Error diagnosis method and device and electronic equipment
By constructing a full-parameter error model and considering multiple error terms, the problem of difficulty in converging accuracy when assembly errors exist in DH kinematics error model is solved, and the robot accuracy is improved and error analysis is achieved.
Patent Information
- Application Number
- CN202411127632.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-15
AI Technical Summary
The existing DH kinematic error model is difficult to converge the robot accuracy to an effective threshold when the robot assembly error exists, resulting in low absolute positioning accuracy.
By constructing a full parameter error model, considering the bias error, deflection error, joint angle error, joint flexibility error and link flexibility error between each two adjacent joints, the coordinates of the calibration component when the robot is at different joint angles are obtained. Combined the target full parameter error model to show the error, and adjust the robot based on this to improve accuracy.
It effectively improves the absolute positioning accuracy of the robot, can quickly diagnose and analyze the source of errors, provides targeted repair solutions, and improves the quality and efficiency of design and production.
Smart Images

Figure CN118990616A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of system modeling and parameter identification of industrial robots, and in particular to an error diagnosis method, device and electronic equipment. Background Art
[0002] Positioning accuracy is an important indicator for evaluating robot performance and an important guarantee for the robot to complete its operational tasks. With the gradual complexity of the robot application environment, people have put forward higher requirements for the robot's motion accuracy.
[0003] In the related art, the Denavit-Hartenberg (DH) kinematic error model is usually used for robot parameter calibration. However, the DH kinematic error model is established when the axes of the robot's joints are horizontal or vertical relative to the base, and does not take into account possible parallelism or verticality errors. In the actual production process, due to the existence of assembly errors, the axes of the robot's joints are not necessarily horizontal or vertical relative to the base in the robot's zero position state. Therefore, when such assembly errors exist, it is difficult to converge the robot's accuracy to an effective threshold using the DH kinematic error model for robot parameter calibration, that is, the robot's absolute positioning accuracy is low. Summary of the invention
[0004] The purpose of the embodiments of the present application is to provide an error diagnosis method, device and electronic device to solve the problem that it is difficult to make the robot accuracy converge to an effective threshold using the DH kinematic error model. The specific technical solution is as follows:
[0005] In a first aspect, an embodiment of the present application provides an error diagnosis method, the method comprising:
[0006] Acquire the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles, and use them as the first coordinates corresponding to each of the first preset joint angles;
[0007] Fitting the correspondence between the first preset joint angle and the first coordinate by an original full-parameter error model to obtain a target full-parameter error model, and using the model parameters in the target full-parameter error model as the first error of the robot and displaying the first error;
[0008] Among them, the original full-parameter error model is a model with full-parameter error as model parameter, robot joint angle as independent variable, and coordinates of the calibration component at the end of the robot as dependent variable. The full-parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between every two adjacent joints.
[0009] In some embodiments, the method further comprises:
[0010] Fitting the corresponding relationship between the first preset joint angle and the first coordinate through the original DH kinematic error model to obtain a target DH kinematic error model, and using the model parameters in the target DH kinematic error model as the second error of the robot and displaying the second error;
[0011] The original DH kinematic error model is a model with DH kinematic error as model parameter, robot joint angle as independent variable, and coordinates of calibration components at the end of the robot as dependent variable. The DH kinematic error includes offset error of each rod length of the robot, zero position error of each joint of the robot, and stiffness error of each joint of the robot.
[0012] The step of fitting the corresponding relationship between the first pre-joint angle and the first coordinate by using an original full-parameter error model to obtain a target full-parameter error model includes:
[0013] In response to the diagnostic instruction for the second error, the corresponding relationship between the first preset joint angle and the first coordinate is fitted by an original full-parameter error model to obtain a target full-parameter error model.
[0014] In some embodiments, the calibration component is a target ball, the first coordinate is the coordinate of the target ball in the laser follower coordinate system obtained after the target ball is captured by a laser follower, and the method further includes:
[0015] For each of the first preset joint angles, respectively, controlling the robot to move to the first preset joint angle with the target ball fixed at the end, and controlling the laser follower to capture the target ball;
[0016] If the coordinates returned by the laser follower are not received within a preset time, determining the conversion relationship between the robot flange coordinate system and the base coordinate system according to the first preset joint angle;
[0017] According to the conversion relationship between the base coordinate system and the laser follower coordinate system, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the flange coordinate system and the base coordinate system, the coordinates of the target sphere in the laser follower coordinate system are determined as the second coordinates;
[0018] Adjusting the orientation of the laser follower so that the field of view of the laser follower covers the second coordinate;
[0019] The coordinates obtained by the laser follower capturing the target sphere are acquired as the first coordinates corresponding to the first preset joint angle.
[0020] In some embodiments, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system are obtained in advance by the following method:
[0021] Obtaining the coordinates of the target ball fixed to the end of the robot in the laser follower coordinate system when the robot is at different second preset joint angles, respectively as the third coordinates corresponding to each of the second preset joint angles; and, for each of the second preset joint angles, obtaining the conversion relationship of the flange coordinate system relative to the base coordinate system, as the conversion relationship of the flange coordinate system corresponding to each of the second preset joint angles relative to the base coordinate system;
[0022] According to the transformation relationship between the flange coordinate system corresponding to each of the second preset joint angles and the base coordinate system, and the third coordinates corresponding to each of the second preset joint angles, the transformation relationship between the tool coordinate system and the flange coordinate system, and the transformation relationship between the base coordinate system and the laser follower coordinate system are fitted.
[0023] In some embodiments, the method further comprises:
[0024] For each of the first errors, determining a maximum accuracy that can be achieved by the robot by adjusting the first error alone as a calibration residual corresponding to each of the first errors;
[0025] Determine a first error whose corresponding calibration residual satisfies a preset condition as a third error;
[0026] The displaying of the first error comprises:
[0027] The third error is displayed in a first display manner, and the fourth error is displayed in a second display manner, wherein the fourth error is an error in the first error except the third error.
[0028] In some embodiments, an error schematic diagram for each error configuration is obtained in advance;
[0029] The displaying of the third error in a first display manner includes:
[0030] For each of the third errors, obtaining an error schematic diagram configured for the third error as a target error schematic diagram for the third error;
[0031] The third errors, the target error diagrams of the third errors, and the calibration residuals corresponding to the third errors are displayed.
[0032] In some embodiments, the method comprises:
[0033] Displaying a first interface for indicating error diagnosis progress;
[0034] In response to a first instruction for the first interface, executing the step of fitting the correspondence between the first preset joint angle and the first coordinate by using an original DH kinematics error model to obtain a target DH kinematics error model;
[0035] In response to a second instruction for the first interface, the step of fitting the correspondence between the first preset joint angle and the first coordinate through an original full-parameter error model to obtain a target full-parameter error model is executed.
[0036] In some embodiments, the method further comprises:
[0037] In response to a third instruction for the first interface, initializing the robot and the laser follower;
[0038] In response to a fourth instruction for the first interface, the current accuracy of the robot is determined and displayed.
[0039] In a second aspect, an embodiment of the present application further provides a calibration error diagnosis device, the device comprising:
[0040] A coordinate acquisition module, used to acquire the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles, as the first coordinates corresponding to each of the first preset joint angles;
[0041] A first error determination module is used to fit the corresponding relationship between the first preset joint angle and the first coordinate through an original full-parameter error model to obtain a target full-parameter error model, and use the model parameters in the target full-parameter error model as the first error of the robot and display the first error;
[0042] Among them, the original full-parameter error model is a model with full-parameter error as model parameter, robot joint angle as independent variable, and coordinates of the calibration component at the end of the robot as dependent variable. The full-parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between every two adjacent joints.
[0043] In some embodiments, the apparatus further comprises:
[0044] A second error determination module is used to fit the corresponding relationship between the first preset joint angle and the first coordinate through an original DH kinematic error model to obtain a target DH kinematic error model, and use the model parameters in the target DH kinematic error model as the second error of the robot and display the second error;
[0045] The original DH kinematic error model is a model with DH kinematic error as model parameter, robot joint angle as independent variable, and coordinates of calibration components at the end of the robot as dependent variable. The DH kinematic error includes offset error of each rod length of the robot, zero position error of each joint of the robot, and stiffness error of each joint of the robot.
[0046] The first error determination module is specifically used to:
[0047] In response to the diagnostic instruction for the second error, the corresponding relationship between the first preset joint angle and the first coordinate is fitted by an original full-parameter error model to obtain a target full-parameter error model.
[0048] In some embodiments, the calibration component is a target ball, the first coordinate is the coordinate of the target ball in the laser follower coordinate system obtained after the target ball is captured by a laser follower, and the device further includes:
[0049] A control module, for controlling the robot to move to the first preset joint angle with the target ball fixed at the end thereof, and controlling the laser follower to capture the target ball;
[0050] a first determination module, configured to determine a conversion relationship between the robot flange coordinate system and the base coordinate system according to the first preset joint angle if the coordinates returned by the laser follower are not received within a preset time;
[0051] A conversion module, used to determine the coordinates of the target sphere in the laser follower coordinate system as the second coordinate according to the conversion relationship between the base coordinate system and the laser follower coordinate system, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the flange coordinate system and the base coordinate system;
[0052] An adjustment module, used for adjusting the orientation of the laser follower so that the field of view of the laser follower covers the second coordinate;
[0053] The coordinate acquisition module is further used to acquire the coordinates obtained by the laser follower capturing the target ball as the first coordinates corresponding to the first preset joint angle.
[0054] In some embodiments, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system are obtained in advance by the following method:
[0055] Obtaining the coordinates of the target ball fixed to the end of the robot in the laser follower coordinate system when the robot is at different second preset joint angles, respectively as the third coordinates corresponding to each of the second preset joint angles; and, for each of the second preset joint angles, obtaining the conversion relationship of the flange coordinate system relative to the base coordinate system, as the conversion relationship of the flange coordinate system corresponding to each of the second preset joint angles relative to the base coordinate system;
[0056] According to the transformation relationship between the flange coordinate system corresponding to each of the second preset joint angles and the base coordinate system, and the third coordinates corresponding to each of the second preset joint angles, the transformation relationship between the tool coordinate system and the flange coordinate system, and the transformation relationship between the base coordinate system and the laser follower coordinate system are fitted.
[0057] In some embodiments, the apparatus further comprises:
[0058] A second determination module is used to determine, for each of the first errors, a maximum accuracy that can be achieved by the robot when only the first error is adjusted, as a calibration residual corresponding to each of the first errors;
[0059] A third determination module is used to determine a first error corresponding to the calibration residual that satisfies a preset condition as a third error;
[0060] The first error determination module is specifically used to:
[0061] The third error is displayed in a first display manner, and the fourth error is displayed in a second display manner, wherein the fourth error is an error in the first error except the third error.
