An industrial robot precision compensation method and device for precision micro-operation

Abstract

This invention belongs to the field of industrial robot-assisted manufacturing and precision operation technology, and discloses a method and device for precision micro-operation of industrial robots. The method includes: establishing a robot differential kinematic error model coupled to a tool coordinate system; automatically acquiring data on the robot's end-effector pose and corresponding robot joints using a low-cost probe and planar calibration block; constructing and solving a set of error identification equations including tool errors to obtain a corrected robot kinematic model; proposing a virtual tool coordinate algorithm to convert the calibration results for the probe into actual target point control of the micro-operation actuator; and using the corrected robot kinematic model and its corresponding compensation program to control the robot to complete high-precision micro-operations. This invention can improve the absolute positioning accuracy of robots from hundreds of micrometers to tens of micrometers, and is low-cost and highly automated, showing broad application prospects in the field of robot-assisted precision micro-operations.

Description

Technical Field

[0001] This invention belongs to the field of industrial robot-assisted manufacturing and precision operation technology, specifically relating to a precision compensation method and device for industrial robots oriented towards precision micro-operations. Background Technology

[0002] With the rapid development of precision manufacturing, biomedicine, microelectronics and cutting-edge physics experiments, the demand for high-precision, high-flexibility picking, assembly and positioning operations on tiny parts or targets (with dimensions in the millimeter or even micrometer range) is becoming increasingly urgent.

[0003] Industrial robots, with their advantages of large workspace, multiple degrees of freedom, high flexibility, and relatively low cost, offer an ideal solution to the aforementioned problems. Mounting micro-manipulation actuators (such as vacuum suction heads and micro-grippers) on the end effector of industrial robots promises to achieve efficient, flexible, and automated precision micro-manipulation. However, this technological approach faces a core bottleneck: the absolute positioning accuracy of industrial robots is generally low (typically at the millimeter or hundreds of micrometer levels), while precision micro-manipulation requires sub-millimeter or even tens of micrometer levels of accuracy. A significant contradiction exists between the low precision of robots and the high precision requirements of micro-manipulation.

[0004] Existing robot accuracy compensation methods are mainly divided into online compensation and offline calibration. Online compensation often relies on expensive external real-time measurement equipment such as laser trackers, which is extremely costly and has complex algorithms, making it difficult to popularize in industrial applications. Offline calibration methods, on the other hand, identify and compensate for robot kinematic parameter errors in advance through mathematical models, which is less costly and has greater potential for industrial applications. However, existing offline calibration research has the following shortcomings: 1) Deficiencies exist in the handling of tool coordinate system installation errors: In existing studies, calibration tools (such as probes) are usually directly mounted on the robot's end effector flange, and their calibration process is independent of the micro-manipulator that ultimately performs the operation. Furthermore, most studies fail to effectively incorporate tool coordinate system installation errors into the robot's kinematic error model for unified identification, still treating tool parameters as additional parameters to the robot's DH (Hyper-Kinematic) parameters, which is incompatible with the standard DH model. This leads to difficulties in convergence and poor compensation accuracy in the parameter identification process.

[0005] 2) Lack of methods for connecting with precision micro-manipulation tasks: Existing robot accuracy calibration research mostly focuses on improving the robot's own positioning accuracy. However, there is a difference between the coordinates of the probe tool used in calibration and the coordinates of the actuator's action point required during operation. How to establish a precise transmission relationship between the two and achieve seamless conversion of compensation results is a gap in existing offline calibration technology, making it difficult to directly translate the accuracy improvement after calibration into an improvement in actual operational accuracy.

[0006] Therefore, there is an urgent need to develop a high-precision industrial robot precision compensation method and device that can be deeply integrated with the end effector, so as to break through the application bottleneck of industrial robots in the field of high-precision micro-manipulation. Summary of the Invention

[0007] To address the problem that industrial robots cannot meet the requirements of high-precision micro-operations due to low absolute positioning accuracy, this invention provides a precision compensation method and device for industrial robots designed for precision micro-operations, which can improve the absolute positioning accuracy of the robot from hundreds of micrometers to tens of micrometers (±50). It is low in cost and highly automated, and has broad application prospects in the field of robot-assisted precision micro-operations.

