A method for docking large rigid components based on fixed-end images

Abstract

The present invention discloses a method for docking large rigid components based on the fixed-end image. The large rigid components include a fixed end and a mobile end; the mobile end is installed on a docking mobile vehicle through a posture adjustment platform; a camera is installed on the docking mobile vehicle, and the lens is directly facing the docking surface of the fixed end; the specific steps of the method are as follows: 1. Calibrating the internal parameters of the vision system; 2. Calibrating the external parameters of the vision system; 3. Calibrating the posture of the mobile end; 4. Positioning and guiding the docking; the proposed method solves the problem that the docking pose cannot be solved when the existing vision technology cannot simultaneously collect the images of the fixed end and the mobile end in the docking scenario.

Description

Technical Field

[0001] The present invention belongs to the field of industrial automation. Specifically, it relates to a method for docking large rigid components based on fixed-end images. Background Art

[0002] The assembly and docking of rigid body equipment is one of the important links in the fields of aerospace, industrial manufacturing, etc. Currently, most docking work is still manually completed by workers, resulting in high labor intensity and low efficiency. At the same time, it is difficult to control the relative pose between the two rigid bodies during the docking process during manual operation, thus greatly affecting the docking accuracy.

[0003] In response to the problems existing in manual operation, the use of laser measurement technology or indoor GPS for positioning and guiding technology has been proposed. However, there are still problems with the above two methods: Laser measurement technology has high precision and strong anti-interference ability, but the equipment is expensive. Indoor GPS has low precision and cannot meet the requirements of high-precision docking and assembly.

[0004] Compared with these two methods, vision technology reduces the cost and operation complexity while ensuring accuracy. Therefore, the use of vision technology to guide the docking of rigid body assemblies has been increasingly widely applied, and many patents on the application of vision technology in the docking of large rigid bodies have been published.

[0005] For example, for the Chinese invention patent with the application number: CN201610309533.7, "A Method for Measuring the Position Deviation of Single and Binocular Vision for Section Docking", two independent monocular vision systems are used to collect the artificial targets at the mobile end and the fixed end respectively, and the docking pose is obtained by combining the relationship between the two monocular cameras calibrated in advance and the three-dimensional coordinate points of the artificial targets.

[0006] For the Chinese invention patent with the application number: CN202010760878.0, "An Automated Missile Horizontal Loading System Based on Vision Alignment and Its Operation Method", a photogrammetry system is used to calibrate the relationship between the artificial target and the circular hole on the docking surface in advance, and a binocular vision system is used to simultaneously obtain the target image information on the mobile end and the fixed end to calculate the docking pose.

[0007] For the Chinese invention patent with the application number: CN 201910339187.0, "A Vision Measurement System and Method for the Relative Pose of Docking and Assembly of Large Components", two binocular vision systems are used for rough positioning and fine positioning respectively to improve the docking accuracy.

[0008] For the Chinese invention patent with the application number: CN 202011156320.8, "An Automatic Docking Method for Large Rigid Components Based on Monocular Vision", a monocular vision system is used to simultaneously collect the artificial target images at the mobile end and the fixed end, and the docking pose of the two cylinder sections can be calculated.

[0009] The above patents can all complete the process of relative pose measurement and guided automatic assembly docking of two rigid body equipments under normal circumstances. However, when the on-site conditions cannot ensure that the vision system can simultaneously collect images of the fixed end and the mobile end, the above methods cannot meet the docking requirements of the positioning and guiding components. Summary of the Invention

[0010] Aiming at the problem that the existing vision technology cannot solve the docking pose when it cannot simultaneously collect the fixed-end image and the mobile-end image, the present invention provides a method for docking large rigid body components based on the fixed-end image.

