Truss robot hand-eye calibration method and device and computer equipment

Abstract

The invention relates to a truss robot hand-eye calibration method. A camera assembly and a robot gripper are installed on the truss robot, a calibration part is arranged on the robot gripper, and the method comprises the steps that based on truss robot coordinates corresponding to the camera assembly at a target photographing position, a target image collected by the camera assembly for a calibration plate when the truss robot is located at the truss robot coordinates is obtained; the calibration piece is controlled to be aligned with at least three feature points on the calibration plate, and target calibration coordinates corresponding to the feature points are determined; for each feature point, obtaining an end effector coordinate matched with the feature point according to a target calibration coordinate corresponding to the feature point and a preset installation offset vector; determining a three-dimensional coordinate corresponding to each feature point based on the target image; and obtaining calibration parameters of the camera assembly according to the three-dimensional coordinates corresponding to the feature points and the coordinates of the end effector. The method can improve the calibration efficiency.

Description

Technical Field

[0001] This application relates to the field of industrial technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for hand-eye calibration of a gantry robot. Background Technology

[0002] Hand-eye calibration is a key technology in the field of robot vision. Its purpose is to determine the spatial transformation relationship between the coordinate system of the robot's end effector (or tool) and the camera coordinate system. For gantry robots, accurate hand-eye calibration is a prerequisite for achieving high-precision grasping and placement operations under vision guidance.

[0003] Currently, in hand-eye calibration methods for gantry robots, the multi-pose method based on AX=XB requires the camera to move multiple times to complete data acquisition. However, for large gantry robots with long travel distances and relatively slow movement speeds, the entire calibration process is time-consuming and inefficient. Summary of the Invention

[0004] Therefore, it is necessary to provide a gantry robot hand-eye calibration method, device, computer equipment, computer-readable storage medium, and computer program product that can reduce calibration time and improve calibration efficiency in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a hand-eye calibration method for a gantry robot, wherein the gantry robot is equipped with a camera assembly and a robot gripper, and the robot gripper is provided with a calibration component, and the method includes:

[0006] Based on the coordinates of the gantry robot corresponding to the camera component at the target image capture position, obtain the target image captured by the camera component for the calibration board when the gantry robot is at the gantry robot coordinates;

[0007] By controlling the calibration component to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined.

[0008] For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector;

[0009] Determine the three-dimensional coordinates of each feature point based on the target image;

[0010] The calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

[0011] In one embodiment, the camera assembly includes at least one camera, and the calibration plate is placed at the center of the target working area; the method further includes:

[0012] For each camera, the gantry robot is controlled to move along the Y-axis so that the camera reaches the center point of the Y-axis range of motion.

[0013] By controlling the gantry robot to move along the X-axis, so that the camera's field of view includes the calibration plate and the target working area, the camera's shooting position is obtained;

[0014] Determine the target shooting location that includes each of the aforementioned shooting locations.

[0015] In one embodiment, determining the target calibration coordinates corresponding to each of the feature points by controlling the calibration member to align with at least three feature points on the calibration plate includes:

[0016] When the Z-axis turntable of the gantry robot rotates to the first angle, the calibration component is controlled to align with at least three feature points on the calibration plate to obtain the first calibration coordinates corresponding to each feature point.

[0017] When the Z-axis turntable of the gantry robot rotates to the second angle, the calibration component is controlled to align with at least three feature points on the calibration plate to obtain the second calibration coordinates corresponding to each feature point; the difference between the second angle and the first angle is 180 degrees.

[0018] For each feature point, the average of the first calibration coordinate and the second calibration coordinate is determined as the target calibration coordinate corresponding to the feature point.

[0019] In one embodiment, obtaining the end effector coordinates matching the feature point based on the target calibration coordinates corresponding to the feature point and a preset installation offset vector includes:

[0020] Based on the target calibration coordinates corresponding to the feature point, the first rotation matrix matching the first angle, and the preset installation offset vector, the first end coordinates corresponding to the feature point are obtained;

[0021] Based on the first rotation matrix and the preset diagonal matrix, a second rotation matrix matching the second angle is obtained;

[0022] Based on the target calibration coordinates, the second rotation matrix, and the installation offset vector, the second end coordinates corresponding to the feature point are obtained;

[0023] The average of the first end coordinates and the second end coordinates is determined as the mean coordinates;

[0024] By swapping the coordinate values ​​in the X-axis direction and the Y-axis direction of the mean coordinates, the end effector coordinates corresponding to the feature point are obtained.

[0025] In one embodiment, obtaining the calibration parameters of the camera assembly based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point includes:

[0026] The coordinates of the end effector corresponding to each of the aforementioned feature points are analyzed to obtain the coordinates of the first centroid.

[0027] The coordinates of the second centroid are obtained by analyzing the three-dimensional coordinates of each feature point.

[0028] For each feature point, the coordinates of the end effector corresponding to the feature point are transformed using the first centroid coordinates to obtain the first transformed coordinates;

[0029] The second centroid coordinates are used to transform the three-dimensional coordinates corresponding to the feature point to obtain the second transformed coordinates;

[0030] The calibration parameters of the camera component are obtained based on the first and second transformed coordinates of each feature point.

