A dual-robot synchronous pose calculation method based on penetration air-coupled ultrasonic imaging detection

Abstract

The application discloses a kind of based on the synchronous pose calculation method of double manipulator of penetration air coupling ultrasonic imaging detection, it includes S1: position layout is carried out to bidirectional curved surface detection device;S2: utilize laser sensor to scan the component to be detected, establishes the three-dimensional mathematical model of the space profile of the component to be detected;S3: trajectory point interpolation is carried out to three-dimensional mathematical model, obtains the coordinate of three-dimensional mathematical model interpolation point;S4: the normal vector of interpolation point, X direction vector and Y direction vector are calculated, obtain the pose of the component to be detected;S5: the workpiece coordinate system where the component to be detected is located is transformed, and the pose matrix of two side manipulators is calculated using the pose of the component to be detected;S6: the angle trajectory and pose of two side manipulators respectively are solved using inverse solution method.The application can accurately obtain the real profile of the detected curved surface component, complete the linkage detection of double side manipulator, improve the accuracy of detection.

Description

Technical Field

[0001] This invention relates to the fields of automation control and ultrasonic testing technology, and in particular to a method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging. Background Technology

[0002] Today, the main development trend in industrial manufacturing is intelligent manufacturing. To improve efficiency and reduce costs, intelligent manufacturing using industrial robots has become the mainstream of automation development. Ultrasonic automation equipment that combines robotic arms with ultrasonic testing technology not only improves testing efficiency but also possesses high precision. It offers greater flexibility in testing various materials and sizes of parts, and is therefore widely used in the field of ultrasonic testing. To meet the automated testing requirements of complex curved surface components using penetrating air-coupled ultrasonic imaging, a dual-robotic-arm penetrating air-coupled ultrasonic imaging testing platform has been established. By scanning the workpiece's actual contour position within the system, the scanning trajectory of the dual robotic arms is planned, enabling 14-axis linkage penetrating air-coupled ultrasonic imaging testing.

[0003] Penetration-coupled ultrasonic imaging (PCI) is widely used in aerospace, rail transportation, and other fields. However, PCI methods are currently mostly designed for planar or rotating bodies. For large and complex curved surfaces, PCI methods present the following problems: it is difficult to obtain the true contour of the workpiece being inspected, thus affecting the accuracy of the inspection results; it is difficult to obtain the motion trajectories of the two robotic arms simultaneously, making it difficult to verify whether the line connecting the centers of the excitation and receiving probes is on the same straight line as the normal to the workpiece; and it is difficult to ensure that the distance between the end faces of the excitation and receiving probes remains constant, thus affecting the inspection results. Summary of the Invention

[0004] This invention provides a method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging detection, in order to overcome the above-mentioned technical problems.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] A method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging detection includes the following steps:

[0007] S1: Position the bidirectional curved surface detection device, mount a laser sensor on the main robot arm, the laser sensor and the main robot arm are kept relatively stationary, mount a receiving probe on the slave robot arm, the receiving probe and the slave robot arm are kept relatively stationary;

[0008] S2: The laser sensor moves along the Y-axis of the main manipulator base with the main manipulator base coordinate system to scan the component to be detected and establish a three-dimensional mathematical model of the spatial contour of the component to be detected.

[0009] S3: The laser sensor is used to scan within the spatial contour of the component to be detected, and trajectory point interpolation is performed on the three-dimensional mathematical model. The scanning result is sent to the receiving probe to obtain the coordinates of the interpolation points in the three-dimensional mathematical model.

[0010] S4: Calculate the normal vector, X-direction vector, and Y-direction vector of the interpolation point in the workpiece coordinate system where the component to be detected is located, to obtain the pose of the component to be detected;

[0011] S5: Transform the workpiece coordinate system where the component to be detected is located, and use the pose of the component to be detected to calculate the pose matrix of the master manipulator and the slave manipulator relative to their respective manipulator base coordinate systems;

[0012] S6: Based on the pose matrices of the master robot and the slave robot relative to their respective robot base coordinate systems, the angular trajectories and poses of the master robot and the slave robot are obtained using the inverse kinematics method.

[0013] Furthermore, the specific method for the laser sensor described in S2 to scan the component to be detected and establish a three-dimensional mathematical model of the component to be detected, moving along the Y-axis of the main manipulator base with the main manipulator base coordinate system, is as follows:

[0014] S2.1: An encoder is installed in the bottom Y-axis slider of the main manipulator base. The bottom Y-axis slider drives the main manipulator to move along the Y-axis track direction of the coordinate system of the main manipulator base to scan the component to be detected.