[0011] The specific technical solution of the present invention is as follows:

[0012] A method for docking large rigid body components based on the fixed-end image, the large rigid body components include a fixed end and a mobile end; the mobile end is installed on the docking mobile vehicle through a pose adjustment platform; the camera is installed on the docking mobile vehicle, and the lens is facing the docking surface of the fixed end; the specific implementation steps of the docking method are as follows:

[0013] Step 1: Calibrate the internal parameters of the vision system;

[0014] Step 2: Calibrate the external parameters of the vision system;

[0015] Use the calibration board to calibrate the external parameters of the camera, and obtain the rotation and translation relationship between the camera coordinate system and the pose adjustment platform coordinate system is the rotation matrix from the camera coordinate system to the pose adjustment platform coordinate system, is the translation vector from the camera coordinate system to the pose adjustment platform coordinate system;

[0016] Step 3: Calibrate the pose of the mobile end;

[0017] Step 3.1: Install the mobile end feature identifier in front of the docking end face of the mobile end to ensure that the mobile end feature identifier is within the camera's field of view and the imaging is clear;

[0018] Step 3.2: Establish the mobile end coordinate system, and obtain the coordinates of the mobile end feature identifier in the mobile end coordinate system according to the relationship between the feature points in the mobile end feature identifier and the mobile end coordinate system

[0019] Step 3.3: According to the coordinates of the mobile end feature identifier in the mobile end coordinate system Combined with the internal parameters of the vision system obtained in Step 1, use the PnP algorithm to calculate the transformation matrix between the mobile end coordinate system and the camera coordinate system Obtain the coordinates P of the mobile end feature identifier in the camera coordinate system m-c ;

[0020] Step 3.4: Combine the calibration results of the external parameters of the vision system in Step 2 Solve for the coordinates P of the mobile feature identifier in the coordinate system of the pose adjustment platform m-b ;

[0021] Step 4: Positioning and guiding for docking;

[0022] Step 4.1: Establish a fixed-end coordinate system and preset the pose relationship between the mobile coordinate system and the fixed-end coordinate system in the docking state

[0023] Step 4.2: According to the relationship between the feature points in the fixed-end feature identifier and the fixed-end coordinate system, obtain the coordinates of the fixed-end feature identifier in the fixed-end coordinate system

[0024] Step 4.3: Utilize the coordinates of the fixed-end feature identifier in the fixed-end coordinate system and the internal parameters of the vision system obtained in Step 1, and use the PnP algorithm to calculate the transformation matrix between the fixed-end coordinate system and the camera coordinate system

[0025] Step 4.4: According to the pose relationship preset in Step 4.1 Solve for the coordinates of the mobile feature identifier in the coordinate system of the pose adjustment platform in the docking state

[0026] Step 4.5: According to the coordinates P of the mobile feature identifier in the coordinate system of the pose adjustment platform in Step 3.4 m-b , the coordinates of the mobile feature identifier in the coordinate system of the pose adjustment platform in the docking state in Step 4.4 Calculate the relative pose R b , t b , and this data is the docking pose;

[0027] Step 4.6: Resolve the docking pose to the six degrees of freedom of the pose adjustment platform to guide the docking and pose adjustment.

[0028] Furthermore, the specific implementation process of the above Step 4.1 is as follows:

[0029] Step 4.1.1: Establish a fixed-end coordinate system, obtain the structural relationship between the feature points on the mobile feature identifier and the fixed end in the docking state through simulation means, and obtain the coordinates of all feature points in the mobile feature identifier in the fixed-end coordinate system during docking

[0030] Step 4.1.2: Calculate the conversion relationship between the coordinate and the coordinate , which is the conversion relationship between the fixed-end coordinate system and the mobile coordinate system in the docking state Its formula is:

[0031] Further, the mobile - end feature identifier is fixed in front of the mobile - end docking surface by an L - shaped bracket; one end of the L - shaped bracket is fixed to the mobile end, and the mobile - end feature identifier is set at the other end.