[0008] To achieve the above objectives, the present invention provides the following solution: A precision compensation method for industrial robots designed for precision micro-manipulation, the method comprising: Establish a differential kinematic error model for an industrial robot with a coupled tool coordinate system; The end-effector pose and corresponding joint angle data of the industrial robot are automatically collected using a probe and a planar calibration block; and the theoretical position of the end-effector of the calibration tool is calculated based on the joint angle data. Based on the theoretical position, the nominal plane equation of the calibration plane is fitted, and combined with the differential kinematic error model of the industrial robot, an overdetermined set of equations is constructed, which includes the kinematic parameter error of the industrial robot and the plane equation parameter error. Solve the overdetermined system of equations to identify the kinematic parameter errors of the industrial robot and obtain the corrected kinematic model of the industrial robot. Based on the virtual tool coordinate method, the action point of the micro-operator is converted into the virtual probe point. The industrial robot moves according to the virtual probe position, and the error is compensated by the corrected industrial robot kinematic model to complete the micro-operation.

[0009] Preferably, the method for establishing a differential kinematic error model of an industrial robot with a coupled tool coordinate system includes three parts: Error propagation formula part: ; in, This refers to the small pose error of the robot's end effector in the end-effector coordinate system. That is, the coordinate system { i The small pose error transfer matrix from coordinate system {6} to coordinate system {6}. ,when hour, ,when hour, It is the identity matrix; Link i The mapping matrix from geometric parameter errors to small pose errors in the link coordinate system. Link i Geometric parameter errors; Error transformation part of the physical constraint-oriented method: ; in, , , This indicates the nominal position of the robot's end effector in the robot's coordinate system. , , This represents the actual position of the robot's end effector in the robot's coordinate system. For a 6-DOF industrial robot, a 3×30 matrix represents this position. The 30×1 matrix ∆ is derived from the error propagation formula. Horizontal splicing and Formed by vertical splicing; Kinematic modeling section: The kinematic coordinate system of the industrial robot and its tool is established according to the MDH method; the spatial pose relationship between the tool coordinate system and the industrial robot body conforms to the MDH standard, and the tool coordinate system is relative to the first kinematic coordinate system of the industrial robot. n The pose of joint -1 can be determined by the industrial robot's first joint. n -1 joints are described using the MDH method, where... n This refers to the number of degrees of freedom of an industrial robot. According to the kinematic coordinate system, the industrial robot MDH parameters of the coupling tool parameters are extracted. The number of industrial robot MDH parameter groups of the coupling tool parameters is consistent with the number of joints of the industrial robot.

[0010] Preferably, the method for automatically acquiring the end-effector pose of an industrial robot and its corresponding joint angle data using a probe and a planar calibration block, and calculating the theoretical position of the calibration tool end-effector based on the joint angle data, includes: An industrial robot carrying a calibration tool automatically approaches and touches three mutually orthogonal planes of a calibration block in various pre-set postures and positions. The moment the calibration tool touches the plane, a signal is sent back to the industrial robot, which immediately stops moving. The robot obtains the joint angle data of its six joints at the corresponding moment and calculates the theoretical position of the end of the calibration tool.

[0011] Preferably, the method for constructing an overdetermined system of equations that includes kinematic parameter errors of the industrial robot and planar equation parameter errors includes: Using the theoretical position, the nominal plane equation of the calibration plane in the industrial robot base coordinate system is fitted, and the plane equation parameter error is introduced to describe the actual plane equation. The expression for the actual end position derived from the differential kinematic error model of the industrial robot is substituted into the actual plane equation after introducing parameter errors, and linearized by discarding higher-order infinitesimals, thereby obtaining the overdetermined system of equations.

[0012] The actual plane equation after introducing parameter errors is: ; in, a , b , c These are the nominal plane parameters; Substituting the expression for the actual end position into the actual plane equation after introducing parameter errors, and discarding higher-order infinitesimals, we obtain the following system of equations: .

[0013] Preferably, the method for converting the action point of the micro-manipulation actuator into the virtual probe point based on the virtual tool coordinate method includes: When planning the operation path, a series of target points are generated for the action point of the micro-operator. P r ; Through spatial pose transformation T Convert all target points to points corresponding to the coordinates of the virtual tool. P n The virtual tool coordinates are the center of the sphere of the calibration tool.

[0014] Preferably, the method for completing micro-operations by moving the industrial robot according to the virtual probe position and compensating for errors using a corrected industrial robot kinematic model includes: In actual operation, the control and computing unit sends the converted virtual tool points to the industrial robot controller. P n The modified kinematic model of the industrial robot is used to calculate the small position error of the end effector of the industrial robot tool, and to drive the industrial robot to perform motion and compensation motion. When the industrial robot moves the virtual tool coordinates to P n At that time, the actual point of action of the micro-operator at its end will be located at the originally planned operating position. P r On the surface, the industrial robot sequentially executes the instructions for adsorption, movement, and placement to complete the target ball picking and placement operations.

