Monocular vision TCP calibration method and equipment for industrial robot

Abstract

The invention provides a monocular vision TCP calibration method and device for an industrial robot. The method comprises the steps that a tool coordinate system # imgabs0 # to be calibrated and a TCP calibrator coordinate system Cdev are determined; moving the robot to enable the tail end of the tool to be located at the calibration area of the TCP calibrator; measuring the X and Y coordinate deviations of the tool-flange transformation matrix # imgabs 1 #; the X and Y coordinates of a tool-flange transformation matrix # imgabs2 # are calibrated; the Z coordinate deviation of the tool-flange transformation matrix # imgabs3 # is measured; the Z coordinate of a tool-flange transformation matrix # imgabs4 # is calibrated; carrying out TCP precision verification, and judging whether a precision requirement is met or not; and TCP calibration is completed. The monocular vision TCP calibration method is simple in principle, convenient to operate, low in hardware cost, high in calibration precision and high in calibration speed.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial robots, and in particular to a monocular vision TCP calibration method and equipment for industrial robots. Background Art

[0002] With the rapid development of industrial robot technology, industrial robots are increasingly being used in high-precision production operations. Industrial robots are often viewed as carriers of specialized tools and equipment, used for operations such as welding, cutting, gluing, grinding, milling, and handling. In these operations, robots operate according to pre-programmed programs, which define the position coordinates of the tool's operating point to establish the tool coordinate system. The position of the tool's operating point relative to the flange coordinate system is usually referred to as the tool center point, or TCP. The robot's path accuracy actually refers to the path accuracy of its TCP. When the robot's TCP coincides with the actual tool operating point, the tool can operate along the planned path. When the robot's TCP deviates from the actual tool operation, the tool's actual operating point will deviate from the planned path, and the robot's path accuracy will deteriorate.

[0003] In actual production operations, the robot's TCP inevitably exhibits positional deviations. First, due to errors in tool production and installation, using the nominal tool operating point as the TCP is inherently subject to deviation. Second, TCP deviations are inevitable during certain operations, such as wear and tear of welding wire during welding, bumps and deformation of dispensing needles during dispensing, and normal tool replacement and disassembly. If the TCP is not promptly calibrated or the motion program is updated, and the original TCP data is used, the robot's path accuracy will deteriorate, making it unsuitable for high-precision operations. Finally, the current robotics industry widely uses point-to-point teaching methods such as the four-point and six-point methods to calibrate the TCP. This method is time-consuming and labor-intensive, making it unsuitable for high-speed, automated operations. During the TCP calibration process, factors such as human error, robot body error, and flexible deformation combine to produce a low-accuracy TCP, making it suitable only for single-piece pin-type tools.

[0004] To address this, automation equipment and robot manufacturers have proposed various TCP calibration methods, or calibration systems. Currently, mainstream robot TCP calibration methods typically involve the robot moving a closed path within the tool coordinate system and using sensors to measure the path deviation between the actual TCP and the theoretical TCP. Existing TCP calibration equipment often uses photoelectric sensors, such as those in patents CN 115229779 A and CN 115122316 A, which employ four sets of photoelectric sensors. Furthermore, patent CN108940745A utilizes two sets of CCDs as TCP calibration sensors for dispensing needle calibration. By improving the calibration method, calibration can be completed with a single CCD, further streamlining the overall calibration process and cost.

[0005] The above-mentioned TCP calibration methods can meet market needs. However, due to the large number of sensors used and their good performance, the TCP calibration equipment itself is often expensive and costly. The price of a TCP calibration device is comparable to that of a complete robot. Therefore, there is still a lot of room for optimization in the above-mentioned TCP calibration methods and the cost of the calibration equipment. Summary of the Invention

[0006] The object of the present invention is to solve at least one of the technical drawbacks.

[0007] To this end, the purpose of the present invention is to propose a monocular vision TCP calibration method and equipment for industrial robots to solve the problems mentioned in the background technology and overcome the shortcomings of the existing technology.

[0008] To achieve the above objectives, an embodiment of the present invention provides a monocular vision TCP calibration method for an industrial robot, comprising:

[0009] Step S1: Determine the tool coordinate system to be calibrated And TCP calibrator coordinate system C dev , including: when the TCP calibrator is installed and debugged, the TCP calibrator coordinate system C is obtained according to the actual installation position of the TCP calibrator and the attitude angle fed back by the attitude sensor dev To base coordinate system C base The transformation matrix After installing the tool, get the tool coordinate system to be calibrated and its flange coordinate system C flan The transformation matrix

[0010] Step S2, moving the robot so that the tool end is located in the calibration area of the TCP calibrator;

[0011] Step S3, measuring tool-flange transformation matrix The X and Y coordinate deviations of the robot are recorded by the TCP calibrator's monocular camera in the tool coordinate system. The position of the tool end point before and after rotation around the Z axis is calculated to calculate the center and radius of the TCP error circle, and then measure the deviation e of the X and Y coordinates of the TCP. x 、e y ;

[0012] Step S4, calibration tool-flange transformation matrix X, Y coordinates of

[0013] Step S5, measuring tool-flange transformation matrix Z coordinate deviation;

[0014] Step S6, calibration tool-flange transformation matrix The Z coordinate of

[0015] Step S7, TCP accuracy check, to determine whether the accuracy requirements are met, if yes, go to step S8; otherwise go to step S2; wherein, the calibrated TCP coordinates are obtained By point Data update tool - flange transformation matrix And re-establish the tool coordinate system, record it as Cool; The value of is updated to point Repeat steps S2 to S6 to perform TCP recalibration;

[0016] Step S8: Complete TCP calibration.

