Digital measurement method for assembly clearance of assembly with large depth-to-width ratio

Abstract

The invention relates to a digital measurement method for an assembly clearance of a large depth-to-width ratio assembly, and the method comprises the following steps: scanning to obtain point cloud data P at a cold plate and an assembly in a TR assembly assembly region, preliminarily dividing the point cloud data into a TR assembly point cloud Pz and a cold plate point cloud P1 through clustering, extracting the features of a mounting hole and the features of a positioning plane from the point cloud Pz and the point cloud P1, and calculating the assembly clearance of the TR assembly according to the features of the mounting hole and the features of the positioning plane; and aligning the corresponding theoretical models of the TR assembly and the cold plate to the point cloud data according to the extracted hole features and the positioning plane features, calculating the gap distribution between the transceiving assembly and the cold plate according to the relative poses of the theoretical models, converting the gap distribution into scalar quantities, and displaying the scalar quantities on the models in a color difference diagram form. In a word, the theoretical model is used for recovery of the assembly clearance through an efficient feature extraction algorithm and a registration method, so that calculation of the complete assembly clearance in an invisible area is achieved, and clearance calculation precision and data processing efficiency are greatly improved.

Description

Technical Field

[0001] The invention belongs to the technical field of automatic assembly of radar antenna electronic products, and in particular relates to a digital measurement method for assembly gaps of components with large aspect ratios. Background Art

[0002] As modern electronic products develop towards multi-functions and high integration, the corresponding assembly processes have also put forward higher requirements in terms of assembly accuracy and reliability. In electronic products such as radar antennas, in order to reduce costs and make full use of space, the space for internal component assembly will be very compact. At the same time, during the assembly process, due to reasons such as part errors, assembly force and connector accuracy, assembly gaps will be generated between components. The generation of this gap will seriously hinder the heat dissipation efficiency of the TR component during operation and reduce the working life of the radar antenna.

[0003] Since the components are in a fitted state and the space is closed after assembly, they still need to be inspected for normal use. Traditional assembly gap measurement methods usually rely on manual contact measurement, such as feeler gauges or vernier calipers. However, these methods can only measure the edge gap of the assembly area and cannot obtain the overall gap distribution between components. At the same time, contact measurement is inefficient and may affect the assembly accuracy, making it impossible to make timely adjustments to the assembly process.

[0004] With the advancement of digital measurement technology, line laser measuring instruments can be used to digitally measure component assembly gaps. The assembly gaps can be analyzed by analyzing and processing the scanned point cloud data. However, due to the extremely large depth-to-width ratio of the gaps, some gaps that reach 40:1 are also called large depth-to-width ratio assembly gaps. Line laser measuring instruments are still unable to obtain measurement data inside these large depth-to-width ratio gaps, resulting in difficulty in digitally measuring the entire gap. Therefore, a digital measurement method for assembly gaps of components with large depth-to-width ratios is urgently needed, which can quickly, accurately and conveniently detect the data inside the large depth-to-width ratio gaps, thereby truly reflecting the actual conditions of these gaps. Summary of the invention

[0005] In view of this, the present invention proposes a digital measurement method for the assembly gap of components with a large aspect ratio, which is applied to the field of automation of radar antenna electronic products. It can solve the existing technical problems that only the edge of the gap can be measured and the internal measurement data between components with a large aspect ratio cannot be obtained.

[0006] In order to achieve the above technical objectives, the specific technical solutions adopted by the present invention are:

[0007] A digital measurement method for assembly gap of components with large aspect ratio, comprising the following steps:

[0008] S1, scan and obtain the point cloud data P of the cold plate and components in the TR component assembly area;

[0009] S2. Preliminarily split the point cloud data into TR component point clouds P by clustering z And cold plate point cloud P l ;

[0010] S3, respectively point cloud P z and P l Extract mounting hole features and positioning plane features;

[0011] S4, aligning the corresponding TR component and the cold plate theoretical model to the point cloud data according to the extracted hole features and positioning plane features;

[0012] S5. Calculate the gap distribution between the transceiver assembly and the cold plate according to the relative posture of the theoretical model;

[0013] S6, converting the gap distribution into a scalar and displaying it on the model in the form of a color difference diagram;

[0014] In step S3, for the cold plate and TR component obtained after segmentation, two positioning holes and one positioning plane are respectively extracted as subsequent registration features, and core data is used to participate in the calculation to improve the registration accuracy and algorithm efficiency;

[0015] In step S4, for the TR component point cloud data and the cold plate point cloud data obtained in step S3, the theoretical model is roughly aligned with the point cloud data through the extracted plane and hole features, and then finely aligned through the ICP algorithm.

