Optical processing robot platform full-link error calibration method, computer equipment and computer readable storage medium

Abstract

The present application relates to the technical field of robot error calibration, and particularly relates to a kind of optical processing robot platform full-link error calibration method, computer equipment and computer readable storage medium, optical processing robot platform full-link error calibration method includes after establishing the kinematic error model of entire optical processing robot platform, respectively, empty load calibration and load calibration are carried out, the mechanical arm movement structure parameter error and hand-eye relationship error are obtained by empty load calibration, and compensation and correction are carried out respectively;Tool end point position deviation, tool coordinate system pose deviation and external reference coordinate system pose deviation are obtained by load calibration;Through step-by-step, hierarchical full-link error calibration method, the identification problem caused by multi-parameter strong coupling is overcome, the calibration accuracy and numerical stability are improved, and the two technical problems of "multi-module pose error coupling" and "missing load working condition error calibration" specific to existing optical processing robot platform are solved.

Description

Technical Field

[0001] This invention belongs to the field of robot error calibration technology, and particularly relates to a full-link error calibration method for an optical processing robot platform, a computer device, and a computer-readable storage medium. Background Technology

[0002] The calibration of robot end-effector positioning accuracy can be achieved in two ways: one is by measuring the robot's end-effector pose using external measuring equipment, and the other is by robot self-calibration technology, which involves adding additional constraint modules or various sensors, combined with the robot's own motion characteristics, to calibrate various robot errors. Patent CN115533888A discloses a one-time calibration method based on kinematic models and robotic arm posture constraints to acquire parameter errors and optimize control parameters. Patent CN117840986A discloses a graded calibration compensation method and system for robot positioning errors, which uses Robert's line representation to model the robot, establish a geometric parameter error identification model, solve the geometric parameter error identification model, and compensate for geometric parameter errors.

[0003] Optical processing platforms based on industrial robots require high positioning accuracy to achieve deterministic high-precision optical mirror processing. Inaccurate positioning and alignment of the optical processing robot tool with the workpiece will lead to increased processing cycle time and decreased surface accuracy, directly affecting the efficiency and accuracy of optical processing. Industrial robots themselves are characterized by high repeatability positioning accuracy but poor absolute positioning accuracy. Industrial robots for optical processing applications have more payload modules and more complex error propagation links. Error calibration based on this situation deals with more complex error parameters, and direct calibration methods using motion posture constraints cannot meet the calibration requirements for complex error parameter deviations.

[0004] Commonly used robot calibration methods, such as one-time overall calibration based on external measuring equipment or self-calibration based on the robot's own constraints, are mostly geared towards robots with simple structures and are performed under conditions of no load, ideal environment, and simple trajectory. For optical processing robot platforms carrying multiple sensing and processing modules, existing methods have the following inherent drawbacks when faced with processing tasks involving large diameters and complex surfaces: (1) Error parameter coupling and complex calibration model: The positioning error of the optical processing platform is caused by the coupling and superposition of multiple sources of errors, such as the kinematic parameter error of the robot body, the hand-eye relationship error between the vision sensor and the robot end effector, and the installation error between the processing tool and the end effector. When a one-time calibration method is used to try to identify all error parameters at the same time, various errors interfere with each other in the mathematical model, which leads to the deterioration of the condition number of the parameter identification equation (i.e., the Jacobian matrix tends to be ill-conditioned), making it difficult to obtain stable and accurate solution results.

[0005] (2) Disconnect between calibration status and actual working conditions: The current calibration of optical processing robots with load processing tools is achieved by using three-coordinate calibration of the processing tool's geometric data to calculate the position of feature points, and then establishing the relationship between the robot's end-effector coordinate system and the pose of the feature points through TCP calibration. However, the actual tool installation process inevitably introduces new pose deviations, and there are also positional deviations between the ideal feature points and the actual motion execution points. Under these parameter conditions, the true geometric relationship during loaded operation cannot be accurately reflected, resulting in a decrease in the accuracy of the calibration results in practical applications. Existing methods fail to provide an error calibration process that covers the entire operation chain from "visual sensing positioning" to "tool end-effector execution". It is difficult to systematically classify and decouple the pose errors between various modules such as the robot body, vision module, and processing tool, as well as their relationship with the external world coordinate system. Summary of the Invention

[0006] In view of this, the present invention provides a hierarchical and step-by-step full-link error calibration method for optical processing robot platforms, which can solve the two major technical problems unique to existing optical processing robot platforms: "multi-module pose error coupling" and "lack of error calibration under load conditions".

[0007] To achieve the above objectives, the technical solution created by this invention is implemented as follows: This invention provides a method for end-to-end error calibration of an optical processing robot platform. The optical processing robot platform includes a robotic arm, a vision camera, a calibration plate, and a worktable. The vision camera is fixedly mounted at the end of the robotic arm. The method for end-to-end error calibration of the optical processing robot platform includes the following steps: S1. Establish a first model, which is the kinematic error model of the entire optical processing robot platform; S2. Perform no-load calibration; during no-load calibration, the end effector of the robotic arm is unloaded; the no-load calibration includes the following steps: S21. Establish a second model based on the first model. The second model is a model of the error of the mechanical arm motion structure parameters and the error of the hand-eye relationship. S22. The first solution data is obtained by solving the model of the error of the mechanical arm motion structure parameter and the hand-eye relationship error. The first solution data includes the optimal estimate of the error of the mechanical arm motion structure parameter and the optimal estimate of the hand-eye relationship error. S23. The calculated data is used to compensate for the mechanical arm's motion structure parameters and correct the homogeneous coordinate matrix data of the hand-eye relationship; S3. Perform loading calibration; during loading calibration, an optical processing tool is loaded onto the end effector of the robotic arm; the loading calibration includes the following steps: S31. Based on the no-load calibration, establish a third model, which is an error model for the tool end point position deviation, the tool coordinate system pose deviation, and the external reference coordinate system pose deviation. S32. Through the third model, the second solution data is obtained, which includes the tool end point position deviation, the tool coordinate system pose deviation, and the external reference coordinate system pose deviation.

[0008] Furthermore, during the no-load calibration, the flange tool at the end of the robotic arm is unloaded; during the loading calibration, the flange tool at the end of the robotic arm is loaded with an optical processing tool; the optical processing tool is a magnetorheological polishing tool.

[0009] Furthermore, the first model is: ; in, , is the differential position error vector of the end effector of the robotic arm in the base coordinate system; = , is the differential attitude error vector of the end effector of the robotic arm in the base coordinate system; , , and These represent the error vectors for the length of each joint link of the robotic arm, the link offset error vector, the link torsion angle error vector, and the joint angle error vector, respectively. , , , and , , , This is the corresponding Jacobian matrix; The combined matrix is ​​a 6×4N Viacoder matrix. This is a vector containing all DH parameter errors.