[0017] Preferably, in any of the above solutions, in step S1, the tool coordinate system to be calibrated is Through the tool-flange transformation matrix Establish;

[0018] in, Describe the tool operating point in the flange coordinate system C flan The pose P under tool , P tool =(x,y,z,α,β,γ);

[0019] Among them, P tool Where x, y, and z are Cartesian coordinates, describing the position of the tool operating point in the flange coordinate system; α, β, and γ are Euler angles, describing the posture of the tool operating point in the flange coordinate system.

[0020] Preferably, in step S2, the transformation matrix from the TCP calibrator coordinate system to the base coordinate system is The calibration point is expressed as a Cartesian point P in the robot base coordinate system calib, where P calib =(x calib ,y calib , z calib , α calib , β calib , γ calib );

[0021] Among them, x calib 、y calib 、z calib is the Cartesian coordinate of the TCP calibrator coordinate system in the robot base coordinate system; α calib , β calib , γ calib is the Euler angle of the TCP calibrator coordinate system in the robot base coordinate system;

[0022] Transformation Matrix and the Cartesian point P calib According to the installation position of the TCP calibrator, the robot controller will plan the movement path so that the TCP moves to the Cartesian point P calib .

[0023] Preferably, when performing TCP calibration, the monocular camera will take a picture and record the coordinates p of the tool end point in the pixel coordinate system. pix =(x pix ,y pix ), according to the pixel coordinate system C pix To TCP calibrator coordinate system C dev The transformation matrix The displacement Δx in pixel coordinates pix , Δy pix Converted into displacement in the TCP calibrator coordinate system;

[0024] Assume that the resolution of each pixel in the viewfinder is ρ, then the displacement l in the TCP calibrator coordinate system is x , l y , l z Expressed as:

[0025]

[0026] Wherein, the displacement Δx pix , Δy pix and l x 、l y 、l z are all vectors.

[0027] Preferably, in step S3, the tool-flange transformation matrix is measured. The X-coordinate deviation includes:

[0028] (1) The robot moves to the Cartesian point in the tool coordinate system say is the measurement point 0; the monocular camera takes a picture of the tool and records the pixel coordinates of the tool end point in the viewfinder, which is recorded as

[0029] (2) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 1; after the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0030] (3) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 2; after the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0031] (4) According to pixel coordinates Calculate the displacement l of the tool end point after two rotations in the TCP calibrator coordinate system x1 、l x2 .

[0032] Preferably, in any of the above solutions, in step S4, the displacement l of the tool end point is obtained according to the measurement. x1 、l x2 、l y1 、l y2 , combined equations, we get the following system of equations:

[0033]

[0034] Solve to get the tool-flange transformation matrix The X and Y coordinate deviation e x , e y , that is, the X and Y coordinate deviation of TCP;

[0035] Modify the TCP coordinates of the tool coordinate system established in the robot controller for Complete calibration tool - flange transformation matrix Calibration of the X and Y coordinates, where

[0036] Preferably, in step S6, the tool-flange transformation matrix is measured and obtained: Z coordinate deviation e z ;

[0037] Modify the TCP coordinates of the tool coordinate system established in the robot controller for Complete calibration tool - flange transformation matrix Calibration of the Z coordinate;

[0038] in,

[0039] Preferably, in any of the above solutions, in step S7, if the tool-flange transformation matrix measured during the second calibration is The X, Y, Z coordinate deviation e x , e y , e z If the following conditions are met, the TCP accuracy is considered to meet the requirements and the TCP calibration is completed; otherwise, repeat this step;

[0040]

[0041] Among them, ε x , ε y , ε z The tool-flange transformation matrix is X, Y, Z coordinate accuracy.

[0042] On the other hand, the present invention further provides a monocular vision TCP calibration device for an industrial robot, characterized in that the monocular vision TCP calibration device for an industrial robot is used to perform the method described in any one of claims 1 to 8, wherein the monocular vision TCP calibration device for an industrial robot is installed in the front area of the robot body, and the TCP calibration device includes: a TCP calibrator and a TCP calibrator mounting base;

[0043] The TCP calibrator includes: a monocular camera, a backlight source, a calibration controller, a posture sensor, a communication element and a cable; wherein the monocular camera and the backlight source are used to measure the TCP deviation; the calibration controller is used to process sensor data, interact with the robot controller, and solve the calibration parameters; the posture sensor is used to measure the posture angle of the TCP calibrator.

[0044] Preferably, the TCP calibration equipment workflow includes:

[0045] (1) Install and debug the TCP calibrator: Before installing the TCP calibrator, return it to zero. The zero point of the attitude sensor is the attitude angle of the robot base. Then, fix the TCP calibrator on the TCP calibrator mounting base with bolts, connect its power supply and the communication cable between it and the robot control cabinet, and wait until the robot controller can interact with the TCP calibrator normally and all sensors can work normally.

