A TCP calibration method and system for a welding gun robot based on machine vision

Abstract

The application provides a TCP calibration method and system of a welding gun robot based on machine vision, comprising: rotating an identifier so that the axis of one of the cylinders points to the preset X axis of the tool coordinate system; obtaining the pose matrix of the flange center of the end of the mechanical arm in the robot coordinate system when the robot reaches the preset position; obtaining the laser point cloud of the surface of the identifier when the robot reaches the preset position, and obtaining the pose matrix of the tool positioning point in the robot coordinate system by using the laser point cloud to identify the axes of the two cylinders in the identifier; and converting the pose matrix of the tool positioning point and the flange center in the robot coordinate system to obtain the calibration position of the tool positioning point; compared with the traditional four-point calibration method of manual operation, the method has high automation degree and improves the calibration efficiency of the tool coordinate system; the accuracy of identifying the cylinder axis by using the laser point cloud is higher than the accuracy of identifying the straight line from the spherical mark point to the spherical center, and therefore the calibration accuracy of the application is higher.

Description

Technical Field

[0001] This invention belongs to the field of robotic arm technology, specifically relating to a TCP calibration method and system for a welding torch robot based on machine vision. Background Technology

[0002] With the development of technology, robotic arms have been widely used in welding, handling, spraying, and sorting. When performing these functions, corresponding tools need to be installed at the end effector of the robotic arm. To accurately execute these functions, a positioning point, the tool center point (TCP), needs to be determined within the tool. This TCP is used to calculate the tool's spatial posture at any time. The accuracy of the TCP's positioning is crucial, directly affecting the tool's positioning precision.

[0003] Currently, the tool coordinate system of TCP is mainly determined by manually controlling a robotic arm to align the tool's center point (TCP) with a fixed reference point from four different positions. However, when the tool performs its functions, slight collisions inevitably occur, affecting the TCP's position. This often requires manual intervention for repositioning, which is inefficient and prone to introducing human error that affects positioning accuracy. Therefore, a new method exists that uses a laser point cloud recognition card attached to a feature marker sphere at the TCP for calibration. This method directly obtains the straight-line equation from the marker point on the feature marker sphere to the sphere's center using a laser scanning device, performing TCP calibration without manual intervention, saving time and increasing efficiency. However, because the straight-line equation from the point cloud recognition sphere's marker point to the sphere's center is inaccurate, the final calibration result may be inaccurate. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this invention proposes a TCP calibration method for welding torch robots based on machine vision, comprising:

[0005] The rotating marker makes the axis of one of the cylinders point to the preset X-axis of the tool coordinate system;

[0006] Obtain the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position;

[0007] The laser point cloud on the surface of the marker is obtained when the robot reaches the preset position. The laser point cloud is used to identify the axis of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system.

[0008] The calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system;

[0009] The marker comprises two perpendicularly intersecting cylinders located at the tip of the welding torch, which serves as the tool positioning point.

[0010] Preferably, the marker has a small hole, which is detachably connected to the tip of the welding gun. The bottom center of the small hole is the intersection of the axes of the two cylinders in the marker.

[0011] Preferably, the marker is inserted into the tip of the welding gun through a small hole, and the tip of the welding gun coincides with the intersection of the axes of the two cylinders in the marker.

[0012] Preferably, the material of the inner wall of the small hole is an elastic material.

[0013] Preferably, the two cylinders in the marker have different diameters.

[0014] Preferably, acquiring the laser point cloud on the surface of the marker when the robot reaches the preset position includes:

[0015] When the robot reaches the preset position, it uses a surface laser sensor to scan the marker to obtain a laser point cloud on the surface of the marker.

[0016] Preferably, obtaining the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position includes:

[0017] When the robot reaches the preset position, the robot outputs the coordinates and orientation of the flange center in the robot coordinate system, and obtains the pose matrix of the flange center based on the coordinates and orientation of the flange center in the robot coordinate system.

[0018] Preferably, the step of obtaining the pose matrix of the tool positioning point in the robot coordinate system using the axes of the two cylinders in the laser point cloud identification marker includes:

[0019] By using the axes of the two cylinders in the laser point cloud recognition marker, the axial coordinates of the preset X and Y axes of the tool coordinate system in the robot coordinate system are obtained, and the coordinates of the origin of the tool coordinate system in the robot coordinate system are obtained by using the intersection of the axes.