[0010] Furthermore, the second model is: ; in, Indicates the sequence number of the pose change; Represents two pose transformations , The difference between the product of the measured value and the coordinate matrix of the final coordinate system; This represents the difference between the product of the base coordinate system coordinate matrix and the Jacobian matrix of the motion structure parameter error during the two pose transformations. This indicates hand-eye relationship error; , indicating the error of the motion structure parameters of the robotic arm; , representing the difference in coordinate deviation between two different poses; , , These are the displacement deviations in the x, y, and z directions caused by the coordinate system transformation deviation resulting from the hand-eye relationship calibration; , , These are the rotational deviations in the x, y, and z directions, respectively, of the coordinate system transformation deviations caused by hand-eye relationship calibration.

[0011] Furthermore, the process of obtaining the first solution data includes: For each different combination of robotic arm posture changes (i,j), write an equation as in the second model; by planning the robotic arm to execute m sets of posture changes, ensuring that each posture combination is linearly independent, m equations can be obtained; where m≥6; By simultaneously solving the m equations, we can construct a system of equations concerning the unknown parameters. and The overdetermined linear equations are obtained in matrix form as follows: ; The matrix form is abbreviated as: ; It is a column vector consisting of m coordinate difference vectors; The design matrix is ​​composed of coefficient matrices; Let be the vector of error parameters to be solved; Using the least squares solution ,in, The solution obtained Thus, the error of the motion structure parameters of the robotic arm is obtained. The optimal estimate and the hand-eye relationship error The optimal valuation.

[0012] Furthermore, the third model is: ; The absolute coordinates measured by the vision camera; These are the calculated values ​​after compensation and correction based on the second model; The product of the measured values ​​transformed by the reference coordinate system, the base coordinate system, and the TCP coordinate system represents... The coefficient extraction matrix; The product of the measured values ​​transformed from the reference coordinate system and the base coordinate system with the coordinate matrix in the end coordinate system represents... The coefficient extraction matrix; The coordinate matrix of the base coordinate system represents The coefficient extraction matrix; This represents the pose deviation of the tool coordinate system. ; For the deviation of the tool end point, ; For the pose deviation of the external reference coordinate system, .

[0013] Furthermore, the process of obtaining the second solution data includes: A system of overdetermined equations is constructed using n poses, where n ≥ 12; the system of overdetermined equations is as follows: ; The overdetermined system of equations is abbreviated as: ; It is a column vector consisting of n coordinate difference vectors; The design matrix is ​​composed of coefficient matrices; Let be the vector of error parameters to be solved; Using the least squares solution ,in, The solution is obtained This leads to the tool tip position deviation. Tool coordinate system pose deviation and pose deviation of external reference coordinate system .

[0014] Furthermore, the end-to-end error calibration method for the optical processing robot platform also includes the following steps: S24. After performing no-load calibration, calculate the first residual; if the first residual is less than the first threshold, then perform load calibration; if the first residual is greater than or equal to the first threshold, then repeat steps S21 to S23. And / or the optical processing robot platform end-to-end error calibration method further includes the following steps: S33. After loading calibration, calculate the second residual; if the second residual is less than the second threshold, the calibration ends; if the first residual is greater than or equal to the first threshold, repeat steps S31 and S32.

[0015] The present invention also provides a computer device, comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the full-link error calibration method for the optical processing robot platform described above.

[0016] The present invention also provides a computer-readable storage medium storing computer instructions for causing the computer to execute the full-link error calibration method for the optical processing robot platform described in any of the preceding claims.

[0017] Compared with the prior art, the present invention can achieve the following beneficial effects: This invention provides a step-by-step, hierarchical, end-to-end error calibration method for optical processing robot platforms. By decomposing the complex system calibration problem into two sequential stages—"no-load calibration" and "loaded calibration"—it mathematically isolates and solves step-by-step errors related to the robot body and hand-eye relationship, tool installation, and external reference, overcoming the identification challenges caused by strong coupling of multiple parameters and improving calibration accuracy and numerical stability. This achieves effective decoupling and step-by-step identification of error sources. After completing the basic geometric parameter calibration in the no-load stage, the errors introduced by tool load and installation are directly calibrated in the loaded stage, ensuring that the final compensation model accurately reflects the geometric relationships under actual robot working conditions. This ensures that the calibration results are consistent with actual processing conditions. Finally, through the step-by-step calibration of robot kinematic parameters, hand-eye relationship, tool parameters, and the external coordinate system, a complete and high-precision error compensation model is constructed from visual perception to the tool end effector, thereby comprehensively improving the absolute positioning accuracy and operational reliability of the optical processing robot platform. This establishes a calibration accuracy benchmark covering the entire optical processing robot chain. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the basic process of the end-to-end error calibration method for the optical processing robot platform in a specific embodiment of the present invention; Figure 2 This is a partial flowchart illustrating the end-to-end error calibration method for the optical processing robot platform in a specific embodiment of the present invention. Figure 3 This is a schematic diagram of the end-to-end error calibration method for the optical processing robot platform in a preferred embodiment of the present invention; Figure 4 This is a schematic diagram showing the relationship between the device and the orientation in the no-load calibration process according to a specific embodiment of the present invention; Figure 5 This is a schematic diagram showing the relationship between the device and the orientation of the loading calibration process in a specific embodiment of the present invention; Figure 6 A schematic diagram of a computer device for the end-to-end error calibration method of an optical processing robot platform in an embodiment of the present invention.

[0019] Figure label: 11. Vision camera; 12. Calibration plate and worktable; 13. Robotic arm; 14. Optical processing tools; 15. External measuring instrument; 21. Origin of the robot arm base coordinate system; 22. Origin of the robot arm end effector coordinate system; 23. Origin of the vision camera coordinate system; 24. Fixed reference point; 25. Optical processing tool TCP; 26. Origin of the external measuring instrument coordinate system; 31. First pose relationship; 32. Second pose relationship; 33. Third pose relationship; 34. Fourth pose relationship; 35. Fifth pose relationship; 36. Sixth pose relationship; 37. Seventh pose relationship; 41. Computer equipment; 42. External devices; 43. Processing unit; 44. Bus; 45. Network adapter; 46. I / O interface; 47. Display; 48. System memory; 49. Random access memory; 50. Cache; 51. Storage system; 52. Program / utility; 53. Program module. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0022] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0023] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0024] like Figure 1 The diagram shows a basic flowchart of the end-to-end error calibration method for an optical processing robot platform provided in a specific embodiment of the present invention. As can be seen from the diagram, the present invention provides an end-to-end error calibration method for an optical processing robot platform, wherein the optical processing robot platform includes a robotic arm, a vision camera, a calibration plate, and a worktable; the vision camera is fixedly installed at the end of the robotic arm; the end-to-end error calibration method for the optical processing robot platform includes the following steps: S1. Establish a first model, which is the kinematic error model of the entire optical processing robot platform; specifically, the first model is: ; in, , is the differential position error vector of the end effector of the robotic arm in the base coordinate system; = , is the differential attitude error vector of the end effector of the robotic arm in the base coordinate system; , , and These represent the error vectors for the length of each joint link of the robotic arm, the link offset error vector, the link torsion angle error vector, and the joint angle error vector, respectively. , , , and , , , This is the corresponding Jacobian matrix; The combined matrix is ​​a 6×4N Viacoder matrix. This is a vector containing all DH parameter errors; all DH parameters include link length, link offset, link torsion angle, and joint angle.