[0046] (2) Run the TCP calibration module: Run the TCP calibration module in the robot control system on the teach pendant or PC. When running the TCP calibration module, the user moves the robot under the guidance of the graphical interface so that the tool end is located in the calibration area of the TCP calibrator. When the monocular camera of the TCP calibrator can normally identify the tool end, the controller drives the robot to perform movement according to the established program. The monocular camera in the TCP calibrator will collect the position information of the tool end during movement, and the calibration controller will calculate the TCP calibration value in real time. The TCP calibration value will be transmitted to the robot controller through the communication link.

[0047] (3) After receiving the TCP calibration data, the robot controller automatically updates the TCP parameters of the current tool and completes the TCP calibration.

[0048] The monocular TCP calibration method and device for industrial robots according to embodiments of the present invention utilize a single 2D industrial camera as a sensor for measuring TCP deviation. The 2D industrial camera captures images of the tool before and after its movement, and a calibration algorithm is used to calculate the TCP deviation, thereby completing TCP calibration. The monocular TCP calibration method of the present invention features a simple principle and convenient operation. The accompanying monocular TCP calibration hardware for industrial robots offers low cost, high calibration accuracy, and rapid calibration speed. It is compatible with a wide range of tool types, possessing significant practical value and broad market prospects.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The monocular vision TCP calibration method can measure the TCP deviation with high precision and complete the automatic calibration of TCP.

[0051] 2. The monocular vision TCP calibration method can adapt to tools of various shapes and structures, and the calibrated TCP position (including tool end point, tool operation point, and tool external point) can be selected according to user needs.

[0052] 3. The monocular vision TCP calibration method has a simple principle, the TCP calibration equipment has low cost, and it can calibrate the TCP of multiple robots.

[0053] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0055] Figure 1Flowchart of a monocular vision TCP calibration method for an industrial robot according to an embodiment of the present invention;

[0056] Figure 2 To determine the tool coordinate system to be calibrated according to an embodiment of the present invention And TCP calibrator coordinate system C dev Schematic diagram of;

[0057] Figure 3 is a diagram showing the actual position of the tool end point after the TCP is moved to the calibration point according to an embodiment of the present invention;

[0058] Figure 4 is a schematic diagram of the actual position of a tool and the tool position calculated in the controller after the tool coordinate system is established according to an embodiment of the present invention;

[0059] Figure 5 The robot according to the embodiment of the present invention is in the tool coordinate system Schematic diagram of rotation around the Z axis;

[0060] Figure 6 A schematic diagram of the position of an imaging machine in a tool viewfinder when measuring an X-coordinate deviation according to an embodiment of the present invention;

[0061] Figure 7 for Figure 6 Schematic diagram of the projection distribution of the tool end point position in the XY plane of the base coordinate system;

[0062] Figure 8 is a schematic diagram of the position of the imaging machine in the tool viewfinder when measuring the Y coordinate deviation according to an embodiment of the present invention;

[0063] Figure 9 for Figure 8 Schematic diagram of the projection distribution of the tool end point position in the XY plane of the base coordinate system;

[0064] Figure 10 is the measurement tool-flange transformation matrix according to an embodiment of the present invention Schematic diagram of the Z coordinate deviation;

[0065] Figure 11 A schematic diagram of the position of an imaging machine in a tool viewfinder when measuring a Z coordinate deviation according to an embodiment of the present invention;

[0066] Figure 12 for Figure 11 Schematic diagram of the projection distribution of the tool end point position at the middle measurement point on the XZ plane of the base coordinate system;

[0067] Figure 13 is a schematic diagram of a tip tool according to an embodiment of the present invention being mounted on a flange of a test robot in this example;

[0068] Figure 14 is a schematic diagram of the results of the four-point method in the robot controller according to an embodiment of the present invention;

[0069] Figure 15 Schematic diagram of a monocular vision TCP calibration device for an industrial robot according to an embodiment of the present invention;

[0070] Figure 16 1 is a flowchart of the workflow of a monocular vision TCP calibration device for an industrial robot according to an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0072] like Figure 1 As shown, the monocular vision TCP calibration method for an industrial robot according to an embodiment of the present invention includes:

[0073] Step S1: Determine the tool coordinate system to be calibrated And TCP calibrator coordinate system C dev .

[0074] Specifically, when the TCP calibrator is installed and debugged, the TCP calibrator coordinate system C is obtained according to the actual installation position of the TCP calibrator and the attitude angle fed back by the attitude sensor. dev To base coordinate system C base The transformation matrix After installing the tool, get the tool coordinate system to be calibrated and its flange coordinate system C flan The transformation matrix

[0075] like Figure 2 As shown in the figure, after the TCP calibrator is installed and debugged, the TCP calibrator coordinate system C can be obtained according to the actual installation position of the TCP calibrator and the attitude angle fed back by the attitude sensor. dev To base coordinate system C base The transformation matrix

[0076] After the tool is installed, the tool coordinate system to be calibrated is obtained and its flange coordinate system C flan The transformation matrix Tool coordinate system to be calibrated is the tool-flange transformation matrix Established. Among them, Describe the tool operating point in the flange coordinate system C flan The pose P under tool .

[0077] P tool =(x,y,z,α,β,γ) (1)

[0078] P tool Where x, y, and z are Cartesian coordinates, describing the position of the tool operating point in the flange coordinate system. α, β, and γ are Euler angles, describing the posture of the tool operating point in the flange coordinate system.

[0079] It should be noted that P tool The representation includes but is not limited to the above form, and the posture can also be represented by quaternions.