[0015] S2.2: The encoder is used to trigger the laser sensor at equal intervals to collect point cloud data of the component to be detected;

[0016] S2.3: Using a surface fitting method, the point cloud data is extracted as the original data for fitting by taking points at equal intervals;

[0017] S2.4: Based on the spatial shape of the original data, in the workpiece coordinate system, with the Z-axis as the dependent variable and the X-axis and Y-axis as independent variables, call the polyfit function in MATLAB to fit the original data into a three-dimensional mathematical model of the component to be inspected.

[0018] Furthermore, in S5, the specific method for transforming the workpiece coordinate system where the component to be detected is located, and calculating the pose matrix of the master robot and the slave robot relative to their respective robot base coordinate systems using the pose of the component to be detected, is as follows:

[0019] S5.1: The pose matrix of the main manipulator relative to the main manipulator base coordinate system is calculated by transforming the pose points of the component to be tested in the workpiece coordinate system. The calculation method is as follows:

[0020]

[0021] In the formula, MA P is the pose matrix of the main manipulator relative to the coordinate system of the main manipulator base. This represents the transformation matrix between the base coordinate system and the world coordinate system of the main manipulator. S P is the pose matrix of the component to be tested in the workpiece coordinate system;

[0022] S5.2: The pose matrix of the slave manipulator relative to the main manipulator base coordinate system is obtained by translating the pose matrix of the main manipulator relative to the main manipulator base coordinate system by the thickness of the component to be detected along its Z vector direction, and then rotating it 180° around its own X-axis. The calculation method is as follows:

[0023]

[0024] In the formula, MA_B P is the pose matrix of the slave manipulator relative to the coordinate system of the master manipulator base, rA ** Let rA be a third-order attitude transformation matrix. 11 rA 21 rA 21 Let rA be the X-direction vector of any point on the three-dimensional mathematical model. 12 rA 22 rA 32 Let rA be the Y-direction vector of any point on the three-dimensional mathematical model. 13 rA 23 rA 33 δ is the normal vector of any point on the three-dimensional mathematical model; XA, YA, and ZA are the interpolated position coordinates, and δ is the thickness of the workpiece.

[0025] S5.3: The above MA_B P is converted into the pose matrix of the slave manipulator relative to the slave manipulator base coordinate system, and the calculation method is as follows:

[0026]

[0027] In the formula, MBP is the pose matrix of the slave manipulator relative to the coordinate system of the slave manipulator base. The transformation matrix between the robot arm base coordinate system and the world coordinate system is... Let be the transformation matrix between the main manipulator's base coordinate system and the world coordinate system. MA_B P is the pose matrix of the slave manipulator relative to the coordinate system of the master manipulator base.

[0028] Furthermore, S2 also includes:

[0029] S2.5: The accuracy of the three-dimensional mathematical model of the component to be tested is determined using the coefficient of determination. The method for calculating the coefficient of determination is as follows:

[0030]

[0031] Among them, R 2 Z is the coefficient of determination. i The Z-axis coordinate of the current point on the component to be detected is [value]. The average value of the Z-axis coordinates of all points on the component to be detected. The data for the current point is obtained by fitting the three-dimensional mathematical model.

[0032] A threshold N for the determination coefficient is set. When the calculated value of the determination coefficient is greater than the threshold N, the three-dimensional mathematical model of the component to be detected is determined to meet the requirements.

[0033] Furthermore, the angular trajectory mentioned in S6 includes a graph showing the change in angle and angular velocity of each axis of the master manipulator and the slave manipulator over time.

