Airborne equipment test point automatic positioning method based on fuselage point cloud features

Abstract

The invention innovatively provides an airborne equipment test point automatic positioning method based on fuselage point cloud features, and belongs to the field of intelligent testing. The method comprises the following steps: firstly, acquiring the position of each test point in a fuselage coordinate system, and then realizing aircraft landing gear point cloud extraction by using a laser point cloud drawing and mapping cutting technology; and rotation and translation vectors are calculated in combination with the barycentric coordinates of the undercarriage, and the process of converting the position of each test point from a fuselage coordinate system to a laser radar coordinate system is completed, so that accurate positioning is realized. According to the invention, the aircraft landing gear is taken as an important reference basis, and the automatic positioning work of the test point can be quickly and accurately completed at any position only by obtaining the position relationship between the test point and the nose landing gear at one time, so that the transformation of a test activity main body from a person to test equipment is promoted, and new vitality is injected for the development of the field of equipment support.

Description

Technical Field

[0001] The present invention relates to the field of intelligent testing, and particularly to an automatic positioning method for test points of airborne equipment based on the body point cloud features. Background Art

[0002] With the rapid development of technologies such as unmanned driving and the Internet of Things, people's lives have been greatly enriched and many conveniences have been brought.

[0003] Currently, in unmanned equipment, to achieve in-situ testing in the field, a common method is to install a test antenna at the test point. By radiating test signals and receiving interference signals, the warning function of the equipment, interference effect, etc. can be evaluated.

[0004] In the above process, the selection of the test point is particularly important. The reason is that when using the radiation method for in-situ testing, due to the directivity of the antenna and the attenuation characteristics of electromagnetic waves in space, both the distance and azimuth angle between the test antenna and the equipment antenna need to be determined in order to obtain core performance indicators such as the sensitivity and radiation power of the equipment through corresponding conversions. In an ideal situation, the aircraft can accurately park at a fixed position according to the ground markings, and the test personnel can quickly complete the numerical conversion of the signal link loss based on the pre-calibrated test points. However, in actual operation, affected by multiple factors such as the field environmental conditions and task urgency, there are often certain deviations in the parking position of the aircraft. At this time, the test personnel need to re-use measurement tools to determine the test points. This process takes a long time and is prone to errors due to human factors; even if the measurement is accurate, its effectiveness is limited to the current test flight and cannot provide a reference for subsequent measurement processes. Summary of the Invention

[0005] The present invention innovatively proposes an automatic positioning method for test points based on the characteristics of the aircraft body. This method takes the aircraft landing gear as an important reference basis and combines the point cloud data provided by the lidar to automatically and accurately complete the precise positioning of test points at any position, thereby providing strong technical support for the intelligent transformation of field maintenance and support work.

[0006] To achieve the above object, the technical solution of the present invention is as follows: An automatic positioning method for test points of airborne equipment based on the body point cloud features, which includes the following three parts: S1. Test point selection and recording; S2. Point cloud drawing and mapping clipping; S3. Test point position calculation.

[0007] Furthermore, the test point selection and recording in step S1 specifically includes: S11. According to the position and orientation of the airborne antenna, the tester selects test points one by one to ensure that a clear and stable positional relationship can be established between the test antenna and the airborne antenna; S12. Take the nose landing gear of the aircraft as the origin , and construct a plane rectangular coordinate system ; Preferably, the backward direction of the central axis of the fuselage is defined as the positive direction of the U axis; S13. With the help of measuring tools, accurately measure the coordinates of each test point in the coordinate system , denoted as .

[0008] Further, the point cloud drawing and mapping clipping described in step S2 means that after the point cloud is drawn in the software, the mapping relationship between the selected point coordinates on the screen window and the actual point cloud coordinates is realized by using the mapping clipping technology, so that the tester can conveniently select the region of interest (ROI) by using computer peripherals such as a mouse; Preferably, the ROI is the three landing gears of the aircraft; Further, step S2 specifically includes: S21. Point cloud drawing; S22. Step-by-step dimensionality reduction and mapping clipping.