[0020] The cross product of the X-axis and Y-axis axial coordinates is used to obtain the Z-axis axial coordinate of the tool coordinate system in the robot coordinate system;

[0021] The pose matrix of the tool positioning point in the robot coordinate system is obtained by using the coordinates of the tool coordinate system at the origin in the robot coordinate system and the axial coordinates of the X, Y and Z axes.

[0022] The preset X-axis of the tool coordinate system points to the axis of one cylinder in the marker, and the preset Y-axis of the tool coordinate system points to the axis of another cylinder.

[0023] Preferably, the formula for calculating the calibration position of the tool positioning point is:

[0024] M = KT

[0025] Where M represents the pose matrix of the tool positioning point in the robot coordinate system; T represents the pose matrix of the flange center in the robot coordinate system; and K represents the calibration position of the tool positioning point.

[0026] Preferably, after obtaining the calibrated position of the tool positioning point, the method further includes:

[0027] After the robot reaches the preset position, its pose is changed multiple times to obtain the pose matrix of the tool positioning point in the robot coordinate system under each pose. Then, the corresponding tool positioning point calibration position is obtained by using the pose matrix of the tool positioning point and the flange center in the robot coordinate system under each pose. The optimal tool positioning point calibration position is obtained by calculating multiple tool positioning point calibration positions using the least squares method.

[0028] Based on the same inventive concept, this invention also proposes a TCP calibration system for a welding torch robot based on machine vision, comprising:

[0029] Control module: The rotation marker makes the axis of one of the cylinders point to the preset X-axis of the tool coordinate system;

[0030] Position Analysis Module: Used to obtain the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position; obtain the laser point cloud on the surface of the marker when the robot reaches the preset position, and use the laser point cloud to identify the axis of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system;

[0031] Position determination module: The calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system.

[0032] Preferably, in the control module, the marker has a small hole, which is detachably connected to the tip of the welding gun. The bottom center of the small hole is the intersection of the axes of the two cylinders in the marker.

[0033] Preferably, the marker is inserted into the tip of the welding gun through a small hole, and the tip of the welding gun coincides with the intersection of the axes of the two cylinders in the marker.

[0034] Preferably, the material of the inner wall of the small hole is an elastic material.

[0035] Preferably, the two cylinders in the marker have different diameters.

[0036] Preferably, in the position analysis module, acquiring the laser point cloud on the surface of the marker when the robot reaches the preset position includes:

[0037] When the robot reaches the preset position, it uses a surface laser sensor to scan the marker to obtain a laser point cloud on the surface of the marker.

[0038] Preferably, obtaining the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position includes:

[0039] When the robot reaches the preset position, the robot outputs the coordinates and orientation of the flange center in the robot coordinate system, and obtains the pose matrix of the flange center based on the coordinates and orientation of the flange center in the robot coordinate system.

[0040] Preferably, the step of obtaining the pose matrix of the tool positioning point in the robot coordinate system using the axes of the two cylinders in the laser point cloud identification marker includes:

[0041] By using the axes of the two cylinders in the laser point cloud recognition marker, the axial coordinates of the preset X and Y axes of the tool coordinate system in the robot coordinate system are obtained, and the coordinates of the origin of the tool coordinate system in the robot coordinate system are obtained by using the intersection of the axes.

[0042] The cross product of the X-axis and Y-axis axial coordinates is used to obtain the Z-axis axial coordinate of the tool coordinate system in the robot coordinate system;

[0043] The pose matrix of the tool positioning point in the robot coordinate system is obtained by using the coordinates of the tool coordinate system at the origin in the robot coordinate system and the axial coordinates of the X, Y and Z axes.

[0044] The preset X-axis of the tool coordinate system points to the axis of one cylinder in the marker, and the preset Y-axis of the tool coordinate system points to the axis of another cylinder.

[0045] Preferably, in the position determination module, the formula for calculating the calibration position of the tool positioning point is:

[0046] M = KT

[0047] Where M represents the pose matrix of the tool positioning point in the robot coordinate system; T represents the pose matrix of the flange center in the robot coordinate system; and K represents the calibration position of the tool positioning point.