[0031] In one embodiment, obtaining the calibration parameters of the camera component based on the first and second transformed coordinates corresponding to each of the feature points includes:

[0032] Based on the first and second transformed coordinates of each feature point, the covariance matrix is ​​obtained;

[0033] Singular value decomposition is performed on the covariance matrix, and the camera rotation matrix of the camera component is obtained based on the decomposition result;

[0034] Based on the camera rotation matrix, the first centroid, and the second centroid, the translation vector of the camera component is obtained;

[0035] The calibration parameters of the camera component are obtained based on the camera rotation matrix and translation vector.

[0036] Secondly, this application provides a hand-eye calibration device for a gantry robot, wherein a camera assembly and a robot gripper are mounted on the gantry robot, and a calibration component is provided on the robot gripper; the device includes:

[0037] The processing module is used to obtain the target image captured by the camera component against the calibration plate when the gantry robot is at the gantry robot coordinates corresponding to the camera component at the target image position.

[0038] The first determining module is used to determine the target calibration coordinates corresponding to each of the feature points by controlling the calibration component to align with at least three feature points on the calibration plate.

[0039] The first analysis module is used to obtain the end effector coordinates that match the feature point for each feature point, based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector.

[0040] The second determining module is used to determine the three-dimensional coordinates corresponding to each of the feature points based on the target image;

[0041] The second analysis module is used to obtain the calibration parameters of the camera assembly based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

[0042] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0043] Based on the coordinates of the gantry robot corresponding to the camera component in the target image position, the target image captured by the camera component against the calibration plate when the gantry robot is at the gantry robot coordinates is obtained.

[0044] By controlling the calibration component in the gantry robot to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined.

[0045] For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector;

[0046] Determine the three-dimensional coordinates of each feature point based on the target image;

[0047] The calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

[0048] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0049] Based on the coordinates of the gantry robot corresponding to the camera component in the target image position, the target image captured by the camera component against the calibration plate when the gantry robot is at the gantry robot coordinates is obtained.

[0050] By controlling the calibration component in the gantry robot to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined.

[0051] For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector;

[0052] Determine the three-dimensional coordinates of each feature point based on the target image;

[0053] The calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

[0054] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0055] Based on the coordinates of the gantry robot corresponding to the camera component in the target image position, the target image captured by the camera component against the calibration plate when the gantry robot is at the gantry robot coordinates is obtained.

[0056] By controlling the calibration component in the gantry robot to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined.

[0057] For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector;

[0058] Determine the three-dimensional coordinates of each feature point based on the target image;

[0059] The calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

[0060] The aforementioned gantry robot hand-eye calibration method, apparatus, computer equipment, computer-readable storage medium, and computer program product acquire the target image captured by the camera assembly against a calibration plate when the gantry robot is at the target image capture position, based on the gantry robot coordinates corresponding to the camera assembly at the target image capture position. By controlling the calibration component to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined. For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates and a preset installation offset vector. The three-dimensional coordinates corresponding to each feature point are determined based on the target image, and the calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates of each feature point and the end effector coordinates. Therefore, this application avoids multiple camera assembly movements by using the gantry robot coordinates corresponding to the camera assembly at the target image capture position, saving calibration time and improving calibration efficiency. By combining the analysis of at least three feature points on the calibration plate and the target image captured against the calibration plate, calibration accuracy can be improved, further increasing the production efficiency of the gantry robot. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart illustrating a hand-eye calibration method for a gantry robot in one embodiment;

[0063] Figure 2 This is a schematic diagram of the gantry robot in one embodiment;

[0064] Figure 3 This is a schematic diagram of a calibration plate in one embodiment;

[0065] Figure 4 This is a flowchart illustrating the hand-eye calibration method for a gantry robot in another embodiment;

[0066] Figure 5 This is a structural block diagram of a gantry robot hand-eye calibration device in one embodiment;

[0067] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0069] It should be noted that the terms "comprising" and "having," and any variations thereof, as used in this application, are intended to cover non-exclusive inclusion. The term "multiple" as used in this application refers to two or more. The term "and / or" as used in this application refers to one of the solutions, or any combination of multiple solutions.

[0070] For gantry robots, most existing calibration methods use a single camera, collecting multiple sets of data by moving the robot or performing a single calibration using calibration tools. The core of this approach lies in establishing the relationship between the camera coordinate system and the robot's base coordinate system. However, single-camera calibration methods are unsuitable for large gantry robots because their working range far exceeds the field of view of a single camera. Typically, two or more cameras are needed to cover the entire working area, but there is a lack of efficient, accurate, and unified calibration processes using dual-camera (or multi-camera) systems. Furthermore, multi-pose methods based on AX=XB require the robot to move and take photos multiple times. For large gantry robots with long travel distances and relatively slow movement speeds, the entire calibration process is extremely time-consuming, severely impacting production efficiency.

[0071] Typically, gantry robots have physical or software-defined travel limits on their X and Y axes. Traditional multi-pose calibration methods require the robot to perform a wide range of multi-angle movements, which may lead to insufficient and diverse pose data acquisition due to hitting these travel limits, resulting in calibration failure or decreased accuracy. Furthermore, when using probes for pointing calibration, the probe installation may have eccentricity errors, meaning the probe tip is not strictly positioned on the rotation center axis of the robot's end flange. This eccentricity causes deviations in the coordinates recorded by the gantry robot when pointing to the same point on the calibration plate in different poses, thus affecting the final calibration accuracy. Existing methods usually ignore or struggle to compensate for this error.