[0062] In some embodiments, an error schematic diagram for each error configuration is obtained in advance;
[0063] The first error determination module is specifically used to:
[0064] For each of the third errors, obtaining an error schematic diagram configured for the third error as a target error schematic diagram for the third error;
[0065] The third errors, the target error diagrams of the third errors, and the calibration residuals corresponding to the third errors are displayed.
[0066] In some embodiments, the apparatus comprises:
[0067] An interface display module, used to display a first interface for indicating the progress of error diagnosis;
[0068] an instruction execution module, configured to execute, in response to a first instruction for the first interface, the step of fitting the correspondence between the first preset joint angle and the first coordinate by using an original DH kinematics error model to obtain a target DH kinematics error model;
[0069] In response to a second instruction for the first interface, the step of fitting the correspondence between the first preset joint angle and the first coordinate through an original full-parameter error model to obtain a target full-parameter error model is executed.
[0070] In some embodiments, the instruction execution module is further configured to:
[0071] In response to a third instruction for the first interface, initializing the robot and the laser follower;
[0072] In response to a fourth instruction for the first interface, the current accuracy of the robot is determined and displayed.
[0073] In a third aspect, an embodiment of the present application further provides an electronic device, including:
[0074] Memory, used to store computer programs;
[0075] The processor is used to implement any of the above-mentioned error diagnosis methods when executing the program stored in the memory.
[0076] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any of the above-mentioned error diagnosis methods is implemented.
[0077] In a fifth aspect, an embodiment of the present application further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the error diagnosis methods described above.
[0078] Beneficial effects of the embodiments of the present application:
[0079] In the technical solution provided by the embodiment of the present application, the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles are obtained, and are respectively used as the first coordinates corresponding to each first preset joint angle; the original full parameter error model is fitted according to the obtained multiple first preset joint angles and the first coordinates to obtain the target full parameter error model, and the model parameters of the target full parameter model are used as the first error of the robot and displayed, so as to facilitate the user to adjust the robot based on the first error to improve the absolute positioning accuracy of the robot. Among them, since the original full parameter error model is a model with full parameter error as the model parameter, robot joint angle as the independent variable, and the coordinates of the calibration component at the end of the robot as the dependent variable, and the full parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between each two adjacent joints, which are not considered when constructing the DH kinematic error model, therefore, when the DH kinematic error model cannot make the accuracy of the robot converge to the effective threshold, the technical solution provided by the embodiment of the present application can effectively improve the accuracy of the robot.
[0080] Of course, implementing any product or method of the present application does not necessarily require achieving all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0081] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other embodiments can also be obtained based on these drawings.
[0082] Figure 1 A schematic diagram of a robot provided in an embodiment of the present application;
[0083] Figure 2 A first flow chart of the error diagnosis method provided in the embodiment of the present application;
[0084] Figure 3 A schematic diagram of the definition of the coordinate systems of each joint of the DH kinematic error model provided in an embodiment of the present application;
[0085] Figure 4 A schematic diagram of error transmission between adjacent joints of a robot provided in an embodiment of the present application;
[0086] Figure 5 A schematic diagram of the definition of each joint coordinate system of the full parameter error model provided in an embodiment of the present application;
[0087] Figure 6 A second flow chart of the error diagnosis method provided in the embodiment of the present application;
[0088] Figure 7 A schematic diagram of a process for obtaining a first coordinate provided in an embodiment of the present application;
[0089] Figure 8 Another schematic diagram of a process for obtaining a first coordinate provided in an embodiment of the present application;
[0090] Fig. 9 A schematic diagram of determining the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system, provided in an embodiment of the present application;
[0091] Fig.10 Another schematic diagram of determining the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system, provided in an embodiment of the present application;
[0092] Fig.11 A first schematic diagram of a first interface provided in an embodiment of the present application;
[0093] Fig.12 A second schematic diagram of the first interface provided in an embodiment of the present application;
[0094] Fig.13 A third schematic diagram of the first interface provided in an embodiment of the present application;
[0095] Fig.14 A schematic diagram of an error diagnosis system provided in an embodiment of the present application;
[0096] Fig.15 A schematic diagram of the structure of an error diagnosis device provided in an embodiment of the present application;
[0097] Fig.16 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0098] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field based on the present application belong to the scope of protection of the present application.
[0099] In order to improve the absolute positioning accuracy of robots and expand the application scenarios of robots in high-precision fields, robots need to be calibrated after production before they can be put into use. Robot calibration refers to the establishment of a kinematic model that matches the robot by identifying the actual structural parameters of the robot. Robot calibration mainly includes three steps: model establishment, error measurement, and parameter identification. Model establishment can be understood as the construction of a kinematic model that reflects the mapping relationship between the robot's joint angles (i.e., joint space coordinates) and the coordinates of the calibration components at the end of the robot (i.e., the Cartesian space coordinates of the end).
[0100] In the related art, the accuracy of the robot is usually calibrated by constructing a DH kinematic error model, but the DH kinematic error model is constructed based on the DH parameter method, and only 12 errors such as rod length offset, joint zero position, and joint stiffness are considered. With the increasing requirements for the absolute positioning accuracy of the robot, adjusting the accuracy of the robot based solely on the DH kinematic error model may not achieve the absolute positioning accuracy expected by the user. In this regard, technical personnel in this field have also considered introducing errors other than DH parameters to improve the absolute positioning accuracy of the robot. For example, a generalized kinematic error calibration method for an industrial robot considers position-related errors, pitch errors, harmonic rotation errors, out-of-plane errors, and backlash errors of gears, and constructs a geometric error model based on the DH model of the robot to be identified using differential kinematics, and constructs a joint error model based on the DH model of the robot to be identified using a polynomial fitting method; the generalized kinematic error model is determined based on the joint error model and the geometric error model, and then the generalized kinematic error model is parameter identified by the implicit loop method to obtain the identified generalized kinematic error model. However, since this method uses the Chebyshev function to represent the joint error model, the identification parameters are the Chebyshev coefficients, and the identification results have no clear physical meaning, the identification results obtained according to this method cannot clearly identify the specific cause of the error, and it is difficult to use it as a basis for error diagnosis to adjust and repair the robot.
[0101] In addition, the DH kinematic error model is based on the theoretical assumption that the robot axis is "straight", that is, when using the DH kinematic error model to calibrate the robot's parameters, the robot's joint axes are assumed to be horizontal or vertical relative to the base, and possible parallelism or verticality errors are not considered. In the actual production process, due to the existence of assembly errors, the robot's joint axes may not necessarily be horizontal or vertical relative to the base. Therefore, when such assembly errors exist, it is difficult to converge the robot's accuracy to an effective threshold when using the DH kinematic error model to calibrate the robot's parameters.
[0102] For example, Figure 1 As shown, Figure 1 A schematic diagram of a robot provided in an embodiment of the present application. In theory, Figure 1 Axis 2 is parallel to axis 3, that is, if dotted line 1 represents axis 2, then theoretically, after assembly, axis 3 should be parallel to axis 2, that is, if represented by lines, axis 3 should be as shown by dotted line 2, and dotted line 2 is parallel to dotted line 1. However, due to assembly errors, axis 3 is actually as shown by dotted line 3, and there is an angle θ between dotted line 3 and dotted line 2. That is, after assembly, there is an angle θ between axis 2 and axis 3, and they are not parallel. At this time, if the DH kinematic error model is used for Figure 1 It is difficult to calibrate the parameters of the robot shown in the figure so that the accuracy of the robot converges to an effective threshold.
[0103] To solve the above problems, the present application provides an error diagnosis method, which fully considers the bias error, deflection error, joint angle error and other errors of each joint of the robot, and constructs a full-parameter error model. Based on the full-parameter error model, it can effectively decouple various kinematic errors, realize rapid diagnosis of calibration problems, and accurately analyze the source of errors. At the same time, the error diagnosis method provided by the embodiment of the present application can also analyze the size of each error term determined, and evaluate the influence of each error term on the absolute positioning accuracy of the robot, realize accurate analysis of calibration errors, and provide a useful reference for robot fault repair. The technical solution provided by the embodiment of the present application can provide targeted problem troubleshooting and design optimization means for robot structural design, production and installation, and improve the quality and efficiency of design and production.
[0104] The error diagnosis method provided in the embodiment of the present application is applied to a host computer, which may be a desktop computer, a portable computer, an intelligent mobile terminal, etc. The error diagnosis method provided in the embodiment of the present application is described below.
[0105] See also Figure 2 , Figure 2 The first flow chart of the error diagnosis method provided in the embodiment of the present application includes the following steps:
[0106] Step S21: acquiring the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles, and using them as the first coordinates corresponding to each first preset joint angle;
[0107] Step S22: Fit the correspondence between the first preset joint angle and the first coordinate through the original full-parameter error model to obtain a target full-parameter error model, and use the model parameters in the target full-parameter error model as the first error of the robot and display the first error.
[0108] In the technical solution provided by the embodiment of the present application, the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles are obtained, and are respectively used as the first coordinates corresponding to each first preset joint angle; the original full parameter error model is fitted according to the obtained multiple first preset joint angles and the first coordinates to obtain the target full parameter error model, and the model parameters of the target full parameter model are used as the first error of the robot and displayed, so as to facilitate the user to adjust the robot based on the first error to improve the absolute positioning accuracy of the robot. Among them, since the original full parameter error model is a model with full parameter error as the model parameter, robot joint angle as the independent variable, and the coordinates of the calibration component at the end of the robot as the dependent variable, and the full parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between each two adjacent joints, which are not considered when constructing the DH kinematic error model, therefore, when the DH kinematic error model cannot make the accuracy of the robot converge to the effective threshold, the technical solution provided by the embodiment of the present application can effectively improve the accuracy of the robot.
[0109] In the above step S21, the joint angle is a parameter used to describe the motion state of the robot. The user can control the robot to move to the motion state described by the joint angle by inputting a specified joint angle; the first coordinate is the coordinate of the robot end calibration component in the Cartesian space coordinate system when the robot moves to the first preset joint angle; the calibration component fixed to the robot end can be a target sphere. When the calibration component is a target sphere, the first coordinate can be the coordinate of the target sphere captured by a laser follower in the laser follower coordinate system.
[0110] It can be understood that the purpose of setting the first preset joint angle is to control the movement of the robot end to different motion states to obtain multiple sets of first preset joint angles and first coordinates, and then, based on these multiple sets of first preset joint angles and first coordinates, realize the fitting of the target full-parameter error model (see the description in step S22 below), and then determine the first error of the robot. Therefore, in order to ensure the accuracy and completeness of the determined first error as much as possible, more first preset joint angles and first coordinates should be used, and the adopted first preset joint angles and first coordinates should be universal. Therefore, when configuring the first preset joint angles, more first preset joint angles can be set, such as the number of first preset joint angles can be set to 50 groups, 60 groups, 70 groups, etc., and the first preset joint angles should cover the entire workspace of the robot as much as possible, that is, all joint angles to which the robot may move.