[0025] S2. Perform no-load calibration; during no-load calibration, the end effector of the robotic arm is unloaded, specifically, the flange tool at the end effector of the robotic arm is unloaded; the no-load calibration includes the following steps: S21. Based on the first model, establish a second model, which is a model of the error in the mechanical arm's motion structure parameters and the error in the hand-eye relationship; specifically, the second model is: ; in, Indicates the sequence number of the pose change; Represents two pose transformations , The difference between the product of the measured value and the coordinate matrix of the final coordinate system; This represents the difference between the product of the base coordinate system coordinate matrix and the Jacobian matrix of the motion structure parameter error during the two pose transformations. This indicates hand-eye relationship error; , indicating the error of the motion structure parameters of the robotic arm; , representing the difference in coordinate deviation between two different poses; specifically, representing the differential error in hand-eye relationship under two different poses i and j of the robotic arm. Differential error of robotic arm pose (or The combined effect of these factors allows for the determination of the coordinates of a fixed reference point in the camera coordinate system. (or The difference in coordinate deviation caused by ); where, This represents the observation point in the camera coordinate system at pose i, caused by the combined hand-eye relationship error and the robot arm pose error. Coordinate deviation; This represents the coordinate deviation caused by the corresponding error at pose j; This refers to the difference in coordinate deviation between the two different poses, which is ignored as a higher-order small quantity in error modeling; , , These are the displacement deviations in the x, y, and z directions caused by the coordinate system transformation deviation resulting from the hand-eye relationship calibration; , , These are the rotational deviations in the x, y, and z directions, respectively, of the coordinate system transformation deviations caused by hand-eye relationship calibration.

[0026] S22. Using the model of the mechanical arm motion structure parameter error and the hand-eye relationship error, first solution data is obtained. The first solution data includes the optimal estimate of the mechanical arm motion structure parameter error and the optimal estimate of the hand-eye relationship error. Specifically, the process of obtaining the first solution data includes: For each different combination of robotic arm posture changes (i,j), write an equation as in the second model; by planning the robotic arm to execute m sets of posture changes, ensuring that each posture combination is linearly independent, m equations can be obtained; where m≥6; By simultaneously solving the m equations, we can construct a system of equations concerning the unknown parameters. and The overdetermined linear equations are obtained in matrix form as follows: ; The matrix form is abbreviated as: ; It is a column vector consisting of m coordinate difference vectors; The design matrix is ​​composed of coefficient matrices; Let be the vector of error parameters to be solved; Using the least squares solution ,in, The solution obtained Thus, the error of the motion structure parameters of the robotic arm is obtained. The optimal estimate and the hand-eye relationship error The optimal valuation.

[0027] S23. The calculated data is used to compensate for the mechanical arm's motion structure parameters and correct the homogeneous coordinate matrix data of the hand-eye relationship; S3. Perform loading calibration; during loading calibration, an optical processing tool is loaded at the end of the robotic arm; specifically, during loading calibration, an optical processing tool is loaded onto the flange tool at the end of the robotic arm; the optical processing tool can be a magnetorheological polishing tool or other optical processing tools.

[0028] The loading calibration includes the following steps: S31. Based on the no-load calibration, establish a third model, which is an error model for the tool end-point position deviation, the tool coordinate system pose deviation, and the external reference coordinate system pose deviation; specifically, the third model is: ; The absolute coordinates measured by the vision camera; These are the calculated values ​​after compensation and correction based on the second model; The product of the measured values ​​transformed by the reference coordinate system, the base coordinate system, and the TCP coordinate system represents... The coefficient extraction matrix; The product of the measured values ​​transformed from the reference coordinate system and the base coordinate system with the coordinate matrix in the end coordinate system represents... The coefficient extraction matrix; The coordinate matrix of the base coordinate system represents The coefficient extraction matrix; This represents the pose deviation of the tool coordinate system. ; For the deviation of the tool end point, ; For the pose deviation of the external reference coordinate system, .

[0029] S32. Using the third model, second solution data is obtained, which includes tool end-point position deviation, tool coordinate system pose deviation, and external reference coordinate system pose deviation; specifically, the process of obtaining the second solution data includes: A system of overdetermined equations is constructed using n poses, where n ≥ 12; the system of overdetermined equations is as follows: ; The overdetermined system of equations is abbreviated as: ; It is a column vector consisting of n coordinate difference vectors; The design matrix is ​​composed of coefficient matrices; Let be the vector of error parameters to be solved; Using the least squares solution ,in, The solution is obtained This leads to the tool tip position deviation. Tool coordinate system pose deviation and pose deviation of external reference coordinate system .

[0030] This invention provides a step-by-step, hierarchical, end-to-end error calibration method for optical processing robot platforms. By decomposing the complex system calibration problem into two sequential stages—"no-load calibration" and "loaded calibration"—it mathematically isolates and solves step-by-step errors related to the robot body and hand-eye relationship, tool installation, and external reference, overcoming the identification challenges caused by strong coupling of multiple parameters and improving calibration accuracy and numerical stability. This achieves effective decoupling and step-by-step identification of error sources. After completing the basic geometric parameter calibration in the no-load stage, the errors introduced by tool load and installation are directly calibrated in the loaded stage, ensuring that the final compensation model accurately reflects the geometric relationships under actual robot working conditions. This ensures that the calibration results are consistent with actual processing conditions. Finally, through the step-by-step calibration of robot kinematic parameters, hand-eye relationship, tool parameters, and the external coordinate system, a complete and high-precision error compensation model is constructed from visual perception to the tool end effector, thereby comprehensively improving the absolute positioning accuracy and operational reliability of the optical processing robot platform. This establishes a calibration accuracy benchmark covering the entire optical processing robot chain.