[0072] In view of this, such as Figure 1 As shown, this application provides a hand-eye calibration method for a gantry robot. The gantry robot is equipped with a camera assembly and a robot gripper, and the robot gripper is provided with calibration components. Taking the application of this method to a controller in a gantry robot as an example, the method includes the following steps:

[0073] S102, based on the coordinates of the gantry robot corresponding to the camera component at the target image capture position, obtain the target image captured by the camera component against the calibration plate when the gantry robot is at the gantry robot coordinates.

[0074] The camera assembly includes at least one camera, and the target image capture position includes the corresponding shooting position for each camera. Each camera has a different mounting position on the crossbeam of the gantry robot. Taking an assembly including a first camera and a second camera as an example... Figure 2 The diagram illustrates the structure of a gantry robot, showing that the first and second cameras are mounted on different crossbeams. When the gantry robot moves along the X-axis or Y-axis, the first or second camera can move synchronously with it.

[0075] The coordinates of a gantry robot refer to the coordinates of the gantry robot in the base coordinate system {B} when the camera assembly is moved to the target imaging position. These coordinates can be characterized using coordinate values ​​on the X, Y, and Z axes. Gantry robots are an important branch of industrial robots and can include mechanical structures (such as guide rails and sliders) and end effectors.

[0076] A calibration board is a calibration board of a specific size placed in the center of the target working area. There are no restrictions on the type of calibration board. For example, a calibration board can be an A1 calibration board, a checkerboard calibration board, or a calibration board with a specific coding pattern (such as ArUco markings).

[0077] In some cases, the calibration plate can be appropriately raised according to the height of the camera in the camera assembly to obtain a clear image. To simulate real-world conditions, a single layer of workpiece can also be placed under the calibration plate, while avoiding interference from multiple layers of workpieces on point cloud recognition.

[0078] In one embodiment, the camera assembly includes at least one camera, and for each camera, a calibration image is acquired by the camera relative to the calibration plate when the gantry robot is in the gantry robot coordinates; the calibration images are fused to obtain a target image acquired by the camera assembly relative to the calibration plate.

[0079] Understandably, because each camera is positioned differently on the gantry robot, the calibration images captured by each camera for the calibration plate will be different.

[0080] In some cases, by viewing the target image acquired by the camera components in real time, it is determined whether the calibration board meets the calibration conditions. The calibration conditions indicate that the calibration board is unobstructed and the image is clear. If the calibration conditions are not met, the calibration board is adjusted to make it meet the calibration conditions.

[0081] S104, by controlling the calibration component to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined.

[0082] The type of calibration component is not limited, as long as it can accurately indicate the target interface of the feature points on the calibration board. For example, the calibration component can be a calibration pointer. Alternatively, the calibration component can be a laser pointer, allowing alignment of the feature points on the calibration board by observing the center of the laser point. Or, the calibration component can be a dedicated tool center point (TCP) calibration tool with a spherical shape.

[0083] Among these features, at least three feature points must be non-collinear, meaning at least three feature points are not on the same straight line on the calibration board. The position of the feature points on the calibration board depends on the type of calibration board. For example, when the calibration board is an A1 calibration board, the center of the circle can be set as a feature point; when the calibration board is a checkerboard calibration board, the corner points can be set as feature points.

[0084] Taking the A1 calibration board as an example, such as Figure 3 As shown, a schematic diagram of a calibration board is provided. Taking the number of feature points as three, and the positions of the three feature points located in the upper left, lower left and upper right of the A1 calibration board as an example, the first center 301, the second center 302 and the third center 303 on the A1 calibration board can be set as non-collinear feature points.

[0085] In some cases, mathematical methods such as least squares can be used for fitting, thereby selecting multiple non-collinear features on the calibration board to obtain more stable and noise-resistant calibration results.

[0086] S106, For each feature point, obtain the end effector coordinates that match the feature point based on the target calibration coordinates corresponding to the feature point and the preset rotation matrix.

[0087] When using a calibration component to point at the calibration plate, the controller in the gantry robot outputs the coordinates of the end effector, which are the center of the Z-axis turntable, in coordinate system {E}. Taking the calibration component as a calibration pointer as an example, the tip of the calibration pointer has an unknown but fixed installation offset vector relative to the end effector. Therefore, the end effector coordinates matching the feature point can be obtained based on the calibration coordinates corresponding to the feature point and the preset installation offset vector.

[0088] S108, determine the three-dimensional coordinates of each feature point based on the target image.

[0089] For example, the target image includes a depth map and a texture map. Determining the three-dimensional coordinates of each feature point based on the target image includes: analyzing the texture map to obtain the first coordinate value of each feature point in the X-axis direction and the second coordinate value in the Y-axis direction; analyzing the depth map to obtain the third coordinate value of each feature point in the Z-axis direction; the three-dimensional coordinates are composed of the first coordinate value, the second coordinate value, and the third coordinate value corresponding to the feature point.

[0090] In some embodiments, a multi-exposure strategy can be used to control the camera component to take multiple shots of the calibration board to obtain multiple target images, each target image including a depth map and a texture map; multiple texture maps are fused to obtain a fused texture image; the fused texture image is analyzed to obtain the first coordinate value of each feature point in the X-axis direction and the second coordinate value in the Y-axis direction; multiple depth maps are fused to obtain a fused depth map; the fused depth map is analyzed to obtain the third coordinate value of each feature point in the Z-axis direction; the three-dimensional coordinates are composed of the first coordinate value, the second coordinate value, and the third coordinate value corresponding to the feature points.

