Hole-shaped part repairing method based on multi-process robot workpiece calibration

Abstract

The invention discloses a porous part repairing method based on multi-process robot workpiece calibration, and relates to the technical field of industrial robot machining equipment.The method comprises the steps that camera calibration and hand-eye calibration are conducted on a groove machining operation platform, a laser cladding operation platform and a grinding and polishing operation platform respectively; the groove machining operation platform moves to the position of the to-be-repaired hole-shaped part, and recognition and groove machining are conducted on the to-be-repaired hole-shaped part; the laser cladding operation platform moves to the position of the to-be-repaired hole-shaped part, and recognition and laser cladding are conducted on the to-be-repaired hole-shaped part; and the grinding and polishing operation platform moves to the position of the to-be-repaired hole-shaped part, and the to-be-repaired hole-shaped part is recognized, ground and polished. The repairing efficiency of the hole-shaped part is improved.

Description

Technical Field

[0001] The present application relates to the technical field of industrial robot processing equipment, and in particular to a method for repairing hole-shaped parts based on multi-process robot workpiece calibration. Background Art

[0002] Hole-shaped parts are widely used in the supporting components of shaft parts of vehicle transmission systems, large industrial machinery and other equipment. They are often subjected to high stress and impact loads of shaft parts. Fatigue cracks are prone to occur after long-term use. The formation and development of such internal cracks seriously threaten the integrity of the part structure and the safety of the equipment.

[0003] There are two commonly used repair methods for the internal cracks of porous parts: manual repair and industrial robot repair. Manual repair has problems such as low efficiency and poor consistency, and it is impossible to repair quickly and in a timely manner. The application of mobile industrial robots improves the repair efficiency and ensures the consistency after repair. However, since the repair process involves multiple processes, how to identify the position to be repaired and achieve coordination between industrial robots in different processes has become another difficult problem in the repair of cracks in porous parts. The traditional method has the problems of low precision and long time consumption. In addition, the internal space of porous parts is relatively small, and it is difficult for industrial vision cameras to enter the interior for direct observation, and it is impossible to accurately locate the position of the crack, which makes the repair efficiency of porous parts low. Summary of the invention

[0004] The purpose of the present application is to provide a method for repairing hole-shaped parts based on multi-process robot workpiece calibration to solve the problem of low repair efficiency of hole-shaped parts.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] The present application provides a hole-shaped part repair method based on multi-process robot workpiece calibration, which is implemented based on a multi-process robot repair system. The multi-process robot repair system includes: a groove processing work platform, a laser cladding work platform and a grinding and polishing work platform. The groove processing work platform, the laser cladding work platform and the grinding and polishing work platform are each provided with an industrial robot, and the end of each of the industrial robots carries a camera. The hole-shaped part repair method based on multi-process robot workpiece calibration includes:

[0007] Performing camera calibration and hand-eye calibration on the groove processing work platform, the laser cladding work platform and the grinding and polishing work platform respectively;

[0008] The groove processing work platform moves to the position of the hole-shaped part to be repaired, and performs groove processing on the hole-shaped part to be repaired, including:

[0009] A first end face image is acquired by a corresponding camera; the first end face image is an image of an unprocessed end face of the hole-shaped part to be repaired, and the end face is an end face of a crack of the hole-shaped part to be repaired with two ArUco codes attached thereto;

[0010] Preprocessing the first end face image to obtain a preprocessed first end face image; the preprocessing includes: scaling, grayscale, Gaussian filtering and binarization;

[0011] Determining a processing feed direction of the hole-shaped part to be repaired based on the preprocessed first end surface image;

[0012] Performing an opening operation on the preprocessed first end face image to obtain the first end face image after opening operation; the opening operation includes: corrosion processing and expansion operation;

[0013] A polygonal approximation method is used to detect the first end face image after the opening operation to obtain the contour of the hole-shaped part to be repaired; the contour of the hole-shaped part to be repaired includes: a first contour and a second contour; the first contour is the contour of the circular hole of the hole-shaped part to be repaired, and the second contour is the contour of a circular mark point, and the circular mark point is attached to the end face and corresponds to the crack;

[0014] Determine a first circumscribed circle and a second circumscribed circle; the first circumscribed circle is the minimum circumscribed circle of the first contour, and the second circumscribed circle is the minimum circumscribed circle of the second contour;

[0015] Determine the pixel coordinates of an intersection point of a line connecting the center of the first circumscribed circle and the center of the second circumscribed circle and the circular hole;

[0016] Convert the pixel coordinates of the intersection point from the pixel coordinate system to the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection point;

[0017] Solve the six joint angles of the industrial robot of the groove processing work platform by inverse kinematics to obtain the groove processing posture of the industrial robot of the groove processing work platform;

[0018] Based on the robot coordinates of the intersection, the industrial robot of the groove processing work platform is controlled to reach the intersection, and the industrial robot of the groove processing work platform walks a set distance in the processing feed direction in the groove processing posture to complete the groove processing process;

[0019] Sending the first pose of the two ArUco codes and the groove processing pose to the laser cladding work platform and the grinding and polishing work platform;