[0034] Beneficial Effects: This invention discloses a method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging. It utilizes a laser sensor to acquire point cloud data of the complex curved surface being inspected, and employs a surface fitting method to establish a three-dimensional mathematical model of the complex curved surface component. This method can more accurately obtain the true contour of the inspected curved surface component, effectively reducing errors caused by operational issues during clamping and calibration. Based on the transformation of the workpiece coordinate system and the base coordinate system, the pose matrix and angular trajectory of both manipulators are obtained, enabling simultaneous acquisition of the motion trajectories of both manipulators and completing the coordinated detection of both manipulators. During the detection process, the angular trajectory data of both manipulators can more easily and accurately ensure that the probe axes on both manipulators are collinear and the spacing remains constant, improving the accuracy of the detection. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart of a method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging detection disclosed in an embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the penetrating air-coupled ultrasonic imaging detection principle disclosed in an embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of the penetrating air-coupled ultrasonic imaging detection process disclosed in an embodiment of the present invention;

[0039] Figure 4 This is a schematic diagram of the penetrating air-coupled ultrasonic imaging detection system disclosed in an embodiment of the present invention;

[0040] Figure 5 This is a schematic diagram of the attitude interpolation effect disclosed in an embodiment of the present invention;

[0041] Figure 6 This is a physical diagram of the detection process disclosed in an embodiment of the present invention.

[0042] In the diagram: 1. Main robot arm; 2. Laser sensor; 3. Main robot arm base; 4. Slave robot arm; 5. Receiving probe; 6. Component to be tested. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] This embodiment provides a method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging detection, such as... Figure 1 As shown, it includes the following steps:

[0045] S1: Position the bidirectional curved surface detection device, mount a laser sensor 2 on the main robot arm 1, the laser sensor 2 and the main robot arm 1 remain relatively stationary, mount a receiving probe 5 on the slave robot arm 4, the receiving probe 5 and the slave robot arm 4 remain relatively stationary;

[0046] S2: The laser sensor 2 moves along the Y-axis of the main manipulator base 3 with the manipulator base 3 to scan the component 6 to be detected, establishing a three-dimensional mathematical model of the spatial contour of the component 6 to be detected, such as... Figure 5 As shown, in this embodiment, the spatial contour of the component to be detected is a slowly varying function, and the three-dimensional mathematical model is a bivariate quadratic function.

[0047] S3: The laser sensor 2 is used to scan within the spatial contour of the component 6 to be detected, and the trajectory point interpolation of the three-dimensional mathematical model is performed. The scanning result is sent to the receiving probe 5 to obtain the coordinates of the interpolation points in the three-dimensional mathematical model.

[0048] S4: Calculate the normal vector, X-direction vector, and Y-direction vector of the interpolation point in the workpiece coordinate system where the component to be detected 6 is located, to obtain the pose of the component to be detected 6;

[0049] The normal vector, X-direction vector, and Y-direction vector of the interpolation point are calculated as follows:

[0050]

[0051]

[0052]

[0053] In the formula, Let be the normal vector of the interpolation point. Let X be the X-direction vector of the interpolation point. The Y-direction vector of the interpolation point;

[0054] like Figure 4As shown in the figure, {W} represents the world coordinate system, {MA} and {MB} represent the coordinate system of the main robot base and the coordinate system of the slave robot base, respectively. The origin of the coordinate system of the main robot base is the center point of the bottom base of the main robot 1. The movement direction of the main robot 1 is the Y-axis direction of the coordinate system of the main robot base, the direction perpendicular to the ground is the Z-axis direction, and the X-axis direction is perpendicular to the YZ plane. The origin of the coordinate system of the slave robot base is the center point of the bottom base of the slave robot 4. The movement direction of the slave robot 4 is the Y-axis direction of the coordinate system of the main robot base, the direction perpendicular to the ground is the Z-axis direction, and the X-axis direction is perpendicular to the YZ plane. {S} represents the workpiece coordinate system. The origin of the workpiece coordinate system is the center point of the component to be detected. The Z-axis direction of the workpiece coordinate system is parallel to the laser emission direction of the laser sensor 2. The X-axis direction of the workpiece coordinate system is perpendicular to the ground, and the Y-axis direction of the workpiece coordinate system is perpendicular to the XZ plane.

[0055] S5: Transform the workpiece coordinate system where the component to be tested 6 is located, and use the pose of the component to be tested 6 to calculate the pose matrix of the master robot 1 and the slave robot 4 relative to their respective robot base coordinate systems.

[0056] S6: Based on the pose matrices of the master robot 1 and the slave robot 4 relative to their respective robot base coordinate systems, the angular trajectories and poses of the master robot 1 and the slave robot 4 are obtained using the inverse kinematics method.