[0091] Therefore, by using a multi-exposure strategy to take multiple photos of the calibration board and fuse the data, random noise can be effectively suppressed, data gaps can be filled, and the accuracy of coordinate analysis can be improved.

[0092] For example, determining the three-dimensional coordinates of each feature point based on the target image includes: analyzing the target image to obtain point cloud data corresponding to the calibration board; and extracting the three-dimensional coordinates of each feature point from the point cloud data.

[0093] In some embodiments, the camera assembly includes multiple cameras. For each camera, the target image acquired by that camera is analyzed to obtain the three-dimensional coordinates corresponding to each feature point. Thus, by utilizing the target image captured by each camera, parameter calibration of each camera can be achieved.

[0094] S110: Based on the three-dimensional coordinates of each feature point and the coordinates of the end effector, the calibration parameters of the camera assembly are obtained.

[0095] Among them, the calibration parameters of the camera component are used to characterize the transformation relationship from the camera coordinate system {C} to the gantry robot base coordinate system {B}, i.e., the extrinsic parameter matrix. This matrix is ​​a 4x4 homogeneous transformation matrix used to transform the coordinates of any point in the camera coordinate system. Convert to robot base coordinate system ,satisfy:

[0096]

[0097] in, and These are the homogeneous coordinate representations (4*1 vectors) of the same point in the base coordinate system and the camera coordinate system, respectively. Includes rotation matrix Translation vector ,satisfy:

[0098]

[0099] As can be seen from the above, to solve for... It is necessary to obtain at least three sets of non-collinear spatial point pairs. Therefore, the calibration parameters of the camera components can be obtained based on the steps in S110.

[0100] The method described in the above embodiments obtains the target image captured by the camera component against the calibration plate when the gantry robot is at the target image capture position, based on the gantry robot coordinates corresponding to the camera component at the target image capture position. By controlling the calibration component to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined. For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates and a preset installation offset vector. The three-dimensional coordinates corresponding to each feature point are determined based on the target image. Finally, the calibration parameters of the camera component are obtained based on the three-dimensional coordinates of each feature point and the end effector coordinates. Therefore, this application avoids multiple movements of the camera component by using the gantry robot coordinates corresponding to the camera component at the target image capture position for calibration, saving calibration time and improving calibration efficiency. By combining the analysis of at least three feature points on the calibration plate and the target image captured against the calibration plate, calibration accuracy can be improved, further increasing the production efficiency of the gantry robot.

[0101] In one embodiment, the target image location can be determined through a testing process. Specifically, the camera assembly includes at least one camera, and a calibration plate is placed at the center of the target working area. The method includes:

[0102] S11: For each camera, control the gantry robot to move along the Y-axis so that the camera reaches the center point of the Y-axis range of motion.

[0103] S12: By controlling the gantry robot to move along the X-axis, the camera's field of view includes the calibration plate and the target working area, thus obtaining the camera's image position.

[0104] S13: Determine the target photo location that includes each photo location.

[0105] By employing the method described in the above embodiments, the camera is moved to the center point of its Y-axis range of motion, allowing it to have the largest and most centrally located field of view in the Y-axis direction. By ensuring that the camera's field of view encompasses the entire calibration plate and the target working area, calibration based on this image capture position can improve the production efficiency of the gantry robot. Furthermore, by determining the corresponding image capture position for each camera, calibration accuracy can be improved when performing calibration based on the target image capture position that includes each image capture position.

[0106] In one embodiment, determining the target calibration coordinates corresponding to each feature point by controlling the calibration component to align with at least three feature points on the calibration plate includes the following steps:

[0107] S21: When the Z-axis turntable of the gantry robot rotates to the first angle, the calibration component is aligned with at least three feature points on the calibration plate to obtain the first calibration coordinates corresponding to each feature point.

[0108] S22: When the Z-axis turntable of the gantry robot rotates to the second angle, the calibration component is aligned with at least three feature points on the calibration plate to obtain the second calibration coordinates corresponding to each feature point.

[0109] Here, the first calibration coordinate and the second calibration coordinate refer to the coordinates in the robot's base coordinate system {B}. The difference between the second angle and the first angle is 180 degrees. For example, the first angle is 0 degrees and the second angle is 180 degrees; or the first angle is 90 degrees and the second angle is 270 degrees, or other settings are also possible.

[0110] S23: For each feature point, the average of the first calibration coordinate and the second calibration coordinate is determined as the target calibration coordinate corresponding to the feature point.

[0111] Therefore, by averaging the coordinates of the corresponding points obtained from the two pointers, the installation eccentricity error can be eliminated, and the accuracy of the determined target calibration coordinates can be improved.

[0112] For example, taking the centers of the upper left, lower left, and upper right circles on the calibration board as feature points, with a first angle of 0 degrees and a second angle of 180 degrees, the first calibration coordinate corresponding to the first feature point is represented as P_LU0, and the second calibration coordinate is represented as P_LU180. Therefore, the target calibration coordinate corresponding to the first feature point is P_LU_final = (P_LU0 + P_LU180) / 2. The second feature point has the same first calibration coordinate represented as P_LD0 and second calibration coordinate represented as P_LD180. Therefore, the target calibration coordinate corresponding to the second feature point is P_LD_final = (P_LD0 + P_LD180) / 2. The third feature point has the same first calibration coordinate represented as P_RU0 and second calibration coordinate represented as P_RU180. Therefore, the target calibration coordinate corresponding to the third feature point is P_RU_final = (P_RU0 + P_RU180) / 2.

