Skin member buckling deformation detection method based on minimum bounding box constraint

Abstract

The invention provides a skin member buckling deformation detection method based on minimum bounding box constraint. The method comprises the following steps: acquiring actual measurement point cloud data and CAD model point cloud data of a wing skin member; noise points are filtered out; calculating a minimum bounding box of the actually measured point cloud data and the CAD model point cloud data; the center coordinate of the minimum bounding box and the plane of the bottom of the box are used as constraints to achieve accurate registration, and the purpose of calculating the buckling deformation amount is achieved by calculating the distance from one point in the point cloud data set of the CAD model to the normal of the point in the point cloud data set of the CAD model to the intersection point of the three-point fitting plane closest to the normal in the actually measured point cloud data set. According to the skin member buckling deformation detection method based on the minimum bounding box constraint, the wing skin after deformation and the CAD model can be correctly registered, the buckling deformation amount can be correctly calculated, and the influence of buckling deformation on the registration result and the mismatching problem existing in the prior art are effectively avoided.

Description

Technical Field

[0001] The invention relates to the field of wing skin deformation detection, and in particular to a method for detecting warping deformation of a skin component based on minimum bounding box constraints. Background Art

[0002] After the wing skin components are formed, due to the influence of residual stress during the processing, warping and deformation will occur on all sides after the fixture is released or the components are demolded. The magnitude of the deformation will affect the stress distribution during the subsequent assembly process. Therefore, it is necessary to calculate the warping deformation of the measured skin on all sides. With the rapid development of laser 3D scanning technology, it is now possible to quickly scan and obtain the measured scanning data of the skin surface. The calculation of the skin warping deformation first requires the registration of the measured point cloud data with the theoretical CAD model data. The accuracy of the registration directly affects the accuracy of the deformation calculation results. However, since the positioning reference of the skin may change during the forming or processing process, the registration process can only rely on the skin's own characteristics for registration. How to accurately achieve the registration of the two data is the key to the calculation of the skin warping deformation.

[0003] Although the point cloud registration technology has been studied in detail, there are still many defects in the registration of point clouds after warping. In the skin registration process, it is easy to fall into the problem of local optimality due to inaccurate corresponding point search. At the same time, if only the best fit matching criterion is used, the calculated deformation distribution will not match the actual deformation. Therefore, the existing model registration method is difficult to be well applied to the precise registration process of skin data. Summary of the invention

[0004] The purpose of the present invention is to address the defects of the existing wing skin registration algorithm and propose a skin component warping deformation detection method based on minimum bounding box constraints. This method can effectively solve the problem that the wing skin cannot be accurately registered due to warping deformation on all sides after manufacturing.

[0005] In order to achieve the above technical objectives, the present invention provides the following technical solutions:

[0006] A method for detecting warping deformation of a skin component based on a minimum bounding box constraint specifically comprises the following steps:

[0007] S1. Obtaining measured point cloud data and CAD model point cloud data of the wing skin;

[0008] S2, filtering out noise points deviating from the subject in the acquired measured point cloud data by a point cloud filtering method;

[0009] S3, calculating the minimum bounding box of the measured point cloud data and the CAD model point cloud data after the noise points are filtered out by the principal component analysis method;

[0010] S4, in the iterative registration process, the center coordinates of the two minimum bounding boxes and their bottom plane equations calculated in step S3 are introduced as constraints to solve the optimal transformation matrix;

[0011] S5. According to the characteristics of the warping deformation of the skin data, the distance from a point in the CAD model point cloud data to its normal and the intersection of the three-point fitting plane closest to the normal in the measured point cloud data are calculated to evaluate the warping deformation.

[0012] Furthermore, step S1 is specifically as follows:

[0013] A 3D scanner is used to obtain the surface 3D data of the wing skin to obtain the measured point cloud dataset P; the CAD model surface is sampled and the sampled data is discretized to obtain the CAD model point cloud dataset Q.

[0014] Furthermore, in step S2, the point cloud filtering method is a radius filtering method.

[0015] Further, step S3 includes:

[0016] S31, using principal component analysis to calculate and obtain the centroid and covariance matrix of point cloud data; the point cloud data includes measured point cloud data and CAD model point cloud data;

[0017] S32, solving eigenvalues ​​and eigenvectors through the calculated covariance matrix, and the solved eigenvector is the main direction of the point cloud data;

[0018] S33, using the obtained main direction and centroid to transform the point cloud data to the origin, and the main direction coincides with the direction of the coordinate system, so as to establish a point cloud bounding box transformed to the origin;

[0019] S34. By calculating the maximum and minimum values ​​of the transformed point cloud data on the three coordinate axes, the vertex coordinates of the minimum bounding box can be obtained, and then the bottom surface equation of the bounding box can be solved.