[0015] The present invention also provides a precision compensation device for industrial robots designed for precision micro-operations. The device is used to implement the aforementioned method and includes: A six-degree-of-freedom industrial robot; A connector is fixedly mounted on the end flange of the industrial robot, and the connector is used to jointly support the calibration tool and the micro-manipulation actuator; A calibration tool that is detachably mounted on the connector; A micro-operation actuator detachably mounted on the connector; A calibration block.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention treats the calibration tool coordinate system as an extension of the robot's sixth joint and couples it with the existing DH parameters of the industrial robot. Without adding new parameter sets, it uniformly incorporates them into the error model for identification and compensation. The calibration results are then seamlessly applied to actual operation using the virtual tool coordinate method. This improves the absolute positioning accuracy of the robot carrying micro-manipulation actuators in the workpiece coordinate system from the millimeter level to the sub-millimeter level (experimentally verified to be within ±0.05mm), meeting the accuracy requirements of most precision micro-operations.

[0017] The calibration process of this invention only requires a high-precision probe and a marble gauge block, which are low-cost devices, eliminating the need for expensive laser trackers. This greatly reduces the deployment and application costs of the device and facilitates industrial promotion. Attached Figure Description

[0018] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the overall process of a precision compensation method for industrial robots designed for precision micro-operations, as described in an embodiment of the present invention. Figure 2 This is a schematic diagram of the hardware structure of the industrial robot precision compensation device according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the establishment of the coordinate system for the robot MDH kinematic model with coupled tool coordinates in an embodiment of the present invention. Figure 4 This is a diagram illustrating the accuracy of the error model in an embodiment of the present invention. Figure 5 This is a diagram showing the initial points and fitting plane of the robot calibration experiment in an embodiment of the present invention. Figure 6This is a diagram showing the corrected points and fitted plane after error identification and compensation in an embodiment of the present invention. Figure 7 This is a comparison chart of experimental results for random pose position detection based on LHS in an embodiment of the present invention; In the diagram: 1. Vibration isolation platform; 2. Calibration block; 3. Calibration tool; 4. Connector; 5. Micro-manipulator; 6. Industrial robot; 7. Industrial robot base. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] To further understand the invention's content, features, and technical effects, the following specific embodiments are provided, along with accompanying drawings. Figure 1 -Appendix Figure 7 Detailed explanation is as follows: Example 1 See attached document Figure 2 This invention provides a precision compensation device for industrial robots designed for precision micro-operations, comprising: Industrial Robot 6: The ABB IRB 1200 six-axis serial industrial robot 6 is selected. It has a payload of 7kg, an open secondary development interface, and a working range and load capacity sufficient to carry the connector 4, calibration tool 3, and micro-manipulator 5 in this device.

[0023] Connector 4: Fixedly mounted on the end flange of the sixth axis of the industrial robot 6. Connector 4 has at least two mounting interfaces, used for mounting the calibration tool 3 and the micro-manipulator 5, respectively. The spatial relationship of each mounting interface on connector 4 is a known fixed value, which is ultimately confirmed by a high-precision 3D scanner.

[0024] Calibration tool 3: A Makino Precision Control JS-PT3D cable communication trigger probe is selected. During accuracy calibration, calibration tool 3 is installed on the corresponding interface of connector 4. Its coordinate system is considered an extension or inherent part of the sixth joint coordinate system of industrial robot 6, and is coupled with the original DH parameters of industrial robot 6 without adding new parameter sets. Ultimately, it is incorporated into the differential kinematic error model for unified identification and compensation. Process: As follows... Figure 3 The kinematic coordinate system of the industrial robot and its tools is established according to the MDH method. The spatial pose relationship between the tool coordinate system and the industrial robot body conforms to the MDH standard, and the tool coordinate system is relative to the ( )th kinematic coordinate system of the industrial robot. n -1) The pose of the joint can be determined by the ( ) joint of the industrial robot n -1) Joints are described using the MDH method ( n (This refers to the number of degrees of freedom of an industrial robot).

[0025] As shown in Table 1, the industrial robot MDH parameters of the coupling tool parameters are extracted according to the kinematic coordinate system. The number of industrial robot MDH parameter groups of the coupling tool parameters is consistent with the number of joints of the industrial robot.

[0026] Micro-manipulation actuator 5: In this embodiment, a vacuum suction head is selected for picking up and placing ICF target pellets with a diameter of approximately 2 mm. During the picking operation, the industrial robot 6 moves the vacuum suction head above the target object using a virtual coordinate method, controlling the distance within the effective working range of the vacuum suction head. Then, the industrial robot 6 controls the vacuum suction head to pick up the target object. During the placement operation, the industrial robot 6 moves the vacuum suction head carrying the target object above the target position using a virtual coordinate method. Then, the industrial robot 6 controls the vacuum suction head to release the suction effect on the target object, completing the placement of the target object. In actual operation, the micro-manipulation actuator 5 is installed on another corresponding interface of the connector 4.