[0031] The preferred embodiment of the optical processing robot platform end-to-end error calibration method of the present invention further includes the following steps: S24. After performing no-load calibration, calculate the first residual; if the first residual is less than the first threshold, perform loading calibration; if the first residual is greater than or equal to the first threshold, repeat steps S21 to S23; specifically, the first residual refers to the parameter obtained from the calibration after completing the no-load calibration step (i.e., solving for ΔX and ΔE), which is then substituted into the robot kinematics model and hand-eye transformation model to recalculate the predicted position of the fixed reference point in the base coordinate system, and compared with the theoretical fixed coordinate value of the point, or the optimal coordinate value fitted by multiple sets of data; the residual can be the root mean square (RMS) of the pose error of all observation points; the first threshold is a value that is preset according to the basic positioning accuracy requirements of the optical processing robot platform for the robot body and vision system, and is set to 1 / 3 of the end-effector repeatability accuracy; if the first residual is less than the threshold, the no-load calibration step is considered to have converged and the result is reliable, and the loading calibration step can be entered.

[0032] And / or the optical processing robot platform end-to-end error calibration method further includes the following steps: S33. After loading calibration, calculate the second residual; if the second residual is less than the second threshold, the calibration ends; if the first residual is greater than or equal to the first threshold, repeat steps S31 and S32. Specifically, the second residual refers to the predicted position of the tool end effector (TCP) in the world coordinate system after completing the loading calibration step (i.e., solving for ΔP, ΔB, and ΔW), using all the parameters obtained from calibration (including ΔX, ΔE, ΔP, ΔB, and ΔW), and comparing it with the actual measured value of a high-precision external measuring device (such as a laser tracker). This residual can also be calculated as the root mean square (RMS) of the pose error of all measurement points; the second threshold is a more stringent value preset according to the final requirement of the absolute positioning accuracy of the tool end effector in the optical processing task, set as 1 / 3 of the system positioning accuracy required for the expected processing accuracy convergence; if the second residual is less than this threshold, the full-link calibration is considered complete and the accuracy meets the standard; otherwise, the loading calibration step needs to be repeated to further optimize the parameters using the current residual information.

[0033] The present invention also provides an electronic device, the electronic device comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the optical processing robot platform end-to-end error calibration method described in any one of the preceding claims.

[0034] The present invention also provides a computer-readable storage medium storing a computer program that can be executed by a processor to perform the full-link error calibration method for the optical processing robot platform described in any of the above-described embodiments.

[0035] During the step-by-step error calibration process, the vision camera identifies and outputs reference point coordinate information under different poses, obtaining multiple sets of reference point coordinate data in the robot arm end-effector coordinate system. Through the coordinate changes from the robot arm end-effector to the base, multiple sets of coordinate data in the robot arm base coordinate system are obtained. Based on the measurement and attitude constraint process described below, the motion structure parameter model based on differential kinematics, the step-by-step geometric position error link mathematical model, and the mathematical formula for calculating the compensation amount of step-by-step error influencing factors, the error factors of the optical processing robot module components are described, and the error factor deviation of the absolute positioning error of the optical processing robot is calibrated.

[0036] Based on the constraints of the equipment's spatial position, the coordinates of a fixed reference point should theoretically remain unchanged in the robot arm's base coordinate system. However, due to errors in the robot arm's motion control, the coordinate data of the reference point in the base coordinate system obtained from camera recognition and coordinate transformation will slightly deviate from the ideal coordinate data with each posture change. Furthermore, due to deficiencies in the calibration technology between the robot arm and each loading module, hand-eye relationship pose errors and pose errors between the robot arm and each loading module also exist within the equipment system.

[0037] In a more specific implementation, differential kinematics is first used to describe the minute deviations generated by each joint during the robot arm's motion control process, as well as the minute deviations in the pose between the loading module and the robot arm's end effector, thus establishing a kinematic error model based on the industrial robot. This model is then used to describe the transformation relationship between the coordinate systems of adjacent joints in the robot's kinematic chain. After differential motion, it changes to: Then, this differential motion process can be described by micro-translation and micro-rotation relative to the reference coordinate system: ; in: : is the differential change of the pose matrix T.

[0038] : Represents the transformation matrix for differential translation. These represent the differential translations along the X, Y, and Z axes of the current coordinate system.

[0039] : Represents the transformation matrix for differential rotation, It is a unit vector representing the direction of the differential rotation axis; Let be the differential rotation angle about axis k.

[0040] Similarly, differential motion can also be described by micro-translations and micro-rotations relative to the current coordinate system: ; in: ; I: is a 4x4 identity matrix.

[0041] : A 4x4 matrix describing differential translation and rotation, the specific form of which will be given below.

[0042] The descriptions of differential translation and differential rotation are transformed into formulas for transformation matrices. This transformation is to more clearly describe the factors influencing the error in the mathematical model of the geometric position error link based on step-by-step calibration presented later. ; in: ; Simplifying, we get: ; Next, based on the partial differential equations of the DH robot motion parameter model and coordinate transformation, we derive the kinematic error model of the absolute geometric position error of the robotic arm under no-load and the deviation of the robot's structural motion parameter error factors.

[0043] Below are the representations of differential translation vectors and differential rotation vectors: ; ; in: ; ; in: : The differential position error vector of the robot's end effector in the base coordinate system.

[0044] = : The differential attitude error vector of the robot end effector in the base coordinate system (consistent with the definition of the differential rotation vector mentioned above).

[0045] , , , : These represent the error vectors of the DH parameters (link length, link offset, link torsion angle, and joint angle) of each joint of the robot.

[0046] The complete kinematic error model of the robot is derived through comprehensive derivation: ; in: , , , and , , , This is the corresponding Jacobian matrix, which describes the effect of changes in various parameters on the position of the robotic arm's end effector.

[0047] : is the combined global Jacobian matrix (6×4N dimensions), which comprehensively reflects the differential influence of all DH parameter errors on the robot's end-effector pose (position and attitude).

[0048] : This is a vector containing all DH parameter errors. This model establishes a linear mapping relationship between the robot's structural parameter errors and its end-effector absolute geometric position errors, which is the theoretical basis for subsequent error calibration.

[0049] Next, the first step of error calibration is carried out. In a specific embodiment of the present invention, a model of the error of the mechanical arm motion structure parameters and the hand-eye relationship error is established based on the kinematic error model to describe the error transmission link of the first step error calibration process and solve the error of the motion structure parameters and the hand-eye relationship pose error.

[0050] like Figure 2 The figure shows a partial flowchart of the full-link error calibration method for the optical processing robot platform in a specific embodiment of the present invention. As can be seen from the figure, the first step of error calibration mainly calibrates the errors of the industrial robot's motion structure parameters and the hand-eye relationship error.