[0009] Further, the point cloud drawing described in step S21 means that the point cloud information collected by the lidar is rendered and drawn using OPENGL. The specific process includes: S211. Take the lidar as the coordinate origin , establish the original point cloud coordinate system , where is the horizontal plane, and the Z axis is perpendicular to upward. Thus, all the point clouds in the lidar field of view can be obtained in this coordinate system. Denote the coordinates of a certain point N as ; Use the pcl::compute3DCentroid function in the point cloud PCL library to calculate the coordinates at its centroid position, denoted as ; S212. Translate the point cloud origin to ; For the translated point cloud data, it is necessary to select the maximum absolute value in the X, Y, and Z directions , and then perform normalization processing, which realizes the conversion of the original point cloud into a format suitable for display in the OPENGL interval ( ); S213. Use the perspective projection matrix provided by OPENGL , project the normalized point cloud data onto the screen interface for display; Therefore, the coordinates of point U in OPENGL It can be expressed by formula (1) as: (1) Furthermore, the step-by-step dimensionality reduction and mapping clipping described in step S22 means that since the window and the point cloud have inconsistent dimensions, the point cloud obtained after the processing in step S21 needs to be temporarily reduced in dimension in two steps to cooperate with the window point selection to obtain the landing gear point cloud; preferably, it can be implemented according to the following process: S221. In , let the coordinates z value be 0, then the two-dimensional point cloud distribution on the horizontal plane is obtained; if the coordinates of any point after dimensionality reduction are , then after perspective projection according to formula (1), its Z coordinate component is a fixed value, denoted as ; S222. Use the mouse to select a point A in the point cloud in the window, and record its coordinates in the window as ; Preferably, the origin of the window coordinate system is located at the upper left corner of the window, while the origin of the OPENGL coordinate system is set at the center of the window. Let the length and width of the window be , respectively. Then the first 2 dimensions of the coordinates of point A in OPENGL are: (2) S223. Since point A is selected from a point on the interface after perspective projection, there is ; according to the perspective projection matrix , the actual normalized coordinates of point A before perspective projection can be restored: (3) Combining formulas (1) to (3), the coordinates of point A mapped to are: (4) Take the first 2 dimensions of , which are the x , y coordinates of the window selected point A mapped to the original point cloud; S224. Restore the coordinates value, and then set the coordinates x value to 0, that is, the point cloud distribution in the vertical plane ( ) is obtained. After perspective projection by formula (1), the mapping relationship with the window selected point can be established again, that is, the clipping in the third dimension is realized, so as to complete the three-dimensional extraction of the landing gear point cloud in the coordinate system .

[0010] Further, the calculation of the test point positions in step S3 refers to the process of converting the test point coordinates obtained in step S13 from the airframe coordinate system to the lidar coordinate system . Specifically, it includes: S31. Calculate the translation vector ; S32. Calculate the rotation vector ; S33. Implement test point positioning: According to the translation vector and the rotation vector , the transformation matrix can be calculated; According to the coordinates of each test point in in step S13, then through the matrix inverse transformation , the position of the test point in the lidar coordinate system can be obtained to achieve positioning.

[0011] The calculation of the translation vector described in step S31, the process is as follows: S311. Regarding the point clouds of the three landing gears as mass points, an isosceles triangle can be formed, where the vertex corresponds to the position of the nose landing gear ; S312. In the coordinate system , calculate the difference between the lengths of any two sides of the isosceles triangle. When the difference is the smallest, the common point of the corresponding two sides is the vertex , and the coordinate of in can be obtained. This coordinate is the translation vector ; S313. After translating to , the resulting coordinate system is denoted as , and this coordinate system shares the origin with the airframe coordinate system .