[0057] This embodiment is for... Figure 4 The 14-axis dual-manipulator linkage detection system shown in this embodiment employs a method for synchronous pose calculation of the two manipulators. In this method, a laser sensor is used to collect point cloud data of the complex curved surface being detected. A three-dimensional mathematical model of the complex curved surface component is established using a surface fitting method. This allows for a more accurate determination of the true contour of the surface component being detected, effectively reducing errors caused by operational issues during clamping and calibration. Based on the transformation of the workpiece coordinate system and the base coordinate system, the pose matrix and angular trajectory of both manipulators are obtained. This enables the simultaneous acquisition of the motion trajectories of both manipulators, completing the linkage detection of both manipulators. During the detection process, the angular trajectory data of both manipulators allows for a simpler and more accurate method to ensure that the probe axes on both manipulators are collinear and the spacing remains constant, thus improving the accuracy of the detection.

[0058] like Figure 2The diagram shown is a schematic of the principle of penetrating air-coupled ultrasonic imaging disclosed in this embodiment. Penetrating air-coupled ultrasonic imaging can be understood as a cross-sectional view of the object under test. The probe scans the surface of the workpiece with an "S" shaped trajectory and collects a set of A signals at certain distance intervals. The points form lines, and the lines cover the entire workpiece. Penetrating air-coupled ultrasonic imaging can reflect the condition of different depths inside the test piece and intuitively make qualitative judgments and locate defects.

[0059] Application number "202321649491.3" discloses a bidirectional curved surface detection device. This invention is based on a detection method for a bidirectional curved surface detection device disclosed in that patent.

[0060] like Figure 3 The diagram shows a schematic of a penetrating air-coupled ultrasonic imaging detection process. The following conditions should be met during detection: the distance d between the end faces of the excitation probe and the receiving probe of the laser sensor remains unchanged; the distance d1 between the excitation probe and the workpiece surface remains unchanged and is perpendicular to any point on the workpiece surface; the axes of the two probes are always on the same straight line.

[0061] like Figure 4 and Figure 6 As shown, the component 6 to be tested is fixed vertically in the middle of the system. The main robot 1 and the slave robot 4 are respectively located on both sides of the component 6. The robot base 3 can translate along the Y-axis of the robot base coordinate system. During testing, the main robot holds a laser sensor and emits a detection laser perpendicularly along the normal to the workpiece surface. The slave robot holds a receiving probe and receives ultrasonic signals along the incident axis, while the distance between the robot and the workpiece surface remains constant. The two robots work together to perform contour scanning on each row of the workpiece to be tested, repeating the scan cyclically.

[0062] In a specific embodiment, the laser sensor 2 in S2 moves along the Y-axis of the main manipulator base 3 to scan the component 6 to be detected and establish a three-dimensional mathematical model of the component 6 to be detected. The specific method is as follows:

[0063] S2.1: An encoder is installed in the bottom Y-axis slider of the robot base 3. The bottom Y-axis slider drives the main robot 1 to move along the Y-axis track of the coordinate system of the main robot base to scan the component 6 to be detected. During scanning, the measurement line should be kept as vertical as possible. When the surface is large, multiple scans and data splicing are required.

[0064] S2.2: The encoder triggers the laser sensor at equal intervals to collect point cloud data of the component to be detected. The component to be detected is placed perpendicular to the ground, and the laser line must also be perpendicular to the ground. The main robot 1 vertically holds the laser sensor 2, and the base of the main robot 1 moves parallel to the slider along the Y-axis. During the laser scanning process, the slave robot 4 remains stationary.

[0065] S2.3: Using a surface fitting method, the point cloud data is extracted as the original data for fitting by taking points at equal intervals;

[0066] S2.4: Based on the spatial shape of the original data, a surface approximation method is used. In the workpiece coordinate system, the Z-axis is used as the dependent variable, and the X-axis and Y-axis are used as independent variables. The polyfit function in MATLAB is called to fit the original data into a bivariate quadratic function form Z = F(x,y), thus obtaining the three-dimensional mathematical model of the component 6 to be detected. In this embodiment, as shown... Figure 5 As shown, when the contour of the component 6 to be detected is a slowly varying surface, the three-dimensional mathematical model of the component 6 to be detected is:

[0067] Z = F(x, y) = a + bx + cy + dx 2 +exy+fy 2

[0068] =2.061e2+1.496e-2x-3.303y+1.006e-06x 2 +3.412e-5xy+1.194e-3y 2

[0069] In a specific embodiment, the method for transforming the workpiece coordinate system where the component to be detected 6 is located in step S5, and calculating the pose matrix of the master robot 1 and the slave robot 4 relative to their respective robot base coordinate systems using the pose of the component to be detected 6, is as follows:

[0070] S5.1: The pose matrix of the main manipulator 1 relative to the coordinate system of the main manipulator base is calculated by transforming the pose points of the component 6 to be tested in the workpiece coordinate system. The calculation method is as follows:

[0071]

[0072] In the formula, MA P is the pose matrix of the main manipulator 1 relative to the coordinate system of the main manipulator base. This represents the transformation matrix between the base coordinate system and the world coordinate system of the main manipulator. S P is the pose matrix of the component to be tested in the workpiece coordinate system;

[0073] in, MA Let P be a fourth-order square matrix, which, when expanded, is represented as:

[0074]

[0075] Where rA 11 rA 21 rA 21 Let rA be the X-direction vector of any point on the three-dimensional mathematical model. 12 rA 22 rA 32 Let rA be the Y-direction vector of any point on the three-dimensional mathematical model. 13 rA 23 rA 33 XA, YA, and ZA are the normal vectors of any point on the 3D mathematical model; XA, YA, and ZA are the interpolated position coordinates.

[0076] S5.2: The pose matrix of the main manipulator 1 relative to the main manipulator base coordinate system is obtained by translating the thickness of the component to be detected along its Z vector direction, and then rotating it 180° around the X-axis of the pose matrix of the main manipulator base coordinate system. The calculation method is as follows:

[0077]

[0078] In the formula, MA_B P is the pose matrix of the slave manipulator 4 relative to the coordinate system of the main manipulator base, rA ** Let rA be the third-order attitude transformation matrix. 11 rA 21 rA 21 Let rA be the X-direction vector of any point on the three-dimensional mathematical model. 12 rA 22 rA 32 Let rA be the Y-direction vector of any point on the three-dimensional mathematical model. 13 rA 23 rA 33 δ is the normal vector of any point on the three-dimensional mathematical model; XA, YA, and ZA are the interpolated position coordinates, and δ is the thickness of the workpiece.

[0079] S5.3: The above MA_B P is converted into the pose matrix of the slave manipulator 4 relative to the slave manipulator base coordinate system, and the calculation method is as follows:

[0080]

[0081] In the formula, MBP is the pose matrix of the slave manipulator 4 relative to the coordinate system of the slave manipulator base. The transformation matrix between the robot arm base coordinate system and the world coordinate system is... Let be the transformation matrix between the main manipulator's base coordinate system and the world coordinate system. MA_B P is the pose matrix of the slave manipulator 4 relative to the coordinate system of the master manipulator base.

[0082] In a specific embodiment, S2 further includes:

[0083] S2.5: The accuracy of the three-dimensional mathematical model of the component to be tested is determined using the coefficient of determination. The method for calculating the coefficient of determination is as follows:

[0084]

[0085] Among them, R 2 Z is the coefficient of determination. i Let Z be the Z-axis coordinate of the current point on the component (6) to be detected, where Z is the average Z-axis coordinate of all points on the component (6) to be detected. The data for the current point is obtained by fitting the three-dimensional mathematical model.

[0086] A threshold N for the determination coefficient is set. When the calculated value of the determination coefficient is greater than the threshold N, the three-dimensional mathematical model of the component to be detected is determined to meet the requirements.