[0032] Further, the above - mentioned fixed - end feature identifier is a fixed - end target with feature points bonded to the fixed - end docking surface, or the inherent feature points on the fixed - end docking surface, and the inherent feature points are regular - shaped protrusions or depressions or screws;

[0033] The mobile - end feature identifier is a mobile - end target, or an inherent feature set at the front end of the mobile end.

[0034] Further, in step 4.4, the coordinates The specific solving process is as follows:

[0035] Step 4.4.1: Use the result obtained in step 4.1 Calculate the coordinates of the feature points of the mobile - end feature identifier in the fixed - end coordinate system in the docking state as:

[0036] Step 4.4.2: Use the result obtained in step 4.3 Calculate the coordinates of the feature points of the mobile - end feature identifier in the pose - adjustment platform coordinate system in the docking state as:

[0037]

[0038] Step 4.4.3: Substitute formula (1) into formula (2) to get The specific calculation formula is:

[0039] Further, in step 3.3, the specific calculation formula of P m-c is: In step 3.4, the specific calculation formula of P m-b is: In step 4.5, the specific calculation formulas of R b and t b are:

[0040] Further, the specific solving process of the above - mentioned step 4.5 is:

[0041]

[0042] Among them, r 11 to r 33 represent the rotation matrix R b, t1 to t3 represent the translation vector t b ; Using the relationship between Euler angles and rotation matrices, and the attitude adjustment sequence, calculate the pitch angle θ, yaw angle ψ, and roll angle φ based on the attitude adjustment platform coordinate system.

[0043] Furthermore, the internal parameters of the vision system in step 1 above include: the ratio of the unit pixel size values in the X and Y axis directions to the focal length in the camera coordinate system (f x , f y ), and the coordinates (u0, v0) of the intersection point of the optical axis of the camera and the image plane in the camera coordinate system;

[0044] The specific calibration process is as follows:

[0045] Step 1.1: Place the internal parameter calibration board within the camera field of view of the vision system;

[0046] Step 1.2: Move the internal parameter calibration board within the camera field of view and traverse the entire camera field of view to collect multiple images of the internal parameter calibration board;

[0047] Step 1.3: Process the multiple images of the internal parameter calibration board collected to obtain the visual coordinates (u, v) of each feature point in the internal parameter calibration board;

[0048] Step 1.4: Establish a world coordinate system, and use the known actual position relationships of each feature point of the internal parameter calibration board to obtain the world coordinates (X w , Y w , Z w ) of each feature point of the internal parameter calibration board;

[0049] Step 1.5: Perform data fitting on the visual coordinates of each feature point obtained in step 1.3 and the world coordinates of each feature point obtained in step 1.4 to solve for the internal parameters of the vision system;

[0050] The specific formula for the internal parameters of the vision system is:

[0051]

[0052] In the formula, M is the conversion relationship between the vision coordinate system and the world coordinate system.

[0053] Furthermore, the specific calibration process of step 2 above is as follows:

[0054] Step 2.1: Fix the external parameter calibration board to the attitude adjustment platform to determine the attitude adjustment platform coordinate system;

[0055] Step 2.2: Control the pose adjustment platform to move the external parameter calibration board in one direction by at least two positions. In this process, the external parameter calibration board rotates and transforms into B1 in the pose adjustment platform coordinate system and A1 in the vision coordinate system. Let the conversion relationship from the camera coordinate system to the pose adjustment platform coordinate system be X, thus constructing the first typical hand-eye calibration equation: A1X = XB1; When the vision system and the control pose adjustment platform are relatively fixed, is the rotation matrix from the camera coordinate system to the pose adjustment platform coordinate system, is the translation vector from the camera coordinate system to the pose adjustment platform coordinate system;

[0056] Step 2.3: Control the pose adjustment platform to move in another direction by at least two positions. In this process, the external parameter calibration board rotates and transforms into B2 in the pose adjustment platform coordinate system and A2 in the camera coordinate system. Let the conversion relationship from the camera coordinate system to the pose adjustment platform coordinate system be X, thus constructing the second typical hand-eye calibration equation: A2X = XB2;

[0057] Step 2.4: Solve for X according to the two hand-eye calibration equations to obtain Thereby completing the external parameter calibration of the vision system.

