced to obtain streamlined point cloud data.

2. The method for three-dimensional reconstruction of an aircraft radome blank according to claim 1, characterized in that: In step S1: The rough workpiece is a large composite panel formed by coating glass fiber reinforced material on the surface of a complex curved mold and curing it; The rotary tooling consists of a rotary table and a positioning fixture. The rotary table is used to provide rotary motion and cooperate with the measuring robot to complete the measurement of all areas to be processed. Marking points are arranged on the lower edge of the rotary table. The marking points are used as corresponding points for registration and splicing of multi-frame scanning data during scanning. The positioning clamp is used to fix the blank workpiece, and fix different blank workpieces at the specified position according to the installation reference, so that the blank workpiece can be clamped on the rotary fixture in a hoisting manner; The measuring robot is a six-axis robot installed next to the rotary table. It is used to clamp the 3D scanner to scan and collect the rough workpiece according to the partition planning path, and the rotating axis of the rotary table is used as the seventh axis of the measuring robot to cooperate with the measuring robot for linkage. The three-dimensional scanner is a scanning instrument with high reflective points, which is clamped at the execution end of the measuring robot and is used to obtain the spatial coordinate information of the blank workpiece using the laser triangulation principle. The three-dimensional scanner is marked with N high reflective marking points, and N ≥ 32; The global positioning camera is hoisted on the side of the rotary tooling, and the acquisition direction of the global positioning camera is facing the blank workpiece. It is used to capture the spatial coordinate information of the marking points, monitor the marking points on the rotary worktable and the high-reflective points on the 3D scanner, obtain the spatial position relationship between the 3D scanner and the blank workpiece in real time, and complete the stitching of multiple frame point clouds during the scanning process.

3. The method for three-dimensional reconstruction of an aircraft radome blank according to claim 2, characterized in that: In step S1, the spatial positioning method of the blank workpiece includes: using a crane to place the combination of the blank workpiece and the complex curved mold on the table of the rotary table, adjusting the position of the mold by a positioning block, using a positioning clamp to fix the mold on the rotary table, and different molds have different clamping positions, and using radial T-slots reserved on the table of the rotary table and the mounting surface of the blank workpiece, clamping the blank workpiece in the T-slot, and positioning it by reference blocks in the X and Y directions, so that the blank workpiece coincides with the rotation center of the rotary table, and then using T-screws and pressure plates to fix the blank workpiece on the table of the rotary table to complete the positioning of the blank workpiece.

4. The method for three-dimensional reconstruction of an aircraft radome blank according to claim 1, characterized in that: In step S2, the path planning measurement method includes: when performing measurement trajectory planning, the surface measurement area needs to be divided, and sector measurement is adopted according to the characteristics of the specific blank workpiece. After the current sector measurement is completed, the blank workpiece is rotated to perform the next sector measurement until the entire blank workpiece is measured.

5. The method for three-dimensional reconstruction of an aircraft radome blank according to claim 1, characterized in that: In step S5, the step of three-dimensional reconstruction of the model includes: a. Based on edge feature points and three-dimensional coordinate data, select non-uniform rational B-spline curve; b. Select non-uniform rational B-spline curve function to perform curve and surface fitting; c. Generate a three-dimensional model of the blank by fitting curves and surfaces.

6. A model building device for the three-dimensional reconstruction method of aircraft radome processing blank according to any one of claims 1 to 5, characterized in that: include: The first processing unit is used to partition the surface of the rough workpiece design model, plan the measurement path of the partitioned surface in a circular order, and complete the full coverage of the measurement path of the rough workpiece surface; The second processing unit is used to control the operation of the rotary tooling and control the 3D scanner to automatically scan and measure the area to be processed of the blank workpiece, turn on the global positioning camera, obtain the spatial position relationship between the 3D scanner and the blank workpiece in real time, and splice the obtained multi-frame point cloud data according to the spatial position of the marking point to generate edge feature points; A third processing unit is used to filter and simplify the point cloud data of the rough workpiece; The fourth processing unit is used to perform curve and surface fitting based on edge feature points and three-dimensional point cloud data to reconstruct a three-dimensional model of the blank workpiece.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for three-dimensional reconstruction of an aircraft radome processing blank as described in any one of claims 1 to 5 are implemented.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for three-dimensional reconstruction of an aircraft radome processing blank as described in any one of claims 1 to 5 are implemented.

Images