[0048] Preferably, after obtaining the calibrated position of the tool positioning point, the method further includes:

[0049] After the robot reaches the preset position, its pose is changed multiple times to obtain the pose matrix of the tool positioning point in the robot coordinate system under each pose. Then, the corresponding tool positioning point calibration position is obtained by using the pose matrix of the tool positioning point and the flange center in the robot coordinate system under each pose. The optimal tool positioning point calibration position is obtained by calculating multiple tool positioning point calibration positions using the least squares method.

[0050] Compared with the closest existing technology, the present invention has the following beneficial effects:

[0051] This invention proposes a TCP calibration method and system for a welding torch robot based on machine vision. The method involves rotating a marker so that the axis of one of the cylinders points to a preset X-axis in the tool coordinate system; obtaining the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position; obtaining the laser point cloud on the marker surface when the robot reaches the preset position; using the laser point cloud to identify the axes of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system; and converting the pose matrices of the tool positioning point and the flange center in the robot coordinate system to obtain the calibration position of the tool positioning point. The laser point cloud can directly identify the axes of the two cylinders in the marker, thereby obtaining the origin and three axial coordinates of the tool coordinate system for TCP position calibration. Compared to the traditional manual four-point calibration method, this method, except for manually installing the marker at the tip of the welding torch, allows the robot to complete all other steps via software control, resulting in a high degree of automation and improved calibration efficiency of the welding torch tool coordinate system. Furthermore, the accuracy of laser point cloud identification of the cylinder axis is higher than the accuracy of identifying the straight line from the spherical marker point to the center of the sphere, thus the calibration result of this invention has higher accuracy. Attached Figure Description

[0052] Figure 1 A schematic diagram of a TCP calibration method for a welding torch robot based on machine vision provided by the present invention;

[0053] Figure 2 This is a schematic diagram of the front view structure of a TCP calibration device for a welding torch robot based on machine vision.

[0054] Figure 3 This is a top view schematic diagram of a TCP calibration device for a welding torch robot based on machine vision.

[0055] Figure 4 A schematic diagram of the structure of a TCP calibration system for a welding torch robot based on machine vision provided by the present invention;

[0056] Among them, 1. large cylinder; 2. small cylinder; 3. small hole. Detailed Implementation

[0057] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0058] Example 1:

[0059] This invention provides a TCP calibration method for a welding torch robot based on machine vision, such as... Figure 1 As shown, it includes:

[0060] S1. Rotate the marker so that the axis of one of the cylinders points to the preset X-axis of the tool coordinate system;

[0061] S2. Obtain the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position;

[0062] S3. Obtain the laser point cloud on the surface of the marker when the robot reaches the preset position, and use the laser point cloud to identify the axis of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system.

[0063] S4. The calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system.

[0064] In step S1, the rotating marker is used to align the axis of one of the cylinders with the preset X-axis of the tool coordinate system, specifically including:

[0065] The marker comprises two perpendicularly intersecting cylinders located at the tip of the welding torch, which serves as the tool positioning point. The two cylinders have different diameters to distinguish the X-axis and Y-axis during calibration. Figure 2 As shown, these are referred to as the large cylinder and the small cylinder, respectively. It is recommended that the diameter of the large cylinder be 1.5 times the diameter of the small cylinder. The specific diameter is not fixed and should be set by the staff for ease of installation during calibration.

[0066] The marker has a small hole through which it can be detachably connected to the tip of a welding torch. The center of the bottom of the small hole is the intersection of the axes of the two cylinders in the marker. Figure 3 As shown, the marker is inserted into the tip of the welding torch through a small hole, with the welding torch tip coinciding with the intersection of the axes of the two cylinders within the marker. The diameter of the small hole is slightly smaller than the diameter of the welding torch tip, which is made of welding wire. The inner wall of the small hole is made of an elastic material. This design of the small hole diameter and inner wall material facilitates a secure insertion of the marker into the welding torch tip. The marker is made of a high-density, rigid material to ensure easy installation on the welding torch tip while minimizing deformation.

[0067] Rotate the marker attached to the tip of the welding torch to align the axis of the large cylinder with the preset X-axis of the tool coordinate system.

[0068] In step S2, the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position is obtained, specifically including:

[0069] Once the robot reaches the preset position, the coordinates and orientation of the flange center in the robot coordinate system are used to obtain the pose matrix T of the flange center.