[0080] Tool-flange transformation matrix and P tool It can be obtained through mechanical design theoretical value or four-point calibration. Under the influence of factors such as installation, use, and loss, the position P of the tool operation point tool There is usually a deviation from the actual value, resulting in the tool coordinate system to be calibrated With the actual tool coordinate system C tool If the deviation between the two is too large, the subsequent calibration steps may not be carried out normally. Therefore, the TCP calibration method of the present invention is for the coordinate system of the tool to be calibrated. There are certain requirements for accuracy.

[0081] In summary, the calibration TCP in the embodiment of the present invention refers to the calibration tool operating point in the flange coordinate system C flan Pose P tool and the transformation matrix The tool coordinate system to be calibrated Corrected to the actual tool coordinate system C tool .

[0082] Step S2: The robot is moved so that the tool end is located at the calibration area of the TCP calibrator.

[0083] Specifically, this step is to move the TCP to the calibration point, that is, to move the robot so that the tool tip is located at the calibration area of the TCP calibrator.

[0084] Among them, TCP refers to the tool coordinate system The origin of the tool is usually the tool end point or tool operation point (TCP can also be a non-tool end point, which is defined by the user as needed). For the convenience of description, the TCP in the examples of the present invention refers specifically to the tool end point.

[0085] The TCP is described as a Cartesian point in the flange coordinate system Its Cartesian coordinates are the same as P tool The Euler angles are all 0.

[0086] The calibration point refers to the TCP calibrator coordinate system C dev The origin of the tool is defined at the geometric center of the TCP calibrator. This point is the intersection of the monocular camera's focal plane and the extended line of the viewfinder center. Here, the monocular camera can capture the clearest tool outline and ensure that the tool image is always within the viewfinder during subsequent calibration movements.

[0087] According to the transformation matrix from the TCP calibrator coordinate system to the base coordinate system in step S1 The calibration point can be expressed as a Cartesian point P in the robot base coordinate system calib .

[0088] P calib =(x calib ,y calib , z calib , α calib , β calib , γ calib ) (2)

[0089] Among them, x calib、 y calib 、z calib is the Cartesian coordinate of the TCP calibrator coordinate system in the robot base coordinate system, α calib , β calib , γ calib is the Euler angle of the TCP calibrator coordinate system in the robot base coordinate system. and the Cartesian point P calib The robot controller will plan the movement path so that the TCP moves to the Cartesian point P. calib .

[0090] like Figure 3 As shown, after moving TCP to the calibration point, due to the tool coordinate system There is a deviation, so the actual tool tip point is not in the center of the viewfinder.

[0091] When performing TCP calibration, the monocular camera will take pictures and record the coordinates p of the tool end point in the pixel coordinate system. pix =(x pix ,y pix ). According to the pixel coordinate system C pix To TCP calibrator coordinate system C dev The transformation matrix The displacement Δx in pixel coordinates can be pix , Δy pix Converted into displacement in the TCP calibrator coordinate system. Assuming the resolution of each pixel in the viewfinder is ρ (unit: mm), the displacement l in the TCP calibrator coordinate system is x , l y , l z Expressed as:

[0092]

[0093] Since the relative position of the monocular camera in the TCP calibrator is fixed, the pixel coordinate system C pix To TCP calibrator coordinate system C dev The transformation matrix is a known constant whose data is stored in the calibration controller. pix , Δy pix and l x 、l y 、l z are all vectors.

[0094] Step S3, measuring tool-flange transformation matrix The X and Y coordinate deviations of the robot are recorded by the TCP calibrator's monocular camera in the tool coordinate system. The position of the tool end point before and after rotation around the Z axis is calculated to calculate the center and radius of the TCP error circle, and then measure the deviation e of the X and Y coordinates of the TCP. x 、e y .

[0095] Specifically, step S3 will use an example to introduce the measurement tool-flange transformation matrix Method and process for determining the X and Y coordinate deviations.

[0096] like Figure 4 As shown, Figure 4 The solid outline is the actual position of the tool, and the dotted line is the tool position calculated in the controller after the tool coordinate system is established. In the embodiment of the present invention, the tool TCP is the tool end point. Due to the deviation of TCP, the actual tool end point position is different from the Cartesian point in the flange coordinate system in the controller. The actual tool end point is called the actual TCP, and the tool coordinate system is The origin is the error TCP, which is the TCP to be calibrated later. Then the actual TCP of the tool and the error TCP are in the flange coordinate system C flan There is a deviation e x , e y , e z .

[0097] like Figure 5 As shown, due to the deviation of TCP on the X and Y axes, x , e y , then when the robot is in the tool coordinate system When the tool is rotated around the Z axis, the actual tool end point will move along an arc. The circle where this arc is located is called the TCP error circle. The center of the arc is the TCP to be calibrated in the robot controller, and the radius of the arc is

[0098] Measurement tool-flange transformation matrix Method for X, Y coordinate deviation: Use a monocular camera to record the robot in the tool coordinate system The position of the tool end point before and after rotation around the Z axis can be used to calculate the center and radius of the TCP error circle based on the geometric properties of the arc, and then the deviation e of the X and Y coordinates of the TCP can be measured. x 、e y .