[0020] Further, step S4 includes:

[0021] S41. Establish a mathematical model for registration, and the formula is expressed as:

[0022]

[0023] Among them, R(x) and T(x) are rotation matrices and translation matrices respectively, p i is the i-th measured point cloud data, q jis the jth CAD model point cloud data; x = (θ x ,θ y ,θ z ,t x ,t y ,t z ) T Represents the spatial rotation angle and spatial translation;

[0024] S42. Define the centroid distance threshold d cen , calculate the distance from the centroid of the measured point cloud data to the centroid of the discrete point cloud data of the CAD model after transformation to the origin position; if the distance is greater than the distance threshold d cen , then rotate the measured point cloud data 180° around the X axis with its bounding box center as the center; then calculate the distance between the two centroids at this time, if it is still greater than the distance threshold d cen , then rotate 180° around the Y axis; then calculate the distance between the two centroids at this time, if it is still greater than the distance threshold d cen , then rotate 180° around the Z axis;

[0025] S43. According to the nearest point search method, the search distance threshold is set to find the corresponding points between the measured point cloud data and the CAD model point cloud data; the initial transformation matrix is ​​calculated using the SVD algorithm based on the two corresponding points found, and then the initial transformation matrix is ​​optimized using the iterative nearest point algorithm. During the optimization process, the bottom surfaces of the two bounding box data are ensured to be on the same plane, so as to calculate the optimal transformation matrix.

[0026] Further, step S5 includes:

[0027] S51, finding the normal vector of the CAD model point cloud data, and specifying that the normal vector direction points to the center point of its minimum bounding box;

[0028] S52, solving the normal line equation corresponding to each point according to the normal direction and point coordinates of each point in the CAD model point cloud data;

[0029] S53, calculating the three points in the measured point cloud data that are closest to the normal line corresponding to each point in the CAD model data, and solving the plane equation formed by the three points;

[0030] S54, calculating the intersection of each point in the CAD model point cloud data with its normal and the plane solved in S53 and the distance between the two points, and considering the two point data as a pair of matching point pairs;

[0031] S55. Define search distance threshold d th , if the distance calculated in step S54 is greater than d th , it is considered that there is no matching point in the CAD model point cloud data and it is discarded;

[0032] S56. The matching point pairs calculated after steps S54 and S55 are used as the correct matching point pairs for deformation, and the distance relationship between the matching point pairs reflects the warping deformation of the skin.

[0033] Based on the above technical solution, the present invention has the following beneficial effects:

[0034] 1) The present invention is suitable for wing skin registration after warping deformation on all sides, and can effectively avoid the influence of warping deformation on the registration result;

[0035] 2) The present invention can accurately calculate the magnitude of the skin warping deformation, thus avoiding the problem of point mismatching in the deformation calculation process of the existing method. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of the method for detecting warping deformation of skin components based on minimum bounding box constraints of the present invention;

[0037] Figure 2 This is a flowchart of the warping deformation calculation of the present invention. DETAILED DESCRIPTION

[0038] In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should fall within the scope of protection of the present application.

[0039] Although the steps in the present invention are arranged with numbers, they are not used to limit the order of the steps. Unless the order of the steps is clearly stated or the execution of a certain step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" used in this article involves and covers any and all possible combinations of one or more of the associated listed items.

[0040] like Figure 1 As shown, a method for detecting warping deformation of a skin component based on minimum bounding box constraints proposed by the present invention is provided, which specifically includes the following steps:

[0041] S1. Obtaining measured point cloud data and CAD model point cloud data of the wing skin;

[0042] As a preferred implementation of step S1, in this embodiment, a three-dimensional scanner is used to obtain the three-dimensional surface data of the wing skin to obtain a measured point cloud data set P; the CAD model surface is sampled, and the sampled data is discretely processed to obtain a CAD model point cloud data set Q.

[0043] S2, filtering out noise points deviating from the subject in the acquired measured point cloud data by a point cloud filtering method;

[0044] As a preferred implementation of step S2, a radius filtering algorithm is adopted in this embodiment, which specifically includes the following steps:

[0045] S21, calculating the number of neighboring points within a set radius for each point in the measured point cloud data set;

[0046] S22, if the number of neighboring points within the radius meets the set threshold, the point is retained, otherwise the point is determined to be an outlier noise point and is filtered out;

[0047] S23. For CAD model point cloud data, no outlier noise points are generated during the sampling process, so radius filtering is not required.