[0070] In step S3, the laser point cloud on the surface of the marker is obtained when the robot reaches the preset position. The pose matrix of the tool positioning point in the robot coordinate system is obtained by identifying the axes of the two cylinders in the marker using the laser point cloud. Specifically, this includes:

[0071] A 3D laser scanning device is placed under the robot's robotic arm to ensure that the device can illuminate the surface of the marker when emitting a surface laser, thus acquiring the laser point cloud of the marker's surface as comprehensively as possible. However, according to the point cloud recognition algorithm for cylinders, only 30% of the point cloud is needed to identify a cylinder. Therefore, it is not necessary for the surface laser sensor to acquire the complete laser point cloud. After the robot reaches the preset position, the surface laser sensor scans the marker at the i-th pose to obtain the laser point cloud on the marker's surface.

[0072] By using laser point cloud recognition to identify the axes of the two cylinders in the i-th pose of the robot, the axial coordinates of the preset X-axis and Y-axis of the tool coordinate system in the robot coordinate system are obtained. The preset X-axis of the tool coordinate system is the axis pointing of the large cylinder in the marker, and the preset Y-axis of the tool coordinate system is the axis pointing of the small cylinder.

[0073] Specifically, the equation of the axis of the large cylinder is: Then, when the robot is in the i-th pose, the preset X-axis coordinate of the tool coordinate system in the robot coordinate system is (m ix n ix p ix ), where x, y, z are parameters in the equation of the line, and t ix Let x be the parameter in the equation of the straight line of the axis of the large cylinder when the robot is in the i-th pose; i0 y i0 , z i0 Let be the coordinates of a point on the axis of the large cylinder when the robot is in the i-th pose;

[0074] The equation of the axis of the small cylinder is: Then, when the robot is in the i-th pose, the preset Y-axis coordinate of the tool coordinate system in the robot coordinate system is (m iy n iy p iy ), where x, y, z are parameters in the equation of the line, and t iy Let x be the parameter in the equation of the straight line of the small cylinder axis when the robot is in the i-th pose; i1 y i1 , z i1 Let be the coordinates of a point on the axis of the small cylinder when the robot is in the i-th pose.

[0075] The coordinates of the origin of the tool coordinate system in the robot coordinate system are obtained by using the intersection of the axes. This is achieved by simultaneously solving the equations of the axes of the large and small cylinders, thus obtaining the intersection point of the axes as (x...). i y i , z i The intersection of the axes is the coordinate of the origin of the tool coordinate system in the robot coordinate system.

[0076] The cross product of the X-axis and Y-axis coordinates yields the Z-axis coordinate of the tool coordinate system within the robot coordinate system:

[0077] (m iz n iz p iz )=(m ix n ix p ix )×(m iy n iy p iy )

[0078] Using the coordinates of the tool coordinate system at the origin in the robot coordinate system, as well as the axial coordinates of the X, Y, and Z axes, the pose matrix M of the tool's positioning point in the robot coordinate system is obtained:

[0079] Among them, (m ix n ix p ix (m) represents the preset X-axis coordinate of the tool coordinate system in the robot coordinate system when the robot is in the i-th pose; iy n iy p iy (m) represents the axial coordinate of the Y-axis in the robot coordinate system, preset in the tool coordinate system when the robot is in the i-th pose; iz n iz p iz (x) represents the axial coordinate of the Z-axis of the tool coordinate system in the robot coordinate system when the robot is in the i-th pose; (x) i y i , z i Let (x, y) be the coordinates of the origin of the tool coordinate system in the robot coordinate system when the robot is in the i-th pose. Converted to homogeneous coordinates, (x, y) i y i , z i ,1).

[0080] In step S4, the calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system. The formula for calculating the calibration position of the tool positioning point is:

[0081] M = KT

[0082] Where M represents the pose matrix of the tool positioning point in the robot coordinate system; T represents the pose matrix of the flange center in the robot coordinate system; and K represents the calibration position of the tool positioning point.

[0083] After the robot reaches the preset position, its pose is changed multiple times to obtain the pose matrix of the tool positioning point in the robot coordinate system under each pose. Then, the corresponding tool positioning point calibration position is obtained by using the pose matrix of the tool positioning point and the flange center in the robot coordinate system under each pose. The optimal tool positioning point calibration position is obtained by calculating the tool positioning point calibration position under multiple poses using the least squares method, which improves the positioning accuracy.