[0012] Further, the calculation of the rotation vector described in step S32, the process is as follows: S321. For the isosceles triangle, its centroid G is located on the perpendicular line from the vertex to the base. Therefore, the coordinate of point G in has the characteristic ; If the coordinate of point G in does not satisfy , it indicates that the same point G is in two co - origin They have different coordinates in the coordinate system, so there is a rotational deviation between the two coordinate systems and rotational correction is required. S322. Set the angle variable , increment it in steps of 0.5°, and rotate clockwise around the point. Each time it rotates, record the current and into the array PL until it rotates one full circle. S323. Traverse the array PL to find the angle when the coordinates of point G satisfy and are closest to 0. This is the angle difference between the two coordinate systems; the rotation vector corresponding to this angle is ; The present invention proposes an automatic positioning method for airborne equipment test points based on the body point cloud features. This method uses the aircraft landing gear as an important reference basis. Only by obtaining the positional relationship between the test points and the front landing gear once, it can combine the point cloud data provided by the lidar, and use the step-by-step dimensionality reduction mapping and cropping technology and coordinate transformation technology to realize the transformation of the positioning work of airborne equipment test points from manual to automatic and from inefficient to efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0014] Figure 1 is the general flowchart of the present invention; Figure 2 is the point cloud map of the actual test scene in the embodiment of the present invention; Figure 3 is the lidar coordinate system , test point selection, and body coordinate system distribution schematic diagram in the embodiment of the present invention; Figure 4 is the point cloud mapping and cropping result in the embodiment of the present invention; Figure 5 is the display diagram of the positioning results of 4 test points in the embodiment of the present invention; Figure 6 is the software architecture design diagram involved in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than limiting the invention. Additionally, it should be noted that for ease of description, only the parts related to the relevant invention are shown in the drawings. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and embodiments.

[0016] Attached Figure 1 FIG. is the overall flowchart of an automatic positioning method for test points of airborne equipment based on fuselage point cloud features provided by an embodiment of the present invention. Combining this drawing, the specific steps of the present invention include: S1. Test point selection and recording.

[0017] In this embodiment, first, the front landing gear of the aircraft is regarded as the origin to construct a plane rectangular coordinate system. Then, the positions of each test point are measured and the position information is recorded. The plane rectangular coordinate system is like a map that can accurately locate the position of the test point on the aircraft fuselage.

[0018] It should be noted that the front landing gear of the aircraft is a fixed and easily recognizable position. Using this as the origin to construct the coordinate system can provide a unified and stable reference standard for the position of the test point. It provides a clear position reference for subsequent test work and ensures the accuracy and consistency of the test points.

[0019] S2. Point cloud drawing and mapping clipping.

[0020] This embodiment involves point cloud drawing. The specific process starts from the data collected by the lidar and converts it into a format that can be displayed on the screen. A three-dimensional coordinate system is established with the lidar as the origin, and the centroid position of all points in the entire scene is calculated. The origin of the point cloud is translated to the centroid, and the point cloud is normalized to make it suitable for the display range of the screen. Finally, the normalized point cloud data is projected onto the screen using a perspective projection matrix. It can be seen that the centroid calculation and translation operations can make the point cloud data more concentrated. Normalization and perspective projection can make the point cloud data adapt to the requirements of screen display and facilitate user interaction.

[0021] This embodiment also realizes the purpose of step-by-step dimensionality reduction and mapping clipping, which is to reverse the corresponding position in the actual point cloud from the selected point on the screen. Specifically, first, the coordinate value of the point cloud in a certain direction is set to zero, thereby reducing the three-dimensional point cloud to a two-dimensional plane. Then, a point is selected on the screen window by the mouse, and using the inverse transformation of the perspective projection matrix, the selected point on the screen is restored to the position in the actual point cloud. Finally, this process is repeated to complete the clipping in the third dimension, thereby extracting the point cloud of the landing gear.

[0022] As an implementation of this embodiment, the point cloud drawing and mapping clipping in step S2 complete the extraction of the aircraft landing gear point cloud, which specifically includes the following steps: S21. Point cloud drawing.

[0023] The point cloud drawing in step S21 refers to using OPENGL to complete the rendering and drawing of the point cloud information collected by the lidar. The specific process includes: S211. Taking the lidar as the coordinate origin , establish a point cloud coordinate system , and record the coordinates of a certain point N therein as ; use the pcl::compute3DCentroid function in the point cloud PCL library function to calculate the centroid position in the entire scene, denoted as ; S212. Translate the point cloud origin to , select the maximum absolute value of the point cloud in the X, Y, and Z directions, and then perform normalization processing to convert the original point cloud into the format displayed in the OPENGL interval; S213. Use the perspective projection matrix provided by OPENGL to project the normalized point cloud data onto the screen interface for display; among them, the coordinates of point N in OPENGL are: (1).