[0058] The beneficial effects of the present invention are as follows:

[0059] 1. The present invention uses the vision system to pre-calibrate the pose relationship between the mobile end and the pose adjustment platform, and then real-time collects the feature marks of the fixed end to obtain the docking pose, guiding the rigid body components of the mobile end to automatically dock, solving the problem that the docking pose cannot be solved when the existing method cannot collect the images of the fixed end and the mobile end simultaneously in the docking scenario.

[0060] 2. The present invention is also applicable to general docking scenarios where the images of the fixed end and the mobile end can be obtained simultaneously. The pose of the mobile end is pre-calibrated using the vision system installed near the mobile end, and then the image of the fixed end is collected in real time to obtain the docking pose.

[0061] 3. The present invention uses a monocular vision measurement system to guide the docking pose of large rigid body components. Compared with binocular and multiocular vision systems, it has a simple structure, a low complexity in the calibration process, higher stability, and is easier to operate and maintain. Description of the Drawings

[0062] Figure 1 is the implementation schematic diagram of the docking process of the present invention.

[0063] Figure 2 is the implementation schematic diagram of step 4.1 in the method of the present invention.

[0064] Figure 3 is the schematic diagram of the internal parameter calibration board and the external parameter calibration board;

[0065] Figure 4 Schematic diagrams of the mobile target and the fixed target

[0066] The reference signs in the drawings are as follows:

[0067] 1 - Fixed end, 2 - Fixed target, 3 - L-shaped bracket, 4 - Mobile end, 5 - Camera, 6 - Pose adjustment platform, 7 - Mobile docking vehicle, 8 - Transport vehicle Specific implementation manners

[0068] The method of the present invention will be further described in detail below with reference to the drawings and embodiments

[0069] When the diameters of the fixed end and the mobile end are quite different, or when there is insufficient side space between the fixed end and the mobile end, collecting images of both ends by one camera may result in unclear imaging and the inability to solve the pose. The present invention uses vision technology to collect the fixed end image in real time to obtain the docking pose, thereby guiding the mobile end and realizing automatic docking

[0070] The main implementation steps of the method of the present invention include: calibration of the internal parameters of the vision system, calibration of the external parameters of the vision system, and calibration of the mobile end pose for positioning and guiding docking

[0071] As Figure 1 shown, when implementing this method, the following preparations need to be made

[0072] 1. Prepare a monocular vision system, which specifically includes a camera 1, a lens, and a light source. When implementing the method of the present invention, the monocular vision system is installed on the mobile docking vehicle 7 that drives the mobile end 4 to move

[0073] 2. Select a feature identifier on the docking surface of the fixed end. This feature identifier can be the fixed target 2 adhered to the docking surface of the fixed end, or a protrusion, a pit, or a screw with a regular shape on the docking surface of the fixed end; in this embodiment, the feature identifier is the fixed target, and in this embodiment, the fixed end is installed on the transport vehicle 8

[0074] 3. Prepare a mobile end feature identifier; in this embodiment, the mobile end feature identifier is the mobile target. When in use, the mobile target can be installed on the mobile end 4 through an L-shaped bracket 3, and the mobile end 4 is installed on the mobile docking vehicle 7 through the pose adjustment platform 6

[0075] When the vision system works, the fixed target is always adhered to the docking surface of the fixed end. The mobile target is installed on the mobile end during the preliminary calibration of the mobile end pose and needs to be removed when starting to guide docking

[0076] The specific implementation process of the method of the present invention is as follows

[0077] Step 1: Calibration of the internal parameters of the vision system

[0078] The internal parameters of the vision system are calibrated using a calibration board. The internal parameters of the vision system include: the ratio of the unit pixel size values in the X and Y axis directions in the camera coordinate system to the focal length (f x , f y ), and the coordinates (u0, v0) of the intersection point of the optical axis of the camera and the image plane in the camera coordinate system;