[0020] The laser cladding operation platform moves to the position of the hole-shaped part to be repaired, and identifies and laser clads the hole-shaped part to be repaired, including:

[0021] A second end surface image is acquired by a corresponding camera; the second end surface image is an image of the end surface of the hole-shaped part to be repaired after the groove processing step;

[0022] Preprocessing the second end surface image to obtain a preprocessed second end surface image;

[0023] Performing pose estimation on the preprocessed second end face image to obtain a second pose of each ArUco code; the second pose is the pose of the ArUco code in the camera coordinate system of the camera of the laser cladding work platform;

[0024] Based on the first posture, the second posture and the groove processing posture, a laser cladding posture is determined; the laser cladding posture is the posture of the end of the industrial robot of the laser cladding work platform;

[0025] Based on the laser cladding posture, the industrial robot of the laser cladding work platform performs a laser cladding and groove filling process on the hole-shaped part to be repaired;

[0026] The grinding and polishing operation platform moves to the position of the hole-shaped part to be repaired, and identifies and grinds and polishes the hole-shaped part to be repaired, including:

[0027] A third end surface image is acquired by a corresponding camera; the third end surface image is an image of the end surface of the hole-shaped part to be repaired after the laser cladding and groove filling process;

[0028] Preprocessing the third end surface image to obtain a preprocessed third end surface image;

[0029] Performing pose estimation on the preprocessed third end face image to obtain a third pose of each ArUco code; the third pose is the pose of the ArUco code in the camera coordinate system of the camera of the grinding and polishing work platform;

[0030] Based on the first posture, the third posture and the groove processing posture, a grinding and polishing posture is determined; the grinding and polishing posture is the posture of the end of the industrial robot of the grinding and polishing operation platform;

[0031] Based on the grinding and polishing posture, the industrial robot of the grinding and polishing work platform performs a cladding layer grinding and polishing process on the hole-shaped part to be repaired.

[0032] Optionally, camera calibration, including:

[0033] The camera is calibrated using Zhang Zhengyou's chessboard camera calibration method to obtain the camera's intrinsic parameter matrix and extrinsic parameter matrix.

[0034] Optionally, hand-eye calibration, including:

[0035] Determine the industrial robot base coordinate system, camera coordinate system, tool coordinate system and checkerboard calibration plate coordinate system;

[0036] Determine the transformation matrix from the camera coordinate system to the tool coordinate system, the transformation matrix from the tool coordinate system to the industrial robot base coordinate system, the transformation matrix from the checkerboard calibration plate coordinate system to the industrial robot base coordinate system, and the transformation matrix from the checkerboard calibration plate coordinate system to the camera coordinate system;

[0037] Based on each transformation matrix, construct the hand-eye calibration equation;

[0038] The dual quaternion method is used to solve the hand-eye calibration equation to obtain the hand-eye relationship matrix, and then the transformation matrix from the camera coordinate system to the industrial robot base coordinate system is obtained.

[0039] Optionally, determining the processing feed direction of the hole-shaped part to be repaired based on the preprocessed first end surface image includes:

[0040] Performing pose estimation on the preprocessed first end face image to obtain the first pose of each ArUco code; the first pose is the pose of the ArUco code in the camera coordinate system of the camera of the groove processing work platform;

[0041] Based on the first position of the two ArUco codes, determining the normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform;

[0042] The normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform is transformed into the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the processing feed direction.

[0043] Optionally, determining a first circumscribed circle and a second circumscribed circle, wherein the first circumscribed circle is a minimum circumscribed circle of the first contour, and the second circumscribed circle is a minimum circumscribed circle of the second contour, comprises:

[0044] calculating a length of the first contour and a length of the second contour;

[0045] When a preset length condition is met, calculating the roundness of the first contour and the roundness of the second contour; the preset length condition is that the length of the first contour meets a first length threshold and the length of the second contour meets a second length threshold;

[0046] When a preset roundness condition is met, the minimum circumscribed circle of the first contour is determined as the first circumscribed circle, and the minimum circumscribed circle of the second contour is determined as the second circumscribed circle; the preset roundness condition is that the roundness of the first contour satisfies a first roundness threshold and the roundness of the second contour satisfies a second roundness threshold.

[0047] Optionally, the pixel coordinates of the intersection are converted from a pixel coordinate system to an industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection, including:

[0048] According to the intrinsic parameter matrix of the camera of the groove processing work platform, the pixel coordinates of the intersection are converted from the pixel coordinate system to the camera coordinate system of the camera of the groove processing work platform to obtain the camera coordinates of the intersection;

[0049] According to the hand-eye calibration result of the groove processing work platform, the camera coordinates of the intersection are converted from the camera coordinate system of the camera of the groove processing work platform to the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection.

[0050] Optionally, determining a laser cladding posture based on the first posture, the second posture and the groove processing posture includes:

[0051] Determine a first pose transformation matrix according to the first pose and the second pose;

[0052] The laser cladding posture is determined according to the first posture transformation matrix and the groove processing posture.