[0027] Calibration block 2: A high-precision marble block is used as calibration block 2, with a surface flatness of 5. Within this range, it serves as the benchmark for calibration.

[0028] Example 2 This invention provides a precision compensation method for industrial robots designed for precision micro-manipulation, and particularly relates to a low-cost planar constraint-based precision compensation method for industrial robots designed for precision manipulation of small target objects, comprising the following steps: Step 1: Establish the differential kinematic error model of the industrial robot 6 with the coupled tool coordinate system. The coordinate system of the calibration tool 3 strictly conforms to the DH model standard and is regarded as an extension or inherent part of the sixth joint coordinate system of the industrial robot 6. Step 2: Control the industrial robot 6 so that the calibration tool 3 mounted on the end of the industrial robot 6 contacts different surfaces of a calibration block 2 in different poses (at this time, the detachable micro-manipulation actuator 5 is not installed), record the joint angle data of the industrial robot 6 at each contact, and calculate the theoretical position of the end of the calibration tool 3 based on the joint angle data. Step 3: Fit the nominal plane equation of the calibration plane based on the theoretical position, and combine it with the industrial robot 6 differential kinematic error model to construct an overdetermined set of equations that includes the kinematic parameter error of the industrial robot 6 and the plane equation parameter error. Step 4: Solve the overdetermined system of equations to identify the kinematic parameter errors of industrial robot 6 (including the coupling tool parameter errors) and obtain the corrected kinematic model of industrial robot 6. Step 5: Before performing a precision micro-operation task, based on the spatial pose transformation relationship between the actual point of action of the micro-operation actuator 5 and the end of the calibration tool 3, the target point of the micro-operation to be performed is converted into a virtual tool point corresponding to the calibration tool 3. Step Six: Control the industrial robot 6 to move to the target position using the virtual tool points, and use the corrected kinematic model of the industrial robot 6 to perform error compensation to complete high-precision micro-operation.

[0029] In this embodiment, the step one of establishing the differential kinematic error model of the industrial robot with coupled tool coordinate system specifically includes: The coordinate system of the calibration tool 3 installed at the flange end of the industrial robot 6 strictly conforms to the MDH model standard and is regarded as an extension or inherent part of the coordinate system of the sixth joint of the industrial robot 6. Based on the MDH modeling method, a kinematic model of industrial robot 6 with coupled tool coordinate system is established. The number of DH parameter sets in this model is still the same as the number of DH parameter sets in the body of industrial robot 6, which is different from the traditional model that includes tool coordinate system. Based on the principle of differential transformation, the mapping relationship between the end-effector pose error of industrial robot 6 and the geometric parameter errors of each link (geometric parameter errors of coupling calibration tool 3) is derived, thus obtaining the differential kinematic error model of industrial robot 6.

[0030] In this embodiment, the step two, in which the industrial robot 6 is controlled to make the calibration tool 3 contact the calibration block 2 in different poses, is accomplished by an automated calibration program. The automated calibration program can automatically traverse multiple preset points and automatically record the corresponding data to improve calibration efficiency.

[0031] In this embodiment, the process of constructing the overdetermined system of equations in step three specifically includes: Using the theoretical position, the nominal plane equation of the calibration plane in the 6-base coordinate system of the industrial robot is fitted, and the plane equation parameter error is introduced to describe the actual plane equation. The expression for the actual end position derived from the 6-differential kinematic error model of the industrial robot is substituted into the actual plane equation after introducing parameter errors, and linearized by discarding higher-order infinitesimals, thus obtaining the overdetermined system of equations.

[0032] In this embodiment, in step four, the overdetermined equations are solved using a nonlinear optimization method based on the Levenberg-Marquardt algorithm, or by using the least squares method, to accurately identify the kinematic parameter errors of the industrial robot 6.

[0033] In this embodiment, in step five, the calibration tool 3 and the micro-manipulation actuator 5 are jointly installed on the end flange of the industrial robot 6 through the same connector 4.

[0034] In this embodiment, the method further includes performing automated calibration and control through a control and calculation unit, which is used to realize motion control, automated calibration, error parameter identification, error compensation, and establishment of the workpiece coordinate system of the industrial robot 6.

[0035] In this embodiment, the workpiece coordinate system of the industrial robot 6 is established by the 3-2-1 measurement method, that is, the origin and axis of the workpiece coordinate system are determined by measuring the positions of three orthogonal planes on the workpiece.