[0024] S22. Step-by-step dimensionality reduction and mapping clipping.

[0025] The step-by-step dimensionality reduction and mapping clipping in step S22 realizes the mapping relationship between the selected point coordinates in the screen window and the actual point cloud coordinates to obtain the landing gear point cloud. The specific process is as follows: S221. In , let the coordinate z value of each point be 0, then the point cloud can be reduced to a horizontal coordinate system for distribution; if the coordinates of any point after dimensionality reduction are , then after perspective projection according to formula (1), its Z coordinate component is a fixed value, denoted as ; S222. Use the mouse to select a point A in the point cloud in the interface window, and record its coordinates in the window as ; assume that the length and width of the window are respectively, and the OPENGL coordinate system is located in the center of the window, then the first 2 dimensions of the coordinates of point A in OPENGL are: (2) S223. Since point A is selected from a point on the interface after perspective projection, there is ; According to the perspective projection matrix , the actual normalized coordinates of point A before perspective projection can be restored : (3) Combining equations (1) to (3), the coordinates of point A mapped to are : (4) Take the first 2 dimensions of , which are the coordinates of point A mapped by the window selection after ; S224. Restore the z values of each point in the point cloud, and then set the x values of each point coordinate to 0, that is, the distribution of the point cloud reduced to the vertical plane is obtained. After OPENGL perspective projection, a mapping relationship is established with the window selection point again, that is, the clipping in the third dimension is realized, so as to complete the three-dimensional extraction of the landing gear point cloud in the coordinate system .

[0026] The step-by-step dimensionality reduction of this embodiment can simplify the calculation and reduce the complexity of data processing. The mapping relationship between the screen selection point and the actual point cloud coordinates enables the user to directly complete the point cloud extraction through interface operations. The three-dimensional point cloud is gradually reduced to a two-dimensional plane by mathematical methods, and then the position information in the original point cloud is restored through inverse transformation, so as to achieve accurate point cloud extraction.

[0027] S3. Calculation of test point positions.

[0028] In this embodiment, in order to convert the test points from the body coordinate system to the lidar coordinate system, the translation vector between the two coordinate systems is calculated first. The specific method is to regard the three landing gear point clouds as mass points, form an isosceles triangle, and find the position of the front landing gear corresponding to the vertex. By calculating the case where the difference in the side lengths of the isosceles triangle is the smallest, the position of the front landing gear can be determined, and then the translation vector can be obtained.

[0029] Since there may be a rotational deviation between the body coordinate system and the lidar coordinate system, a rotation vector needs to be calculated for correction. The specific method is to find the centroid of the isosceles triangle and adjust the coordinate system through rotation operations so that the positions of the centroid in the two coordinate systems are the same. The last step is to construct a transformation matrix based on the translation vector and the rotation vector to convert the test points from the body coordinate system to the lidar coordinate system. Through matrix inverse transformation, the position of the test points in the lidar coordinate system can be obtained.

[0030] The calculation of the test point positions in step S3 of this embodiment refers to the process of converting the test point coordinates obtained in step S1 from the airframe coordinate system to the lidar coordinate system . Specifically, it includes: S31. Calculate the translation vector .

[0031] The calculation of the translation vector described in step S31 is as follows: S311. Treat the point clouds of the three landing gears as mass points to form an isosceles triangle, where the vertex corresponds to the position of the nose landing gear ; S312. In the coordinate system , calculate the difference between the lengths of any two sides of the isosceles triangle. When the difference is the smallest, the common point of the corresponding two sides is . Then, the coordinates of in can be obtained. This coordinate is the translation vector ; S313. After translating to , the resulting coordinate system is denoted as . This coordinate system shares the origin with the airframe coordinate system .

[0032] S32. Calculate the rotation vector .