[0053] Optionally, determining the laser cladding posture according to the first posture transformation matrix and the groove processing posture includes:

[0054] The first posture transformation matrix is ​​transformed from the camera coordinate system of the camera of the laser cladding work platform to the industrial robot base coordinate system of the industrial robot of the laser cladding work platform to obtain the transformed first posture transformation matrix;

[0055] The laser cladding posture is determined according to the converted first posture transformation matrix and the groove processing posture.

[0056] Optionally, determining a grinding and polishing posture based on the first posture, the third posture and the groove processing posture includes:

[0057] Determining a second posture transformation matrix according to the first posture and the third posture;

[0058] The grinding and polishing posture is determined according to the second posture transformation matrix and the groove processing posture.

[0059] Optionally, determining a grinding and polishing posture according to the second posture transformation matrix and the groove processing posture includes:

[0060] The second posture transformation matrix is ​​transformed from the camera coordinate system of the camera of the grinding and polishing operation platform to the industrial robot base coordinate system of the industrial robot of the grinding and polishing operation platform to obtain a transformed second posture transformation matrix;

[0061] The grinding and polishing posture is determined according to the converted second posture transformation matrix and the groove processing posture.

[0062] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0063] The present application discloses a method for repairing hole-shaped parts based on multi-process robot workpiece calibration. It aims at the identification and positioning problems between industrial robots in different processes of hole-shaped part crack repair, and proposes an effective method for solving the problem of locating the crack position in the hole-shaped parts; it improves the repair efficiency. Compared with the traditional industrial robot teaching processing method, the method proposed in the present application realizes the automation of the industrial robot repair process and reduces the dependence on the manual teaching method; it improves the consistency of the repair effect, and the collaborative positioning between industrial robots in different processes reduces human intervention and avoids the deviation that may be introduced by manual operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0065] Figure 1 A schematic flow chart of a method for repairing a hole-shaped part based on multi-process robot workpiece calibration provided in one embodiment of the present application;

[0066] Figure 2 It is a schematic diagram of the specific implementation process of the hole-shaped parts repair method based on multi-process robot workpiece calibration;

[0067] Figure 3 A schematic diagram of the process of repairing hole-shaped parts for a multi-process robot repair system to achieve multi-process robot workpiece calibration;

[0068] Figure 4 This is a diagram of the hand-eye relationship;

[0069] Figure 5 This is a schematic diagram of ArUco code;

[0070] Figure 6 This is the end view with ArUco code and circular marking points;

[0071] Figure 7 Schematic diagram of ArUco recognition and pose estimation;

[0072] Figure 8 This is a schematic diagram of the camera imaging model;

[0073] Fig. 9 To identify circular holes and circular landmarks and obtain a schematic diagram of the location to be repaired. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0075] The purpose of this application is to provide a method for repairing hole-shaped parts based on multi-process robot workpiece calibration, aiming to improve the repair efficiency of hole-shaped parts.

[0076] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0077] In an exemplary embodiment, Figure 1-Figure 3 As shown, the hole-shaped part repair method based on multi-process robot workpiece calibration in this embodiment is implemented based on a multi-process robot repair system. The multi-process robot repair system includes: a groove processing work platform, a laser cladding work platform and a grinding and polishing work platform. The groove processing work platform, the laser cladding work platform and the grinding and polishing work platform are each provided with an industrial robot. The end of each industrial robot carries a camera, which is an industrial vision binocular camera. The hole-shaped part to be repaired is fixed, and the three platforms are dynamically moved to the part position in turn. The hole-shaped part repair method based on multi-process robot workpiece calibration includes:

[0078] Step 1: Perform camera calibration and hand-eye calibration on the groove processing platform, laser cladding platform and grinding and polishing platform respectively.

[0079] As an optional implementation, camera calibration includes:

[0080] Step 101: Use Zhang Zhengyou's chessboard camera calibration method to calibrate the camera to obtain the camera's intrinsic parameter matrix and extrinsic parameter matrix.

[0081] Specifically, the expressions of the internal parameter matrix and the external parameter matrix are:

[0082]

[0083]

[0084] Among them, K is the internal parameter matrix; f x is the focal distance along the x direction; f y is the focal distance along the y direction; (c x ,c y ) is the center point of the camera's optical axis, which represents the projection position of the center point on the image plane in the camera coordinate system; [R|t] is the external parameter matrix; r 11 is the element of the first row and first column of the rotation matrix; r 12 is the element in the first row and second column of the rotation matrix; r 13 is the element of the first row and third column of the rotation matrix; t1 is the first element of the translation matrix; r 21 is the element of the 2nd row and 1st column of the rotation matrix; r 22 is the element in the 2nd row and 2nd column of the rotation matrix; r 23 is the element of the 2nd row and 3rd column of the rotation matrix; t2 is the 2nd element of the translation matrix; r 31 is the element in the 3rd row and 1st column of the rotation matrix; r 32 is the element in the 3rd row and 2nd column of the rotation matrix; r 33 is the element of the 3rd row and 3rd column of the rotation matrix; t3 is the 3rd element of the translation matrix.

[0085] As an optional implementation, the hand-eye calibration includes:

[0086] Step 111: Determine the industrial robot base coordinate system, the camera coordinate system, the tool coordinate system and the checkerboard calibration plate coordinate system.