[0036] Example 3 like Figure 1 As shown, the present invention also provides a precision compensation method for industrial robots for precision micro-operations, comprising the following steps: Step 1: Establish the industrial robot 6 differential kinematic error model with coupled tool coordinate system. The tool coordinate system strictly conforms to the DH model standard and is regarded as an extension or inherent part of the industrial robot 6 sixth joint coordinate system. like Figure 3 As shown, when using calibration tool 3 for calibration, the center of the probe's end sphere is considered as an extension of the sixth joint of industrial robot 6, and coupled with the original DH parameters of industrial robot 6. No new parameter sets are added, and the kinematic model is established strictly according to the MDH method. The industrial robot DH parameter table for coupled tool coordinates is shown in Table 1: Table 1 The method for establishing a differential kinematic error model of an industrial robot with a coupled tool coordinate system consists of three parts: Error propagation formula part: Based on the principle of differential transformation, the small pose error vector of the six-end effector of the industrial robot is derived. A linear mapping relationship between the error vectors of the geometric parameters (coupling tool geometric parameters) of all links: in, This refers to the small pose error of the robot's end effector in the end-effector coordinate system. That is, the coordinate system { i The small pose error transfer matrix from coordinate system {6} to coordinate system {6}. ,when hour, ,when hour, It is the identity matrix. , , Let X be the direction cosine (unit projection) of the unit vector of the X-axis of the target coordinate system and the X, Y, and Z axes of the reference coordinate system. , , Let X be the direction cosine (unit projection) of the unit vector of the Y-axis of the target coordinate system and the X, Y, and Z axes of the reference coordinate system. , , Let Z be the direction cosine (unit projection) of the unit vector of the Z-axis of the target coordinate system and the X, Y, and Z axes of the reference coordinate system. , , This represents the position of the origin of the target coordinate system in the reference coordinate system. It is a homogeneous transformation matrix; Link i The mapping matrix from geometric parameter errors to small pose errors in the link coordinate system. That is , That is , Link i The DH parameter represents the distance along the link. i X of the coordinate system i Direction, Link i Z in the coordinate system of -1 i -1 and connecting rod i Z of the coordinate system i The distance between them Link i The DH parameter represents the distance along the link. i X of the coordinate system i Looking at the direction, the connecting rod i Z in the coordinate system of -1 i -1 and connecting rod iZ of the coordinate system i The angle between them These are the rotational parameters about the corresponding axes when the axes of the two joints are parallel. Link i Geometric parameter errors.

[0037] Error transformation (actual end-effector position expression) part of the physical constraint-oriented method: ; For a 6-DOF industrial robot, a 3×30 matrix is ​​used. The 30×1 matrix ∆ is derived from the error propagation formula. Horizontal splicing and Formed by vertical splicing.

[0038] In the formula, , , This indicates the nominal position of the robot's end effector in the robot's coordinate system. , , This indicates the actual position of the robot's end effector in the robot's coordinate system. , , These are the error mapping row vectors corresponding to the three position error components x, y, and z in the robot's end effector coordinate system. They are obtained by combining the error transfer matrix of each link coordinate system and the link geometric parameter error mapping matrix. , , This is used to describe the influence of errors in various geometric parameters of the robot on the three directional components of the end-effector position error; among them... Multiplying this by the error parameter vector yields the position error in the x-direction of the final coordinate system. Multiplying this by the error parameter vector yields the position error in the y-direction of the final coordinate system. Multiplying the error by the error parameter vector yields the position error in the z-direction of the final coordinate system: , , .

[0039] Kinematic modeling section: The kinematic coordinate system of the industrial robot and its tools is established according to the MDH method. The spatial pose relationship between the tool coordinate system and the industrial robot body conforms to the MDH standard, and the pose of the tool coordinate system relative to the (n-1)th joint of the industrial robot can be described by the (n-1)th joint of the industrial robot using the MDH method (n is the number of degrees of freedom of the industrial robot).

[0040] The industrial robot MDH parameters of the coupling tool parameters are extracted according to the kinematic coordinate system. The number of industrial robot MDH parameter sets of the coupling tool parameters is consistent with the number of joints of the industrial robot.

[0041] This step embeds the installation error of calibration tool 3 into the entire kinematic model, laying the foundation for subsequent unified identification.

[0042] Step 2: Automated Data Acquisition In this embodiment, 80 points are collected on each plane, and a total of 240 sets of data are collected on the three planes. The entire process requires no manual intervention.

[0043] Step 3: Construct the overdetermined error equation system The end-effector position error of an industrial robot 6 can be expressed in the end-effector coordinate system as: In the formula, the 3×30 matrix With a 30×1 matrix From step 1 respectively Horizontal splicing and Formed by vertical splicing: , , .