[0111] In an embodiment of the present application, the method of fitting the target full-parameter error model based on multiple sets of first preset joint angles and first coordinates can be: based on multiple sets of first preset joint angles and first coordinates, multiple nonlinear equations containing full-parameter errors are established, and the optimal solution is calculated by the least squares error method. That is, the Gauss-Newton method is used to solve the nonlinear least squares problem to identify the full-parameter error of the robot. For example, the constructed full-parameter error model can be expressed in the form of a formula as F=f(q,x1,x2…,xn), where F represents the first coordinate, q represents the first preset joint angle, x1,x2…,xn represent various full-parameter errors, and the multiple nonlinear equations containing full-parameter errors established can be F1=f(q1,x1,x2…,xn); F2=f(q2,x1,x2…,xn);…; F m =f(q m ,x1,x2…,xn), where m is the number of the first preset joint angles configured; q1,q2,…,q m Indicates the first preset joint angle of each group; F1, F2, ..., F m Indicates the first coordinates corresponding to each set of first preset joint angles.
[0112] In some embodiments, the robot is composed of a robot body and a controller. The host computer can establish communication with the laser follower and the robot controller, thereby controlling the laser follower to adjust the direction to capture the target ball, and controlling the controller to make the robot move to different first preset joint angles. The specific method of establishing communication can be found in the following Fig.11 and Fig.12 It is understandable that the robot mentioned in this application may include components such as a driver and a sensor in addition to the robot body and the controller. This embodiment is only described by taking the robot consisting of the robot body and the controller as an example, which does not serve as a limitation.
[0113] In some embodiments, the robot in the embodiments of the present application can be understood as a six-degree-of-freedom serial robot that meets the Pieper criterion. According to the Pieper criterion, when the configuration of a robot satisfies "the axes of three consecutive revolute joints intersect at the same point", it indicates that the robot has a closed form kinematic inverse solution. For example, Figure 3 As shown, Figure 3 The robot shown is a six-degree-of-freedom serial robot that complies with the Pieper criterion. It can be seen that Figure 3 The robot shown has six links, namely L1-L6; and six joints, among which the axes of joints 4, 5, and 6 intersect at one point; Figure 3 The coordinate systems shown in are used to represent the positions of the joints of the robot.
[0114] In the above step S22, the full parameter error model can be a model with full parameter error as a model parameter, robot joint angle as an independent variable, and the coordinates of the calibration component at the end of the robot as a dependent variable. The full parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between every two adjacent joints of the robot.
[0115] The original full parameter error model is a full parameter error model in which the full parameter error is an unknown quantity; the original full parameter error model is fitted based on the first preset joint angles and the first coordinates obtained in the above step S21, and the corresponding relationship between the robot joint angles and the coordinates where the robot terminal calibration component is located can be obtained. At this time, the full parameter error in the fitted corresponding relationship is a known quantity, and the corresponding relationship between the robot joint angles and the coordinates where the robot terminal calibration component is located can be regarded as the target full parameter error model. Therefore, the model parameters (i.e., the full parameter error) in the target full parameter model can be used as the first error of the robot and displayed.
[0116] The following describes how to construct a full parameter error model and how to display the first error.
[0117] It can be understood that the full parameter error model is realized by constructing the conversion relationship of the coordinate systems between adjacent joints, that is, a conversion relationship of the coordinate systems between adjacent joints of the robot when the full parameter error is introduced is pre-constructed. For a robot, when the coordinates of a certain joint are determined, the coordinates of its adjacent joints are determined based on the pre-constructed conversion relationship of the coordinate systems between adjacent joints. Then, the coordinates of the adjacent joints of the adjacent joints are determined based on the determined coordinates of the adjacent joints and the pre-constructed conversion relationship of the coordinate systems between adjacent joints. In this way, the coordinates of each joint of the robot can be obtained by recursion, and then the correspondence between the joint angles of the robot and the coordinates of the calibration components at the end of the machine can be obtained. It is just that the full parameter error in this correspondence is an unknown quantity. At this time, this correspondence can be regarded as a full parameter error model.
[0118] Below, the construction method of the conversion relationship between the coordinate systems of adjacent joints of the robot under the condition of introducing the full parameter error is explained. It can be understood that the full parameter error includes bias error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error. In the process of obtaining the coordinates of the next joint (i.e., the adjacent joint of the previous joint) according to the coordinate reasoning of the previous joint, the order of introducing the bias error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error is not limited. The following content only takes the introduction of the bias error first, the deflection error, the joint angle error, the joint flexibility error, and the connecting rod flexibility error as an example to explain the construction method of the full parameter error model:
[0119] Without considering the error and only considering the theoretical link length between adjacent joints, the relationship between the coordinates of the current joint (i.e., the adjacent joint of the previous joint) and the coordinates of the previous joint can be constructed as shown in the following formula (1):
[0120] p i =p i-1 +R i-1 l i-1 ,i (1)
[0121] Among them, p i Indicates the coordinates of the current joint; p i-1 Represents the coordinates of the previous joint; R i-1 is the rotation matrix; l i-1,i represents the position offset. i and R i-1 It can be obtained based on the DH parameters and the coordinate transformation principle of space rigid body.
[0122] Introduce the offset error x of the current joint relative to the previous joint in the x, y, and z directions i ,y i 、z i After that, the relationship between the coordinates of the current joint and the coordinates of the previous joint can be constructed as shown in the following formula (2):
[0123]
[0124] Among them, p i Indicates the coordinates of the current joint; p i,x Indicates the direction of the x-axis of the current joint after the offset error along the x-direction is introduced; p i,y Indicates the y-axis direction of the current joint after the offset error along the y direction is introduced; p i,z Indicates the direction of the z-axis of the current joint after the offset error along the z-direction is introduced; x i ,y i 、z i They represent the offset error of the current joint relative to the previous joint in the x, y, and z directions respectively.
[0125] When the deflection error is introduced, the relationship between the coordinates of the current joint and the coordinates of the previous joint can be constructed as shown in the following formula (3):
[0126]
[0127] in,
[0128]
[0129] Among them, rolli Indicates the deflection error of the current joint relative to the previous joint around the x direction; pitch i Indicates the deflection error of the current joint relative to the previous joint in the y direction; yaw i Represents the deflection error of the current joint relative to the previous joint around the z direction; R i,roll , R i,pitch , R i,yaw Represents the rotation matrix after the deflection error is introduced.
[0130] When the joint angle error is introduced, the relationship between the coordinates of the current joint and the coordinates of the previous joint can be constructed as shown in the following formula (4):
[0131]
[0132] Among them, R i,jnt Represents the coordinates of the current joint after the joint rotation is introduced; θ i represents the joint rotation angle of the current joint relative to the previous joint; w i Represents the axis unit vector of the current joint.
[0133] When the joint flexibility error (i.e., the rotation deviation caused by the joint flexibility) is introduced, the relationship between the coordinates of the current joint and the coordinates of the previous joint can be constructed as shown in the following formula (5):
[0134]
[0135] Among them, c i,jnt represents the joint flexibility coefficient, c i,jnt =c i,jnt ω i ·τ i ; τ i Represents the force matrix of the current joint; R i,comp,jnt Indicates the coordinates of the current joint after the rotation deviation caused by joint flexibility is introduced.
[0136] When the link flexibility error (i.e., the rotation deviation caused by the link flexibility) is introduced, the relationship between the coordinates of the current joint and the coordinates of the previous joint can be constructed as shown in the following formula (6):
[0137]
[0138] in,
[0139]
[0140] Among them, c i,x 、c i,y 、c i,z Represents the connecting rod flexibility coefficient.
[0141] It can be understood that based on the above formulas (1) to (6), the transformation relationship between the coordinates of each joint of the robot can be fitted, and then the full parameter error model (T all-error ). For example, Figure 4 As shown, T i-1 and T i is the coordinate system of the adjacent joints of the robot, T i-1 is the coordinate system of the previous joint, T i is the coordinate system of the next joint, T i,2 , ..., T i,6 Respectively represent T after introducing corresponding errors i changes in Figure 4 (a) indicates that at T i-1 Based on the error T i (Right now Figure 4 T i,1 ) and T i-1 The conversion relationship between T i-1 With T i The length of the connecting rod is l i Schematic diagram of the original coordinate transformation as an example. Figure 4 (b) shows the introduction of bias error (i.e., introduction of displacement bias) Δl i In the case of T i (Right now Figure 4 T i,2 ) and T i-1 An illustration of the conversion relationship between them. Figure 4 (c) indicates the introduction of bias error Δl i With deflection errors ΔRx, ΔRy, ΔRz, T i (Right now Figure 4 T i,3 ) and T i-1 The conversion relationship between them is shown in figure 2. Figure 4 (d) indicates the introduction of joint angle error (such as R θi ), T i (Right now Figure 4 T i,4 ) and T i-1 The conversion relationship between them is shown in figure 2. Figure 4 (e) indicates that the moment is Torq x When the rotation deviation caused by joint flexibility (i.e., joint flexibility error) is introduced, T i (Right now Figure 4 T i,5 ) and T i-1 The conversion relationship between them is shown in figure 2. Figure 4(f) indicates the case where the rotation deviation caused by the connecting rod flexibility (i.e., connecting rod flexibility error) is introduced. i (Right now Figure 4 T i,6 ) and T i-1 Schematic diagram of the conversion relationship between, where the rotation deviation θ x The formula θ x =K x Torq x Calculated, K x is the flexibility coefficient; Torq x is the torque.
[0142] The definition of each joint coordinate system of the full parameter error model is different from the definition of each joint coordinate system of the DH kinematic error model. In some embodiments, each joint coordinate system of the full parameter error model can be defined as Figure 5 The form shown, Figure 5 The coordinate systems shown in are used to indicate the position of each joint of the robot. For example, the coordinate system shown by x1, y1, and z1 is used to indicate the position of joint 1 of the robot.
[0143] by Figure 5 Taking the definition of the joint coordinate system shown in the figure as an example, when the full parameter error model is used to diagnose the error of the robot, in addition to the error types that can be identified by the DH kinematic error model (that is, the contents shown in Table 2 below), the error types that can be identified are shown in Table 1 below:
[0144] Table 1. Error types additionally identified by the full parameter error model relative to the DH kinematic error model
[0145]
[0146] In the above Table 1, the first column represents the mathematical description of the full parameter error identified by the full parameter error model, and the second column represents the description of the physical meaning of the full parameter error identified by the full parameter error model. From the content of the first row in Table 1, it can be seen that the mathematical description of the full parameter error is "the rotation angle of axis 2 relative to axis 1 around the x direction", and its physical meaning is "there is a perpendicularity error between axis 2 and axis 1"; from the content of the second row in Table 1, it can be seen that the mathematical description is "the rotation angle of axis 3 relative to axis 2 around the x direction", and its physical meaning is "there is a parallelism error between axis 3 and axis 2". The meanings of other rows in Table 1 are the same, and will not be repeated here.