[0016] Furthermore, in step S1, after the TR component is assembled to the antenna frame, the assembled TR component and the cold plate area are scanned by a line laser scanner mounted on a robotic arm. During the scanning, the scanner is located directly above the TR component and the cold plate, and the scanning distance is within the effective measurement distance range of the line laser scanner.

[0017] Furthermore, in step S1, the scanning path is determined according to the direction of the cold plate, and the scanning range covers all TR components and the exposed area of ​​the cold plate.

[0018] Furthermore, in step S2, the original point cloud data obtained by scanning is preprocessed to remove outliers in the point cloud to obtain the point cloud P c , the point cloud is segmented by using the radius search-based growing clustering strategy. c Segmentation into cold plate point cloud P l And component point cloud P z .

[0019] Furthermore, in step S3, for the TR component, the plane p is extracted from the TR component point cloud by the RANSAC algorithm. m; Then, the points within a certain threshold of the plane formed in the previous step are used as input point cloud data, and the boundary point set S in the point cloud data is extracted by the region growing method based on curvature e ; Then in the boundary point set S e The RANSAC method is used to extract S e The circular features in the figure are used as the features of the two holes h1 and h2. The structures of the segmented point cloud data of the cold plate and TR component are similar, and the feature extraction of the cold plate structure is no longer introduced in detail. Finally, the plane data corresponding to the cold plate structure and the corresponding two positioning hole data are obtained.

[0020] Furthermore, the coarse registration matrix of the TR component theoretical model obtained by the coarse registration calculation in step S4 is W1, the coarse registration matrix of the cold plate theoretical model is W1', the registration matrix of the TR component theoretical model obtained by the fine registration calculation is W2, and the fine registration matrix of the cold plate theoretical model is W2'. The transformations W2·W1 and W2'·W1' are applied to the TR component theoretical model and the cold plate theoretical model respectively, so that the TR component theoretical model and the cold plate theoretical model have a relative posture consistent with that in the actual assembly state.

[0021] Furthermore, in step S5, the fitting plane of the TR component and the cold plate theoretical model is downsampled into a point cloud L z and L l , with L l As a benchmark, calculate L z Each point to point cloud L l The closest distance is taken as the gap value at this point.

[0022] Furthermore, in step S6, the gap value is converted into a corresponding color through color mapping according to the gap range, and a cloud diagram of the gap distribution between the TR component and the cold plate is generated.

[0023] By adopting the above technical solution, the present invention can also bring the following beneficial effects:

[0024] 1. The present invention mentions a digital measurement method for the assembly gap of components with a large aspect ratio. Through an efficient feature extraction algorithm and a registration method, a theoretical model is used to complete and restore the gap data, thereby overcoming the problem of incomplete gap data caused by spatial occlusion and narrow viewing angle. The overall gap distribution in the gap area between components with a large aspect ratio can be obtained through the fused data after completion, thus breaking through the problem of unmeasurable internal gaps caused by missing data.

[0025] 2. The digital measurement method for assembly gaps of components with large aspect ratios mentioned in the present invention is very suitable for assembly gaps with large aspect ratios, especially when most of the gap data on the fitting surface cannot be obtained. The overall scanning and analysis process is fully automated by using a scanner mounted on a robotic arm, which greatly improves the degree of automation and gap detection efficiency. Not only is the accuracy greatly improved, but the scope of use of line laser scanner measurements can also be expanded, which has a very positive significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 It is a flow chart of a digital measurement method of assembly gap of a component with a large aspect ratio mentioned in the present invention;

[0028] Figure 2 It is a schematic diagram of the positional relationship between the linear laser scanner and the radar antenna in a specific embodiment of the present invention when in use;

[0029] Figure 3 It is a schematic diagram of the positional relationship between the TR component and the antenna cold plate in a specific embodiment of the present invention;