[0079] Step 1.1: Place the internal parameter calibration board as shown in Figure 1 the camera's field of view to obtain a clear image;

[0080] Step 1.2: Move the internal parameter calibration board within the camera's field of view and traverse the entire field of view of the camera to collect multiple images of the internal parameter calibration board;

[0081] Step 1.3: Process the multiple images of the internal parameter calibration board collected to obtain the visual coordinates (u, v) of each feature point on the internal parameter calibration board;

[0082] Step 1.4: Establish a world coordinate system, and use the known actual position relationships of each feature point on the internal parameter calibration board to obtain the world coordinates (X w , Y w , Z w ) of each feature point on the internal parameter calibration board;

[0083] Step 1.5: Perform data fitting on the visual coordinates of each feature point obtained in Step 1.3 and the world coordinates of each feature point obtained in Step 1.4 to solve for the internal parameters of the vision system;

[0084] The specific formula for the internal parameters of the vision system is:

[0085]

[0086] In the formula, M is the conversion relationship between the vision coordinate system and the world coordinate system.

[0087] Step 2: Calibrate the external parameters of the vision system.

[0088] The external parameters of the camera are calibrated using an external parameter calibration board to obtain the rotation and translation relationship between the camera coordinate system and the pose adjustment platform coordinate system.

[0089] The specific process of calibrating the external parameters of the vision system is as follows:

[0090] Step 2.1: Fix the external parameter calibration board to the pose adjustment platform to determine the pose adjustment platform coordinate system;

[0091] Step 2.2: Control the pose adjustment platform to move the external parameter calibration board in one direction to at least two positions. In this process, the external parameter calibration board rotates and transforms to B1 in the pose adjustment platform coordinate system and rotates and transforms to A1 in the vision coordinate system. Let the transformation relationship from the camera coordinate system to the pose adjustment platform coordinate system be X, thus constructing the first typical hand-eye calibration equation: A1X = XB1; When the vision system and the control pose adjustment platform are relatively fixed, is the rotation matrix from the camera coordinate system to the pose adjustment platform coordinate system, is the translation vector from the camera coordinate system to the pose adjustment platform coordinate system;

[0092] Step 2.3: Control the pose adjustment platform to move in another direction to at least two positions. In this process, the external parameter calibration board rotates and transforms to B2 in the pose adjustment platform coordinate system and rotates and transforms to A2 in the camera coordinate system. Let the transformation relationship from the camera coordinate system to the pose adjustment platform coordinate system be X, thus constructing the second typical hand-eye calibration equation: A2X = XB2;

[0093] Step 2.4: Solve for X according to the two hand-eye calibration equations to obtain thus completing the external parameter calibration of the vision system.

[0094] Step 3: Mobile terminal pose calibration;

[0095] When the camera can only collect fixed-end images in real time and calculate the docking pose, it is necessary to calibrate the pose of the mobile terminal in advance to obtain the pose relationship between the mobile terminal and the pose adjustment platform;

[0096] The specific process of mobile terminal pose calibration is as follows:

[0097] Step 3.1: Install the mobile terminal target on the docking surface of the mobile terminal using an L-shaped bracket, ensuring that the mobile terminal target is within the camera's field of view and the imaging is clear;

[0098] Step 3.2: Establish the mobile terminal coordinate system, and obtain the coordinates of the mobile terminal target in the mobile terminal coordinate system according to the relationship between the feature points of the mobile terminal target and the mobile terminal coordinate system

[0099] Step 3.3: According to the coordinates of the mobile terminal target in the mobile terminal coordinate system Combined with the internal parameters of the vision system obtained in Step 1, use the PnP algorithm to calculate the transformation matrix between the mobile terminal coordinate system and the camera coordinate system Obtain the coordinates P of the feature points of the mobile terminal target in the camera coordinate system m-c , P m-c The specific calculation formula is:

[0100] Step 3.4: Combine the calibration results of the external parameters of the vision system in Step 2 Solve for the coordinates P of the feature points of the mobile target in the coordinate system of the pose adjustment platform m-b , P m-b The specific calculation formula for is:

[0101] Step 4: Positioning and guiding docking

[0102] Collect the images of the fixed target with known coordinates on the fixed end in real time. Combine the data obtained from the vision system calibration and the mobile end pose calibration to solve the final moving pose of the mobile end relative to the fixed end, and guide the pose adjustment platform to adjust the pose.