[0048] S3, calculating the minimum bounding box of the measured point cloud data and the CAD model point cloud data after the noise points are filtered out by the principal component analysis method;

[0049] As a preferred implementation mode of step S3, in this embodiment, step S3 is specifically as follows:

[0050] S31, using principal component analysis to calculate and obtain the centroid and covariance matrix of point cloud data; the point cloud data here includes measured point cloud data and CAD model point cloud data;

[0051] Taking the measured point cloud data as an example, the calculation process of step S31 is as follows: for the measured point cloud data set P, there are m data elements in total, P i (x i ,y i ,z i )(i≤m) is any point element, and its position average, i.e., the centroid w, can be calculated:

[0052]

[0053] The covariance matrix Cov can be further constructed:

[0054]

[0055] S32, solve the eigenvalue and eigenvector through the calculated covariance matrix, and the solved eigenvector is the main direction of the point cloud data; that is, orthogonal normalization of each eigenvector can obtain the orthogonal matrix Z Cov =[α 1 ,α 2 ,α 3 ] T , where α 1 ,α 2 ,α 3 There are three main directions;

[0056] S33, using the obtained main direction and centroid to transform the point cloud data to the origin, and the main direction coincides with the direction of the coordinate system, that is, the three main directions obtained in step S32 correspond to the positive directions of the three coordinate axes respectively, and a point cloud bounding box transformed to the origin can be established;

[0057] S34. By calculating the maximum and minimum values ​​of the transformed point cloud data on the three coordinate axes, the vertex coordinates of the minimum bounding box can be obtained, and then the bottom surface equation of the bounding box can be solved.

[0058] S4, in the iterative registration process, the center coordinates of the two minimum bounding boxes and their bottom plane equations calculated in step S3 are introduced as constraints to solve the optimal transformation matrix;

[0059] As a preferred implementation of step S4, in this embodiment, step S4 specifically includes:

[0060] S41. Establish a mathematical model for registration, and the formula is expressed as:

[0061]

[0062] Among them, R(x) and T(x) are rotation matrices and translation matrices respectively, p i is the i-th measured point cloud data, q j is the jth CAD model point cloud data; x = (θ x ,θ y ,θ z ,t x ,t y ,t z ) T Represents the spatial rotation angle and spatial translation;

[0063] In this embodiment, the rotation matrix and translation matrix from the measured point cloud data to the CAD model point cloud data are respectively:

[0064]

[0065] T(x)=(t x ,t y ,tz ) T ;

[0066] S42. Define the centroid distance threshold d cen , calculate the distance from the centroid of the measured point cloud data to the centroid of the discrete point cloud data of the CAD model after transformation to the origin position; if the distance is greater than the distance threshold d cen , then rotate the measured point cloud data 180° around the X axis with its bounding box center as the center; then calculate the distance between the two centroids at this time, if it is still greater than the distance threshold d cen , then rotate 180° around the Y axis; then calculate the distance between the two centroids at this time, if it is still greater than the distance threshold d cen , then rotate 180° around the Z axis;

[0067] S43. According to the nearest point search method, the search distance threshold is set to find the corresponding points between the measured point cloud data and the CAD model point cloud data; the initial transformation matrix is ​​calculated using the SVD algorithm (matrix singular value decomposition algorithm) based on the two corresponding points found, and then the initial transformation matrix is ​​optimized using the iterative nearest point algorithm. During the optimization process, the bottom surfaces of the two bounding box data are ensured to be on the same plane, so as to calculate the optimal transformation matrix.

[0068] After step S4, the measured point cloud data and the CAD model point cloud data are aligned.

[0069] S5. According to the characteristics of skin data warping deformation, the distance from a point in the CAD model point cloud data to its normal line and the intersection point of the three-point fitting plane closest to the normal line in the measured point cloud data are calculated to evaluate the warping deformation amount;

[0070] As a preferred implementation of step S5, Figure 2 As shown, step S5 specifically includes:

[0071] S51, finding the normal vector of the CAD model point cloud data, and specifying that the normal vector direction points to the center point of its minimum bounding box;

[0072] S52, solving the normal line equation corresponding to each point according to the normal direction and point coordinates of each point in the CAD model point cloud data;

[0073] S53, calculating the three points in the measured point cloud data that are closest to the normal line corresponding to each point in the CAD model data, and solving the plane equation formed by the three points;

[0074] S54, calculating the intersection of each point in the CAD model point cloud data with its normal and the plane solved in S53 and the distance between the two points, and considering the two point data as a pair of matching point pairs;

[0075] S55. Define search distance threshold dth , if the distance calculated in step S54 is greater than d th , it means that the plane fitted by the three points found in the measured point cloud data in S53 deviates too much from the actual surface, that is, it is considered that there is no matching point for this point in the CAD model point cloud data and it is discarded;

[0076] S56. The matching point pairs calculated after steps S54 and S55 are used as the correct matching point pairs for deformation, and the distance relationship between the matching point pairs reflects the warping deformation of the skin.

[0077] In summary, the method proposed in the present invention can correctly align the deformed wing skin and the CAD model, and can correctly calculate the amount of warping deformation, effectively avoiding the influence of warping deformation on the alignment result and the mismatching problem existing in the prior art.