[0030] Figure 4 It is a distribution cloud diagram of the assembly gap of the TR assembly in a specific embodiment of the present invention;

[0031] Among them: 1. Robotic arm; 2. Line laser scanner; 3. Radar antenna; 4. TR component; 5. Cold plate. DETAILED DESCRIPTION

[0032] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0033] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0034] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present invention, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0035] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention. The drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0036] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0037] In one embodiment of the present invention, the present invention provides a digital measurement method for assembly gap of components with large aspect ratio, comprising the following steps:

[0038] Step S1, scanning and obtaining point cloud data of the TR component 4 and the cold plate 5;

[0039] Point cloud data acquisition, such as Figure 2 As shown, the point cloud data of the measured object TR component 4 and cold plate 5 are obtained by a robot arm 1 (KUKA KR20R1810) equipped with a line laser scanner 2 (LJ-X8000). The scanning path of the robot arm 1 is determined according to several parameters such as the direction of the cold plate 5, the width of the line laser, the effective scanning distance of the scanner and the direction of the radar antenna 3.

[0040] The running trajectory of the robot arm 1 can be planned according to the digital model corresponding to the radar antenna 3, and can also be determined according to the number and area of ​​the TR components 4 actually assembled on site.

[0041] Converting line laser 2D data into 3D point cloud data. Since the data directly measured by the line laser only contains the elevation data of the measured surface, in order to obtain the complete point cloud data of TR component 4 and cold plate 5, it is necessary to convert the 2D line laser data into 3D point cloud data.

[0042] In this embodiment, the line laser scanner 2 acquires the surface profile of the object to be measured at a fixed sampling frequency h, and the running speed of the robot 1 is v. Therefore, the scanning resolution along the path of the robot 1 is v / h. Assume that the path of the robot 1 is consistent with the path of the cold plate 5 and is a zigzag type route parallel to the y-axis. The x coordinate of each point in each frame of the line laser has been calibrated. Assume that the position of the robot 1 at time t0 is (x0, y0), then the y-direction position of the point cloud data acquired after time t is (x0, y0+v*t). According to the above formula, the elevation data of each frame of the line laser can be fused with the current plane coordinate value to form three-dimensional point cloud data.

[0043] Step S2, clustering and segmenting the point cloud data into TR component 4 point cloud and cold plate 5 point cloud;

[0044] After obtaining the overall point cloud data, it is necessary to segment the TR component 4 and the cold plate 5 data. In this embodiment, since the cold plate 5 and the TR component 4 have a regular structure with a clear boundary, the point cloud is segmented using a region growing method based on geometric continuity, as follows:

[0045] Randomly select seed points in the point cloud, use the search radius r as the threshold to search for points in the adjacent space, and include them in the same area. Then, use these newly added points as new seed points and repeat the above process until there are no more points that can be added to the area.

[0046] When all points are assigned to a certain area, the segmentation of the point cloud data is completed, and finally the point cloud data p of TR component 4 are obtained. z And the point cloud data of cold plate 5 l .

[0047] The search radius r of the seed point needs to be larger than the scanning resolution v / h of the line laser but smaller than the minimum empirical value of the assembly gap, so as to avoid the failure of point cloud data segmentation caused by the threshold being lower than the scanning resolution or larger than the assembly gap.

[0048] Step S3, extracting positioning holes and plane features from the point cloud data of the TR component 4 and the cold plate 5 respectively;

[0049] The structure of the segmented point cloud data of the cold plate 5 and the TR component 4 is similar, and the main structure is composed of the upper surface plane of the cold plate 5 and the TR component 4 and two positioning holes located in the plane.

[0050] The feature extraction is explained by taking TR component 4 as an example. First, the RANSAC (random sampling consensus) algorithm is used to extract the point cloud p of TR component 4. z Processing is performed to extract the plane features in the point cloud as the main plane p m .

[0051] Take the distance P m Points within a certain threshold range are taken as input point cloud P m ', and further extract the boundary features.