[0033] The calculation of the rotation vector described in step S32 is as follows: S321. For the isosceles triangle, its centroid G is located on the perpendicular line from the vertex to the base. Then, the coordinates of point G in the airframe coordinate system have the characteristic ; if the coordinates of point G in do not satisfy , it indicates that the same point G has different coordinates in two coordinate systems with a common origin . There is a rotational deviation between the two coordinate systems, and rotation correction is required; S322. Rotate step by step clockwise around point. Each time it rotates, record the current rotation angle and the coordinates of point G, and store them in the array PL until it rotates one full circle; S323. Traverse the array PL to find the point G coordinates that satisfy and The angle corresponding to when it is closest to 0 , which is the angle difference between the two coordinate systems ; The rotation vector corresponding to this angle is .

[0034] S33. Implement test point positioning: According to the translation vector and the rotation vector , the transformation matrix can be calculated; The coordinates of each test point recorded in step S1 , through the matrix inverse transformation , can obtain the position of the test point in , that is, the position in the lidar coordinate system , to achieve positioning.

[0035] It can be seen that in this embodiment, the translation vector is determined by an isosceles triangle, avoiding a complex coordinate matching process. Ensuring the accurate relative position relationship between the two coordinate systems. Using geometric relationships and distance calculations, the translation deviation between the two coordinate systems is found, thereby determining the translation vector. Through rotation correction, the angular deviation between the two coordinate systems is eliminated. Ensuring that the test points will not be displaced due to rotation errors during the conversion process. Moreover, through rotation operations and the matching of the centroid positions, the angular deviation between the two coordinate systems is adjusted, thereby achieving precise alignment. The use of the transformation matrix makes the conversion between coordinate systems efficient and accurate.

[0036] In this embodiment, an Advantech EPC-301 embedded industrial computer and a LIVOX AVIA lidar are used to form a test platform, which is placed about 15 m away from the aircraft's center of gravity; After the lidar irradiates the aircraft fuselage to obtain point cloud data, the actual test scene point cloud map shown in the appendix is drawn, which includes objects such as the aircraft, the ground, and buildings. Combining this scene, the automatic positioning of the test points is gradually realized according to steps S1 to S3. The specific process is as follows: Figure 2 (1) For the test point selection described in step S1, in this embodiment, a total of 4 test points are selected based on the local equipment of the aircraft, denoted as , and their distribution schematic diagram in the fuselage coordinate system is as shown in , and the specific position and angle information of each point are recorded in Table 1. Table 1 is the position and angle information of each test point in the fuselage coordinate system Figure 3 .

[0037] Table 1

[0038] ​(2) After obtaining the body point cloud data, the software will perform the point cloud drawing and mapping clipping described in step S2 to obtain the point cloud of the landing gear area; the specific constraint intervals are as follows: (5) That is, in the X direction, the interval [12, 45, 14.25] is removed, in the Y direction, the interval [1.72, 3.75] is removed, and in the Z direction, the interval [-0.18, 0.73] is retained. The interval unit is m. The landing gear point cloud obtained by clipping is as Figure 4 shown; considering the three landing gears as mass points, an isosceles triangle O’BC can be formed; point O’ represents the nose landing gear, and points B and C represent the two rear landing gears respectively.

[0039] (3) Taking the coordinates of each point listed in Table 1 as known quantities and performing the test point position calculation method described in step S3, the precise coordinates of the 4 test points in this embodiment can be obtained in only a very short time, that is, the point positioning work is completed, and the results are shown in the column of "Measured Coordinates / m" in Table 2. Table 2 gives the test point positioning and error comparison results.

[0040] Table 2

[0041] Figure 5 is a diagram showing the automatic positioning results of the test points of the airborne equipment based on the body point cloud features provided by the embodiment of the present invention. It can be clearly observed in the figure the distribution of the four test points around the landing gear. At the same time, the coordinates of each point in are also listed in the Figure 5 table; In addition, Figure 5 the table also provides angle information, and its meaning is specifically illustrated by taking the point as an example: At the beginning, the test platform is located at and its positive direction, that is, the test platform direction, is ; after calculating the position (15.45, 9.73) of in the coordinate system through step S3, the platform first rotates counterclockwise by 32°, making the positive direction point to , and then advances to . After that, in order to make the platform face the equipment antenna, it is necessary to initially make the test platform face the nose landing gear . Therefore, it needs to rotate counterclockwise by 239° again around its own axis. In this way, the sum of the two rotation amounts is 271°, which explains the origin of 32°, 239°, and 271° in the table.