[0087] In a specific embodiment, the angular trajectories of the two robotic arms in S5 include the image of the change of the angle of each axis of the two robotic arms over time and the image of the change of the angular velocity of each axis of the two robotic arms over time.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for synchronous pose calculation of dual robotic arms based on penetrating air-coupled ultrasonic imaging detection, characterized in that, Includes the following steps: S1: Position the bidirectional curved surface detection device, clamp a laser sensor (2) on the main robot (1), the laser sensor (2) and the main robot (1) remain relatively stationary, clamp a receiving probe (5) on the slave robot (4), the receiving probe (5) and the slave robot (4) remain relatively stationary; S2: The laser sensor (2) moves along the Y-axis of the coordinate system of the main manipulator base (3) to scan the component to be detected (6) and establish a three-dimensional mathematical model of the spatial contour of the component to be detected (6); S3: The laser sensor (2) is used to scan inside the spatial contour of the component to be detected (6), and the trajectory point interpolation of the three-dimensional mathematical model is performed. The scanning result is sent to the receiving probe (5) to obtain the coordinates of the interpolation points in the three-dimensional mathematical model. S4: Calculate the normal vector, X-direction vector and Y-direction vector of the interpolation point in the workpiece coordinate system where the component to be tested (6) is located, and obtain the pose of the component to be tested (6); S5: Transform the workpiece coordinate system where the component to be tested (6) is located, and use the pose of the component to be tested (6) to calculate the pose matrix of the master manipulator (1) and the slave manipulator (4) relative to their respective manipulator base coordinate systems; S6: Based on the pose matrices of the master manipulator (1) and the slave manipulator (4) relative to their respective manipulator base coordinate systems, the angle trajectory and pose of the master manipulator (1) and the slave manipulator (4) are obtained by inverse kinematics. The laser sensor (2) mentioned in S2 moves along the Y-axis of the main manipulator base (3) to scan the component to be detected (6) and establish a three-dimensional mathematical model of the component to be detected (6). The specific method is as follows: S2.1: An encoder is installed in the bottom Y-axis slider of the main manipulator base (3). The bottom Y-axis slider drives the main manipulator (1) to move along the Y-axis track direction of the coordinate system of the main manipulator base to scan the component to be detected (6). S2.2: The encoder is used to trigger the laser sensor (2) at equal intervals to collect the point cloud data of the component to be detected (6); S2.3: Using a surface fitting method, the point cloud data is extracted as the original data for fitting by taking points at equal intervals; S2.4: Based on the spatial shape of the original data, in the workpiece coordinate system, with the Z-axis as the dependent variable and the X-axis and Y-axis as independent variables, call the polyfit function in MATLAB to fit the original data into a three-dimensional mathematical model of the component to be detected (6). In S5, the workpiece coordinate system where the component to be detected (6) is located is transformed, and the pose matrix of the master robot (1) and the slave robot (4) relative to their respective robot base coordinate systems is calculated using the pose of the component to be detected (6): S5.1: The pose matrix of the main manipulator (1) relative to the main manipulator base coordinate system is calculated by transforming the pose points of the component to be tested (6) in the workpiece coordinate system. The calculation method is as follows: In the formula, Let be the pose matrix of the main manipulator (1) relative to the coordinate system of the main manipulator base. This represents the transformation matrix between the base coordinate system and the world coordinate system of the main manipulator. Let be the pose matrix of the component to be detected in the workpiece coordinate system; S5.2: The pose matrix of the main manipulator (1) relative to the coordinate system of the main manipulator base is obtained by translating the thickness of the component to be detected along its Z vector direction and then rotating it 180° around its own X axis. The calculation method is as follows: In the formula, Let be the pose matrix of the slave manipulator (4) relative to the coordinate system of the master manipulator base. Here is a third-order attitude transformation matrix, where Let X be the X-direction vector of any point on the three-dimensional mathematical model. Let Y be the Y-direction vector of any point on the three-dimensional mathematical model. Let XA be the normal vector of any point on the 3D mathematical model; XA, YA, and ZA are the interpolated position coordinates. The thickness of the workpiece; S5.3: The above The pose matrix of the manipulator (4) relative to the manipulator base coordinate system is converted into the following calculation method: In the formula, Let be the pose matrix of the slave manipulator (4) relative to the slave manipulator base coordinate system. The transformation matrix between the robot arm base coordinate system and the world coordinate system is... Let be the transformation matrix between the main manipulator's base coordinate system and the world coordinate system. Let be the pose matrix of the slave manipulator (4) relative to the coordinate system of the master manipulator base.

2. The method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging detection according to claim 1, characterized in that, S2 also includes: S2.5: The accuracy of the three-dimensional mathematical model of the component to be tested is determined using the coefficient of determination. The method for calculating the coefficient of determination is as follows: in, As the coefficient of determination, The Z-axis coordinate of the current point on the component (6) to be detected is... The average value of the Z-axis coordinates of all points on the component to be tested (6) is given. The data for the current point is obtained by fitting the three-dimensional mathematical model. A threshold N for the determination coefficient is set. When the calculated value of the determination coefficient is greater than the threshold N, the three-dimensional mathematical model of the component to be detected is determined to meet the requirements.

3. The method for synchronous pose calculation of dual manipulators based on penetrating air-coupled ultrasonic imaging detection according to claim 1, characterized in that, The angular trajectory described in S6 includes a graph of the changes in angle and angular velocity of each axis of the master manipulator (1) and the slave manipulator (4) over time.

Images