[0051] Specifically, the main structure of the equipment for the first step of error calibration is an ABB6700 robotic arm with an unloaded end flange tool. An industrial camera is fixedly connected to the end via tooling, forming an "eye in the hand" vision sensing module. The robotic arm faces a 2m flat worktable, on which a high-precision 20µm checkerboard calibration plate is fixed. Fixed points are constrained on the calibration plate, which serve as the identification reference points during the first step of error parameter identification and calibration.

[0052] like Figure 4 The diagram shows the device and pose relationship for the first step of error calibration, i.e., no-load calibration, in a specific embodiment of the present invention. As can be seen from the diagram, during the no-load calibration process, the end effector of the robotic arm 13 is connected to the vision camera 12. The robotic arm 13 can achieve different posture changes and positioning control. The workspace of the robotic arm 13 is arranged with a calibration plate and a worktable 12. Specifically, 21 is the origin of the robotic arm base coordinate system, 22 is the origin of the robotic arm end effector coordinate system, 23 is the origin of the vision camera coordinate system, and 24 is a fixed reference point located on the calibration plate. The first pose relationship 31 is the pose transformation relationship between the robotic arm end effector and the robotic arm base; the second pose relationship 32 is the pose transformation relationship between the vision camera and the robotic arm end effector; the third pose relationship 33 is the pose position of the fixed reference point in the vision camera coordinate system; and the fourth pose relationship 34 is the pose position of the fixed reference point in the robotic arm base coordinate system.

[0053] Based on the equipment used in the first step of error calibration, the ABB6700 robotic arm vision sensing module is used to identify the reference points pre-constrained on the calibration plate. This reference point The coordinates are fixed in the world coordinate system {W}; the vision sensing module, i.e., the vision camera 11, is pre-calibrated with the homogeneous transformation matrix of the pose relationship between the actual camera and the robotic arm end effector. hand-eye relationship By using coordinate system transformation, the coordinate data of the reference point in the coordinate system {B} of the robotic arm's end effector are obtained. Using a pre-planned robotic arm pose change model, the homogeneous matrix of the transformation relationship between the robotic arm end effector and the base coordinate system in the current pose is obtained. The coordinates of the reference point in the base coordinate system {B} can be further calculated based on the robot's internal model and visual observations. .

[0054] Based on the first error calibration device, this invention pre-plans the changes in the robotic arm's motion posture and obtains a dataset of reference point coordinates in the base coordinate system under multiple posture changes. This dataset is calculated using a mathematical model based on the relationship between the robotic arm's structural parameter errors and the geometric position error of the hand-eye relationship, thus solving for the errors in the robotic arm's motion structural parameters of the optical processing robot platform. Pose error in hand-eye coordinate system The actual input data.

[0055] Specifically, in the process of establishing a model for the error of the robotic arm's motion structure parameters and the hand-eye relationship error based on the kinematic error model, the actual value of the fixed reference point in the robotic arm's base coordinate system is... The ideal value of the fixed reference point in the coordinate system of the robot arm base is The meanings of each parameter are as follows: The coordinates of the fixed point under the robot arm base are obtained by coordinate transformation after identifying the fixed point after each posture change; The coordinates of the fixed point in the camera coordinate system after each attitude change; This refers to the ideal location of the reference point in the base coordinate system. This represents the deviation between the actual location and the ideal location.

[0056] Expand = ; The coordinate system transformation deviations caused by hand-eye relationship calibration, specifically the displacement and rotation deviations in the x, y, and z directions, are expressed as follows: ; The key derivations and conclusions of the model for establishing the error of the robotic arm's motion structure parameters and the hand-eye relationship error are expressed as follows: ; .

[0057] Then, through no fewer than 6 sets of posture changes, the coordinate data of 12 reference points recognized by the visual camera are obtained. Based on the conclusion formula of the model of mechanical arm motion structure parameter error and hand-eye relationship error, the mathematical model for solving the hand-eye relationship pose error and structure parameter error in the first step of error calibration can be derived: ; in, Indicates the sequence number of the pose change; Represents two pose transformations , The difference between the product of the measured value and the coordinate matrix of the final coordinate system; This represents the difference between the product of the base coordinate system coordinate matrix and the Jacobian matrix of the motion structure parameter error during the two pose transformations. This indicates hand-eye relationship error; , indicating the error of the motion structure parameters of the robotic arm; , representing the difference between two coordinate deviations, can be regarded as a high-order infinitesimal quantity that can be omitted from calculation; The difference between higher-order quadratic terms is ignored during the calculation. The measured coordinate data is substituted into the solution model, and the equations are solved simultaneously. The least squares method is used to fit the solution results to obtain the required camera pose deviation for the end-effector coordinate system transformation and the error of the mechanical arm end-effector motion structure parameters from the base, thus completing the first step of error calibration. The solution data is directly used for the mechanical arm structural parameter compensation and the correction of the homogeneous coordinate matrix data of the hand-eye relationship in the first step of error calibration.

[0058] For each different combination of robotic arm posture changes (i,j), an equation as shown above can be written. By planning the robotic arm to perform m sets (m≥6) of posture changes and ensuring that each posture combination is linearly independent, m equations can be obtained.

[0059] By simultaneously solving the m equations, we construct an overdetermined linear system of equations concerning the unknown parameters ΔE and ΔX (n-dimensional, where n is the number of DH parameters of the robotic arm), in matrix form: ; Abbreviated as ; in: : is a column vector consisting of m coordinate difference vectors; : is a design matrix composed of coefficient matrices; : is the error parameter vector to be solved.

[0060] Least squares solution: ; in: This is the optimal calibration result. The solution obtained... This includes the optimal estimates of the hand-eye relationship pose deviation ΔE and the mechanical arm motion structure parameter error ΔX. These results are directly used to compensate and correct the kinematic model and hand-eye transformation matrix in the robot controller, completing the first step of error calibration.

[0061] The second step of error calibration mainly involves the pose error of the tool center point (TCP) of the optical processing tool relative to the end of the robotic arm, and the pose error of the industrial robot base-reference (laser tracker) coordinate system. Regardless of whether the end of the robotic arm is equipped with a polishing tool, milling tool, cleaning nozzle, or measuring probe, as long as it is installed as a rigid tool at the end of the robotic arm, there will be a pose relationship that needs to be calibrated between its end operating point (TCP) and the flange at the end of the robotic arm. The main purpose of the second step of error calibration is to accurately measure and compensate for this pose deviation generated after the installation of various processing tools, thereby ensuring that the robotic arm can accurately control the end of the tool to reach the predetermined spatial position.