[0052] Since the input point clouds are all in the plane, the present invention uses the interior point method based on Euclidean distance to judge the boundary points. The specific steps are as follows:

[0053] For P m 'For any point m(x0,y0) in the inner space, set the neighborhood radius r, and search for all points within the neighborhood whose distance is less than 2r for each point to form a set O;

[0054] In the set O, select any point n(x1,y1) on the plane p m In the circle with radius r passing through points m and n, two circles are obtained, whose center is C 1 and C 2 The coordinate calculation formula is as follows:

[0055]

[0056]

[0057] in,

[0058]

[0059] L 2 =(x0-x1) 2 +(y0-y1) 2

[0060] For all points in set O except points m and n, calculate their distance to the two circle centers C 1 and C 2 distance.

[0061] If the distances of all points are greater than r, that is, they are outside the circle, then point m is determined to be a boundary point.

[0062] If there are points in set O whose distance is less than r, reselect another point as point n and repeat step 3. If there are no points in set O that satisfy step 4, point m is a non-boundary point.

[0063] Return to step 1 for p m 'Judge the other points in the process until all points are traversed.

[0064] After the boundary extraction operation, a point cloud consisting of the outer boundary of TR component 4 and the internal hole boundary is obtained. The RANSAC algorithm is used to extract the point clouds h1 and h2 at the positioning holes in the remaining point clouds using the hole diameter as the diameter of the circle to be fitted.

[0065] The feature point cloud data F extracted through the above steps p Including positioning plane data p m And two positioning hole data h1, h2.

[0066] For the segmented point cloud data of the cold plate 5 and the TR component 4, their structures are similar, and the feature extraction of the cold plate 5 structure will not be introduced in detail. Finally, the plane data corresponding to the cold plate 5 structure and the corresponding two positioning hole data are obtained.

[0067] Step S4, registering the theoretical models of the TR component 4 and the cold plate 5 to the point cloud data respectively through the positioning holes and the plane features;

[0068] According to the above steps, the point cloud data corresponding to the TR component 4 is aligned with the theoretical model using the plane and hole features obtained as constraints.

[0069] Since the structure of the cold plate 5 is basically the same as that of the TR component 4, the same method as above is used for data processing to obtain the corresponding positioning hole data and plane data. Using the obtained cold plate 5 plane and hole features as constraints, the point cloud data corresponding to the cold plate 5 is registered with the theoretical model.

[0070] Taking the registration of the point cloud and theoretical model of TR component 4 as an example, let the theoretical model of TR component 4 be Q and the actual point cloud data be p z , the characteristic data of the theoretical positioning plane and the two holes are Ω, and the corresponding point cloud data are Ψ.

[0071] According to the above data, the theoretical model is registered with the actual point cloud data. To improve the accuracy of data registration, this embodiment designs a precise registration method based on assembly positioning feature constraints and defines the following optimization function:

[0072]

[0073] Among them, q i ∈Q,p i ∈P z ,ω j ∈Ω,ψ j ∈Ψ and N q is the number of points on the theoretical model of TR component 4 involved in the registration, N f is the number of points on the theoretical model of the reference features (planes, holes) involved in the registration. At the same time, the threshold δ=0.005 is determined through actual experiments.

[0074] The above optimization problem can be transformed into a quadratic programming problem under bounded quadratic constraints, and finally the precisely aligned rotation matrix R and translation vector t are obtained.

[0075] The rotation matrix R and translation vector t are applied to the theoretical model of TR component 4 to make its posture the same as the posture of the parts in the actual assembly state, thereby realizing the recovery of the gap data in the unmeasurable area of ​​TR component 4.

[0076] Since the structure of the cold plate 5 is basically the same as that of the TR assembly 4, the same method as above is used for registration processing to obtain the rotation matrix and translation vector of precise registration. The rotation matrix and translation vector are applied to the theoretical model of the cold plate 5 to make its posture the same as the posture of the parts in the actual assembly state, so as to realize the recovery of the gap data of the unmeasurable area of ​​the cold plate 5.

[0077] Step S5, calculating the gap distribution between the TR assembly 4 and the cold plate 5:

[0078] Calculate the assembly gap between the TR component 4 and the cold plate 5 theoretical model;

[0079] First, the theoretical model is converted into point cloud data. Then, within a certain neighborhood range, the neighborhood radius is generally less than 3 mm. The shortest distance from each point in the TR component 4 to all points in the neighborhood range in the cold plate 5 point cloud is cyclically calculated as the gap value of the point.