[0087] Specifically, in the industrial robot visual positioning system, the hand-eye relationship is mainly divided into two categories: eyes on the hand (the camera is fixed at the end of the industrial robot) and eyes outside the hand (the camera is fixed outside the industrial robot). This application adopts the method of fixing the camera at the end of the industrial robot, so the hand-eye relationship is eyes on the hand.

[0088] like Figure 4 As shown, the industrial robot base coordinate system of each industrial robot is represented by {B}, the camera coordinate system of each camera is represented by {C}, the tool coordinate system of each tool is represented by {E}, and the checkerboard calibration plate coordinate system is represented by {S}.

[0089] Tools refer to the processing tools at the end of the industrial robot's robotic arm: welding gun, milling spindle, and grinding spindle.

[0090] Step 112: Determine the transformation matrix from the camera coordinate system to the tool coordinate system, the transformation matrix from the tool coordinate system to the industrial robot base coordinate system, the transformation matrix from the checkerboard calibration plate coordinate system to the industrial robot base coordinate system, and the transformation matrix from the checkerboard calibration plate coordinate system to the camera coordinate system.

[0091] Step 113: Construct a hand-eye calibration equation based on each transformation matrix.

[0092] Specifically, the relationship between the transformation matrices obtained through the posture transformation relationship is:

[0093]

[0094] in, Represents the transformation matrix from the checkerboard calibration plate coordinate system to the industrial robot base coordinate system; Represents the transformation matrix from the tool coordinate system to the industrial robot base coordinate system; Represents the transformation matrix from the camera coordinate system to the tool coordinate system; Represents the transformation matrix from the checkerboard calibration plate coordinate system to the camera coordinate system.

[0095] Move the end of the industrial robot to a position where the two sets of rotation axes are not parallel. Since the relative position of the checkerboard calibration plate coordinate system and the industrial robot base coordinate system is fixed, the transformation matrix The hand-eye calibration equation is unchanged:

[0096]

[0097] Step 114: Use the dual quaternion method to solve the hand-eye calibration equation to obtain the hand-eye relationship matrix, and then obtain the transformation matrix from the camera coordinate system to the industrial robot base coordinate system.

[0098] Specifically, the dual quaternion method is used to solve the hand-eye calibration equation to obtain the hand-eye relationship matrix , and then obtain the transformation matrix from the camera coordinate system to the industrial robot base coordinate system :

[0099]

[0100] Among them, R CB is the rotation matrix from the camera coordinate system to the industrial robot base coordinate system; t CB is the translation matrix from the camera coordinate system to the industrial robot base coordinate system.

[0101] Step 2: The groove processing work platform moves to the position of the hole-shaped part to be repaired, and the hole-shaped part to be repaired is identified and grooved, including:

[0102] Step 201: Capture a first end face image through a corresponding camera; the first end face image is an image of the unprocessed end face of the hole-shaped part to be repaired, and the end face is the end face of the crack of the hole-shaped part to be repaired with two ArUco codes attached. Figure 5 The position of the ArUco code is shown in Figure 6 As shown, two ArUco codes are respectively arranged at two ends of the circular hole of the hole-shaped part to be repaired.

[0103] Step 202: preprocessing the first end face image to obtain a preprocessed first end face image; the preprocessing includes: scaling, grayscale, Gaussian filtering and binarization.

[0104] Specifically, the first end face image is scaled according to a preset ratio to reduce the amount of calculation and speed up the processing speed; the scaled first end face image is grayed to obtain a first gray image; the first gray image is blurred using a Gaussian filter, the first gray image is convolved with a Gaussian kernel, the intensity values ​​of adjacent pixels are averaged, the high-frequency noise in the first gray image is reduced, and the edge of the first gray image is smoother using the OTSU algorithm; the first gray image after the Gaussian filter is binarized, and a threshold value that maximizes the inter-class variance of the first gray image is calculated, pixels with gray values ​​greater than or equal to the threshold are assigned 255 (white), and pixels with gray values ​​less than the threshold are assigned 0 (black), thereby dividing the image pixels into foreground (white) and background (gray) categories.

[0105] Step 203: Determine the processing feed direction of the hole-shaped part to be repaired based on the preprocessed first end surface image.

[0106] As an optional implementation, step 203 includes:

[0107] Step 2031: estimate the pose of the preprocessed first end face image to obtain the first pose of each ArUco code; the first pose is the pose of the ArUco code in the camera coordinate system of the camera of the groove processing work platform.

[0108] Specifically, Figure 7 As shown, the cv::aruco::estimatePoseSingleMarkers() function in OpenCV is used to solve the preprocessed first end face image to obtain the first position of each ArUco code.

[0109] Step 2032: Based on the first position of the two ArUco codes, determine the normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform.

[0110] Specifically, the plane where the two ArUco codes are located coincides with the end face of the hole-shaped part to be repaired, and the normal vectors in the first position of the two ArUco codes are averaged to obtain the normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform.

[0111] Step 2033: transform the normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform into the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the processing feed direction.