[0062] The equipment for the second step of error calibration is mainly based on the equipment used in the first step. A specific optical processing tool, such as a magnetorheological polishing tool, is mounted on the end flange of the robotic arm. The relationship between the tool's end-operating point (i.e., the tool's center point TCP) and the robotic arm's end-effector coordinate system is calibrated. This process yields the homogeneous matrix of the transformation relationship between the tool coordinate system and the robotic arm's end-effector coordinate system. and TCP pose calibration matrix For the working area faced by the robotic arm, while retaining the original reference points, an external reference system W is added, based on an optical laser tracker. The laser tracker is used to measure the coordinates of the tool in the external reference system when the robotic arm loads the tool and changes its posture, and when it contacts the positioning reference point.

[0063] like Figure 5The diagram shows the relationship between the device and the pose of the second step error calibration, i.e., loading calibration, in a specific embodiment of the present invention. As can be seen from the diagram, during the loading calibration process, the end of the robotic arm 13 is connected to the vision camera 11, and the end of the robotic arm 13 carries an optical processing tool 14, specifically an optical processing polishing tool. The robotic arm 13 can achieve different posture changes and positioning control. The workspace of the robotic arm 13 is arranged with a calibration plate, a worktable 12, and an external measuring instrument 15. Specifically, 21 is the origin of the robotic arm base coordinate system, 22 is the origin of the robotic arm end coordinate system, 23 is the origin of the vision camera coordinate system (i.e., the optical center of the vision camera 11), 24 is a fixed reference point located on the calibration plate, 25 is the optical processing tool TCP, and 26 is the origin of the external measuring instrument coordinate system (i.e., the origin of the reference coordinate system during this loading calibration process). The external measuring instrument 15 can measure the pose of the calibrated robotic arm base and the fixed reference point in the external measuring instrument coordinate system. The first pose relationship 31 is the pose transformation relationship between the end effector of the robotic arm and the robotic arm base; the second pose relationship 32 is the pose transformation relationship between the vision camera and the end effector of the robotic arm; the third pose relationship 33 is the pose position of the fixed reference point in the vision camera coordinate system; the fourth pose relationship 34 is the pose position of the fixed reference point in the robotic arm base coordinate system; the fifth pose relationship 35 is the pose transformation relationship between the optical processing tool and the end effector of the robotic arm; the sixth pose relationship 36 is the pose relationship between the robotic arm base and the (external measuring instrument) reference coordinate system; and the seventh pose relationship 37 is the pose position of the fixed reference point in the (external measuring instrument) reference coordinate system.

[0064] In the second step of error calibration, the coordinate systems for fixed positions in the equipment are the robotic arm base coordinate system and the external reference coordinate system. The fixed points are the reference points constrained on the calibration plate. Because the laser tracker has high-precision position coordinate measurement, the measured coordinates of the reference points in the external reference coordinate system {W} are used... As standard reference coordinate data, the camera identifies the constrained reference points in the current robotic arm posture. Homogeneous matrix of hand-eye relationship coordinate transformation The coordinate system transformation yields the coordinates of the reference point in the coordinate system {E} of the robotic arm's end effector. The homogeneous transformation matrix of the robot arm's base-endpoint coordinate system in the current posture. The coordinates calculated in the base coordinate system {B} are obtained based on the internal model and vision. .

[0065] Then, a homogeneous transformation matrix relating the pose of a pre-calibrated laser tracker (which may have errors) to the robotic arm base is used. ,Will By calculating the transformation matrix and converting the coordinates to the base coordinate system {B}, we obtain the coordinate values ​​of a reference point based on external high-precision measurements but after preliminary coordinate transformation. This represents the coordinate data of the reference point in the reference coordinate system under a specific attitude. Based on the equipment used for error calibration in the second step, multiple pose changes of the robotic arm can be performed to obtain two sets of data: high-precision external measurement conversion values. and internal model visual calculation values The difference between the two becomes the input to the error calibration model based on fixed-point constraints. This coordinate dataset is used by the error calibration model based on fixed-point constraints to solve the pose error between the external reference system and the base coordinate system. TCP pose relationship error TCP center point deviation Absolute positioning error of optical processing robot The actual input data.

[0066] A kinematic error model is established, comprising mathematical models for TCP position deviation, TCP pose deviation, and external reference coordinate system pose deviation errors. This model describes the error propagation link in the second step of error calibration. Mathematical formulas for calculating TCP position deviation, TCP pose deviation, and external reference coordinate system pose deviation errors are also established. The second step of error calibration is performed after the calibration and compensation of the robot's kinematic parameter error ΔX and hand-eye relationship error ΔE have been completed in the first step. This step aims to establish error models for the tool end effector (TCP) position deviation ΔP, tool coordinate system pose deviation ΔB, and external reference coordinate system pose deviation ΔW.

[0067] The mathematical model for the second step of error calibration is similar in derivation to that of the first step, both based on differential kinematics and fixed-point constraints. Linearization is achieved by ignoring higher-order terms, ultimately resulting in a system of linear equations in the form of "Measured value - Theoretical value = Coefficient matrix · Error parameter". However, there are fundamental differences between the two in terms of error sources, reference standards, and solution objectives: The source of error differs from the solution objective: the first step of the model The objective is to solve for the robot's kinematic parameter error ΔX and the hand-eye relationship error ΔE. The second step involves the model... The objectives are to determine the tool center point deviation ΔP, the tool coordinate system pose deviation ΔB, and the base-external reference coordinate system pose deviation ΔW.

[0068] Different reference benchmarks: the "theoretical value" in the first step. Calculations based on the uncorrected robot model indicate its "fixed point". It is a theoretically unknown constant. The second step is the "theoretical value". This is calculated based on the robot model that has been calibrated and corrected in the first step, while the "measured value" is... These are high-precision absolute coordinate measurements from a laser tracker, forming a higher-order precision benchmark.

[0069] The second step of error calibration involves modeling a new, independent error source (tool, external reference) after the first step of error calibration, which has already corrected the error between the robot body and the hand-eye relationship. Together, the two error calibration models constitute a complete error calibration chain from the robot body to the tool end effector and from the internal relative reference to the external absolute reference.

[0070] In the second step of error calibration, the reference points on the calibration board are first identified using a vision camera, and the coordinates of the reference points in the end coordinate system are output. and the coordinates of the reference point in the base coordinate system Reference point coordinates in the end coordinate system Combined with the calibrated TCP coordinate system transformation relationship It is used for the positioning and alignment of the robotic arm carrying the loaded processing tool with the reference point.

[0071] The external measuring instrument (laser tracker) base of the calibration system serves as the established reference coordinate system. It is used to calibrate the pose relationship between the laser tracker base coordinate system and the robotic arm base coordinate system, thereby obtaining their homogeneous coordinate transformation matrix. The laser tracker measures the position coordinates of the calibration plate reference point in the reference coordinate system. Coordinates of this location This serves as the ideal position information for the reference point during the second step of error calibration. Upon completion of control positioning and alignment, a laser tracker is used to measure the position coordinates of the tool in its current attitude within the reference coordinate system. .