[0099] Measurement tool-flange transformation matrix The X-coordinate deviation is calculated as follows:

[0100] (1) The robot moves to the Cartesian point in the tool coordinate system say is the measurement point 0. The monocular camera takes a picture of the tool and records the pixel coordinates of the tool end point in the viewfinder, which is recorded as

[0101] (2) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 1. After the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0102] (3) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 2. After the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0103] (4) According to pixel coordinates Calculate the displacement l of the tool end point after two rotations in the TCP calibrator coordinate system x1 、l x2 .

[0104] like Figure 6 As shown, the displacement l in step (4) x1 、l x2 is the measurement tool-flange transformation matrix The X-coordinate deviation is the displacement of the tool end point on the X-axis in the TCP calibrator coordinate system.

[0105]

[0106] like Figure 7 As shown in the figure, the projections of the tool end point positions of measurement point 0, measurement point 1, and measurement point 2 on the XY plane of the base coordinate system are distributed on the TCP error circle. x1 , l x2 , the following expression can be obtained from the geometric relationship.

[0107]

[0108] in θ=atan(e y / e x )

[0109] Measurement tool-flange transformation matrix The Y coordinate deviation method is as follows:

[0110] (1) The robot moves to the Cartesian point in the tool coordinate system say is the measurement point 0. The monocular camera takes a picture of the tool and records the pixel coordinates of the tool end point in the viewfinder, which is recorded as

[0111] (2) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 1. After the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0112] (3) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 2. After the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0113] (4) According to pixel coordinates Calculate the displacement l of the tool end point after two rotations in the TCP calibrator coordinate system y1 、l y2 .

[0114] like Figure 8 As shown, the displacement l in step (4) y1 、l y2 is the measurement tool-flange transformation matrix The Y coordinate deviation is the displacement of the tool end point on the X axis in the TCP calibrator coordinate system.

[0115]

[0116] like Figure 9 As shown in the figure, the projections of the tool end point positions of measurement point 0, measurement point 1, and measurement point 2 on the XY plane of the TCP calibrator coordinate system are distributed on the TCP error circle. x1 , l x2 , the following expression can be obtained from the geometric relationship:

[0117] The following expressions can be obtained from the geometric relationship, where the formulas for r and θ are consistent with those used in calibrating the X-axis deviation.

[0118]

[0119] in θ=atan(e y / e x ).

[0120] Step S4, calibrate tool-flange transformation matrix The X and Y coordinates of the .

[0121] Specifically, after step S3, the displacement l of the tool end point is measured. x1 、l x2 、l y1 、l y2 . Solving the equations simultaneously, we get the following system of equations.

[0122]

[0123] The above equations are nonlinear. Solving the nonlinear equations, we can get the tool-flange transformation matrix The X and Y coordinate deviation e x , e y , that is, the X and Y coordinate deviation of TCP.

[0124] Modify the TCP coordinates of the tool coordinate system established in the robot controller for Complete calibration tool - flange transformation matrix Calibration of the X and Y coordinates.

[0125] Step S5, measuring tool-flange transformation matrix The Z coordinate deviation.

[0126] like Figure 10As shown, the solid outline is the actual position of the tool, and the dotted line is the tool position calculated in the controller after the tool coordinate system is established. In the embodiment of the present invention, the TCP has a deviation e on the Z-axis component. z When the robot is in the tool coordinate system When rotating around the X-axis or Y-axis, the actual movement trajectory of the tool end point is also on a circular arc. The circle where this arc is located is called the TCP error circle. The center of the arc is the TCP to be calibrated in the robot controller, and the radius of the arc is r = |e z |.

[0127] Measurement Tool—Flange Transformation Matrix The method for measuring the Z coordinate deviation is similar to the method for measuring the X and Y coordinate deviations in step S4.

[0128] Measurement Tool—Flange Transformation Matrix The X-coordinate deviation is calculated as follows:

[0129] (1) The robot moves to the Cartesian point in the tool coordinate system say is the measurement point 0. The monocular camera takes a picture of the tool and records the pixel coordinates of the tool end point in the viewfinder, which is recorded as

[0130] (2) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 1. After the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0131] (3) The robot moves to the Cartesian point in the tool coordinate system remember is measurement point 2. After the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point

[0132] (4) According to pixel coordinates Calculate the displacement l of the tool end point after two rotations in the pixel coordinate system z1 、l z2 .

[0133] like Figure 11 As shown, the displacement l in step (4) z1 、l z2 is the measurement tool-flange transformation matrix The Z coordinate deviation is the displacement of the tool end point on the X axis in the pixel coordinate system.

[0134]

[0135] like Figure 12 As shown in the figure, the projections of the tool end point positions of measurement point 0, measurement point 1, and measurement point 2 on the XZ plane of the TCP calibrator coordinate system are distributed on the TCP error circle. z1 , l z2 , the following expression can be obtained from the geometric relationship:

[0136]

[0137] Step S6, calibrate tool-flange transformation matrix The Z coordinate of

[0138] Specifically, after step S5, the tool-flange transformation matrix is measured and obtained Z coordinate deviation e z .

[0139] Modify the TCP coordinates of the tool coordinate system established in the robot controller for Complete calibration tool - flange transformation matrix The Z coordinate of the calibration.

[0140]

[0141] Step S7, TCP accuracy check, to determine whether the accuracy requirements are met, if yes, go to step S8; otherwise go to step S2; wherein, the calibrated TCP coordinates are obtained By point Data update tool - flange transformation matrix And re-establish the tool coordinate system, recorded as C tool ; Set the coordinates The value of is updated to point Repeat steps S2 to S6 to perform TCP recalibration.