[0044] In the formula, , , These are the error mapping row vectors corresponding to the three position error components x, y, and z in the robot's end effector coordinate system. They are obtained by combining the error transfer matrix of each link coordinate system and the link geometric parameter error mapping matrix. , , This is used to describe the influence of errors in various geometric parameters of the robot on the three directional components of the end-effector position error; among them... Multiplying this by the error parameter vector yields the position error in the x-direction of the final coordinate system. Multiplying this by the error parameter vector yields the position error in the y-direction of the final coordinate system. Multiplying the error by the error parameter vector yields the position error in the z-direction of the final coordinate system.

[0045] The above positional errors are those in the end-effector coordinate system of the industrial robot 6. Therefore, the actual position of the end-effector coordinate system of the industrial robot 6 in the base coordinate system can be expressed as: In the calibration method for planar constraints, the nominal position coordinates of the measuring points can be used to fit the nominal position equation of the measured plane in the 6-base coordinate system of the industrial robot using the linear least squares method. The nominal plane parameters of the plane were obtained. a , b , c However, due to the existence of errors, the equation of the actual plane is: Substituting the expression for the actual end position into the actual plane equation after introducing parameter errors, and discarding higher-order infinitesimals, we obtain the following system of equations: .

[0046] By substituting the 240 sets of point data from the three planes obtained through automated data acquisition in step 2 into the linear equations, a set of three-plane joint calibration linear overdetermined equations can be formed.

[0047] Step 4: Error Parameter Identification and Model Correction To verify the accuracy of the error model and identification algorithm, numerical simulation was performed before identification. By pre-setting known error parameters, the nominal error calculated by the model was compared with the actual error calculated by forward kinematics. The results showed a high degree of agreement. Figure 4 As shown, the correctness of the model is proven.

[0048] The Levenberg-Marquardt (LM) nonlinear optimization algorithm was used to solve the above overdetermined equations. Through the solution, all error parameters were accurately identified. These identified error parameters were then compensated into the nominal MDH parameters, resulting in the corrected kinematic model of the industrial robot 6, as shown in Table 2.

[0049] Table 2 Step 5: Machining point conversion based on virtual tool coordinates After calibration, there is no need to disassemble the calibration tool 3; only the micro-operation actuator 5 needs to be installed.

[0050] This invention proposes the concept of virtual tool coordinates. When performing error compensation for coupled tool coordinates, the tool coordinate system is the center of the sphere of the calibration tool 3 (virtual tool coordinates); however, during actual operation, the tool coordinate system is the actual point of action of the micro-manipulation actuator 5 (such as the vacuum suction head end). A fixed spatial pose transformation relationship exists between these two coordinate points. TBefore actual operation, the micro-manipulator 5 is tested using high-precision measuring equipment (such as line laser scanning) to accurately determine the spatial pose transformation relationship between the actual point of action of the micro-manipulator 5 (such as the vacuum suction head end) and the end of the calibration tool 3. T .

[0051] When planning the operation path, a series of target points are first generated for the action point of the micro-operator 5. ,in, cell_x = [ cell_x.x , cell_x.y , cell_x.z ]^T, meaning The X-axis direction is described using the workpiece coordinate system. cell_y = [ cell_y.x , cell_y.y , cell_y.z ]^T, meaning The Y-axis direction is described using the workpiece coordinate system. cell_z = [ cell_z.x , cell_z.y, cell_z.z ]^T, meaning The Z-axis direction is described using the workpiece coordinate system. center = [ center.x , center.y , center.z ]^T, meaning The position is described using the workpiece coordinate system. cell_x.x This represents the component of the X-axis in the target point coordinate system along the Y-axis in the workpiece coordinate system, and so on. Then, through spatial pose transformation... T Convert all target points to points corresponding to the coordinates of the virtual tool. P n Spatial pose transformation T The transformation matrix is ​​4×4 homogeneous, and its specific values ​​are confirmed by a high-precision 3D scanner: First, the micromanipulator 5 and the calibration tool 3 are scanned using a high-precision 3D scanner to obtain the positional relationship between the center of the ball at the end of the calibration tool 3 and the vacuum adsorption port of the micromanipulator 5. Based on this positional relationship, the second transformation matrix is ​​constructed: There are transformation methods: = Step 6: High-precision micro-operation execution In actual operation, the industrial robot moves the virtual tool to the virtual tool location according to the original parameter model (Table 1). P nThen, the joint values ​​are read and substituted into the actual parameter model (Table 2). The actual position of the virtual tool is calculated using the MDH method. The virtual tool points are then... P n Subtracting the actual position from the nominal position, we obtain the minute difference between the nominal and actual positions of the virtual tool in the industrial robot's coordinate system. Finally, the robot performs a secondary movement based on this minute position difference to compensate for the error and accurately move the virtual tool to the preset position. When the industrial robot 6 precisely moves the virtual tool's coordinates to... P n At that time, the actual point of action of the micro-operator 5 at its end will be precisely located at the originally planned operating position. P r Above. Industrial robot 6 sequentially executes commands such as adsorption, movement, and placement to complete high-precision target picking and placement operations.