[0072] When the robotic arm carrying the tool moves to the i-th pose and aligns with the fixed reference point, the laser tracker measures the actual coordinates of the tool's end effector in the world coordinate system {W} as follows: The theoretical coordinates calculated using the calibrated transformation relationships are: : ; : These are the coordinates of TCP in the tool coordinate system {T}. Due to uncalibrated errors ΔP, ΔB, and ΔW, the actual values ​​deviate from the theoretical values.

[0073] Introducing the differential transformation matrix The attitude deviation ΔB of the description tool installation. Describe the pose deviation ΔW of the base coordinate system installation, and the position deviation vector. Description of the deviation of the end point of the tool The complete transformation relationship after considering the errors is the mathematical model for TCP position deviation, TCP pose deviation, and external reference coordinate system pose deviation errors: ; Expanding on the above, and systematically ignoring all second-order and higher error cross terms (including...) The second-order and higher-order terms of the error term are then used. Only the first-order error term is retained, resulting in the linearized model: ; Ignoring the quadratic term of the coordinate transformation pose error, the expression is rearranged to obtain: ; ; ; The product of the measured values ​​transformed by the reference coordinate system, the base coordinate system, and the TCP coordinate system represents... The coefficient extraction matrix; The product of the measured values ​​transformed from the reference coordinate system and the base coordinate system with the coordinate matrix in the end coordinate system represents... The coefficient extraction matrix; The coordinate matrix of the base coordinate system represents The coefficient extraction matrix; The third model, namely the error model for tool end-point position deviation, tool coordinate system pose deviation, and external reference coordinate system pose deviation, is as follows: ; This refers to the pose deviation in the tool coordinate system, specifically the pose deviation between the TCP and the robot arm's end effector coordinate system. The non-zero elements correspond to it; ; This refers to the tool end-point position deviation, specifically the coordinate deviation of the TCP relative to the origin of the robotic arm's end-point coordinate system. The first three elements correspond to it; ; This refers to the pose deviation of the external reference coordinate system, i.e., the pose deviation between the reference coordinate system and the robot arm base coordinate system. The non-zero elements correspond to it; ; In a specific implementation, the hand-eye relationship pose deviation in the optical processing robot platform end-link error calibration method provided by the present invention mainly refers to the displacement and rotation deviation between the pose of the vision camera and the pose of the end-axis of the robotic arm; while the TCP pose deviation mainly refers to the displacement and rotation deviation between the feature point position of the end-axis load tool of the robotic arm and the pose of the end-axis of the robotic arm.

[0074] Since a high-precision external measuring device has been introduced to measure the coordinate data of the reference point, and this data is regarded as the standard value, the mathematical model formula given above, which ignores quadratic terms, can be directly substituted into the data to solve for the error deviation. During the calculation, the difference of higher-order quadratic terms can be ignored. The known measured coordinate data is substituted into the solution model, the equations are solved simultaneously, and the least squares method is used to fit the solution results. The TCP point position deviation, TCP pose deviation, and external reference coordinate system pose deviation are obtained. This completes the second step of error calibration, realizing the calibration of the absolute position error and the deviation of related error factors of the optical processing robot platform. The specific formulas are as follows: Construct an overdetermined system of equations using n poses (n≥12): ; Abbreviated as ; Least squares solution: ; in: This represents the optimal calibration result.

[0075] This invention provides a calibration method and apparatus for the deviation of positioning errors in an industrial robot optical processing platform. Its main objective is to achieve high-precision measurement and calibration of pose deviations and system absolute geometric position errors for multiple load modules of an optical processing industrial robot. The invention employs step-by-step, fixed-point-constrained pose measurement and positioning error measurement based on machine vision. It utilizes differential kinematics to construct a pose error relationship model between modules, accurately calculating the values ​​of pose deviations and system absolute geometric position errors. Through an innovative step-by-step decoupling calibration strategy, combined with rigorous overdetermined equations and least-squares fitting, it systematically solves the calibration problem of multi-source error coupling in optical processing robot platforms, ultimately achieving high-precision identification of end-to-end errors.

[0076] In a specific preferred embodiment, such as Figure 3 As shown, the first step of the no-load error calibration experiment is completed to obtain measured point data. Based on the error mathematical model derived in the invention, the deviation value of the error influencing factors is calculated, and the motion control parameters of the robotic arm are optimized and the posture deviation is corrected to control the error residual after the first step of no-load error calibration within the expected threshold value range. Then, the second step of the loading error calibration process is carried out. According to the experimental data according to the flowchart shown, the error influence deviation is calculated according to the error mathematical model derived in the invention to realize error calibration and parameter optimization, and finally realize the positioning error calibration and alignment positioning accuracy of the optical processing robot equipment. By calculating the residual and comparing it with the threshold, and simultaneously performing iterative compensation optimization, the practicality of the full-link error calibration method of the present invention is further ensured through an optimized closed loop.

[0077] Accordingly, according to embodiments of the present invention, the present invention also provides a computer device, a computer-readable storage medium, and a computer program product.

[0078] Figure 6 This is a schematic diagram of the structure of a computer device 41 provided in an embodiment of the present invention. Figure 6 A block diagram of an exemplary computer device 41 suitable for implementing embodiments of the present invention is shown. Figure 6 The computer device 41 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0079] like Figure 6 As shown, computer device 41 is represented in the form of a general-purpose computing device. Computer device 41 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0080] The components of the computer device 41 may include, but are not limited to: one or more processors or processing units 43, system memory 48, and bus 44 connecting different system components (including system memory 48 and processing unit 43).

[0081] Bus 44 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0082] Computer device 41 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 41, including volatile and non-volatile media, removable and non-removable media.

[0083] System memory 48 may include computer system readable media in the form of volatile memory, such as random access memory 49 (RAM) and / or cache memory 50. Computer device 41 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 51 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 6 Not shown; usually referred to as a "hard drive"). Although Figure 6 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 44 via one or more data media interfaces. System memory 48 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0084] A program / utility 52 having a set (at least one) of program modules 53 may be stored, for example, in system memory 48. Such program modules 53 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 53 typically perform the functions and / or methods described in the embodiments of the present invention.

[0085] Computer device 41 can also communicate with one or more external devices 42 (e.g., keyboard, pointing device, display 47, etc.), and with one or more devices that enable a user to interact with computer device 41, and / or with any device that enables computer device 41 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 46. Furthermore, computer device 41 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 45.