[0147] It can be understood that since the full parameter error model is constructed based on full parameter errors such as bias error and joint angle error, after obtaining the model parameters of the target full parameter model, the model parameters should also indicate the existing bias error, angle error, etc., which is the mathematical description in Table 1. However, it can be seen from the mathematical description in Table 1 that the mathematical description cannot intuitively represent the specific error, especially when the user is not clear about the x, y, and z axis directions of the coordinate system established by each axis. There is no guiding hint on how to adjust the robot to eliminate the error. Therefore, for each identifiable full parameter error, the corresponding physical meaning is given. Based on the physical meaning, the user can be intuitively informed of the specific error so that the user can eliminate the corresponding error.
[0148] Next, the display method of the first error is described.
[0149] In some embodiments, for each first error, the host computer can determine the highest accuracy that the robot can achieve by adjusting only the first error, as the calibration residual corresponding to each first error. For example, there are two first errors, namely error 1 and error 2. For error 1, the host computer can determine the highest accuracy that the robot can achieve by adjusting only error 1 without adjusting error 2, as calibration residual 1 corresponding to error 1; for error 2, the host computer can determine the highest accuracy that the robot can achieve by adjusting only error 2 without adjusting error 1, as calibration residual 2 corresponding to error 2.
[0150] It is understandable that the accuracy of a robot generally refers to the accuracy with which the robot performs tasks. If expressed in numerical values, the smaller the numerical value of the accuracy, the higher the accuracy of the robot. For example, the accuracy of a robot that can reach an accuracy of 0.1 mm is higher than the accuracy of a robot that can only reach an accuracy of 0.2 mm. It can be seen that in the embodiment of the present application, the smaller the numerical value of the calibration residual, the higher the accuracy of the robot. Therefore, the calibration residual calculated in the embodiment of the present application is the highest accuracy that can be achieved when only a certain first error is adjusted, so as to distinguish the influence of each first error on the accuracy of the robot, and can intuitively inform the user of the accuracy that the robot can achieve after adjusting the corresponding first error.
[0151] For example, for Error 1 and Error 2 in the above example, the accuracy of the robot can reach 0.1mm by adjusting Error 1 alone, and the accuracy of the robot can reach 0.2mm by adjusting Error 2 alone. That is, compared with Error 2, Error 1 has a greater impact on the accuracy of the robot. In this regard, if the calibration residual 1 corresponding to Error 1 is recorded as 0.3mm when determining the calibration residual (it can be understood that after adjusting Error 1, the robot accuracy can reach 0.1mm, and it can also reach 0.3mm, if Error 1 is not fully adjusted), and the calibration residual 2 corresponding to Error 2 is recorded as 0.2mm, then the error that has a greater impact on the accuracy of the robot determined from the value of the calibration residual is Error 2. It can be seen that this way of determining the calibration residual may cause the user to incorrectly determine the impact of each first error on the robot. The impact of the robot's accuracy on the robot, and thus complicates the user's calibration process for the robot. For example, if the user expects the robot to achieve an accuracy of 0.1mm, at this time, if the highest achievable accuracy is used as the calibration residual corresponding to the first error, that is, the calibration residual 1 corresponding to error 1 is 0.1mm, and the calibration residual 2 corresponding to error 2 is recorded as 0.2mm, then the user can know that the expected accuracy can be achieved after adjusting error 1 based on the calibration residual corresponding to the error, and can directly adjust error 1; but if the calibration residual 1 is recorded as 0.3mm and the calibration residual 2 is recorded as 0.2mm as mentioned above, the user will get information that only adjusting error 2 cannot achieve the expected accuracy, and it is necessary to adjust error 1 and error 2 at the same time to see if the expected accuracy can be achieved. It can be seen that using the highest achievable accuracy as the calibration residual corresponding to each first error can not only reflect the impact of each error on the robot's accuracy, but also intuitively inform the user of the accuracy that the robot can achieve after adjusting the corresponding first error, which has guiding significance for the user to determine the process of adjusting the first error that needs to be adjusted.
[0152] In addition, in some embodiments, the highest accuracy may not be used as the calibration residual corresponding to the first error. Instead, the calibration residual obtained is only required to be able to distinguish the impact of the first error on the robot's accuracy. For example, in the above example, calibration residual 1 may be recorded as 0.2mm, and calibration residual 2 may be recorded as 0.2mm. In this way, the impact of each first error on the robot's accuracy can still be determined through the calibration residual. However, the calibration residual determined in this way cannot intuitively let the user know the highest accuracy that can be achieved after adjusting each first error. For example, when the user expects the robot to achieve an accuracy of 0.1mm, the numerical value of the calibration error determined based on this method may not be able to intuitively let the user know that the user only needs to adjust error 1 to achieve the expected accuracy.
[0153] After determining the calibration residual corresponding to each first error, the upper computer can determine the first error whose corresponding calibration residual satisfies the preset condition as the third error, that is, for each first error, if there is a calibration residual corresponding to one or more first errors that satisfies the preset condition, then the one or more first errors are used as the third error. Here, the preset condition can be that the value of the calibration residual is less than the preset value, wherein the preset value can be set according to actual needs, such as 0.2mm, 0.5mm, etc., or the preset condition can also be that the calibration residuals corresponding to each first error are arranged in ascending order and are located in the first m positions in the sequence, wherein the value of m can be 1, 2, 3, etc.
[0154] In some embodiments, in the process of determining errors using a full-parameter error model, the serial number corresponding to each error term, the parameter value corresponding to the error term, and the calibration residual corresponding to the error term can be recorded, and these contents can be output in the form of a structure to obtain multiple structures. Afterwards, the multiple structures are arranged in ascending order according to the size of the calibration residual to obtain a set of structure sequences. The meaning of the parameters contained in the n-th structure in the sequence is: the error term that ranks the n-th in terms of impact on calibration accuracy, the parameter value corresponding to the error term, and the calibration residual corresponding to the error term. According to the sorted structure sequence, the primary error term (i.e., the third error) and the parameter value corresponding to the error term can be located, and the maximum accuracy that can be achieved after repair can be quantitatively analyzed, thereby realizing rapid positioning of factors affecting calibration accuracy, error diagnosis, and accuracy analysis. For details, please refer to the following Fig.13 Examples in .
[0155] The first error corresponding to the calibration residual that meets the preset conditions is taken as the third error, and the error other than the third error in each first error is taken as the fourth error. It can be understood that the numerical value of the calibration residual can represent the influence of the first error on the robot accuracy. Therefore, for the preset conditions set, relative to the fourth error, the first error (i.e., the third error) corresponding to the calibration residual that meets the preset conditions has a greater influence on the robot accuracy. Therefore, in order to enable the user to intuitively understand which errors have a greater influence on the robot accuracy and which errors have a smaller influence on the robot accuracy, so as to facilitate the user to determine the error to be adjusted, the third error can be displayed in the first display mode, and the fourth error can be displayed in the second display mode, that is, the third error and the fourth error can be displayed in different display modes. Here, different display modes can be displaying the third error and the fourth error in different colors, displaying the third error and the fourth error in different font sizes, etc.
[0156] In some embodiments, for each error that can be identified by the full-parameter error model, a corresponding error schematic diagram can be pre-configured in the upper computer. After determining the third error, for each third error, the upper computer can obtain the error schematic diagram configured for each third error as a target error schematic diagram. In this way, when displaying each third error, the upper computer can display the target error schematic diagram corresponding to each third error, the third error (such as the mathematical description, physical meaning, etc. of the third error) and the calibration residual corresponding to the third error, so as to more intuitively inform the user of the location and cause of the third error.
[0157] See also Figure 6 , Figure 6 A second flow chart of the error diagnosis method provided in the embodiment of the present application.
[0158] In some embodiments, before determining the first error through the full parameter error model, the DH kinematic error model can also be used to pre-determine the second error of the robot, that is, the robot is first adjusted based on the second error determined by the DH kinematic error model. If the robot accuracy cannot reach the preset threshold after adjustment or the robot accuracy still needs to be further improved, the full parameter error model is used to determine the first error, and the robot is adjusted based on the first error. Figure 6 The method shown may include the following steps:
[0159] Step S61: acquiring coordinates of a calibration component fixed to the end of the robot when the robot is at different first preset joint angles, and using them as first coordinates corresponding to each of the first preset joint angles;
[0160] Step S62: fitting the correspondence between the first preset joint angle and the first coordinate through the original DH kinematic error model to obtain a target DH kinematic error model, and using the model parameters in the target DH kinematic error model as the second error of the robot and displaying the second error;
[0161] Step S63: In response to the diagnostic instruction for the second error, the corresponding relationship between the first preset joint angle and the first coordinate is fitted by the original full-parameter error model to obtain a target full-parameter error model.
[0162] In the technical solution provided in the embodiment of the present application, a DH kinematic error model is constructed, which includes the offset errors of each rod length of the robot, the zero position errors of each joint of the robot, and the stiffness errors of each joint of the robot. After obtaining each first preset joint angle and the first coordinate corresponding to each first preset joint angle, the DH kinematic error model is first used to determine the second error of the robot. When the accuracy of the robot cannot reach the expected accuracy by adjusting the second error, the full parameter error model is used to determine the first error, and then the accuracy of the robot is improved by adjusting the first error. When the accuracy of the robot can reach the expected accuracy by adjusting the second error, the full parameter error model is no longer used to determine the first error. In this way, the amount of calculation can be reduced and the efficiency of calibrating the robot parameters can be improved.
[0163] The above step S61 and Figure 2 The same as step S21 in FIG. 1 and will not be repeated here.
[0164] In the above step S62, the DH kinematic error model can be a model with DH kinematic error as a model parameter, robot joint angle as an independent variable, and the coordinates of the calibration component at the end of the robot as a dependent variable. The DH kinematic error includes the offset error of each rod length of the robot, the zero position error of each joint of the robot, and the stiffness error of each joint of the robot.
[0165] The original DH kinematic error model is a DH kinematic error model in which the DH kinematic error is an unknown quantity; by fitting the original DH kinematic error model based on each first preset joint angle and the first coordinate obtained in the above step S61, the corresponding relationship between the robot joint angle and the coordinate of the robot terminal calibration component can be obtained. At this time, the DH kinematic error in the fitted corresponding relationship is a known quantity, and the corresponding relationship between the robot joint angle and the coordinate of the robot terminal calibration component can be regarded as the target DH kinematic error model. Therefore, the model parameters in the target DH kinematic error model (i.e., the DH kinematic error) can be used as the second error of the robot and displayed. Among them, each joint coordinate system of the DH kinematic error model can be defined as the above Figure 3 The form described.