[0113] By employing the method described in the above embodiments, and by using the Z-axis rotation of the gantry robot, calibration deviations in the X and Y axes of the gantry robot can be eliminated, thereby improving calibration accuracy.

[0114] In one embodiment, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates corresponding to the feature point and a preset installation offset vector, including the following steps:

[0115] S31: Based on the target calibration coordinates corresponding to the feature point, the first rotation matrix matching the first angle, and the preset installation offset vector, obtain the first end coordinates corresponding to the feature point.

[0116] For example, the coordinates of the first end point corresponding to the feature point satisfy:

[0117]

[0118] in, Indicates the coordinates of the first end point. Indicates the target's calibration coordinates. This represents the first rotation matrix, used to characterize the rotation matrix of the end effector. This represents the installation offset vector.

[0119] S32: Based on the first rotation matrix and the preset diagonal matrix, obtain the second rotation matrix that matches the second angle.

[0120] For example, the second rotation matrix satisfies:

[0121]

[0122] in, Denotes the second rotation matrix. This represents a predefined diagonal matrix.

[0123] S33: Based on the target calibration coordinates, the second rotation matrix, and the installation offset vector, obtain the second end coordinates corresponding to the feature points.

[0124] For example, the second end coordinates satisfy:

[0125]

[0126] S34: Determine the mean coordinates of the first and second end coordinates.

[0127] S35: Interchange the coordinate values ​​in the X-axis direction and the Y-axis direction in the mean coordinate to obtain the end effector coordinates corresponding to the feature point.

[0128] By performing averaging, errors can be eliminated and the accuracy of the analysis can be improved. By interchangering the X-axis and Y-axis coordinates in the average coordinates according to the definitional differences between the gantry robot's base coordinate system and the image coordinate system, the accuracy of the analysis of the end effector coordinates corresponding to feature points can be improved.

[0129] For example, mean coordinates satisfy:

[0130]

[0131] Further satisfy:

[0132]

[0133] set up , satisfy:

[0134]

[0135] but:

[0136]

[0137] but satisfy:

[0138]

[0139] As can be seen from the above, the installation offset vector Components in the XY plane and Completely canceled out, final coordinates The true coordinates of the feature points There is only one offset along the tool's Z-axis, which has minimal impact on the calibration of the horizontal gripping surface. Yes An extremely accurate estimate. Furthermore, the mean coordinates... The coordinates along the X-axis and Y-axis are interchanged to obtain the coordinates of the end effector corresponding to the feature point. .

[0140] By employing the method described in the above embodiments, the calibration error caused by the installation offset vector can be eliminated, thereby reducing the influence of the horizontal gripping surface on the calibration results and improving calibration accuracy.

[0141] In one embodiment, the calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates of each feature point and the coordinates of the end effector, including the following steps:

[0142] S41: Analyze the coordinates of the end effector corresponding to each feature point to obtain the coordinates of the first centroid.

[0143] For example, the first centroid coordinates satisfy:

[0144]

[0145] Where n represents the number of feature points.

[0146] S42: Analyze the three-dimensional coordinates of each feature point to obtain the coordinates of the second centroid.

[0147] For example, the second centroid coordinates satisfy:

[0148]

[0149] in, This represents the three-dimensional coordinates corresponding to the i-th feature point.

[0150] S43: For each feature point, use the first centroid coordinates to transform the end effector coordinates corresponding to the feature point to obtain the first transformed coordinates.

[0151] For example, the first transformed coordinates corresponding to the i-th feature point satisfy:

[0152]

[0153] S44: Use the second centroid coordinates to transform the three-dimensional coordinates corresponding to the feature points to obtain the second transformed coordinates.

[0154] For example, the second transformed coordinates corresponding to the i-th feature point satisfy:

[0155]

[0156] S45: Obtain the calibration parameters of the camera component based on the first and second transformed coordinates of each feature point.

[0157] By using the method described in the above embodiments, the coordinates corresponding to the feature points can be transformed according to the centroid, and the coordinates corresponding to the feature points can be moved to a coordinate system with the centroid as the origin, thereby improving the accuracy of the calibration parameter analysis.

[0158] In one embodiment, the calibration parameters of the camera component are obtained based on the first and second transformed coordinates corresponding to each feature point, including the following steps:

[0159] S51. Based on the first and second transformed coordinates of each feature point, the covariance matrix is ​​obtained.

[0160] For example, the covariance matrix H satisfies:

[0161]

[0162] S52, perform singular value decomposition on the covariance matrix, and obtain the camera rotation matrix of the camera component based on the decomposition result.

[0163] For example, the decomposition result satisfies:

[0164]

[0165] The camera rotation matrix R satisfies:

[0166]

[0167] S53, based on the camera rotation matrix, the first centroid, and the second centroid, obtains the translation vector of the camera component.

[0168] For example, the translation vector t satisfies:

[0169]

[0170] S54, based on the camera rotation matrix and camera translation vector, obtains the calibration parameters of the camera components.

[0171] The camera component includes at least one camera, the camera rotation matrix includes the rotation matrix corresponding to each camera, and the camera translation vector includes the translation vector corresponding to each camera.

[0172] By employing the method described in the above embodiments and using singular value decomposition on the covariance matrix, the accuracy of calibration parameter analysis can be improved.

[0173] The calibration parameters of the camera components are obtained based on the camera rotation matrix and camera translation vector. These parameters can be determined using the following methods, as exemplified:

[0174] In one embodiment, obtaining the calibration parameters of the camera component based on the camera rotation matrix and the camera translation vector includes: determining the calibration parameters of the camera component that include the camera rotation matrix and the camera translation vector.