[0112] Specifically, the normal vector The processing feed direction of the hole-shaped parts to be repaired is obtained by transforming the industrial robot base coordinate system of the industrial robot on the groove processing work platform. It is the processing feed direction.

[0113] The principle of camera pinhole imaging is as follows Figure 8 As shown, O w X w Y w Z w is the world coordinate system, O c X c Y c Z c is the camera coordinate system, ouv is the pixel coordinate system, and OXY is the image coordinate system.

[0114] The relationship between the world coordinate system and the pixel coordinate system is:

[0115]

[0116] Where s is the proportionality coefficient.

[0117] Steps 204 to 207 are all completed in the pixel coordinate system.

[0118] Step 204: performing an opening operation on the preprocessed first end face image to obtain the first end face image after the opening operation; the opening operation includes: corrosion processing and dilation operation.

[0119] Specifically, the preprocessed first end surface image is firstly corroded, and then a dilation operation is performed on the corrosion result to ensure the integrity of the circular or approximately circular portion in the image, further smooth the image, and remove noise.

[0120] Step 205: Use a polygonal approximation method to detect the first end face image after the opening operation to obtain the contour of the hole-shaped part to be repaired; the contour of the hole-shaped part to be repaired includes: a first contour and a second contour; the first contour is the contour of the circular hole of the hole-shaped part to be repaired, and the second contour is the contour of the circular mark point, which is attached to the end face and corresponds to the crack. The position of the circular mark point is as follows Figure 6 shown.

[0121] Step 206: Determine a first circumscribed circle and a second circumscribed circle; the first circumscribed circle is the minimum circumscribed circle of the first contour, and the second circumscribed circle is the minimum circumscribed circle of the second contour.

[0122] As an optional implementation, step 206 includes:

[0123] Step 2061: Calculate the length of the first contour and the length of the second contour.

[0124] Specifically, the calculation formula for the length of the contour is:

[0125]

[0126] Where L is the length of the contour; n is the number of contour boundary points; (x i+1 ,y i+1 ) is the pixel coordinate of the i+1th contour boundary point; (x i ,y i ) is the pixel coordinate of the i-th contour boundary point.

[0127] Step 2062: When a preset length condition is met, the roundness of the first contour and the roundness of the second contour are calculated. The preset length condition is that the length of the first contour meets a first length threshold and the length of the second contour meets a second length threshold.

[0128] Specifically, the calculation formula for the roundness of the contour is:

[0129]

[0130]

[0131] Among them, C is the roundness of the contour; S is the area of ​​the contour.

[0132] Step 2063: When a preset roundness condition is met, the minimum circumscribed circle of the first contour is determined as the first circumscribed circle, and the minimum circumscribed circle of the second contour is determined as the second circumscribed circle; the preset roundness condition is that the roundness of the first contour satisfies the first roundness threshold and the roundness of the second contour satisfies the second roundness threshold.

[0133] Step 207: Determine the pixel coordinates of the intersection of the line connecting the center of the first circumscribed circle and the center of the second circumscribed circle and the circular hole.

[0134] Specifically, the standard equation of a circle is:

[0135] (xx c ) 2 +(yy c ) 2 =r 2 (10)

[0136] Among them, (x c ,y c ) is the center of the circle; r is the radius.

[0137] From formula (10), the standard equations of the second circumscribed circle and the first circumscribed circle are:

[0138] (xx c1 ) 2 +(yy c1 ) 2 =r1 2 (11)

[0139] (xx c2 ) 2 +(yy c2 ) 2 =r2 2 (12)

[0140] Among them, (x c1 ,y c1 ) is the center of the second circumscribed circle; r1 is the radius of the second circumscribed circle; (x c2 ,y c2 ) is the center of the first circumscribed circle; r2 is the radius of the first circumscribed circle.

[0141] The line connecting the center of the first circumscribed circle and the center of the second circumscribed circle is a line segment. Any point (x, y) on the line segment can be expressed as:

[0142]

[0143] Among them, the parameter α varies in the range of [0, 1]; (x1, y1) and (x2, y2) are any two points on the line segment.

[0144] Substituting the parametric equation of the line segment into the equation of the first circumscribed circle, we get:

[0145] (x1+α(x2-x1)-x c2 ) 2 +(y1+α(y2-y1)-y c2 ) 2 =r2 2 (14)

[0146] Assume d x =x2-x1,d y =y2-y1,l x =x1-x c2 , l y =y1-y c2 , substituting into formula (14) and rearranging it, we can get:

[0147] α 2 (d x 2 +d y 2 )+2α(l x d x +l y d y )+(l x 2 +l y 2 -r2 2 )=0(15)

[0148] Let a = d x 2 +d y 2 , b=2(l x d x +l y d y ), c=l x 2 +l y 2 -r2 2 , then the quadratic equation about α is:

[0149] aα 2 +bα+c=0(16)

[0150] Using the discriminant Δ = b 2 -4ac determines the situation of the root of the equation. If Δ<0, the line segment has no intersection with the circular hole; if Δ=0, the line segment is tangent to the circular hole and has only one intersection; if Δ>0, the line segment has two intersections with the circular hole. If the discriminant Δ≥0, two solutions for the parameter α can be obtained:

[0151]

[0152] According to the range of α [0, 1], the solution of α is selected to obtain the pixel coordinates of the intersection point A of the line segment and the circle as (u+αd x ,v+αd y ). The intersection point A is Fig. 9 The position to be repaired is the crack position.