[0166] by Figure 3 Taking the definition of the joint coordinate system shown in the figure as an example, when the DH kinematic error model is used to diagnose the robot error, the error types that can be identified are shown in Table 2 below:
[0167] Table 2. Error types for DH kinematic error model identification
[0168]
[0169] In Table 2 above, the first column indicates the error types that can be identified using the DH kinematic error model, and the second column indicates the error types that can be identified using the DH kinematic error model. Figure 3 The robot shown in the figure uses the DH kinematic error model to specifically determine the error. From the content of the first row in Table 2, it can be seen that when the error type is rod length offset, the error term that can be determined is the offset error of the robot links L2, L3, L4, and L6; from the content of the second row in Table 2, it can be seen that when the error type is joint zero position, the error terms that can be determined are θ2, θ3, θ4, and θ5; from the content of the third row in Table 2, it can be seen that when the error type is joint stiffness, the error terms that can be determined are k2, k3, k4, and k5. It can be understood that although the error terms in Table 2 above are divided into multiple error types, the DH kinematic error model is used to simultaneously identify the error terms in Table 2 above, rather than only identifying the error terms of a certain error type.
[0170] In the above step S63, after adjusting the accuracy of the robot based on the second error, if the user still needs to use the full parameter error model to determine the first error, the host computer can execute the content shown in the above step S22 in response to the diagnostic instruction for the second error. For the relevant content of the diagnostic instruction, please refer to the following Fig.13 Description in .
[0171] See also Figure 7 , Figure 7 A schematic diagram of a process for obtaining a first coordinate provided in an embodiment of the present application.
[0172] In some embodiments, considering factors such as measurement accuracy, measurement range, stability, etc., the calibration component fixed to the end of the robot can be a target ball, and a laser follower is used to obtain the first coordinate. Figure 7 The method for obtaining the first coordinates may include the following steps:
[0173] Step S71: for each first preset joint angle, controlling the robot to move to the first preset joint angle with a target ball fixed at the end, and controlling the laser follower to capture the target ball;
[0174] Step S72: if the coordinates returned by the laser follower are not received within the preset time, determining the conversion relationship between the robot flange coordinate system and the base coordinate system according to the first preset joint angle;
[0175] Step S73: according to the conversion relationship between the base coordinate system and the laser follower coordinate system, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the flange coordinate system and the base coordinate system, the coordinates of the target sphere in the laser follower coordinate system are determined as the second coordinates;
[0176] Step S74: adjusting the orientation of the laser follower so that the field of view of the laser follower covers the second coordinate;
[0177] Step S75: Acquire the coordinates obtained by the laser follower capturing the target ball as the first coordinates corresponding to the first preset joint angle.
[0178] In the technical solution provided by the embodiment of the present application, if the coordinates of the target ball returned by the laser follower are not received within the preset time, the host computer can determine the conversion relationship between the current flange coordinate system of the robot and the base coordinate system according to the current first preset joint angle; then, according to the conversion relationship between the tool coordinate system and the flange coordinate system, the conversion relationship between the flange coordinate system and the base coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system, the coordinates of the target ball in the laser follower coordinate system (i.e., the second coordinate) can be calculated; then, the host computer adjusts the direction of the laser follower so that the field of view of the laser follower covers the second coordinate, so that the laser follower can capture the target ball and obtain the first coordinate. In this way, even if there is a large difference between adjacent first preset joint angles causing the laser follower to lose light, the host computer can adjust the direction of the laser follower in time to recapture the target ball.
[0179] In the above step S71, a target ball is fixed to the end of the robot. For each first preset joint angle, after the upper computer controls the robot to move to the first preset joint angle, it can control the laser follower to capture the target ball at the end of the robot; or, in the process of the robot moving to each first preset joint angle, the laser follower automatically follows the movement of the target ball, and when the robot reaches the first preset joint angle, the coordinates of the target ball in the laser follower coordinate system are captured.
[0180] In the above step S72, if the host computer does not receive the coordinates returned by the laser follower within the preset time, the conversion relationship between the robot flange coordinate system and the base coordinate system can be determined according to the current first preset joint angle. The determined conversion relationship between the robot flange coordinate system and the base coordinate system is directly determined according to the current first preset joint angle without considering the error.
[0181] Among them, the specific value of the preset time can be set according to the actual situation. The host computer can start calculating whether the coordinates returned by the laser follower are received within the preset time when the host computer controls the robot to adjust to the first preset joint angle; or, after the host computer controls the robot to adjust to the first preset joint angle, it controls the laser follower to capture the target ball, and starts calculating whether the coordinates returned by the laser follower are received within the preset time from controlling the laser follower to capture the target ball; or, after the host computer controls the robot to adjust to the first preset joint angle, it controls the laser follower to capture the target ball through a control instruction, and receives the information returned by the laser follower to indicate that the control instruction has been received, and starts calculating whether the coordinates returned by the laser follower are received within the preset time from receiving the information returned by the laser follower. In addition, the node at which the host computer starts calculating whether it is within the preset time can also be other time nodes, which is determined according to user needs and is not limited to this.
[0182] In the above step S73, the host computer can determine the coordinates of the target ball in the laser follower coordinate system, that is, the second coordinate, based on the conversion relationship between the tool coordinate system and the flange coordinate system, the conversion relationship between the flange coordinate system and the base coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system. The second coordinate obtained here is actually the theoretical position of the current target ball in the laser follower coordinate system inferred based on the current first preset joint angle and the conversion relationship between each coordinate system. Among them, the conversion relationship between the base coordinate system and the laser follower coordinate system, and the conversion relationship between the tool coordinate system and the flange coordinate system can be pre-determined by the following Fig. 9 The method shown is determined and will not be repeated here.
[0183] In the above step S74, after the host computer obtains the second coordinate, it can adjust the orientation of the laser follower according to the second coordinate so that the field of view of the laser follower covers the second coordinate. For example, the host computer can send an orientation adjustment instruction to the laser follower to control the laser follower to adjust the orientation, so that the field of view of the laser follower covers the second coordinate. It can be understood that the second coordinate is the theoretical position of the current target ball in the laser follower coordinate system obtained by inference. Therefore, the second coordinate is not necessarily the coordinate of the target ball in the laser follower coordinate system under the first preset joint angle, but the second coordinate can be approximately regarded as the coordinate of the target ball in the laser follower coordinate system. For the laser follower, even if there is a millimeter-level deviation, autonomous capture can be achieved. Therefore, after the field of view of the laser follower covers the second coordinate, the target ball can be captured, and the real coordinate of the target ball in the laser follower coordinate system is obtained as the first coordinate corresponding to the current first preset joint angle.
[0184] In the above step S75, after the laser follower captures the target ball, the coordinates of the target ball in the laser follower coordinate system (ie, the first coordinate) can be obtained, and the first coordinate is returned to the host computer as the first coordinate corresponding to the current first preset joint angle.
[0185] Figure 8 Another schematic diagram of a process for obtaining a first coordinate provided in an embodiment of the present application is as follows: Figure 8 Taking the example of the laser follower in the present application as an example, the process of re-capturing the target ball and obtaining the first coordinate after the laser follower loses light is described. Figure 8 As shown, the host computer controls the robot's motion state transformation through the controller (i.e., ①). When the robot transforms from the previous first preset joint angle to the current first preset joint angle, the laser follower loses light, that is, the host computer detects that the coordinates returned by the laser follower are not received within the preset time. At this time, the host computer can determine the conversion relationship between the current robot flange coordinate system and the base coordinate system according to the current first preset joint angle, and then calculate the second coordinate (also called the preset point) (i.e., ②) according to the conversion relationship between the current robot flange coordinate system and the base coordinate system, and the predetermined conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system. After that, the host computer can send a direction adjustment instruction to the laser follower based on the second coordinate, so that the second coordinate can be located within the field of view of the laser follower after the laser follower adjusts its direction (i.e., ③), and then the laser follower captures the target ball and obtains the coordinate of the current target ball in the laser follower coordinate system (i.e., the first coordinate).
[0186] In some embodiments, the following Fig. 9 The method shown in the figure predetermines the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system. Fig. 9 The method shown may include the following steps:
[0187] Step S91: obtaining the coordinates of the target ball fixed to the end of the robot in the laser follower coordinate system when the robot is at different second preset joint angles, respectively as the third coordinates corresponding to each second preset joint angle; and, for each second preset joint angle, obtaining the conversion relationship between the flange coordinate system and the base coordinate system, as the conversion relationship between the flange coordinate system and the base coordinate system corresponding to each second preset joint angle;
[0188] Step S92: According to the transformation relationship between the flange coordinate system corresponding to each second preset joint angle and the base coordinate system, and the third coordinate corresponding to each second preset joint angle, the transformation relationship between the tool coordinate system and the flange coordinate system, and the transformation relationship between the base coordinate system and the laser follower coordinate system are fitted.
[0189] In the technical solution provided in the embodiment of the present application, multiple second preset joint angles are pre-configured, and for each second preset joint angle, the conversion relationship between the flange coordinate system and the base coordinate system corresponding to each second preset joint angle and the third coordinate are obtained; based on the correspondence between the third coordinate and the second preset joint angle, and the conversion relationship between the flange coordinate system and the base coordinate system corresponding to each second preset joint angle, the upper computer can fit the conversion relationship between the tool coordinate system corresponding to the robot relative to the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system.
[0190] In the above step S91, the second preset joint angle is used to describe the motion state to which the robot terminal needs to move. The host computer can control the robot to move to the corresponding motion state through each second preset joint angle, and obtain the third coordinate corresponding to each second preset joint angle respectively. Among them, the third coordinate has the same meaning as the first coordinate, and the details can be found in the description of the above step S21. In addition, for each second preset joint angle, the host computer can solve the conversion relationship between the flange coordinate system and the base coordinate system corresponding to each second preset joint angle.
[0191] It can be understood that the purpose of determining the conversion relationship between the flange coordinate system and the base coordinate system, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system is to determine the second coordinate corresponding to the current joint angle, that is, to determine the theoretical position of the target ball at the end of the robot in the laser follower coordinate system at the current joint angle, so that the target ball can be automatically captured when the field of view of the laser follower covers the second coordinate. Therefore, when determining the conversion relationship between the flange coordinate system and the base coordinate system here, the existing errors can be ignored, and the corresponding conversion relationship between the flange coordinate system and the base coordinate system can be directly obtained from the joint angle.
[0192] In addition, the purpose of setting the second preset joint angle is to obtain multiple sets of second preset joint angles and third coordinates by controlling the end of the robot to move to different motion states, and then, based on these multiple sets of second preset joint angles and third coordinates, the conversion relationship between the tool coordinate system and the laser follower coordinate system is determined. Therefore, when configuring the second preset joint angle, it should be considered that the movement amplitude between adjacent second preset joint angles should be small, so as to avoid the loss of light in the process of obtaining the third coordinate as much as possible. Among them, the number of second preset joint angles can be set to 10 groups, 20 groups, 30 groups, etc., which can be set according to the actual needs of the user.