[0052] This invention provides experimental verification of the above-mentioned embodiment of the precision compensation method and device for industrial robots oriented towards precision micro-operations: Numerical simulation verification of the kinematic parameter error calibration effect of industrial robot 6: Figure 5 The spatial distribution of 240 points collected on three orthogonal planes before calibration and their fitted planes are shown. It can be seen that, due to the precision error of the industrial robot 6, the points that should nominally be on the same plane are actually very scattered.

[0053] Figure 6 The diagram shows the recalculated point distribution after correcting the same 240 sets of joint angle data using the identified error parameters. It is clear that the corrected points converge tightly to their respective fitting planes, indicating a significant improvement in the absolute positioning accuracy of the industrial robot 6.

[0054] Verification of the absolute positioning accuracy of industrial robot 6 equipped with calibration tool 3 in the workpiece coordinate system: To further quantify the effect of the improved accuracy, we conducted a random pose detection experiment based on Latin hypercube sampling (LHS). A workpiece coordinate system was established using the 3-2-1 six-point method based on the three orthogonal planes of calibration block 2. Fifteen test points with random positions and orientations were generated on each plane using Latin hypercube sampling, resulting in a total of 45 sets of points. Subsequently, a high-precision feeler gauge and micrometer (accuracy 0.001 mm) were used to measure the actual distance from the end effector 3 of the industrial robot 6 to the plane. Finally, the actual distance was compared with a preset distance to obtain the absolute positioning error of the end effector 3 of the robot 11 in the normal direction of the plane (workpiece coordinate system Z / Y / X directions) during this measurement. The experimental results are as follows: Figure 7As shown, the absolute positioning error of industrial robot 6 before compensation was around ±0.7mm, and after compensation, its error stabilized within ±0.05mm, improving the accuracy by an order of magnitude.

[0055] In summary, this invention, through its innovative technical solution, effectively overcomes the application obstacles of industrial robots in the field of high-precision micro-manipulation, and provides a low-cost, high-precision, and highly practical solution for precision operation tasks such as ICF target assembly.

[0056] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A precision compensation method for industrial robots designed for precision micro-manipulation, characterized in that, The method includes: Establish a differential kinematic error model for an industrial robot with a coupled tool coordinate system; The end-effector pose and corresponding joint angle data of the industrial robot are automatically collected using a probe and a planar calibration block; and the theoretical position of the end-effector of the calibration tool is calculated based on the joint angle data. Based on the theoretical position, the nominal plane equation of the calibration plane is fitted, and combined with the differential kinematic error model of the industrial robot, an overdetermined set of equations is constructed, which includes the kinematic parameter error of the industrial robot and the plane equation parameter error. Solve the overdetermined system of equations to identify the kinematic parameter errors of the industrial robot and obtain the corrected kinematic model of the industrial robot. Based on the virtual tool coordinate method, the action point of the micro-operator is converted into the virtual probe point. The industrial robot moves according to the virtual probe position, and the error is compensated by the corrected industrial robot kinematic model to complete the micro-operation.