[0103] The specific process of positioning and guiding docking is as follows:

[0104] Step 4.1: Establish the fixed end coordinate system and preset the pose relationship between the mobile end coordinate system and the fixed end coordinate system in the docking state The specific steps are as follows:

[0105] Step 4.1.1: As Figure 2 shown, establish the fixed end coordinate system. According to the structural relationship between the mobile target and the fixed end obtained from the simulation in the docking state, obtain the coordinates of all mobile target feature points in the fixed end coordinate system during docking

[0106] Step 4.1.2: Calculate the conversion relationship between the coordinate and the coordinate which is the conversion relationship between the fixed end coordinate system and the mobile end coordinate system in the docking state Its formula is:

[0107] Step 4.2: According to the relationship between the fixed target feature points and the fixed end coordinate system, obtain the coordinates of the fixed target in the fixed end coordinate system

[0108] Step 4.3: Use the coordinates of the fixed target in the fixed end coordinate system and the internal parameters of the vision system obtained in Step 1, and use the PnP algorithm to calculate the transformation matrix between the fixed end coordinate system and the camera coordinate system

[0109] Step 4.4: According to the pose relationship preset in Step 4.1 Solve for the coordinates of the mobile target in the coordinate system of the pose adjustment platform in the docking state

[0110] Step 4.4.1: Use the results obtained in Step 4.1 The coordinates of the feature points of the mobile target in the fixed-end coordinate system in the docking state are:

[0111] Step 4.4.2: Using the results obtained in Step 4.3 The coordinates of the feature points of the mobile target in the pose adjustment platform coordinate system in the docking state are:

[0112]

[0113] Step 4.4.3: Substitute Equation (1) into Equation (2) to obtain The specific calculation formula of

[0114] Step 4.5: According to the coordinates P m-b of the feature points of the mobile target in the pose adjustment platform coordinate system in Step 3.4, and the coordinates of the marked points of the mobile target in the pose adjustment platform coordinate system in the docking state in Step 4.4, calculate the relative pose R b and t b between the mobile end and the fixed end. This data is the docking pose, and the specific calculation formula is:

[0115] Step 4.6: Solve the docking pose to the six degrees of freedom of the pose adjustment platform to guide the docking pose adjustment. The specific process of this step is:

[0116] The docking pose matrix solved in this step is shown in Equation (3). The upper left three rows and three columns are the rotation matrix, and the right three rows and one column are the translation vector. Using the relationship between the Euler angles and the rotation matrix, according to the pose adjustment sequence, the pose adjustment sequence in this embodiment is yaw angle - pitch angle - roll angle, and the pitch angle θ, yaw angle ψ, and roll angle φ based on the pose adjustment platform coordinate system as shown in Equation (4) are solved. The displacement required to move in each direction is separated using the translation vector in Equation (3).

[0117]

[0118] In addition, it should be emphasized that: the internal parameter calibration board used in Step 1, the external parameter calibration board used in Step 2, the mobile target and the fixed target used in Steps 3 and 4 in this embodiment are similar in form; a feature point array is provided on each of them, and at least two feature points of the mobile target and the fixed target are used as coded marked points for identifying the identity of the target and sorting the non-coded marked points on the target.