[0193] In the above step S92, it can be understood that if q represents the second preset joint angle and base T flange Indicates the conversion relationship between the flange and the base coordinate system. flange Ttool Indicates the relative relationship between the tool coordinate system and the flange coordinate system. laser T base The conversion relationship between the base coordinate system and the laser follower coordinate system is expressed as F, and the third coordinate is expressed as F = g(q, flange T tool , laser T base ). At the same time, due to the formula F = g(q, flange T tool , laser T base ) where q is used to represent the conversion relationship between the flange and the base coordinate system base T flange Therefore, the formula F=g(q, flange T tool , laser T base ) can also be rewritten as base T flange , flange T tool , laser T base The formula with variables as F=h( base T flange , flange T tool , laser T base ).
[0194] Therefore, in the embodiment of the present application, based on the conversion relationship between the flange coordinate system corresponding to each second preset joint angle and the base coordinate system, and the third coordinate corresponding to each second preset joint angle, the host computer can fit the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system.
[0195] Take an example to illustrate, for example, 10 sets of second preset joint angles are preset, and these 10 sets of second preset joint angles are q1, q2, ..., q 10 ; The conversion relationships of the flange coordinate system corresponding to the 10 sets of second preset joint angles relative to the base coordinate system are respectively base T flange1 , base T flange2 , …, base T flange10 ; The third coordinates corresponding to the 10 sets of second preset joint angles are F1, F2, ..., F 10 ;in, base T flange1It indicates the correspondence between the flange coordinate system and the base coordinate system corresponding to q1, and F1 indicates the third coordinate corresponding to q1; base T flange2 It indicates the correspondence between the flange coordinate system and the base coordinate system corresponding to q2, F2 indicates the third coordinate corresponding to q2; the meanings of other characters are similar and will not be repeated here.
[0196] Thus, 10 nonlinear equations can be established, namely F1=g(q1, flange T tool , laser T base ); F2=g(q2, flange T tool , laser T base );…;F 10 =g(q 10 , flange T tool , laser T base ), based on these 10 nonlinear equations, the host computer can fit the transformation relationship between the tool coordinate system and the flange coordinate system, as well as the transformation relationship between the base coordinate system and the laser follower coordinate system.
[0197] Fig.10 A schematic diagram of determining the conversion relationship between the tool coordinate system and the laser follower coordinate system provided in an embodiment of the present application, such as Fig.10 As shown, the host computer calculates the conversion relationship between each second preset joint angle, each third coordinate, and each flange coordinate system relative to the base coordinate system. base T flange Fitting to obtain the transformation relationship between the tool coordinate system and the flange coordinate system flange T tool , the conversion relationship between the base coordinate system and the laser follower coordinate system laser T base ; Based on the transformation relationship between the tool coordinate system and the flange coordinate system flange T tool , the conversion relationship between the flange coordinate system corresponding to each second preset joint angle and the base coordinate system base T flange , and the conversion relationship between the base coordinate system and the laser follower coordinate system laser T base The conversion relationship between the tool coordinate system corresponding to each second preset joint angle and the laser follower coordinate system can also be obtained. laser T tool .
[0198] In some embodiments, in order to provide visual information, facilitate users to understand the execution progress, and independently select the steps to be executed, a first interface for displaying the error diagnosis progress is pre-configured. In response to the first instruction for the first interface, the host computer executes the step of fitting the correspondence between the first preset joint angle and the first coordinate through the DH kinematic error model to obtain the DH kinematic error model. In response to the second instruction for the first interface, the host computer executes the step of fitting the correspondence between the first preset joint angle and the first coordinate through the original full parameter error model to obtain the target full parameter error model. Among them, the second instruction can also be understood as the diagnostic instruction in the above step S63. In response to the third instruction for the first interface, the host computer initializes the robot and the laser follower, such as realizing the communication between the host computer and the robot controller and the laser follower, the zero alignment of each axis structure, the load installation and load information configuration, etc. In response to the fourth instruction for the first interface, the host computer can execute the pre-configured steps for detecting the performance of the robot to detect the performance of the robot to determine the current performance of the robot and display it. Among them, the pre-configured steps for detecting the performance of the robot can include point accuracy test and straightness test, etc.
[0199] The first interface is described below with a specific example. Fig.11 , Fig.11 A first schematic diagram of a first interface provided in an embodiment of the present application. Fig.11 The first interface shown includes 7 items, namely: serial number, enable, step, content, status, progress, and test; wherein, the serial number indicates the serial number of each step in the first interface; enable indicates the execution status of each step in the first interface. In this example, when the execution status of the step is untested, the enable icon corresponding to the step is light-colored; when the status of the step is tested passed, the enable icon corresponding to the step is dark-colored; the step indicates each step to be executed in the first interface; the content indicates the specific content of each step in the first interface; the status indicates the execution status of each step in the first interface; and the progress indicates the execution progress of each step in the first interface. It can be understood that the representation of the serial number, enable, status, etc. provided in this example is only an example and does not serve any limiting role.
[0200] Fig.11 The multiple steps in the above example can be divided into three stages: the first stage is initialization; the second stage is calibration and diagnosis; and the third stage is performance testing. The first stage corresponds to the above example, in response to the third instruction for the first interface, initializing the robot and the laser follower. Fig.11In the example shown, the third instruction for the first interface can be a click operation on the "test" button after the steps corresponding to sequence numbers 1-10, or an operation input by the user to execute all steps in the first interface, such as a "start" button in the first interface, in response to the user clicking the "start" button, all steps in the first interface are executed. The second stage corresponds to the above example, in response to the first instruction and the second instruction for the first interface, obtaining the relevant content of the target DH kinematic error model and the target full parameter error model, Fig.11 In the example shown, the first instruction for the first interface can be understood as a click operation on the "test" button after the steps corresponding to sequence numbers 11-13, or an operation input by the user to execute all steps in the first interface; for a description of the second instruction, see the following Fig.13 The third stage corresponds to the above example, in response to the fourth instruction, determining the current progress of the robot and displaying the corresponding content, Fig.11 In the example shown, the fourth instruction for the first interface may be a click operation on a "test" button after the steps corresponding to sequence numbers 14-18, or an operation input by the user to execute all steps in the first interface.
[0201] Next, combine Fig.12 , and each step included in the above three stages is explained respectively.
[0202] The first stage includes steps corresponding to sequence numbers 1 to 10, which are used to realize the communication between the host computer and the robot controller, the communication between the host computer and the laser follower, the zero position alignment of each axis structure, and the load installation and load information configuration. Specifically, the step "initialize controller" corresponding to sequence number 1 is used to connect the host computer with the robot controller; the step "upload CAPA file" corresponding to sequence number 2 and the step "upload model file" corresponding to sequence number 3 are used to upload basic files to the robot controller; the step "synchronize controller time" corresponding to sequence number 4 is used to synchronize the robot controller time with the host computer time; each axis of the robot is designed with a corresponding zero position groove in the structure, and the step "structural calibration" corresponding to sequence number 5 is used to roughly mark the robot zero position by manually aligning the grooves of each axis; the step "set tool configuration" corresponding to sequence number 6 is used to send instructions to configure the mass and center of mass of the robot's end load into the robot's controller; the step "move to the load installation point" corresponding to sequence number 7 is used The step "Move to the target ball search point" corresponding to sequence number 8 is used to send instructions to make the robot move to the target ball search position to facilitate the operator to install the load; the step "Initialize the laser follower" corresponding to sequence number 9 is used to establish communication between the host computer and the laser follower; the step "Target ball search" corresponding to sequence number 10 (that is, the laser follower automatically searches for the target ball, and tries manual debugging if it fails) is used to enable the laser follower to capture the target ball installed on the end load of the robot; among them, the step corresponding to sequence number 5 also includes one-key return to zero, zero point calibration, factory reset, wait for restart, verify zero point calibration results, set installation method, and advanced option settings (that is, turn off singularity detection).
[0203] The second stage includes steps corresponding to sequence numbers 11 to 13, wherein step "coordinate system calibration" corresponding to sequence number 11 is used for pre-calibration to obtain the conversion relationship between the base coordinate system and the laser follower coordinate system and the conversion relationship between the tool coordinate system and the flange coordinate system. For details, please refer to the above Figure 8 ; The step corresponding to serial number 11 specifically includes the calibration of the tool coordinate system and the world coordinate system, and the calculation of the coordinate system calibration result. The step "kinematic calibration" corresponding to serial number 12 is used to control the robot to adjust to different preset motion states, and then obtain the collected data to realize the subsequent call to the kinematic model and the full parameter model. For example, the step corresponding to serial number 12 is pre-configured with 50 sets of first preset joint angles, and the laser follower can collect the corresponding 50 sets of first coordinates. At the same time, the host computer obtains the corresponding 50 sets of joint angles of the robot, calibrates the robot DH kinematic parameters, the tool coordinate system, and the laser follower coordinate system. For details, please refer to the above Figure 6; The steps corresponding to sequence number 12 specifically include advanced option settings (i.e., turning on joint deformation and vibration suppression), kinematic calibration sampling points, kinematic calibration calculations, writing calibration results to the controller, restoring factory settings, waiting for restart, and setting the installation method, wherein "kinematic calibration calculations" are used to calculate calibration results and calibration residuals. When the calibration residuals meet the expected requirements (such as less than 1mm), the calibration results can be written to the robot's controller, and the parameters just written into the robot controller can be updated by restoring factory settings and restarting. The step "configuration tools and advanced option settings" corresponding to sequence number 13 is used to reconfigure tools and advanced options after the robot controller is restarted. The steps corresponding to sequence number 13 specifically include setting tool configuration, advanced option settings-turning on joint deformation correction and vibration suppression, and advanced option settings-turning off singularity detection. It should be noted that for the step "kinematic calibration calculations" corresponding to sequence number 12.3, regardless of whether calling the DH kinematic error model can make the robot's accuracy meet the expected requirements, the user can call the full parameter error model according to the second instruction for the first interface to determine the first error.
[0204] The third stage includes sequence numbers 14 to 18. The step "cube accuracy verification" corresponding to sequence number 14 includes cube accuracy point sampling and cube accuracy calculation, which is used to test the absolute positioning accuracy of the robot in the local space. For example, the host computer controls the robot to traverse the 8 vertices of the local space cube in a linear motion, and uses a laser follower to collect the coordinates of the target ball in the laser follower coordinate system to verify the absolute positioning accuracy of the robot in the local space; the step "point accuracy verification" corresponding to sequence number 15 includes point accuracy point sampling and point accuracy calculation, which is used to test the absolute positioning accuracy and repeatability of the robot. For example, the host computer can control the robot to traverse 5 national standard test points in a linear motion, and use a laser follower to collect test points to verify the absolute positioning accuracy and repeatability of the robot. Step 16 corresponding to "X-axis linear accuracy verification", step 17 corresponding to "Y-axis linear accuracy verification", and step 18 corresponding to "Z-axis linear accuracy verification" are used to verify the distance accuracy and trajectory accuracy of the robot in the X\Y\Z directions, respectively. For example, the host computer can control the robot to run at full speed in the X, Y, and Z directions in a straight line with a distance of 300 mm in a straight line, and use a laser follower to continuously collect the coordinates of the target ball in the laser follower coordinate system to verify the distance accuracy and trajectory accuracy of the robot in the X\Y\Z directions.