[0175] In one embodiment, the calibration parameters of the camera component are obtained based on the camera rotation matrix and camera translation vector, including: obtaining the calibration error based on the camera rotation matrix, camera translation vector, and the three-dimensional coordinates and end effector coordinates corresponding to each feature point; when the calibration error is greater than a preset error threshold, at least three new feature points are selected again, and the process returns to the step of determining the target calibration coordinates corresponding to each feature point by controlling the calibration component to align at least three feature points on the calibration plate; when the new calibration error is less than or equal to the preset error threshold, the camera rotation matrix and camera translation vector corresponding to the new calibration error are determined as the calibration parameters of the camera component.

[0176] For example, the calibration parameters of the camera component can be obtained when the following conditions are met, wherein:

[0177]

[0178] By introducing calibration errors using the method described in the above embodiments, the accuracy of the calibration results can be analyzed, which can further improve the production efficiency of the gantry robot.

[0179] Understandably, once the three-dimensional coordinates of multiple feature points and the coordinates of the end effector are obtained, the hand-eye calibration problem is transformed into a classic absolute orientation problem. The goal is to find a rigid body transformation (rotation matrix R and translation vector t) that minimizes the sum of squared errors between the transformed point set and the target point set.

[0180] In some embodiments, when the three-dimensional coordinates and end effector coordinates corresponding to multiple feature points are obtained, in addition to directly matching the three-dimensional point set, such as the Singular Value Decomposition (SVD) method, other algorithms that can solve the camera rotation matrix and camera translation matrix based on the correspondence of three-dimensional points can also be used.

[0181] Based on the above, the gantry robot is equipped with at least one camera and a robotic gripper, with a calibration pointer mounted on the gripper. For example... Figure 4 As shown, a hand-eye calibration method for a gantry robot is provided. Taking the application of this method to the controller of a gantry robot as an example, the method may include the following steps:

[0182] S402, for each camera, controls the gantry robot to move along the Y-axis so that the camera reaches the center point of the Y-axis range of motion.

[0183] S404 controls the gantry robot to move along the X-axis so that the camera's field of view includes the calibration plate and the target working area, thus obtaining the camera's image position.

[0184] S406, based on the gantry robot coordinates corresponding to the camera at the shooting position, obtains the target image captured by the camera on the calibration plate when the gantry robot is at the gantry robot coordinates.

[0185] S408, when the Z-axis turntable of the gantry robot is rotated to 0 degrees, the calibration component is aligned with at least three feature points on the calibration plate to obtain the first calibration coordinates corresponding to each feature point.

[0186] S410, when the Z-axis turntable of the gantry robot rotates to 180 degrees, the calibration component is controlled to align with at least three feature points on the calibration plate to obtain the second calibration coordinates corresponding to each feature point; the difference between the second angle and the first angle is 180 degrees.

[0187] S412, for each feature point, the average of the first calibration coordinate and the second calibration coordinate is determined as the target calibration coordinate corresponding to the feature point.

[0188] S414. Based on the target calibration coordinates corresponding to the feature point, the first rotation matrix matching 0 degrees, and the preset installation offset vector, the first end coordinates corresponding to the feature point are obtained.

[0189] S416, based on the first rotation matrix and the preset diagonal matrix, obtain the second rotation matrix that matches 180 degrees.

[0190] S418, based on the target calibration coordinates, the second rotation matrix and the installation offset vector, obtain the second end coordinates corresponding to the feature point.

[0191] S420, the average of the first end coordinates and the second end coordinates is determined as the mean coordinates.

[0192] S422, interchange the coordinate values ​​in the X-axis direction and the Y-axis direction in the mean coordinate to obtain the end effector coordinates corresponding to the feature point.

[0193] S424, determine the three-dimensional coordinates of each feature point based on the target image.

[0194] S426: Based on the three-dimensional coordinates of each feature point and the coordinates of the end effector, the camera calibration parameters are obtained.

[0195] The specific content of S402 to S426 can be found in the aforementioned description.

[0196] Based on the above, this application ensures the optimality (maximum field of view) and repeatability of the image capture point by limiting the camera to its physical or logical center on the Y-axis and to its X-axis position to cover the entire calibration board and the largest target working area. This eliminates reliance on the operator's subjective judgment during the calibration process, ensuring consistency of results across different times and personnel, and providing a procedural guarantee for obtaining a stable extrinsic parameter matrix. Therefore, the obtained extrinsic parameter matrix of the camera can be used to subsequently unify the camera's point cloud data into the gantry robot's base coordinate system, enabling seamless stitching recognition and high-precision grasping of large-scale workpieces.

[0197] In summary, the method provided in this application eliminates the complex process of the traditional AX=XB method, which requires the robot to reciprocate several times. By determining a unique and optimal shooting position for each camera, data acquisition can be completed with only one camera movement. This significantly shortens calibration time, simplifies on-site operation, and is particularly suitable for large, slow-moving gantry robot systems.

[0198] Furthermore, this application determines the camera's shooting position by actively adapting to the physical travel limits of the truss. Specifically, it maximizes the field of view by moving the camera to its travel limit, ensuring that the optimal calibration and working point is found within a limited space. This makes calibration possible in constrained systems, where traditional methods may not be feasible.