Claims

1. A method for docking large rigid components based on the fixed-end image, the large rigid components including a fixed end and a mobile end; the mobile end is installed on a docking mobile vehicle through a posture adjustment platform; characterized in that, The camera is installed on the docking mobile vehicle, and the lens is facing the docking surface of the fixed end; the specific implementation steps of the docking method are as follows: Step 1: Calibrate the internal parameters of the vision system; The internal parameters of the vision system include: the ratios of the unit pixel sizes in the X and Y axis directions in the camera coordinate system to the focal length (f x , f y ), and the coordinates (u0, v0) of the intersection point of the optical axis of the camera and the image plane in the camera coordinate system; The specific calibration process is as follows: Step 1.1: Place the internal parameter calibration board within the camera's field of view of the vision system; Step 1.2: Move the internal parameter calibration board within the camera's field of view and traverse the entire camera's field of view to collect multiple images of the internal parameter calibration board; Step 1.3: Process the multiple images of the internal parameter calibration board collected to obtain the visual coordinates (u, v) of each feature point on the internal parameter calibration board; Step 1.4: Establish a world coordinate system, and use the actual positional relationship of each feature point of the known internal parameter calibration board to obtain the world coordinates (X w , Y w , Z w ) of each feature point of the internal parameter calibration board; Step 1.5: Perform data fitting on the visual coordinates of each feature point obtained in Step 1.3 and the world coordinates of each feature point obtained in Step 1.4 to solve for the internal parameters of the vision system; The specific formula for the internal parameters of the vision system is: In the formula, M is the conversion relationship between the visual coordinate system and the world coordinate system; Step 2: Calibrate the external parameters of the vision system; Use a calibration board to calibrate the external parameters of the camera, and obtain the rotation and translation relationship between the camera coordinate system and the pose adjustment platform coordinate system is the rotation matrix from the camera coordinate system to the pose adjustment platform coordinate system, is the translation vector from the camera coordinate system to the pose adjustment platform coordinate system; Step 2.1: Fix the external parameter calibration board to the pose adjustment platform to determine the pose adjustment platform coordinate system; Step 2.2: Control the pose adjustment platform to move the external parameter calibration board along one direction to at least two positions. In this process, the external parameter calibration board is rotationally transformed into B1 in the pose adjustment platform coordinate system and into A1 in the vision coordinate system. Let the conversion relationship from the camera coordinate system to the pose adjustment platform coordinate system be X, thus constructing the first typical hand-eye calibration equation: A1X = XB1; When the vision system and the control pose adjustment platform are relatively fixed, is the rotation matrix from the camera coordinate system to the pose adjustment platform coordinate system, is the translation vector from the camera coordinate system to the pose adjustment platform coordinate system; Step 2.3: Control the pose adjustment platform to move at least two positions in another direction. In this process, the external parameter calibration board rotates and transforms to B2 in the pose adjustment platform coordinate system and rotates and transforms to A2 in the camera coordinate system; let the conversion relationship from the camera coordinate system to the pose adjustment platform coordinate system be X, thus constructing the second typical hand-eye calibration equation: A2X = XB2; Step 2.4: Solve for X according to the two hand-eye calibration equations to obtain Thereby completing the external parameter calibration of the vision system; Step 3: Calibrate the pose of the mobile end; Step 3.1: Install a mobile end feature identifier in front of the docking end face of the mobile end to ensure that the mobile end feature identifier is within the camera's field of view and the imaging is clear; Step 3.2: Establish a mobile coordinate system, and obtain the coordinates of the mobile feature identifier in the mobile coordinate system according to the relationship between the feature points in the mobile feature identifier and the mobile coordinate system Step 3.3: According to the coordinates of the mobile device feature identifier in the mobile device coordinate system Combined with the internal parameters of the vision system obtained in Step 1, use the PnP algorithm to calculate the transformation matrix between the mobile device coordinate system and the camera coordinate system Obtain the coordinates P of the mobile device feature identifier in the camera coordinate system m-c ; Step 3.4: Combine the calibration result of the external parameters of the vision system in Step 2 Solve for the coordinates P of the mobile device feature identifier in the coordinate system of the posture adjustment platform m-b ; Step 4: Position and guide the docking; Step 4.1: Establish a fixed-end coordinate system and preset the pose relationship between the mobile-end coordinate system and the fixed-end coordinate system in the docking state Step 4.2: Obtain the coordinates of the fixed-end feature identifier in the fixed-end coordinate system according to the relationship between the feature points in the fixed-end feature identifier and the fixed-end coordinate system Step 4.3: Use the coordinates of the fixed-end feature identifier in the fixed-end coordinate system and the internal parameters of the vision system obtained in Step 1, and use the PnP algorithm to calculate the transformation matrix between the fixed-end coordinate system and the camera coordinate system Step 4.4: According to the pose relationship preset in Step 4.1 Solve the coordinates of the mobile - end feature identifier in the coordinate system of the pose - adjustment platform in the docking state Step 4.5: According to the coordinates P of the mobile feature identifier in the pose adjustment platform coordinate system in Step 3.4 m-b and the coordinates of the mobile feature identifier in the pose adjustment platform coordinate system in the docking state in Step 4.4 calculate the relative pose R b and t b of the mobile end and the fixed end. This data is the docking pose; Step 4.6: Solve the docking pose to the six degrees of freedom of the pose adjustment platform to guide the docking pose adjustment.