[0205] In addition to the above three stages, Fig.12It also includes serial numbers 19 to 22. The step corresponding to serial number 19 is "Move to the load unloading point"; the step corresponding to serial number 20 is "Dynamic calibration", which is the main step of dynamic calibration, including switching to automatic mode and powering on, dynamic calibration, and downloading controller parameter files; the step corresponding to serial number 21 is "Restore factory settings", including resetting operation and maintenance time, restoring factory settings, and waiting for restart; the step corresponding to serial number 22 is "Advanced option settings", which is to turn off singularity detection.
[0206] In some embodiments, there is a "Start" button in the first interface, and the user can click the "Start" button of the first interface. In response to the user's click operation, the host computer can execute each step in the software interface (i.e., the first interface) in sequence. It can be understood that when the DH kinematic error model and the full parameter error model are configured at the same time, in the process of executing each step in the software interface in sequence, the host computer can first call the DH kinematic model to determine the second error. When the DH kinematic error model cannot achieve the convergence of the robot's accuracy to the effective threshold, a reminder can be sent to the user in the form of a pop-up window, etc., to instruct the user to proceed to the next step, such as calling the full parameter error model to determine the first error.
[0207] In other embodiments, the user can also click the test button corresponding to each step. In response to the user's click operation, the host computer can execute the steps specified by the user. For example, if the user only wants to test the accuracy of the robot, the user can click Fig.11 The test button after the steps corresponding to sequence numbers 14 to 18 in the figure can cause the host computer to execute the corresponding steps in response to the user's click operation. It can be understood that the host computer can only communicate with the robot and the laser follower after establishing corresponding connections with them, and then interact with commands and data. Therefore, when manually selecting the steps to be executed, the corresponding initialization steps must also be performed in advance.
[0208] Next, combine Fig.13 The first interface shown illustrates the second instruction in the above example (equivalent to the diagnosis instruction in the above step S63).
[0209] See also Fig.13 , Fig.13 A schematic diagram of an interface for calling a full parameter model provided in an embodiment of the present application, where a user clicks Fig.13 After the step "kinematic calibration" corresponding to the sequence number 12 in the first interface shown (i.e. ① executing the kinematic calibration step), in response to the user's click operation, Fig.13A new interface is added to the right side of the first interface as shown in the figure, which is used as the second interface; in response to the user clicking the "debugging" option in the lower left corner of the second interface (i.e. ② selecting the debugging interface), the interface jumps to the debugging interface. There is a "parameter diagnosis" button in the upper left corner of the debugging interface (i.e. ③ one-key error diagnosis). In response to the user clicking the "parameter diagnosis" button, the host computer uses the full parameter error model to determine the first error. For details, please refer to the above Figure 2 Afterwards, the determined first error and calibration error can be displayed in the debugging interface (i.e., ④ diagnosis result display). At the same time, in order to enable the user to more intuitively understand the impact of the existing first error on the robot accuracy and the location of the first error, an error diagram is also configured for the two first errors with the smallest calibration residuals when displayed in the debugging interface.
[0210] like Fig.13 As shown, five first errors are determined based on the full parameter error model, and these five first errors are displayed on the debugging interface. The parameter numbers (i.e., key parameter numbers) corresponding to these five first errors are 29 (rotation angle of axis 4 relative to axis 3 around the z direction), 25 (axis 4 offset along the y direction relative to axis 3), 23 (rotation angle of axis 3 relative to axis 2 around the z direction), 50 (torsional flexibility of axis 3 around the z direction), and 15 (rotation angle of axis 2 relative to axis 1 around the x direction). It can be understood that in this example, corresponding parameter numbers are assigned in advance to each error term that may be determined by the full parameter error model. When it is determined that a certain error term exists, the parameter number corresponding to the error term can be displayed on the debugging interface to inform the user of the current error term. For each first error, the debugging interface also displays its corresponding parameter value (i.e., the size of the key parameter) and the calibration residual (rated accuracy after compensation). For example, the rated accuracy after compensation corresponding to the key parameter number 29 (rotation angle of axis 4 relative to axis 3 around the z direction) is 0.387470 millimeters (mm), and the size of the key parameter is 0.009948 degrees (°); the rated accuracy after compensation corresponding to the key parameter number 25 (offset of axis 4 relative to axis 3 along the y direction) is 0.398398mm, and the size of the key parameter is 0.082109mm. ; The rated accuracy after compensation corresponding to the key parameter number 23 (rotation angle of axis 3 relative to axis 2 around the z direction) is 0.402648mm, and the key parameter size is 0.006918°; the rated accuracy after compensation corresponding to the key parameter number 50 (torsional flexibility of axis 3 around the z direction) is 0.408028mm, and the key parameter size is 0.000004; the rated accuracy after compensation corresponding to the key parameter number 15 (rotation angle of axis 2 relative to axis 1 around the x direction) is 0.416715mm, and the key parameter size is -0.004440°.
[0211] Depend on Fig.13 visible, Fig.13The debugging interface in is divided into two parts. One part is used to express the five determined first errors and the rated accuracy after compensation and the calibration residual corresponding to each error in text form, and the five determined first errors are arranged in ascending order according to the numerical value of the rated accuracy after compensation, so that the user can determine the influence of each determined first error on the absolute positioning accuracy of the robot. The other part is used to display the two first errors with the smallest rated accuracy after calibration among the five determined first errors, and the corresponding error diagrams are configured for these two first errors. Among these two first errors, the first error with the smallest rated accuracy after calibration is used as the primary influencing factor, so that the user can intuitively understand the first error with the greatest influence on the absolute positioning accuracy of the robot, as well as the location and reason of the first error.
[0212] Based on the above examples, the second instruction can be the user clicking the "Parameter Diagnosis" button in the debugging interface, or the second instruction can be a series of operations in which the user clicks "Kinematic Calibration", clicks the "Debug" option, and clicks "Parameter Diagnosis".
[0213] In order to facilitate understanding of the relationship between the first interface, the robot, the laser follower, etc., the present application embodiment provides an error diagnosis system, such as Fig.14 As shown, it includes a software layer, a hardware layer and an algorithm layer. The software layer (i.e., the above-mentioned first interface) is configured in the host computer. In response to the function selection made by the user, the host computer can be used to call the hardware layer and the algorithm layer to perform corresponding tasks. For example, the software layer can control the robot (i.e., the industrial robot) of the hardware layer to adjust to different motion states, and can also call the DH kinematic error model (i.e., the kinematic model) or the full parameter error model (i.e., the full parameter model) in the algorithm layer to diagnose the robot error and display the execution result of the task; the hardware layer includes a robot (i.e., the industrial robot) and a laser follower; it is used to respond The acquisition instructions of the software layer realize data acquisition, obtain the acquired data, and send the acquired data to the software layer. The acquired data can be the above-mentioned first coordinate or third coordinate, etc.; the algorithm layer includes the DH kinematic error model (i.e., the kinematic model) and the full parameter error model (i.e., the full parameter model). The software layer can input data (such as the first coordinate, third coordinate, etc. acquired) to the algorithm layer, and send a call instruction to the algorithm layer. The algorithm layer responds to the call instruction of the software layer. The algorithm layer can use the DH kinematic error model or the full parameter error model indicated by the call instruction to determine the existing error and obtain the calibration result and the diagnosis result.
[0214] In summary, the user can interact with the short hair layer and the hardware layer by operating the software interface (i.e. the first interface mentioned above) and selecting the corresponding function. The error diagnosis system integrates calibration tools to realize the automated calibration process: data collection, parameter identification, and result display. For products that do not meet the calibration standards, the error diagnosis system integrates diagnostic tools to quickly locate the main factors affecting the calibration accuracy and realize error diagnosis and accuracy analysis.
[0215] By applying the technical solution provided in the embodiments of the present application, the following beneficial effects can be achieved:
[0216] (1) This application establishes a robot calibration error diagnosis system involving multiple levels of software, algorithms, and hardware;
[0217] (2) This application establishes a full-parameter error model including robot joint offset error, joint deflection error, joint flexibility error, and link flexibility error, and proposes a calibration error diagnosis method based on the full-parameter error model;
[0218] (3) This application proposes a calibration preset point mode, which uses a small number of pre-calibrated points (i.e., the second preset joint angles) to avoid the problem of laser follower light loss when the robot moves in full space;
[0219] (4) The present application can solve the problem in the related art that the robot calibration progress based on the DH kinematic error model cannot converge to the effective threshold, and can realize the rapid positioning of the calibration error, the accurate analysis of the error size, and the preliminary evaluation of the repair accuracy (i.e., the calibration residual);
[0220] (5) The robot calibration error diagnosis method based on the full parameter model provided in this application can be used as a robot production auxiliary tool to check assembly accuracy and troubleshoot assembly problems;
[0221] (6) The technical solution provided in the embodiment of the present application can provide tool (i.e., the above-mentioned first interface) support for robot structural repair and locate damaged components.
[0222] Corresponding to the above-mentioned error diagnosis method, the embodiment of the present application also provides an error diagnosis device, such as Fig.15 As shown, the device comprises:
[0223] Coordinate acquisition module 151: used to acquire the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles, respectively as the first coordinates corresponding to each first preset joint angle;
[0224] A first error determination module 152 is used to fit the corresponding relationship between the first preset joint angle and the first coordinate through an original full-parameter error model to obtain a target full-parameter error model, and use the model parameters in the target full-parameter error model as the first error of the robot and display the first error;
[0225] Among them, the original full-parameter error model is a model with full-parameter error as model parameter, robot joint angle as independent variable, and coordinates of the calibration component at the end of the robot as dependent variable. The full-parameter error includes offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between every two adjacent joints.
[0226] In the technical solution provided by the embodiment of the present application, the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles are obtained, and are respectively used as the first coordinates corresponding to each first preset joint angle; the original full parameter error model is fitted according to the obtained multiple first preset joint angles and the first coordinates to obtain the target full parameter error model, and the model parameters of the target full parameter model are used as the first error of the robot and displayed, so as to facilitate the user to adjust the robot based on the first error to improve the absolute positioning accuracy of the robot. Among them, since the original full parameter error model is a model with full parameter error as the model parameter, robot joint angle as the independent variable, and the coordinates of the calibration component at the end of the robot as the dependent variable, and the full parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between each two adjacent joints, which are not considered when constructing the DH kinematic error model, therefore, when the DH kinematic error model cannot make the accuracy of the robot converge to the effective threshold, the technical solution provided by the embodiment of the present application can effectively improve the accuracy of the robot.
[0227] In some embodiments, the apparatus further comprises:
[0228] The second error determination module is used to fit the corresponding relationship between the first preset joint angle and the first coordinate through the original DH kinematic error model to obtain the target DH kinematic error model, and use the model parameters in the target DH kinematic error model as the second error of the robot and display the second error;
[0229] The original DH kinematic error model is a model with DH kinematic error as model parameter, robot joint angle as independent variable, and coordinates of the calibration components at the end of the robot as dependent variable. The DH kinematic error includes the offset error of each rod length of the robot, the zero position error of each joint of the robot, and the stiffness error of each joint of the robot.