2. The method according to claim 1, characterized in that, The method for establishing a differential kinematic error model of an industrial robot with a coupled tool coordinate system consists of three parts: Error propagation formula part: ； in, This refers to the small pose error of the robot's end effector in the end-effector coordinate system. That is, the coordinate system { i The small pose error transfer matrix from coordinate system {6} to coordinate system {6}. ,in, , , Let X be the direction cosine of the unit vector along the X-axis of the target coordinate system and the X, Y, and Z axes of the reference coordinate system. , , Let be the direction cosines of the unit vector along the Y-axis of the target coordinate system and the X, Y, and Z axes of the reference coordinate system. , , Let Z be the direction cosines of the unit vector of the target coordinate system's Z-axis and the X, Y, and Z axes of the reference coordinate system. , , This represents the position of the origin of the target coordinate system in the reference coordinate system. Let be a homogeneous transformation matrix, when hour, ,when hour, It is the identity matrix; Link i The mapping matrix from geometric parameter errors to small pose errors in the link coordinate system. That is , That is , Link i The DH parameter represents the distance along the link. i X of the coordinate system i Direction, Link i Z in the coordinate system of -1 i -1 and connecting rod i Z of the coordinate system i The distance between them Link i The DH parameter represents the distance along the link. i X of the coordinate system i Looking at the direction, the connecting rod i Z in the coordinate system of -1 i -1 and connecting rod i Z of the coordinate system i The angle between them These are the rotational parameters about the corresponding axes when the axes of the two joints are parallel. Link i Geometric parameter errors; Error transformation part of the physical constraint-oriented method: ； in, , , This indicates the nominal position of the robot's end effector in the robot's coordinate system. , , This represents the actual position of the robot's end effector in the robot's coordinate system. For a 6-DOF industrial robot, a 3×30 matrix represents this position. The 30×1 matrix ∆ is derived from the error propagation formula. Horizontal splicing and Formed by vertical splicing, , , These are the error mapping row vectors corresponding to the three position error components x, y, and z in the robot's end-effector coordinate system; Kinematic modeling section: The kinematic coordinate system of the industrial robot and its tool is established according to the MDH method. The spatial pose relationship between the tool coordinate system and the industrial robot body conforms to the MDH standard, and the tool coordinate system is relative to the first kinematic coordinate system of the industrial robot. n The pose of joint -1 is determined by the industrial robot's first joint. n -1 joints are described using the MDH method, where... n This refers to the number of degrees of freedom of an industrial robot. According to the kinematic coordinate system, the industrial robot MDH parameters of the coupling tool parameters are extracted. The number of industrial robot MDH parameter groups of the coupling tool parameters is consistent with the number of joints of the industrial robot.

3. The method according to claim 2, characterized in that, The probe and planar calibration block are used to automatically collect the end-effector pose of the industrial robot and its corresponding joint angle data. The method for calculating the theoretical position of the calibration tool tip based on the joint angle data includes: An industrial robot carrying a calibration tool automatically approaches and touches three mutually orthogonal planes of a calibration block in various pre-set postures and positions. The moment the calibration tool touches the plane, a signal is sent back to the industrial robot, which immediately stops moving. The robot obtains the joint angle data of its six joints at the corresponding moment and calculates the theoretical position of the end of the calibration tool.

4. The method according to claim 3, characterized in that, Methods for constructing an overdetermined system of equations that includes kinematic parameter errors and planar equation parameter errors of industrial robots include: Using the theoretical position, the nominal plane equation of the calibration plane in the industrial robot base coordinate system is fitted, and the plane equation parameter error is introduced to describe the actual plane equation. The expression for the actual end position derived from the differential kinematic error model of the industrial robot is substituted into the actual plane equation after introducing parameter errors, and linearized by discarding higher-order infinitesimals, thereby obtaining the overdetermined system of equations.

5. The method according to claim 4, characterized in that, The actual plane equation after introducing parameter errors is: ； in, a , b , c These are the nominal plane parameters; Substituting the expression for the actual end position into the actual plane equation after introducing parameter errors, and discarding higher-order infinitesimals, we obtain the following system of equations: 。 6. The method according to claim 1, characterized in that, The methods for converting the action point of a micro-manipulator into a virtual probe point based on the virtual tool coordinate method include: When planning the operation path, a series of target points are generated for the action point of the micro-operator. P r ; Through spatial pose transformation T Convert all target points to points corresponding to the coordinates of the virtual tool. P n The virtual tool coordinates are the center of the sphere of the calibration tool.

7. The method according to claim 1, characterized in that, The method of using a modified industrial robot kinematic model to compensate for errors and complete micro-operations by moving the industrial robot according to the virtual probe position includes: In actual operation, the control and computing unit sends the converted virtual tool points to the industrial robot controller. P n The modified kinematic model of the industrial robot is used to calculate the small position error of the end effector of the industrial robot tool, and to drive the industrial robot to perform motion and compensation motion. When the industrial robot moves the virtual tool coordinates to P n At that time, the actual point of action of the micro-operator at its end will be located at the originally planned operating position. P r On the surface, the industrial robot sequentially executes the instructions for adsorption, movement, and placement to complete the target ball picking and placement operations.

8. A precision compensation device for industrial robots designed for precision micro-operations, the device being used to implement the method described in any one of claims 1-7, characterized in that, The device includes: A six-degree-of-freedom industrial robot; A connector is fixedly mounted on the end flange of the industrial robot, and the connector is used to jointly support the calibration tool and the micro-manipulation actuator; A calibration tool that is detachably mounted on the connector; A micro-operation actuator detachably mounted on the connector; A calibration block.

Images