[0084] In this solution, the axes of the two cylinders in the marker can be directly identified using laser point cloud, thereby obtaining the origin and three axial coordinates of the tool coordinate system for TCP position calibration. Compared to the traditional manual four-point calibration method, this method only requires manually installing the marker at the tip of the welding torch; all other steps can be completed by software-controlled robots, resulting in a high degree of automation and improved calibration efficiency of the welding torch tool coordinate system. Furthermore, the accuracy of laser point cloud in identifying the cylinder axis is higher than that in identifying the straight line from the spherical marker point to the center of the sphere, thus the calibration results of this invention are more accurate.

[0085] Example 2:

[0086] This invention provides a TCP calibration system for a welding torch robot based on machine vision, such as... Figure 4 As shown, it includes:

[0087] Control module: The rotation marker makes the axis of one of the cylinders point to the preset X-axis of the tool coordinate system;

[0088] Position Analysis Module: Used to obtain the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position; obtain the laser point cloud on the surface of the marker when the robot reaches the preset position, and use the laser point cloud to identify the axis of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system;

[0089] Position determination module: The calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system.

[0090] In the control module, the marker includes two vertically intersecting cylinders located at the tip of the welding torch, which serves as the tool positioning point.

[0091] The marker has a small hole, through which it can be detachably connected to the tip of the welding torch. The bottom center of the small hole is the intersection of the axes of the two cylinders in the marker.

[0092] The marker is inserted into the tip of the welding gun through a small hole, and the tip of the welding gun coincides with the intersection of the axes of the two cylinders in the marker.

[0093] The material of the inner wall of the small hole is an elastic material.

[0094] The two cylinders in the marker have different diameters.

[0095] In the position analysis module, acquiring the laser point cloud on the surface of the marker when the robot reaches the preset position includes:

[0096] When the robot reaches the preset position, it uses a surface laser sensor to scan the marker to obtain a laser point cloud on the surface of the marker.

[0097] The process of obtaining the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position includes:

[0098] When the robot reaches the preset position, the robot outputs the coordinates and orientation of the flange center in the robot coordinate system, and obtains the pose matrix of the flange center based on the coordinates and orientation of the flange center in the robot coordinate system.

[0099] Preferably, the step of obtaining the pose matrix of the tool positioning point in the robot coordinate system using the axes of the two cylinders in the laser point cloud identification marker includes:

[0100] By using the axes of the two cylinders in the laser point cloud recognition marker, the axial coordinates of the preset X and Y axes of the tool coordinate system in the robot coordinate system are obtained, and the coordinates of the origin of the tool coordinate system in the robot coordinate system are obtained by using the intersection of the axes.

[0101] The cross product of the X-axis and Y-axis axial coordinates is used to obtain the Z-axis axial coordinate of the tool coordinate system in the robot coordinate system;

[0102] The pose matrix of the tool positioning point in the robot coordinate system is obtained by using the coordinates of the tool coordinate system at the origin in the robot coordinate system and the axial coordinates of the X, Y and Z axes.

[0103] The preset X-axis of the tool coordinate system points to the axis of one cylinder in the marker, and the preset Y-axis of the tool coordinate system points to the axis of another cylinder.

[0104] In the location determination module, the formula for calculating the calibration position of the tool positioning point is:

[0105] M = KT

[0106] Where M represents the pose matrix of the tool positioning point in the robot coordinate system; T represents the pose matrix of the flange center in the robot coordinate system; and K represents the calibration position of the tool positioning point.

[0107] After obtaining the calibrated position of the tool positioning point, the process further includes:

[0108] After the robot reaches the preset position, its pose is changed multiple times to obtain the pose matrix of the tool positioning point in the robot coordinate system under each pose. Then, the corresponding tool positioning point calibration position is obtained by using the pose matrix of the tool positioning point and the flange center in the robot coordinate system under each pose. The optimal tool positioning point calibration position is obtained by calculating the tool positioning point calibration positions under multiple poses using the least squares method.

[0109] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0111] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0112] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the scope of protection of the claims pending approval.