[0199] Furthermore, this application effectively compensates for the eccentricity error caused by the incomplete coaxial installation of the calibration pointer by rotating the calibration pointer 180 degrees around the tool center and averaging the two measurement results. This makes the robot world coordinates of the obtained calibration plate feature points more accurate, thereby directly improving the accuracy of the final hand-eye calibration extrinsic parameter matrix.

[0200] Furthermore, the method provided in this application can be used to calibrate two cameras in a system, enabling them to share the same robot base coordinate system. By independently calculating precise extrinsic parameters to the robot base for each camera, a solid and reliable foundation is laid for subsequent advanced applications such as visual data stitching and collaborative grasping, ensuring coordinate uniformity and accuracy throughout the large workspace.

[0201] In summary, this application specifically addresses the calibration of large gantry robots with travel limits, providing a method that combines an operational mode of "determining an optimal fixed shooting position for each camera" with a high-precision world coordinate acquisition technology of "rotating the tool 180 degrees and averaging the coordinates of the pointing point." This combined approach not only solves the travel limit problem but also significantly simplifies the calibration process and substantially improves calibration accuracy.

[0202] Furthermore, by moving the camera along one axis to its travel limit to maximize the field of view, and then moving it along another axis to fully capture the calibration plate, the optimal fixed shooting position is found within a confined space. Additionally, an initial measurement of the target point is performed using a pointing tool mounted on a rotatable tool flange. A second measurement is then performed after rotating the tool flange 180 degrees, and the average of the two measurements is taken as the final coordinates to eliminate installation eccentricity errors of the pointing tool. Moreover, by independently performing the above calibration method on multiple cameras sharing the same robot base, accurate extrinsic parameters from each camera to the robot's base coordinate system are obtained, providing coordinate system consistency assurance for large-scale visual stitching applications.

[0203] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0204] Based on the same inventive concept, this application also provides a gantry robot hand-eye calibration device for implementing the aforementioned gantry robot hand-eye calibration method. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the gantry robot hand-eye calibration device provided below can be found in the limitations of the gantry robot hand-eye calibration method described above, and will not be repeated here.

[0205] In one exemplary embodiment, such as Figure 5 As shown, a hand-eye calibration device for a gantry robot is provided. The gantry robot is equipped with a camera assembly and a robot gripper. The robot gripper is provided with calibration components. The device includes: a processing module 502, a first determining module 504, a first analyzing module 506, a second determining module 508, and a second analyzing module 510, wherein:

[0206] Processing module 502 is used to obtain a target image captured by the camera component against a calibration plate when the gantry robot is at the target imaging position, based on the gantry robot coordinates corresponding to the camera component at the target imaging position; a first determining module 504 is used to determine the target calibration coordinates corresponding to each of the feature points by controlling the calibration component to align with at least three feature points on the calibration plate; a first analysis module 506 is used to obtain the end effector coordinates matching the feature point for each feature point based on the target calibration coordinates corresponding to the feature point and a preset installation offset vector; a second determining module 508 is used to determine the three-dimensional coordinates corresponding to each of the feature points based on the target image; and a second analysis module 510 is used to obtain the calibration parameters of the camera component based on the three-dimensional coordinates corresponding to each of the feature points and the end effector coordinates.

[0207] In one embodiment, the camera assembly includes at least one camera, and the calibration plate is placed at the center of the target working area; the processing module 502 is further configured to: for each of the cameras, control the gantry robot to move along the Y-axis direction so that the camera reaches the center point of the Y-axis movement range; control the gantry robot to move along the X-axis direction so that the camera's field of view includes the calibration plate and the target working area, thereby obtaining the camera's image capture position; and determine a target image capture position that includes each of the image capture positions.

[0208] In one embodiment, the first determining module 504 is further configured to: when the Z-axis turntable of the gantry robot rotates to a first angle, control the calibration component to align with at least three feature points on the calibration plate to obtain first calibration coordinates corresponding to each feature point; when the Z-axis turntable of the gantry robot rotates to a second angle, control the calibration component to align with at least three feature points on the calibration plate to obtain second calibration coordinates corresponding to each feature point; the difference between the second angle and the first angle is 180 degrees; for each feature point, the average of the first calibration coordinates and the second calibration coordinates is determined as the target calibration coordinates corresponding to the feature point.

[0209] In one embodiment, the first analysis module 506 is further configured to: obtain the first end coordinates corresponding to the feature point based on the target calibration coordinates corresponding to the feature point, the first rotation matrix matching the first angle, and the preset installation offset vector; obtain the second rotation matrix matching the second angle based on the first rotation matrix and the preset diagonal matrix; obtain the second end coordinates corresponding to the feature point based on the target calibration coordinates, the second rotation matrix, and the installation offset vector; determine the mean of the first end coordinates and the second end coordinates as the mean coordinates; and interchange the coordinate values ​​in the X-axis direction and the Y-axis direction of the mean coordinates to obtain the end effector coordinates corresponding to the feature point.

[0210] In one embodiment, the second analysis module 510 is further configured to: analyze the end effector coordinates corresponding to each of the feature points to obtain first centroid coordinates; analyze the three-dimensional coordinates corresponding to each of the feature points to obtain second centroid coordinates; for each feature point, use the first centroid coordinates to transform the end effector coordinates corresponding to the feature point to obtain first transformed coordinates; use the second centroid coordinates to transform the three-dimensional coordinates corresponding to the feature point to obtain second transformed coordinates; and obtain the calibration parameters of the camera assembly based on the first and second transformed coordinates corresponding to each of the feature points.