[0078] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0079] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A method for detecting warping deformation of skin components based on minimum bounding box constraints, characterized in that: The following steps are involved: S1. Obtaining measured point cloud data and CAD model point cloud data of the wing skin; S2, filtering out noise points deviating from the subject in the acquired measured point cloud data by a point cloud filtering method; S3, calculating the minimum bounding box of the measured point cloud data and the CAD model point cloud data after the noise points are filtered out by the principal component analysis method; S4, in the iterative registration process, the center coordinates of the two minimum bounding boxes and their bottom plane equations calculated in step S3 are introduced as constraints to solve the optimal transformation matrix; S5. According to the characteristics of the warping deformation of the skin data, the distance from a point in the CAD model point cloud data to its normal and the intersection of the three-point fitting plane closest to the normal in the measured point cloud data are calculated to evaluate the warping deformation.

2. The method for detecting warping deformation of skin components based on minimum bounding box constraints according to claim 1, characterized in that: Step S1 is specifically as follows: A 3D scanner is used to obtain the surface 3D data of the wing skin to obtain the measured point cloud dataset P; the CAD model surface is sampled and the sampled data is discretized to obtain the CAD model point cloud dataset Q.

3. The method for detecting warping deformation of skin components based on minimum bounding box constraints according to claim 1, characterized in that: In step S2, the point cloud filtering method is a radius filtering method.

4. The method for detecting warping deformation of skin components based on minimum bounding box constraints according to claim 1, characterized in that: Step S3 specifically includes the following steps: S31, using principal component analysis to calculate and obtain the centroid and covariance matrix of point cloud data; the point cloud data includes measured point cloud data and CAD model point cloud data; S32, solving eigenvalues ​​and eigenvectors through the calculated covariance matrix, and the solved eigenvector is the main direction of the point cloud data; S33, using the obtained main direction and centroid to transform the point cloud data to the origin, and the main direction coincides with the direction of the coordinate system, so as to establish a point cloud bounding box transformed to the origin; S34. By calculating the maximum and minimum values ​​of the transformed point cloud data on the three coordinate axes, the vertex coordinates of the minimum bounding box can be obtained, and then the bottom surface equation of the bounding box can be solved.

5. The method for detecting warping deformation of skin components based on minimum bounding box constraints according to claim 1, characterized in that: Step S4 specifically includes the following steps: S41. Establish a mathematical model for registration, and the formula is expressed as: Among them, R(x) and T(x) are rotation matrices and translation matrices respectively, p i is the i-th measured point cloud data, q j is the jth CAD model point cloud data; x = (θ x ,θ y ,θ z ,t x ,t y ,t z ) T Represents the spatial rotation angle and spatial translation; S42. Define the centroid distance threshold d cen , calculate the distance from the centroid of the measured point cloud data to the centroid of the discrete point cloud data of the CAD model after transformation to the origin position; if the distance is greater than the distance threshold d cen , then rotate the measured point cloud data 180° around the X axis with its bounding box center as the center; then calculate the distance between the two centroids at this time, if it is still greater than the distance threshold d cen , then rotate 180° around the Y axis; then calculate the distance between the two centroids at this time, if it is still greater than the distance threshold d cen , then rotate 180° around the Z axis; S43. According to the nearest point search method, the search distance threshold is set to find the corresponding points between the measured point cloud data and the CAD model point cloud data; the initial transformation matrix is ​​calculated using the SVD algorithm based on the two corresponding points found, and then the initial transformation matrix is ​​optimized using the iterative nearest point algorithm. During the optimization process, the bottom surfaces of the two bounding box data are ensured to be on the same plane, so as to calculate the optimal transformation matrix.

6. The method for detecting warping deformation of skin construction based on minimum bounding box constraint according to claim 1, characterized in that: Step S5 specifically includes the following steps: S51, finding the normal vector of the CAD model point cloud data, and specifying that the normal vector direction points to the center point of its minimum bounding box; S52, solving the normal line equation corresponding to each point according to the normal direction and point coordinates of each point in the CAD model point cloud data; S53, calculating the three points in the measured point cloud data that are closest to the normal line corresponding to each point in the CAD model data, and solving the plane equation formed by the three points; S54, calculating the intersection of each point in the CAD model point cloud data with its normal and the plane solved in S53 and the distance between the two points, and considering the two point data as a pair of matching point pairs; S55. Define search distance threshold d th , if the distance calculated in step S54 is greater than d th , it is considered that there is no matching point in the CAD model point cloud data and it is discarded; S56. The matching point pairs calculated after steps S54 and S55 are used as the correct matching point pairs for deformation, and the distance relationship between the matching point pairs reflects the warping deformation of the skin.

Images