[0086] like Figure 6 As shown, network adapter 45 communicates with other modules of computer device 41 via bus 44. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with computer device 41, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0087] The processing unit 43 executes various functional applications and data processing by running programs stored in the system memory 48, such as implementing the full-link error calibration method for the optical processing robot platform provided in the embodiments of the present invention.

[0088] In a specific embodiment of the present invention, a computer-readable storage medium storing computer instructions is also provided, specifically a non-transient computer-readable storage medium storing a computer program thereon, wherein the program, when executed by a processor, is the end-to-end error calibration method for the optical processing robot platform provided in all embodiments of the present application.

[0089] The computer storage medium of this invention can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0090] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0091] The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof. The computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0092] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described end-to-end error calibration method for an optical processing robot platform.

[0093] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0094] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for end-to-end error calibration of an optical processing robot platform, characterized in that: The optical processing robot platform includes a robotic arm, a vision camera, a calibration plate, and a worktable; the vision camera is fixedly mounted at the end of the robotic arm; the end-to-end error calibration method for the optical processing robot platform includes the following steps: S1. Establish a first model, which is the kinematic error model of the entire optical processing robot platform; S2. Perform no-load calibration; during no-load calibration, the end effector of the robotic arm is unloaded; the no-load calibration includes the following steps: S21. Establish a second model based on the first model. The second model is a model of the error of the mechanical arm motion structure parameters and the error of the hand-eye relationship. S22. The first solution data is obtained by solving the model of the error of the mechanical arm motion structure parameter and the hand-eye relationship error. The first solution data includes the optimal estimate of the error of the mechanical arm motion structure parameter and the optimal estimate of the hand-eye relationship error. S23. The calculated data is used to compensate for the mechanical arm's motion structure parameters and correct the homogeneous coordinate matrix data of the hand-eye relationship; S3. Perform loading calibration; during loading calibration, an optical processing tool is loaded onto the end effector of the robotic arm; the loading calibration includes the following steps: S31. Based on the no-load calibration, establish a third model, which is an error model of the tool end point position deviation, the tool coordinate system pose deviation, and the external reference coordinate system pose deviation. S32. Through the third model, the second solution data is obtained, which includes the tool end point position deviation, the tool coordinate system pose deviation, and the external reference coordinate system pose deviation.

2. The end-to-end error calibration method for an optical processing robot platform according to claim 1, characterized in that: During the no-load calibration, the flange tool at the end of the robotic arm is unloaded; during the loading calibration, the flange tool at the end of the robotic arm is loaded with an optical processing tool; the optical processing tool is a magnetorheological polishing tool.

3. The end-to-end error calibration method for an optical processing robot platform according to claim 1, characterized in that: The first model is: ； in, , is the differential position error vector of the end effector of the robotic arm in the base coordinate system; = , is the differential attitude error vector of the end effector of the robotic arm in the base coordinate system; , , and These represent the error vectors for the length of each joint link of the robotic arm, the link offset error vector, the link torsion angle error vector, and the joint angle error vector, respectively. , , , and , , , This is the corresponding Jacobian matrix; The combined matrix is ​​a 6×4N Viacoder matrix. This is a vector containing all DH parameter errors.

4. The end-to-end error calibration method for an optical processing robot platform according to claim 1, characterized in that: The second model is: ; in, Indicates the sequence number of the pose change; Represents two pose transformations , The difference between the product of the measured value and the coordinate matrix of the final coordinate system; This represents the difference between the product of the base coordinate system coordinate matrix and the Jacobian matrix of the motion structure parameter error during the two pose transformations. This indicates hand-eye relationship error; , indicating the error of the motion structure parameters of the robotic arm; , representing the difference in coordinate deviation between two different poses; , , These are the displacement deviations in the x, y, and z directions caused by the coordinate system transformation deviation resulting from the hand-eye relationship calibration; , , These are the rotational deviations in the x, y, and z directions, respectively, of the coordinate system transformation deviations caused by hand-eye relationship calibration.

5. The end-to-end error calibration method for an optical processing robot platform according to claim 4, characterized in that: The process of obtaining the first solution data includes: For each different combination of robotic arm posture changes (i,j), write an equation as in the second model; by planning the robotic arm to execute m sets of posture changes, ensuring that each posture combination is linearly independent, m equations can be obtained; where m≥6; By simultaneously solving the m equations, we can construct a system of equations concerning the unknown parameters. and The overdetermined linear equations are obtained in matrix form as follows: ； The matrix form is abbreviated as: ; It is a column vector consisting of m coordinate difference vectors; The design matrix is ​​composed of coefficient matrices; Let be the vector of error parameters to be solved; Using the least squares solution ,in, The solution obtained Thus, the error of the motion structure parameters of the robotic arm is obtained. The optimal estimate and the hand-eye relationship error The optimal valuation.

6. The end-to-end error calibration method for an optical processing robot platform according to claim 5, characterized in that: The third model is: ； The absolute coordinates measured by the vision camera; These are the calculated values ​​after compensation and correction based on the second model; The product of the measured values ​​transformed by the reference coordinate system, the base coordinate system, and the TCP coordinate system represents... The coefficient extraction matrix; The product of the measured values ​​transformed from the reference coordinate system and the base coordinate system with the coordinate matrix in the end coordinate system represents... The coefficient extraction matrix; The coordinate matrix of the base coordinate system represents The coefficient extraction matrix; This represents the pose deviation of the tool coordinate system. ； For the deviation of the tool end point, ； For the pose deviation of the external reference coordinate system, 。 7. The end-to-end error calibration method for an optical processing robot platform according to claim 6, characterized in that: The process of obtaining the second solution data includes: A system of overdetermined equations is constructed using n poses, where n ≥ 12; the system of overdetermined equations is as follows: ； The overdetermined system of equations is abbreviated as: ; It is a column vector consisting of n coordinate difference vectors; The design matrix is ​​composed of coefficient matrices; Let be the vector of error parameters to be solved; Using the least squares solution ,in, The solution is obtained This leads to the tool tip position deviation. Tool coordinate system pose deviation and pose deviation of external reference coordinate system .

8. The end-to-end error calibration method for an optical processing robot platform according to claim 1, characterized in that: The full-link error calibration method for the optical processing robot platform also includes the following steps: S24. After performing no-load calibration, calculate the first residual; if the first residual is less than the first threshold, then perform load calibration; if the first residual is greater than or equal to the first threshold, then repeat steps S21 to S23. And / or the optical processing robot platform end-to-end error calibration method further includes the following steps: S33. After loading calibration, calculate the second residual; if the second residual is less than the second threshold, the calibration ends; if the first residual is greater than or equal to the first threshold, repeat steps S31 and S32.

9. A computer device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the full-link error calibration method for the optical processing robot platform as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to execute the full-link error calibration method for the optical processing robot platform as described in any one of claims 1 to 8.

Images