[0142] After the above steps, the calibrated TCP coordinates are obtained. By point Data update tool - flange transformation matrix And re-establish the tool coordinate system, recorded as C tool . Set the coordinates The value of is updated to point Repeat steps S2 to S6 to perform TCP recalibration.

[0143] If the tool-flange transformation matrix measured during the second calibration is The X, Y, Z coordinate deviation e x , e y , ez If the following conditions are met, the TCP accuracy is considered to meet the requirements and the TCP calibration is completed; otherwise, repeat this step.

[0144]

[0145] Among them, ε x , ε y , ε z The tool-flange transformation matrix is The X, Y, and Z coordinate accuracy is user-defined.

[0146] Step S8: Complete TCP calibration.

[0147] Reference below Figure 3 The effects of the present invention are described with reference to specific embodiments, in which the data appearing in the embodiments are all in mm.

[0148] Figure 13 Shown is a tip tool mounted on a test robot flange in this example.

[0149] The coordinate measuring machine can measure the coordinates of the tool tip to the center of the robot flange to obtain the actual tool-flange transformation matrix Actual tool - flange transformation matrix The specific representation is shown in Table 1:

[0150] Table 1

[0151] 1.0000e+00 -5.9899e-05 -1.0006e-04 0.051 6.0099e-05 1.0000e+00 2.0000e-03 0.042 9.9939e-05 -2.0000e-03 1.0000e+00 100.530 0 0 0 1

[0152] The actual tool described above - the flange transformation matrix The data inside the thick box on the right, from top to bottom, is the tool-flange transformation matrix The X, Y, and Z coordinates of the tool are used to determine the tool's TCP. The tool's TCP is a vector containing three elements. The actual tool TCP obtained by the three-coordinate measuring machine is shown in Table 2:

[0153] Table 2

[0154] X Y Z 0.051 0.042 100.530

[0155] The tool coordinate system of the tool is established by the four-point method. The result of the four-point method in the robot controller is as follows: Figure 14 As shown, the TCP of the tool in the robot controller is shown in Table 3:

[0156] Table 3

[0157] X Y Z -0.320 0.240 100.040

[0158] It can be seen that the TCP of the tool coordinate system established by the four-point method deviates from the actual TCP of the tool. The tool-flange transformation matrix to be calibrated obtained by the four-point method is As shown in Table 4:

[0159] Table 4

[0160] 1.0000e+00 -5.9899e-05 -1.0006e-04 -0.320 6.0099e-05 1.0000e+00 2.0000e-03 0.240 9.9939e-05 -2.0000e-03 1.0000e+00 100.040 0 0 0 1

[0161] After adopting the monocular vision TCP calibration method described in the present invention, the TCP deviation that can be measured is shown in Table 5:

[0162] Table 5

[0163] <![CDATA[e x ]]> <![CDATA[e y ]]> <![CDATA[e z ]]> 0.380 -0.230 0.044

[0164] The TCP data of the tool in the controller after monocular vision TCP calibration is shown in Table 6:

[0165] Table 6

[0166] X Y Z 0.060 0.010 100.480

[0167] The calibrated tool-flange transformation matrix is As shown in Table 7:

[0168] Table 7

[0169] 1.0000e+00 -5.9899e-05 -1.0006e-04 0.060 6.0099e-05 1.0000e+00 2.0000e-03 0.010 9.9939e-05 -2.0000e-03 1.0000e+00 100.480 0 0 0 1

[0170] The tool-flange transformation matrix after calibration It can be considered as the actual tool-flange transformation matrix Basically the same, TCP calibration is completed.

[0171] In the embodiment of the present invention, the TCP calibration coordinate accuracy is set to ε x =0.5,ε y =0.5,ε z =0.5.

[0172] like Figure 15 As shown, an embodiment of the present invention further provides a monocular vision TCP calibration device for an industrial robot, which is used to perform the above calibration method.

[0173] The tool to be calibrated is mounted on the flange at the end of the robot body. The industrial robot's monocular vision TCP calibration equipment is located in the front area of the robot body. The TCP calibration equipment includes a TCP calibrator and a TCP calibrator mounting base.

[0174] Specifically, the TCP calibrator integrates a monocular camera, backlight, calibration controller, attitude sensor, communication components, and cables. The monocular camera and backlight are used to measure TCP deviation. The calibration controller processes sensor data, interacts with the robot controller, and calculates calibration parameters. The attitude sensor measures the TCP calibrator's attitude angle.

[0175] The TCP calibrator comes with multiple mounting holes. Besides being mounted on the included TCP calibrator mount, it can also be mounted independently on a workbench, wall, or other locations. The TCP calibrator mount is adjustable in height, and its base includes holes for bolting and screwing. The TCP calibrator also comes with a storage case. When not in use, the TCP calibrator can be fitted with the camera lens cap and the mount retracted to its minimum height for easy storage.

[0176] like Figure 16 As shown, the workflow of the monocular vision TCP calibration device for industrial robots according to an embodiment of the present invention is as follows:

[0177] First, install and debug the TCP calibrator. This means returning the TCP calibrator to zero before installation. The zero point of the attitude sensor is the attitude angle of the robot base. Then, bolt the TCP calibrator to the TCP calibrator mounting base (or elsewhere with reserved mounting holes for the TCP calibrator). Connect the power cord and the communication cable between the TCP calibrator and the robot control cabinet. Wait until the robot controller is able to interact with the TCP calibrator and all sensors are functioning normally.