Claims

1. A TCP calibration method for a welding torch robot based on machine vision, characterized in that, include: The rotating marker makes the axis of one of the cylinders point to the preset X-axis of the tool coordinate system; Obtain the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position; The laser point cloud on the surface of the marker is obtained when the robot reaches the preset position. The laser point cloud is used to identify the axis of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system. The calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system; The marker includes two perpendicularly intersecting cylinders located at the tip of the welding torch, which serves as the tool positioning point. The marker has a small hole, through which it can be detachably connected to the tip of the welding gun. The bottom center of the small hole is the intersection of the axes of the two cylinders in the marker. The marker is inserted into the tip of the welding gun through a small hole, and the tip of the welding gun coincides with the intersection of the axes of the two cylinders in the marker. The two cylinders in the marker have different diameters; The method of obtaining the pose matrix of the tool positioning point in the robot coordinate system using the axes of the two cylinders in the laser point cloud recognition marker includes: By using the axes of the two cylinders in the laser point cloud recognition marker, the axial coordinates of the preset X and Y axes of the tool coordinate system in the robot coordinate system are obtained, and the coordinates of the origin of the tool coordinate system in the robot coordinate system are obtained by using the intersection of the axes. The cross product of the X-axis and Y-axis axial coordinates is used to obtain the axial coordinate of the tool coordinate system in the robot coordinate system along the Z-axis. The pose matrix of the tool positioning point in the robot coordinate system is obtained by using the coordinates of the tool coordinate system at the origin in the robot coordinate system and the axial coordinates of the X, Y and Z axes. The preset X-axis of the tool coordinate system is the axis of one cylinder in the marker, and the preset Y-axis of the tool coordinate system is the axis of another cylinder. The equation of the axis of the large cylinder is: Then the robot is in the first position. The preset X-axis coordinate of the tool coordinate system in the robot coordinate system at each pose is: ; In the formula, These are parameters in the axis equation. For the robot to be in the first The parameters in the axis equation of the large cylinder axis at each pose; For the robot to be in the first The coordinates of a point on the axis of the large cylinder at each pose; The equation of the axis of the small cylinder is: Then the robot is in the first position. The preset Y-axis coordinate of the tool coordinate system in the robot coordinate system at each pose is: ; In the formula, These are parameters in the axis equation. For the robot to be in the first The parameters in the axis equation of the small cylinder axis at each pose; For the robot to be in the first The coordinates of a point on the axis of the small cylinder at each pose.

2. The TCP calibration method for a welding torch robot based on machine vision as described in claim 1, characterized in that, The material of the inner wall of the small hole is an elastic material.

3. The TCP calibration method for a welding torch robot based on machine vision as described in claim 1, characterized in that, The acquisition of the laser point cloud on the surface of the marker when the robot reaches the preset position includes: When the robot reaches the preset position, it uses a surface laser sensor to scan the marker to obtain a laser point cloud on the surface of the marker.

4. The TCP calibration method for a welding torch robot based on machine vision as described in claim 1, characterized in that, The process of obtaining the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position includes: When the robot reaches the preset position, the robot outputs the coordinates and orientation of the flange center in the robot coordinate system, and obtains the pose matrix of the flange center based on the coordinates and orientation of the flange center in the robot coordinate system.

5. The TCP calibration method for a welding torch robot based on machine vision as described in claim 1, characterized in that, The formula for calculating the calibration position of the tool positioning point is: in, The pose matrix representing the tool's positioning point in the robot coordinate system; This represents the pose matrix of the flange center in the robot coordinate system; This indicates the calibrated position of the tool's positioning point.

6. The TCP calibration method for a welding torch robot based on machine vision as described in claim 1, characterized in that, After obtaining the calibrated position of the tool positioning point, the process further includes: After the robot reaches the preset position, its pose is changed multiple times to obtain the pose matrix of the tool positioning point in the robot coordinate system under each pose. Then, the corresponding tool positioning point calibration position is obtained by using the pose matrix of the tool positioning point and the flange center in the robot coordinate system under each pose. The optimal tool positioning point calibration position is obtained by calculating multiple tool positioning point calibration positions using the least squares method.

7. A TCP calibration system for a welding torch robot based on machine vision, used to implement the method as described in claim 1, characterized in that, include: Control module: The rotation marker makes the axis of one of the cylinders point to the preset X-axis of the tool coordinate system; Position Analysis Module: Used to obtain the pose matrix of the flange center at the end of the robotic arm in the robot coordinate system when the robot reaches the preset position; obtain the laser point cloud on the surface of the marker when the robot reaches the preset position, and use the laser point cloud to identify the axis of the two cylinders in the marker to obtain the pose matrix of the tool positioning point in the robot coordinate system; Position determination module: The calibration position of the tool positioning point is obtained by transforming the pose matrix of the tool positioning point and the flange center in the robot coordinate system.

Images