[0042] To ensure the accuracy and reliability of the positioning results, error analysis was also carried out in this embodiment. For this purpose, after completing the test point selection in step S1, 4 marks were set on the ground at each test point. Subsequently, using measuring tools, the coordinates of these 4 marks in the coordinate system were measured and used as standard values for comparison with the measured values obtained in this embodiment. The results are listed in Table 2. It can be seen from Table 2 that the positioning results of the 4 test points obtained in this embodiment are relatively accurate, and the absolute distance error between them and the standard values is controlled within the order of 10 cm, while the relative error does not exceed 1%, fully verifying the accuracy and reliability of the method involved in the present invention.

[0043] To implement the above steps, this application also developed a set of control software with rich functions. Refer to Figure 6 as shown, including an interface layer (user interface: QT framework, OPENGL rendering: VAO, VBO), an application layer (data acquisition - point cloud processing - point selection mapping - cutting and cropping - coordinate transformation - test point calibration), a bottom - layer dependency layer (Ubuntu18.04, ROS Melodic), and a hardware layer (LIDAR, GPU, CPU). This software runs under the UBUNTU18.04 operating system and is built and compiled based on the Qt5.9 development environment. The interface design of the software is clear and has strong interactivity. Its core functions include the drawing of three - dimensional point clouds, mapping and cropping, and the execution of point position parsing and measurement algorithms. These functions together form the basis for realizing the automatic positioning of test points involved in the present invention. In terms of software architecture design, in order to obtain and process lidar data in real time, the software integrates the ROS (Robot Operating System) distributed communication architecture. By establishing nodes and subscribing to lidar - related topics, the software can receive and process point cloud data in real time. At the same time, with the help of the GPU acceleration technology in OPENGL, the software realizes the smooth rendering of 240,000 point cloud data per second in a three - dimensional space, providing important support for subsequent position calculation. In addition, to ensure the efficient operation of each algorithm, the software allocates specific threads for the operations at each stage. By calling independent CPU resources, it ensures that the main interface thread can continuously respond to input and output.

[0044] The beneficial effects of the above embodiments include: 1. Utilizing the high - precision point cloud data provided by the lidar, through mapping and cropping and coordinate system transformation, automatic positioning of test points is achieved; 2. Simplifying the operation process, while significantly reducing the repeatability and error of manual measurement, and remarkably improving the efficiency and accuracy of in - situ field tests; 3. Providing key technical support for the intelligent transformation of maintenance and support work, and promoting the technological progress of the industry; 4. Provide a feasible solution for the test platform with an antenna to complete preliminary alignment and expand the application scope.

[0045] In summary, the embodiments of the present invention show significant beneficial effects in aspects such as accurately positioning test points, improving measurement efficiency, and application expansion.

[0046] The above specifically describes the preferred embodiments of the present invention. However, the present invention is not limited to the described embodiments. Those skilled in the art can make equivalent deformations or substitutions without departing from the spirit of the present invention, and these equivalent deformations and substitutions are all included within the scope defined by the claims of this application.

Claims

1. A method for automatically locating test points of airborne equipment based on fuselage point cloud features, characterized in that: It includes the following 3 steps: S1. Test point selection and recording: The front landing gear of the aircraft is used as the origin , construct a plane rectangular coordinate system ; Measure each test point The coordinates of ; S2, point cloud drawing and mapping clipping; S3. Calculation of test points.

2. The method for automatically locating test points of airborne equipment based on fuselage point cloud features according to claim 1, characterized in that: Step S2, the point cloud drawing and mapping clipping, completes the extraction of the aircraft landing gear point cloud, and specifically includes the following steps: S21, point cloud drawing; S22. Step-by-step dimensionality reduction and mapping pruning.