[0178] Secondly, run the TCP calibration module. Running the TCP calibration module means that the user needs to run the TCP calibration module in the robot control system on a teach pendant or PC. When running the TCP calibration module, the user moves the robot under the guidance of the graphical interface so that the tool tip is located in the calibration area of the TCP calibrator. When the monocular camera of the TCP calibrator can normally identify the tool tip, the user needs to enable TCP calibration, after which the controller drives the robot to perform a series of movements according to the established program. The monocular camera in the TCP calibrator will collect the position information of the tool tip during movement, and the calibration controller will calculate the TCP calibration value in real time. The TCP calibration value will be transmitted to the robot controller through a communication link, and the communication link includes but is not limited to a wired communication link and a wireless communication link.

[0179] Finally, the robot controller automatically updates the TCP parameters of the current tool after receiving the TCP calibration data, completing the TCP calibration.

[0180] In summary, the monocular vision TCP calibration method and equipment for industrial robots in the embodiments of the present invention uses a monocular camera to measure the actual TCP position of a tool and solves the TCP deviation based on geometric properties. In addition, an algorithm and technology are proposed to calculate the tool end point, tool operation point, and user-defined TCP pixel coordinates based on visual recognition of the tool contour.

[0181] It should be noted that the term "monocular vision TCP calibration" and any TCP calibration method or equipment solution used for industrial robots, which has the same or similar meaning as "monocular vision" or "single camera, visual measurement", are protected by the present invention.

[0182] The monocular vision TCP calibration method and apparatus for industrial robots according to embodiments of the present invention utilizes a single 2D industrial camera as a sensor for measuring TCP deviation. The 2D industrial camera captures images of the tool before and after its movement, and a calibration algorithm is used to calculate the TCP deviation, thereby completing TCP calibration. The monocular vision TCP calibration method of the present invention features a simple principle and convenient operation. The accompanying monocular vision TCP calibration hardware for industrial robots offers low cost, high calibration accuracy, and rapid calibration speed. It is compatible with a wide range of tool types, possessing significant practical value and broad market prospects.

[0183] Compared with the prior art, the present invention has the following beneficial effects:

[0184] 1. The monocular vision TCP calibration method can measure the TCP deviation with high precision and complete the automatic calibration of TCP.

[0185] 2. The monocular vision TCP calibration method can adapt to tools of various shapes and structures, and the calibrated TCP position (including tool end point, tool operation point, and tool external point) can be selected according to user needs.

[0186] 3. The monocular vision TCP calibration method has a simple principle, the TCP calibration equipment has low cost, and it can calibrate the TCP of multiple robots.

[0187] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0188] Those skilled in the art will readily understand that the present invention encompasses any combination of the components described in the Summary and Detailed Description of the Invention and the accompanying drawings. Due to space limitations and for the sake of clarity, not all of the various solutions resulting from these combinations are described. Any modifications, equivalent substitutions, and improvements within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

[0189] Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are illustrative and are not to be construed as limiting the present invention. Those skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments without departing from the principles and intent of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A monocular vision TCP calibration method for industrial robots, characterized in that: The steps include: Step S1: Determine the tool coordinate system to be calibrated And TCP calibrator coordinate system C dev , including: when the TCP calibrator is installed and debugged, the TCP calibrator coordinate system C is obtained according to the actual installation position of the TCP calibrator and the attitude angle fed back by the attitude sensor dev To base coordinate system C base The transformation matrix After installing the tool, get the tool coordinate system to be calibrated and its flange coordinate system C flan The transformation matrix Step S2, moving the robot so that the tool end is located in the calibration area of the TCP calibrator; Step S3, measuring tool-flange transformation matrix The X and Y coordinate deviations of the robot are recorded by the TCP calibrator's monocular camera in the tool coordinate system. The position of the tool end point before and after rotation around the Z axis is calculated to calculate the center and radius of the TCP error circle, and then measure the deviation e of the X and Y coordinates of the TCP. x 、e y ; Step S4, calibration tool-flange transformation matrix X, Y coordinates of Step S5, measuring tool-flange transformation matrix Z coordinate deviation; Step S6, calibration tool-flange transformation matrix The Z coordinate of Step S7, TCP accuracy check, to determine whether the accuracy requirements are met, if yes, go to step S8; otherwise go to step S2; wherein, the calibrated TCP coordinates are obtained By point Data update tool - flange transformation matrix And re-establish the tool coordinate system, recorded as C tool ; Set the coordinates The value of is updated to point Repeat steps S2 to S6 to perform TCP recalibration; Step S8: Complete TCP calibration.

2. The monocular vision TCP calibration method for an industrial robot according to claim 1, wherein: In step S1, the tool coordinate system to be calibrated Through the tool-flange transformation matrix Establish; in, Describe the tool operating point in the flange coordinate system C flan The pose P under tool , P tool =(x,y,z,α,β,γ); Among them, P tool Where x, y, and z are Cartesian coordinates, describing the position of the tool operating point in the flange coordinate system; α, β, and γ are Euler angles, describing the posture of the tool operating point in the flange coordinate system.