[0230] The first error determination module 152 is specifically configured to:
[0231] In response to the diagnostic instruction for the second error, the corresponding relationship between the first preset joint angle and the first coordinate is fitted by the original full-parameter error model to obtain a target full-parameter error model.
[0232] In some embodiments, the calibration component is a target ball, the first coordinate is the coordinate of the target ball in the laser follower coordinate system obtained after the target ball is captured by the laser follower, and the device further includes:
[0233] The control module is used to control the robot to move to the first preset joint angle with a target ball fixed at the end for each first preset joint angle, and control the laser follower to capture the target ball;
[0234] A first determination module: used to determine the conversion relationship between the robot flange coordinate system and the base coordinate system according to the first preset joint angle if the coordinates returned by the laser follower are not received within a preset time;
[0235] Conversion module: used to determine the coordinates of the target ball in the laser follower coordinate system as the second coordinate according to the conversion relationship between the base coordinate system and the laser follower coordinate system, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the flange coordinate system and the base coordinate system;
[0236] Adjustment module: used for adjusting the joint angle of the laser follower so that the field of view of the laser follower covers the second coordinate;
[0237] The coordinate acquisition module 151 is also used to acquire the coordinates obtained by the laser follower capturing the target ball as the first coordinates corresponding to the first preset joint angle.
[0238] In some embodiments, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system are obtained in advance by the following method:
[0239] Obtaining the coordinates of the target ball fixed to the end of the robot in the laser follower coordinate system when the robot is at different second preset joint angles, respectively as the third coordinates corresponding to each of the second preset joint angles; and, for each of the second preset joint angles, obtaining the conversion relationship between the flange coordinate system and the base coordinate system, as the conversion relationship between the flange coordinate system corresponding to each of the second preset joint angles and the base coordinate system;
[0240] According to the transformation relationship between the flange coordinate system corresponding to each of the second preset joint angles and the base coordinate system, and the third coordinates corresponding to each of the second preset joint angles, the transformation relationship between the tool coordinate system and the flange coordinate system, and the transformation relationship between the base coordinate system and the laser follower coordinate system are fitted.
[0241] In some embodiments, the apparatus further comprises:
[0242] A second determination module is used to determine, for each first error, a maximum accuracy that can be achieved by the robot when only the first error is adjusted, as a calibration residual corresponding to each first error;
[0243] A third determination module is used to determine a first error corresponding to the calibration residual that satisfies a preset condition as a third error;
[0244] The first error determination module 152 is specifically configured to:
[0245] The third error is displayed in a first display manner, and the fourth error is displayed in a second display manner, where the fourth error is an error in the first error except the third error.
[0246] In some embodiments, an error schematic diagram for each error configuration is obtained in advance;
[0247] The first error determination module 152 is specifically configured to:
[0248] For each third error, obtaining an error schematic diagram configured for the third error as a target error schematic diagram of the third error;
[0249] Display each third error, a target error diagram of each third error, and a calibration residual corresponding to each third error.
[0250] In some embodiments, the apparatus comprises:
[0251] Interface display module: used to display a first interface for indicating the progress of error diagnosis;
[0252] Instruction execution module: used to respond to a first instruction for the first interface, execute the step of fitting the correspondence between the first preset joint angle and the first coordinate through the original DH kinematic error model to obtain the target DH kinematic error model; respond to a second instruction for the first interface, execute the step of fitting the correspondence between the first preset joint angle and the first coordinate through the original full-parameter error model to obtain the target full-parameter error model.
[0253] In some embodiments, the instruction execution module is further used to:
[0254] In response to a third instruction for the first interface, initializing the robot and the laser follower;
[0255] In response to the fourth instruction for the first interface, the current accuracy of the robot is determined and displayed.
[0256] Corresponding to the above error diagnosis method, the present application embodiment also provides an electronic device, such as a host computer, such as Fig.16 As shown, including:
[0257] A memory 161, for storing computer programs;
[0258] The processor 162 is used to implement any of the above-mentioned error diagnosis methods when executing the program stored in the memory 161.
[0259] In addition, the electronic device may further include a communication bus and / or a communication interface, and the processor 162, the communication interface, and the memory 161 communicate with each other via the communication bus.
[0260] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.
[0261] The communication interface is used for communication between the above electronic device and other devices.
[0262] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.
[0263] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0264] In another embodiment provided in the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned error diagnosis methods are implemented.
[0265] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any one of the error diagnosis methods in the above embodiments.
[0266] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a solid-state hard disk (SSD), etc.
[0267] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0268] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic device, and storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.
[0269] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.
Claims
1. An error diagnosis method, characterized in that: The method comprises: Acquire the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles, and use them as the first coordinates corresponding to each of the first preset joint angles; Fitting the correspondence between the first preset joint angle and the first coordinate by an original full-parameter error model to obtain a target full-parameter error model, and using the model parameters in the target full-parameter error model as the first error of the robot and displaying the first error; Among them, the original full-parameter error model is a model with full-parameter error as model parameter, robot joint angle as independent variable, and coordinates of the calibration component at the end of the robot as dependent variable. The full-parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between every two adjacent joints.
2. The method according to claim 1, characterized in that: The method further comprises: Fitting the corresponding relationship between the first preset joint angle and the first coordinate through the original DH kinematic error model to obtain a target DH kinematic error model, and using the model parameters in the target DH kinematic error model as the second error of the robot and displaying the second error; The original DH kinematic error model is a model with DH kinematic error as model parameter, robot joint angle as independent variable, and coordinates of calibration components at the end of the robot as dependent variable. The DH kinematic error includes offset error of each rod length of the robot, zero position error of each joint of the robot, and stiffness error of each joint of the robot. The step of fitting the corresponding relationship between the first pre-joint angle and the first coordinate by using the original full-parameter error model to obtain a target full-parameter error model includes: In response to the diagnostic instruction for the second error, the corresponding relationship between the first preset joint angle and the first coordinate is fitted by an original full-parameter error model to obtain a target full-parameter error model.
3. The method according to claim 1, characterized in that The calibration component is a target ball, the first coordinate is the coordinate of the target ball in the laser follower coordinate system obtained after the target ball is captured by a laser follower, and the method further includes: For each of the first preset joint angles, respectively, controlling the robot to move to the first preset joint angle with the target ball fixed at the end, and controlling the laser follower to capture the target ball; If the coordinates returned by the laser follower are not received within a preset time, determining the conversion relationship between the robot flange coordinate system and the base coordinate system according to the first preset joint angle; According to the conversion relationship between the base coordinate system and the laser follower coordinate system, the conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the flange coordinate system and the base coordinate system, the coordinates of the target sphere in the laser follower coordinate system are determined as the second coordinates; Adjusting the orientation of the laser follower so that the field of view of the laser follower covers the second coordinate; The coordinates obtained by the laser follower capturing the target sphere are acquired as the first coordinates corresponding to the first preset joint angle.
4. The method according to claim 3, characterized in that The conversion relationship between the tool coordinate system and the flange coordinate system, and the conversion relationship between the base coordinate system and the laser follower coordinate system are obtained in advance by the following method: Obtaining the coordinates of the target ball fixed to the end of the robot in the laser follower coordinate system when the robot is at different second preset joint angles, respectively as the third coordinates corresponding to each of the second preset joint angles; and, for each of the second preset joint angles, obtaining the conversion relationship of the flange coordinate system relative to the base coordinate system, as the conversion relationship of the flange coordinate system corresponding to each of the second preset joint angles relative to the base coordinate system; According to the transformation relationship between the flange coordinate system corresponding to each of the second preset joint angles and the base coordinate system, and the third coordinates corresponding to each of the second preset joint angles, the transformation relationship between the tool coordinate system and the flange coordinate system, and the transformation relationship between the base coordinate system and the laser follower coordinate system are fitted.
5. The method according to claim 1, characterized in that The method further comprises: For each of the first errors, determining a maximum accuracy that can be achieved by the robot by adjusting the first error alone as a calibration residual corresponding to each of the first errors; Determine a first error whose corresponding calibration residual satisfies a preset condition as a third error; The displaying of the first error comprises: The third error is displayed in a first display manner, and the fourth error is displayed in a second display manner, wherein the fourth error is an error in the first error except the third error.
6. The method according to claim 5, characterized in that Obtaining in advance an error diagram with configurations for each error; The displaying of the third error in a first display manner includes: For each of the third errors, obtaining an error schematic diagram configured for the third error as a target error schematic diagram for the third error; The third errors, the target error diagrams of the third errors, and the calibration residuals corresponding to the third errors are displayed.
7. The method according to claim 2, characterized in that The method comprises: Displaying a first interface for indicating error diagnosis progress; In response to a first instruction for the first interface, executing the step of fitting the correspondence between the first preset joint angle and the first coordinate by using an original DH kinematics error model to obtain a target DH kinematics error model; In response to a second instruction for the first interface, the step of fitting the correspondence between the first preset joint angle and the first coordinate through an original full-parameter error model to obtain a target full-parameter error model is executed.
8. The method according to claim 7, characterized in that The method further comprises: In response to a third instruction for the first interface, initializing the robot and the laser follower; In response to a fourth instruction for the first interface, the current accuracy of the robot is determined and displayed.
9. A calibration error diagnostic device, characterized in that: The device comprises: A coordinate acquisition module: used to acquire the coordinates of the calibration component fixed to the end of the robot when the robot is at different first preset joint angles, and use them as the first coordinates corresponding to each of the first preset joint angles; An error determination module: used to determine a first conversion relationship between the robot end coordinates and the robot joint angle according to the plurality of the first coordinates and the first preset joint angles; Error determination and display module: fitting the correspondence between the first preset joint angle and the first coordinate through the original full-parameter error model to obtain a target full-parameter error model, and using the model parameters in the target full-parameter error model as the first error of the robot and displaying the first error; Among them, the original full-parameter error model is a model with full-parameter error as model parameter, robot joint angle as independent variable, and coordinates of the calibration component at the end of the robot as dependent variable. The full-parameter error includes the offset error, deflection error, joint angle error, joint flexibility error, and connecting rod flexibility error between every two adjacent joints.
10. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-8 when executing a program stored in a memory.
Citation Information
Patent Citations
Kinematic method for improving absolute positioning precision of industrial robot
CN109746915A
Robot absolute positioning precision calibration method based on kinematics and spatial interpolation
CN112873199A
Absolute precision compensation method and system, equipment and computer readable storage medium
CN113927599A
Building robot error analysis method and verification method based on rigid-flexible coupling
CN114474070A
Industrial robot precision compensation method based on multi-source error modeling
CN114833834A
Cited By
Operation rod zero position calibration method of surgical robot and related device
CN119632687A