[0080] Step S6, convert the gap value into a cloud map and display it on the model:

[0081] The gap values ​​of all points in the previous step are stored in the distance matrix M, the distance matrix M is normalized to the range of [0,1], and mapped to the color space to complete the generation of the color difference cloud map, such as Figure 4 As shown, the color of each point is determined by its gap value to another point cloud.

[0082] In summary, this invention uses the theoretical model to restore the assembly gap through an efficient feature extraction algorithm and a registration method, thereby realizing the calculation of the complete assembly gap in the invisible area and greatly improving the gap calculation accuracy and data processing efficiency.

[0083] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A digital measurement method for assembly gap of components with large aspect ratio, characterized in that: The following steps are involved: S1, scanning and obtaining the point cloud data P of the cold plate (5) and the components in the assembly area of ​​the TR component (4); S2. Preliminary splitting of point cloud data into TR components by clustering (4) Point cloud P z And cold plate (5) point cloud P l ; S3, respectively point cloud P z and P l Extract mounting hole features and positioning plane features; S4, aligning the corresponding theoretical models of the TR component (4) and the cold plate (5) to the point cloud data according to the extracted hole features and positioning plane features; S5, calculating the gap distribution between the transceiver assembly and the cold plate (5) according to the relative posture of the theoretical model; S6, converting the gap distribution into a scalar and displaying it on the model in the form of a color difference diagram; In the step S3, for the cold plate (5) and the TR component (4) obtained after segmentation, two positioning holes and one positioning plane are respectively extracted as subsequent registration features, and the core data is used to participate in the calculation to improve the registration accuracy and algorithm efficiency; In step S4, for the point cloud data of the TR component (4) and the point cloud data of the cold plate (5) obtained in step S3, the theoretical model is roughly aligned with the point cloud data through the extracted plane and hole features, and then finely aligned through the ICP algorithm.

2. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 1, characterized in that: In the step S1, after the TR component (4) is assembled to the antenna frame, the assembled TR component (4) and the cold plate (5) area are scanned by the mechanical arm (1) equipped with a line laser scanner (2). During the scanning, the scanner is located directly above the TR component (4) and the cold plate (5), and the scanning distance is within the effective measurement distance range of the line laser scanner (2).

3. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 2, characterized in that: In step S1, the scanning path is determined according to the direction of the cold plate (5), and the scanning range covers all TR components (4) and the exposed areas of the cold plate (5).

4. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 3, characterized in that: In step S2, the original point cloud data obtained by scanning is preprocessed to remove outliers in the point cloud to obtain a point cloud P c , the point cloud is segmented by using the growth clustering strategy based on radius search. c Segmentation into cold plate (5) point cloud P l And component point cloud P z .

5. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 4, characterized in that: In step S3, for the TR component (4), a plane p is extracted from the point cloud of the TR component (4) by using the RANSAC algorithm. m ; Then, the points within a certain threshold of the plane formed in the previous step are used as input point cloud data, and the boundary point set S in the point cloud data is extracted by the region growing method based on curvature e ; Then in the boundary point set S e The RANSAC method is used to extract S e The circular feature in the figure is used as the features of the two holes h1 and h2.

6. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 5, characterized in that: The coarse registration matrix of the theoretical model of the TR component (4) obtained by the coarse registration calculation in the step S4 is W1, the coarse registration matrix of the theoretical model of the cold plate (5) is W1', the registration matrix of the theoretical model of the TR component (4) obtained by the fine registration calculation is W2, and the fine registration matrix of the theoretical model of the cold plate (5) is W2'. The transformations W2·W1 and W2'·W1' are applied to the theoretical model of the TR component (4) and the theoretical model of the cold plate (5) respectively, so that the theoretical model of the TR component (4) and the theoretical model of the cold plate (5) have a relative posture consistent with that in the actual assembly state.

7. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 6, characterized in that: In step S5, the fitting plane of the theoretical model of the TR component (4) and the cold plate (5) is downsampled into a point cloud L z and L l , with L l As a benchmark, calculate L z Each point on the point cloud L l The closest distance is taken as the gap value at that point.

8. A digital measurement method for assembly gap of components with large aspect ratio as claimed in claim 7, characterized in that: In the step S6, the gap value is converted into a corresponding color through color mapping according to the gap range, and a gap distribution cloud map of the TR component (4) and the cold plate (5) is generated.

Images