2. The method for docking large rigid components based on the fixed-end image according to claim 1, wherein: The specific implementation process of Step 4.1 is: Step 4.1.1: Establish a fixed-end coordinate system, obtain the structural relationship between the feature points on the mobile-end feature identifier and the fixed end in the docking state through simulation means, and obtain the coordinates of all feature points in the mobile-end feature identifier in the fixed-end coordinate system during docking Step 4.1.2: Calculate the coordinates and the coordinates The conversion relationship between them is the conversion relationship between the fixed-end coordinate system and the mobile-end coordinate system in the docking state The formula is as follows:

3. The method for docking large rigid components based on the fixed-end image according to claim 2, characterized in that: The mobile end feature identifier is fixed to the front of the mobile end docking face using an L-shaped bracket; one end of the L-shaped bracket is fixed to the mobile end, and the mobile end feature identifier is set at the other end.

4. The method for docking large rigid components based on the fixed-end image according to claim 3, wherein: The fixed end feature identifier is a fixed end target with feature points adhered to the fixed end docking face, or the inherent feature points on the fixed end docking face, and the inherent feature points are regular-shaped protrusions or depressions or screws; The mobile end feature identifier is a mobile end target, or the inherent feature set at the front end of the mobile end.

5. The method for docking large rigid body components based on the fixed end image according to claim 4, characterized in that: The coordinates in step 4.4 are solved as follows: Step 4.4.1: Using the result obtained in Step 4.1 Calculate that the coordinates of the feature points of the mobile device's feature identifier in the docking state in the fixed device coordinate system are: Step 4.4.2: Using the result obtained in Step 4.3 Calculate the coordinates of the feature points of the mobile device feature identifier in the docking state in the pose adjustment platform coordinate system as follows: Step 4.4.3: Substitute Equation (1) into Equation (2) to obtain The specific calculation formula is as follows:

6. The method for docking large rigid body components based on the fixed end image according to claim 5, characterized in that: The P in step 3.3 m-c has the following specific calculation formula: In step 3.4, P m-b The specific calculation formula is as follows: In step 4.5, R b , t b The specific calculation formulas are as follows:

7. The method for docking large rigid components based on the fixed-end image according to any one of claims 1-6, characterized in that: The specific solution process of Step 4.5 is: Among them, r 11 to r 33 represents the rotation matrix R b , t1 to t3 represent the translation vector t b ; Using the relationship between Euler angles and the rotation matrix, as well as the attitude adjustment sequence, the pitch angle θ, yaw angle ψ, and roll angle φ based on the attitude adjustment platform coordinate system are calculated.

Images