[0153] Step 208: Convert the pixel coordinates of the intersection point from the pixel coordinate system to the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection point.

[0154] As an optional implementation, step 208 includes:

[0155] Step 2081: According to the intrinsic parameter matrix of the camera of the groove processing work platform, the pixel coordinates of the intersection point are converted from the pixel coordinate system to the camera coordinate system of the camera of the groove processing work platform to obtain the camera coordinates of the intersection point.

[0156] Specifically, the camera coordinates of the intersection point (X C ,Y C ,Z C ) is calculated as:

[0157]

[0158]

[0159] Among them, Z C is the depth distance of the camera; f is the focal length of the camera; B is the baseline distance; d1 is the parallax.

[0160] Step 2082: According to the hand-eye calibration result of the groove processing work platform, the camera coordinates of the intersection are converted from the camera coordinate system of the camera of the groove processing work platform to the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection.

[0161] Specifically, the robot coordinates of the intersection point (X B ,Y B ,Z B ) is calculated as:

[0162]

[0163] in, A transformation matrix for converting the camera coordinate system of the camera of the groove processing work platform to the industrial robot base coordinate system of the industrial robot of the groove processing work platform.

[0164] Step 209: Inverse kinematics is used to solve the six joint angles of the industrial robot of the groove processing work platform to obtain the groove processing posture of the industrial robot of the groove processing work platform.

[0165] Step 210: Based on the robot coordinates of the intersection, the industrial robot of the groove processing work platform is controlled to reach the intersection, and the industrial robot of the groove processing work platform walks a set distance in the processing feed direction in the groove processing posture to complete the groove processing process.

[0166] Step 211: Send the first position and groove processing position of the two ArUco codes to the laser cladding work platform and the grinding and polishing work platform.

[0167] Step 3: The laser cladding work platform moves to the location of the hole-shaped part to be repaired, and identifies and laser clads the hole-shaped part to be repaired, including:

[0168] Step 301: Capture a second end surface image through a corresponding camera; the second end surface image is an image of the end surface of the hole-shaped part to be repaired after the groove processing step.

[0169] Step 302: preprocess the second end surface image to obtain a preprocessed second end surface image.

[0170] Step 303: performing pose estimation on the preprocessed second end face image to obtain a second pose of each ArUco code; the second pose is the pose of the ArUco code in the camera coordinate system of the camera of the laser cladding work platform.

[0171] Step 304: Determine the laser cladding posture based on the first posture, the second posture and the groove processing posture; the laser cladding posture is the posture of the end of the industrial robot of the laser cladding work platform.

[0172] As an optional implementation, step 304 includes:

[0173] Step 3041: Determine the first pose transformation matrix according to the first pose and the second pose.

[0174] Specifically, the calculation formula of the first pose transformation matrix is:

[0175]

[0176] in, is the transformation matrix of the first pose; T2 is the second pose; T1 is the first pose.

[0177] Step 3042: Determine the laser cladding posture according to the first posture transformation matrix and the groove processing posture.

[0178] As an optional implementation, step 3042 includes:

[0179] Step 30421: convert the first pose transformation matrix from the camera coordinate system of the laser cladding work platform to the industrial robot base coordinate system of the industrial robot of the laser cladding work platform to obtain the converted first pose transformation matrix.

[0180] Specifically, the calculation formula of the first pose transformation matrix after conversion is:

[0181]

[0182] in, is the first pose transformation matrix after conversion; It is the transformation matrix from the camera coordinate system of the laser cladding work platform to the industrial robot base coordinate system of the laser cladding work platform.

[0183] Step 30422: Determine the laser cladding posture according to the converted first posture transformation matrix and groove processing posture.

[0184] Specifically, the calculation formula of laser cladding posture is:

[0185]

[0186] Among them, T Laser is the laser cladding posture; T groove This is the groove processing posture.

[0187] Step 305: Based on the laser cladding posture, the industrial robot of the laser cladding work platform performs a laser cladding and groove filling process on the hole-shaped part to be repaired.

[0188] Step 4: The grinding and polishing platform moves to the location of the hole-shaped part to be repaired, and identifies and grinds the hole-shaped part to be repaired, including:

[0189] Step 401: Capture a third end face image through a corresponding camera; the third end face image is an image of the end face of the hole-shaped part to be repaired after the laser cladding and groove filling process.

[0190] Step 402: preprocess the third end surface image to obtain a preprocessed third end surface image.

[0191] Step 403: perform pose estimation on the preprocessed third end face image to obtain a third pose of each ArUco code; the third pose is the pose of the ArUco code in the camera coordinate system of the camera of the polishing work platform.

[0192] Step 404: Based on the first posture, the third posture and the groove processing posture, determine the grinding and polishing posture; the grinding and polishing posture is the posture of the end of the industrial robot of the grinding and polishing work platform.