3. The method for automatically locating test points of airborne equipment based on fuselage point cloud features as claimed in claim 2, characterized in that: The point cloud drawing in step S21 refers to rendering and drawing the point cloud information collected by the laser radar using OPENGL, and the specific process includes: S211, laser radar as the coordinate origin , establish the point cloud coordinate system , let the coordinates of a certain point N be ; Use the pcl::compute3DCentroid function in the point cloud PCL library function to calculate the center of mass position of the entire scene, recorded as ; S212, translate the point cloud origin to At, select the maximum absolute value of the point cloud in the X, Y, and Z directions , and then normalize it to convert the original point cloud into the format displayed in the OPENGL interval; S213, using the perspective projection matrix provided by OPENGL , project the normalized point cloud data onto the screen interface for display; where point N is the coordinate in OPENGL for: （1）。 4. The method for automatically locating test points of airborne equipment based on fuselage point cloud features as claimed in claim 2, characterized in that: The step-by-step dimensionality reduction and mapping clipping described in step S22 realize the mapping relationship between the screen window selected point coordinates and the actual point cloud coordinates to obtain the landing gear point cloud. The specific process is as follows: S221, in In the example, let the coordinates of each point be z If the value is 0, the point cloud can be reduced to a horizontal coordinate system. distribution; if the coordinates of any point after dimensionality reduction are , then after perspective projection according to formula (1), its Z coordinate component is a constant value, denoted as ; S222. Use the mouse to select a point A in the point cloud in the interface window, and record its coordinates in the window as ; Assume the window length and width are , the OPENGL coordinate system is located in the center of the window, then point A is in the OPENGL Coordinates in The first two dimensions of are: （2） S223. Since point A is selected from a point on the interface after perspective projection, ; According to the perspective projection matrix , the actual normalized coordinates of point A before perspective projection can be restored : （3） Combining equations (1) to (3), point A is mapped to Coordinates for: （4） Pick The first two dimensions are the window point A after mapping coordinate; S224, restore each point of the point cloud z Value, and then the coordinates of each point x The value is set to 0, which means that the point cloud is reduced to the vertical surface. After the OPENGL perspective projection, the mapping relationship is established again with the window selection point, which realizes the clipping in the third dimension, thus completing the 3D extraction of landing gear point cloud.

5. The method for automatically locating test points of airborne equipment based on fuselage point cloud features as claimed in claim 4, characterized in that: The test point calculation in step S3 refers to converting the test point coordinates obtained in step S1 into the fuselage coordinate system. Convert to LiDAR coordinate system The process includes: S31. Calculate the translation vector ; S32, calculate the rotation vector ; S33, realize the test point positioning: according to the translation vector and the rotation vector , the transformation matrix can be calculated ; The coordinates of each test point recorded in step S1 , through the inverse matrix transformation , we can get the test point at That is, the position in the laser radar coordinate system , to achieve positioning.

6. The method for automatically locating test points of airborne equipment based on fuselage point cloud features as claimed in claim 5, characterized in that: Step S31 of calculating the translation vector , the process is: S311. The three landing gear point clouds are regarded as mass points to form an isosceles triangle, where the vertex corresponds to the position of the front landing gear. ; S312, in the coordinate system , calculate the difference between any two sides of an isosceles triangle. When the difference is the smallest, the common point of the two corresponding sides is , can be obtained exist Coordinates in , which is the translation vector ; S313, will Pan to The coordinate system after , this coordinate system is consistent with the fuselage coordinate system shared origin .

7. The method for automatically locating test points of airborne equipment based on fuselage point cloud features as claimed in claim 6, characterized in that: Step S32 calculates the rotation vector , the process is: S321. For an isosceles triangle, its centroid G is located on the perpendicular line from the vertex to the bottom edge. Then point G is in the fuselage coordinate system. Coordinates in Features ; If point G is Coordinates Dissatisfied , it means that the same point G is at two common origins The coordinates in the coordinate system are different, and there is a rotation deviation between the two coordinate systems, which needs to be rotated and corrected; S322, will Around The point is rotated clockwise in steps, and the current rotation angle is recorded each time it rotates. and the coordinates of point G , stored in the array PL until it rotates one circle; S323, traverse the array PL and find the point G whose coordinates satisfy and The angle closest to 0 , is the angle between the two coordinate systems ; The rotation vector corresponding to this angle is .

Images