3. The monocular vision TCP calibration method for an industrial robot according to claim 1, wherein: In step S2, the transformation matrix from the TCP calibrator coordinate system to the base coordinate system is The calibration point is expressed as a Cartesian point P in the robot base coordinate system calib , where P calib =(x calib ,y calib , z calib , α calib , β calib , γ calib ); Among them, x calib 、y calib 、z calib is the Cartesian coordinate of the TCP calibrator coordinate system in the robot base coordinate system; α calib , β calib , γ calib is the Euler angle of the TCP calibrator coordinate system in the robot base coordinate system; Transformation Matrix and the Cartesian point P calib According to the installation position of the TCP calibrator, the robot controller will plan the movement path so that the TCP moves to the Cartesian point P calib .

4. The monocular vision TCP calibration method for an industrial robot according to claim 3, wherein: When performing TCP calibration, the monocular camera will take pictures and record the coordinates p of the tool end point in the pixel coordinate system. pix =(x pix ,y pix ), according to the pixel coordinate system C pix To TCP calibrator coordinate system C dev The transformation matrix The displacement Δx in pixel coordinates pix , Δy pix Converted into displacement in the TCP calibrator coordinate system; Assume that the resolution of each pixel in the viewfinder is ρ, then the displacement l in the TCP calibrator coordinate system is x , l y , l z Expressed as: Wherein, the displacement Δx pix , Δy pix and l x 、l y 、l z are all vectors.

5. The monocular vision TCP calibration method for an industrial robot according to claim 1, wherein: In step S3, the tool-flange transformation matrix is measured The X-coordinate deviation includes: (1) The robot moves to the Cartesian point in the tool coordinate system say is the measurement point 0; the monocular camera takes a picture of the tool and records the pixel coordinates of the tool end point in the viewfinder, which is recorded as (2) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 1; after the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point (3) The robot moves to the Cartesian point in the tool coordinate system say is measurement point 2; after the robot moves to the target point, the monocular camera takes a picture and records the corresponding pixel coordinates of the tool end point (4) According to pixel coordinates Calculate the displacement l of the tool end point after two rotations in the TCP calibrator coordinate system x1 、l x2 .

6. The monocular vision TCP calibration method for an industrial robot according to claim 1, wherein: In step S4, the displacement l of the tool end point is obtained according to the measurement. x1 、l x2 、l y1 、l y2 , combined equations, we get the following system of equations: Solve to get the tool-flange transformation matrix The X and Y coordinate deviation e x , e y , that is, the X and Y coordinate deviation of TCP; Modify the TCP coordinates of the tool coordinate system established in the robot controller for Complete calibration tool - flange transformation matrix Calibration of the X and Y coordinates, where 7. The monocular vision TCP calibration method for an industrial robot according to claim 1, wherein: In step S6, the tool-flange transformation matrix is measured and obtained Z coordinate deviation e z ; Modify the TCP coordinates of the tool coordinate system established in the robot controller for Complete calibration tool - flange transformation matrix Calibration of the Z coordinate; in, 8. The monocular vision TCP calibration method for an industrial robot according to claim 1, wherein: In step S7, if the tool-flange transformation matrix measured during the second calibration is The X, Y, Z coordinate deviation e x , e y , e z If the following conditions are met, the TCP accuracy is considered to meet the requirements and the TCP calibration is completed; otherwise, repeat this step; Among them, ε x , ε y , ε z The tool-flange transformation matrix is X, Y, Z coordinate accuracy.

9. A monocular vision TCP calibration device for industrial robots, characterized in that: The monocular vision TCP calibration device for an industrial robot is used to perform the method according to any one of claims 1 to 8, wherein the monocular vision TCP calibration device for an industrial robot is installed in the front area of the robot body, and the TCP calibration device includes: a TCP calibrator and a TCP calibrator mounting base; The TCP calibrator includes: a monocular camera, a backlight source, a calibration controller, a posture sensor, a communication element and a cable; wherein the monocular camera and the backlight source are used to measure the TCP deviation; the calibration controller is used to process sensor data, interact with the robot controller, and solve the calibration parameters; the posture sensor is used to measure the posture angle of the TCP calibrator.

10. The monocular vision TCP calibration device for an industrial robot according to claim 9, characterized in that: The TCP calibration equipment workflow includes: (1) Install and debug the TCP calibrator: Before installing the TCP calibrator, return it to zero. The zero point of the attitude sensor is the attitude angle of the robot base. Then, fix the TCP calibrator on the TCP calibrator mounting base with bolts, connect its power supply and the communication cable between it and the robot control cabinet, and wait until the robot controller can interact with the TCP calibrator normally and all sensors can work normally. (2) Run the TCP calibration module: Run the TCP calibration module in the robot control system on the teach pendant or PC. When running the TCP calibration module, the user moves the robot under the guidance of the graphical interface so that the tool end is located in the calibration area of the TCP calibrator. When the monocular camera of the TCP calibrator can normally identify the tool end, the controller drives the robot to perform movement according to the established program. The monocular camera in the TCP calibrator will collect the position information of the tool end during movement, and the calibration controller will calculate the TCP calibration value in real time. The TCP calibration value will be transmitted to the robot controller through the communication link. (3) After receiving the TCP calibration data, the robot controller automatically updates the TCP parameters of the current tool and completes the TCP calibration.

Images