[0193] As an optional implementation, step 404 includes:

[0194] Step 4041: Determine the second pose transformation matrix based on the first pose and the third pose.

[0195] Specifically, the calculation formula of the second posture transformation matrix is:

[0196]

[0197] in, is the transformation matrix of the second pose; T3 is the transformation matrix of the third pose.

[0198] Step 4042: Determine the grinding and polishing posture according to the second posture transformation matrix and the groove processing posture.

[0199] As an optional implementation, step 4042 includes:

[0200] Step 40421: convert the second pose transformation matrix from the camera coordinate system of the camera of the grinding and polishing work platform to the industrial robot base coordinate system of the industrial robot of the grinding and polishing work platform to obtain the converted second pose transformation matrix.

[0201] Specifically, the calculation formula of the converted second posture transformation matrix is:

[0202]

[0203] in, is the second pose transformation matrix after conversion; It is the transformation matrix from the camera coordinate system of the camera of the grinding and polishing work platform to the industrial robot base coordinate system of the grinding and polishing work platform.

[0204] Step 40422: Determine the grinding and polishing posture according to the converted second posture transformation matrix and the groove processing posture.

[0205] Specifically, the calculation formula for the grinding and polishing posture is:

[0206]

[0207] Among them, T Grinding This is the grinding and polishing posture.

[0208] Step 405: Based on the grinding and polishing posture, the industrial robot of the grinding and polishing operation platform performs a cladding layer grinding and polishing process on the hole-shaped part to be repaired.

[0209] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 specification.

[0210] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A hole-shaped part repair method based on multi-process robot workpiece calibration, the hole-shaped part repair method based on multi-process robot workpiece calibration is implemented based on a multi-process robot repair system, and the multi-process robot repair system includes: A groove processing platform, a laser cladding platform and a grinding and polishing platform, each of which is provided with an industrial robot, and each of which carries a camera at the end, is characterized in that the hole-shaped parts repair method based on multi-process robot workpiece calibration includes: Performing camera calibration and hand-eye calibration on the groove processing work platform, the laser cladding work platform and the grinding and polishing work platform respectively; The groove processing work platform moves to the position of the hole-shaped part to be repaired, and performs groove processing on the hole-shaped part to be repaired, including: A first end face image is acquired by a corresponding camera; the first end face image is an image of an unprocessed end face of the hole-shaped part to be repaired, and the end face is an end face of a crack of the hole-shaped part to be repaired with two ArUco codes attached thereto; Preprocessing the first end face image to obtain a preprocessed first end face image; the preprocessing includes: scaling, grayscale, Gaussian filtering and binarization; Determining a processing feed direction of the hole-shaped part to be repaired based on the preprocessed first end surface image; Performing an opening operation on the preprocessed first end face image to obtain the first end face image after opening operation; the opening operation includes: corrosion processing and expansion operation; A polygonal approximation method is used to detect the first end face image after the opening operation to obtain the contour of the hole-shaped part to be repaired; the contour of the hole-shaped part to be repaired includes: a first contour and a second contour; the first contour is the contour of the circular hole of the hole-shaped part to be repaired, and the second contour is the contour of a circular mark point, and the circular mark point is attached to the end face and corresponds to the crack; Determine a first circumscribed circle and a second circumscribed circle; the first circumscribed circle is the minimum circumscribed circle of the first contour, and the second circumscribed circle is the minimum circumscribed circle of the second contour; Determine the pixel coordinates of an intersection point of a line connecting the center of the first circumscribed circle and the center of the second circumscribed circle and the circular hole; Convert the pixel coordinates of the intersection point from the pixel coordinate system to the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection point; Solve the six joint angles of the industrial robot of the groove processing work platform by inverse kinematics to obtain the groove processing posture of the industrial robot of the groove processing work platform; Based on the robot coordinates of the intersection, the industrial robot of the groove processing work platform is controlled to reach the intersection, and the industrial robot of the groove processing work platform walks a set distance in the processing feed direction in the groove processing posture to complete the groove processing process; Sending the first pose of the two ArUco codes and the groove processing pose to the laser cladding work platform and the grinding and polishing work platform; The laser cladding operation platform moves to the position of the hole-shaped part to be repaired, and identifies and laser clads the hole-shaped part to be repaired, including: A second end surface image is acquired by a corresponding camera; the second end surface image is an image of the end surface of the hole-shaped part to be repaired after the groove processing step; Preprocessing the second end surface image to obtain a preprocessed second end surface image; Performing pose estimation on the preprocessed second end face image to obtain a second pose of each ArUco code; the second pose is the pose of the ArUco code in the camera coordinate system of the camera of the laser cladding work platform; Based on the first posture, the second posture and the groove processing posture, a laser cladding posture is determined; the laser cladding posture is the posture of the end of the industrial robot of the laser cladding work platform; Based on the laser cladding posture, the industrial robot of the laser cladding work platform performs a laser cladding and groove filling process on the hole-shaped part to be repaired; The grinding and polishing operation platform moves to the position of the hole-shaped part to be repaired, and identifies and grinds and polishes the hole-shaped part to be repaired, including: A third end surface image is acquired by a corresponding camera; the third end surface image is an image of the end surface of the hole-shaped part to be repaired after the laser cladding and groove filling process; Preprocessing the third end surface image to obtain a preprocessed third end surface image; Performing pose estimation on the preprocessed third end face image to obtain a third pose of each ArUco code; the third pose is the pose of the ArUco code in the camera coordinate system of the camera of the grinding and polishing work platform; Based on the first posture, the third posture and the groove processing posture, a grinding and polishing posture is determined; the grinding and polishing posture is the posture of the end of the industrial robot of the grinding and polishing operation platform; Based on the grinding and polishing posture, the industrial robot of the grinding and polishing work platform performs a cladding layer grinding and polishing process on the hole-shaped part to be repaired.

2. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 1 is characterized in that: Camera calibration, including: The camera is calibrated using Zhang Zhengyou's chessboard camera calibration method to obtain the camera's intrinsic parameter matrix and extrinsic parameter matrix.

3. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 1 is characterized in that: Hand-eye calibration, including: Determine the industrial robot base coordinate system, camera coordinate system, tool coordinate system and checkerboard calibration plate coordinate system; Determine the transformation matrix from the camera coordinate system to the tool coordinate system, the transformation matrix from the tool coordinate system to the industrial robot base coordinate system, the transformation matrix from the checkerboard calibration plate coordinate system to the industrial robot base coordinate system, and the transformation matrix from the checkerboard calibration plate coordinate system to the camera coordinate system; Based on each transformation matrix, construct the hand-eye calibration equation; The dual quaternion method is used to solve the hand-eye calibration equation to obtain the hand-eye relationship matrix, and then the transformation matrix from the camera coordinate system to the industrial robot base coordinate system is obtained.

4. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 1 is characterized in that: Determining a processing feed direction of the hole-shaped part to be repaired based on the preprocessed first end surface image includes: Performing pose estimation on the preprocessed first end face image to obtain the first pose of each ArUco code; the first pose is the pose of the ArUco code in the camera coordinate system of the camera of the groove processing work platform; Based on the first position of the two ArUco codes, determining the normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform; The normal vector of the end face in the camera coordinate system of the camera of the groove processing work platform is transformed into the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the processing feed direction.

5. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 1, characterized in that: Determining a first circumscribed circle and a second circumscribed circle, wherein the first circumscribed circle is the minimum circumscribed circle of the first contour, and the second circumscribed circle is the minimum circumscribed circle of the second contour, comprises: calculating a length of the first contour and a length of the second contour; When a preset length condition is met, calculating the roundness of the first contour and the roundness of the second contour; the preset length condition is that the length of the first contour meets a first length threshold and the length of the second contour meets a second length threshold; When a preset roundness condition is met, the minimum circumscribed circle of the first contour is determined as the first circumscribed circle, and the minimum circumscribed circle of the second contour is determined as the second circumscribed circle; the preset roundness condition is that the roundness of the first contour satisfies a first roundness threshold and the roundness of the second contour satisfies a second roundness threshold.

6. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 2 is characterized in that: The pixel coordinates of the intersection are converted from a pixel coordinate system to an industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection, including: According to the intrinsic parameter matrix of the camera of the groove processing work platform, the pixel coordinates of the intersection are converted from the pixel coordinate system to the camera coordinate system of the camera of the groove processing work platform to obtain the camera coordinates of the intersection; According to the hand-eye calibration result of the groove processing work platform, the camera coordinates of the intersection are converted from the camera coordinate system of the camera of the groove processing work platform to the industrial robot base coordinate system of the industrial robot of the groove processing work platform to obtain the robot coordinates of the intersection.

7. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 1, characterized in that: Determining a laser cladding posture based on the first posture, the second posture and the groove processing posture includes: Determine a first pose transformation matrix according to the first pose and the second pose; The laser cladding posture is determined according to the first posture transformation matrix and the groove processing posture.

8. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 7 is characterized in that: Determining the laser cladding posture according to the first posture transformation matrix and the groove processing posture includes: The first posture transformation matrix is ​​transformed from the camera coordinate system of the camera of the laser cladding work platform to the industrial robot base coordinate system of the industrial robot of the laser cladding work platform to obtain the transformed first posture transformation matrix; The laser cladding posture is determined according to the converted first posture transformation matrix and the groove processing posture.

9. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 1, characterized in that: Determining a grinding and polishing posture based on the first posture, the third posture and the groove processing posture includes: Determining a second posture transformation matrix according to the first posture and the third posture; The grinding and polishing posture is determined according to the second posture transformation matrix and the groove processing posture.

10. The hole-shaped parts repair method based on multi-process robot workpiece calibration according to claim 9, characterized in that: Determining a grinding and polishing posture according to the second posture transformation matrix and the groove processing posture includes: The second posture transformation matrix is ​​transformed from the camera coordinate system of the camera of the grinding and polishing operation platform to the industrial robot base coordinate system of the industrial robot of the grinding and polishing operation platform to obtain a transformed second posture transformation matrix; The grinding and polishing posture is determined according to the converted second posture transformation matrix and the groove processing posture.

Images