[0211] In one embodiment, the second analysis module 510 is further configured to: obtain a covariance matrix based on the first and second transformed coordinates corresponding to each of the feature points; perform singular value decomposition on the covariance matrix and obtain the camera rotation matrix of the camera component based on the decomposition result; obtain the translation vector of the camera component based on the camera rotation matrix, the first centroid, and the second centroid; and obtain the calibration parameters of the camera component based on the camera rotation matrix and the translation vector.

[0212] Each module in the aforementioned gantry robot hand-eye calibration device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0213] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data used in the calibration process. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a hand-eye calibration method for a gantry robot.

[0214] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0215] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0216] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0217] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0218] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0219] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0220] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0221] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for hand-eye calibration of a gantry robot, characterized in that, The gantry robot is equipped with a camera assembly and a robot gripper, and the robot gripper is provided with a calibration component. The method includes: Based on the coordinates of the gantry robot corresponding to the camera component at the target image capture position, obtain the target image captured by the camera component for the calibration board when the gantry robot is at the gantry robot coordinates; By controlling the calibration component to align with at least three feature points on the calibration plate, the target calibration coordinates corresponding to each feature point are determined. For each feature point, the end effector coordinates matching the feature point are obtained based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector; Determine the three-dimensional coordinates of each feature point based on the target image; The calibration parameters of the camera assembly are obtained based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

2. The method according to claim 1, characterized in that, The camera assembly includes at least one camera, and the calibration plate is placed at the center of the target working area; the method further includes: For each camera, the gantry robot is controlled to move along the Y-axis so that the camera reaches the center point of the Y-axis range of motion. By controlling the gantry robot to move along the X-axis, so that the camera's field of view includes the calibration plate and the target working area, the camera's shooting position is obtained; Determine the target shooting location that includes each of the aforementioned shooting locations.

3. The method according to claim 1, characterized in that, The step of controlling the calibration component to align with at least three feature points on the calibration plate and determining the target calibration coordinates corresponding to each feature point includes: When the Z-axis turntable of the gantry robot rotates to the first angle, the calibration component is controlled to align with at least three feature points on the calibration plate to obtain the first calibration coordinates corresponding to each feature point. When the Z-axis turntable of the gantry robot rotates to the second angle, the calibration component is controlled to align with at least three feature points on the calibration plate to obtain the second calibration coordinates corresponding to each feature point; the difference between the second angle and the first angle is 180 degrees. For each feature point, the average of the first calibration coordinate and the second calibration coordinate is determined as the target calibration coordinate corresponding to the feature point.

4. The method according to claim 3, characterized in that, The step of obtaining the end effector coordinates matching the feature point based on the target calibration coordinates corresponding to the feature point and a preset installation offset vector includes: Based on the target calibration coordinates corresponding to the feature point, the first rotation matrix matching the first angle, and the preset installation offset vector, the first end coordinates corresponding to the feature point are obtained; Based on the first rotation matrix and the preset diagonal matrix, a second rotation matrix matching the second angle is obtained; Based on the target calibration coordinates, the second rotation matrix, and the installation offset vector, the second end coordinates corresponding to the feature point are obtained; The average of the first end coordinates and the second end coordinates is determined as the mean coordinates; By swapping the coordinate values ​​in the X-axis direction and the Y-axis direction of the mean coordinates, the end effector coordinates corresponding to the feature point are obtained.

5. The method according to claim 1, characterized in that, The step of obtaining the calibration parameters of the camera assembly based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point includes: The coordinates of the end effector corresponding to each of the aforementioned feature points are analyzed to obtain the coordinates of the first centroid. The coordinates of the second centroid are obtained by analyzing the three-dimensional coordinates of each feature point. For each feature point, the coordinates of the end effector corresponding to the feature point are transformed using the first centroid coordinates to obtain the first transformed coordinates; The second centroid coordinates are used to transform the three-dimensional coordinates corresponding to the feature point to obtain the second transformed coordinates; The calibration parameters of the camera component are obtained based on the first and second transformed coordinates of each feature point.

6. The method according to claim 5, characterized in that, The step of obtaining the calibration parameters of the camera component based on the first and second transformed coordinates corresponding to each of the feature points includes: Based on the first and second transformed coordinates of each feature point, the covariance matrix is ​​obtained; Singular value decomposition is performed on the covariance matrix, and the camera rotation matrix of the camera component is obtained based on the decomposition result; Based on the camera rotation matrix, the first centroid, and the second centroid, the translation vector of the camera component is obtained; The calibration parameters of the camera component are obtained based on the camera rotation matrix and translation vector.

7. A hand-eye calibration device for a gantry robot, characterized in that, The gantry robot is equipped with a camera assembly and a robot gripper, and the robot gripper is provided with a calibration component. The device includes: The processing module is used to obtain the target image captured by the camera component against the calibration plate when the gantry robot is at the gantry robot coordinates corresponding to the camera component at the target image position. The first determining module is used to determine the target calibration coordinates corresponding to each of the feature points by controlling the calibration component to align with at least three feature points on the calibration plate. The first analysis module is used to obtain the end effector coordinates that match the feature point for each feature point, based on the target calibration coordinates corresponding to the feature point and the preset installation offset vector. The second determining module is used to determine the three-dimensional coordinates corresponding to each of the feature points based on the target image; The second analysis module is used to obtain the calibration parameters of the camera assembly based on the three-dimensional coordinates and end effector coordinates corresponding to each feature point.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images