Measurement method for robot error calibration considering thermal error real-time dynamic compensation

Abstract

The invention discloses a robot error calibration measuring method considering real-time dynamic compensation of thermal errors, which comprises the following steps of: simultaneously measuring a standard component unit and a workpiece unit to be measured in the same space through a combined measuring system unit arranged on a robot, processing point cloud data of the standard component unit in measured point cloud, and calculating the point cloud data of the standard component unit; kinematics parameter calibration of the robot unit is realized; and dynamically compensating the parameter error into the kinematic model of the robot unit, and processing the data of the to-be-measured workpiece in the same space according to the compensated kinematic parameters of the robot unit, thereby realizing the real-time dynamic compensation of the kinematic parameters of the robot unit and the measurement of the to-be-measured workpiece unit. The standard component and the workpieces to be measured are scanned at the same time through the composite measuring system, it is guaranteed that the temperature of the robot is consistent with other factors when the standard component and the workpieces to be measured are measured, and the purpose of real-time dynamic compensation of kinematics parameters of the robot when each workpiece to be measured is measured is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of precision measurement, and particularly relates to a measurement method for robot error calibration with real-time dynamic compensation considering thermal errors. Background Art

[0002] With the continuous development of intelligent manufacturing, the demand for robot accuracy is also constantly increasing. Robots are required to have higher precision, flexibility and adaptive capabilities to adapt to increasingly complex and high-precision production tasks. However, when the ambient temperature or the temperature of the robot's own motor changes during long-term operation, the structural parameters of the robot change, resulting in the invalidation of the previous robot error calibration that did not consider temperature changes and unable to guarantee the current accuracy requirements of the robot. Different from the compensation of robot geometric errors and flexible deformation errors, the errors caused by the influence of temperature on the robot have not been widely studied.

[0003] The existing research on the errors caused by the influence of temperature on robots includes:

[0004] Poonyapak et al. proposed a model for predicting the thermal errors of industrial robots using a thermal imaging camera.

[0005] Lubrano et al. fixed 11 platinum resistance thermometers on an interferometer and a robot to measure the thermal errors of the robot.

[0006] Li et al. used the finite element theory to create a thermal distribution and deformation model of the robot system to measure the thermal errors of the robot.

[0007] Jones et al. provided a method for determining the optimal robot path, which can minimize the thermal changes on the surface during the movement of the robot. The high-precision online measurement system proposed by Mekaman has a temperature drift compensation function. By measuring a standard ball placed in a small space outside the space of the workpiece to be measured, the robot calibration work is carried out.

[0008] However, several of the above methods either require a costly thermal imaging camera or need to install additional sensors in the working space. Due to the cumbersome installation of sensors, most methods are difficult to continuously measure on-site, or cannot achieve the characteristics of universality and intelligence from the perspective of practicality; or only measure and calibrate the robot at different angles in a small space to indirectly compensate the space of the workpiece to be measured, without directly compensating in the space of the workpiece to be measured.

[0009] Therefore, there is an urgent need to propose a robot error calibration method considering real-time dynamic compensation of thermal errors to solve the problems that the environmental temperature changes and the self-temperature of the robot changes due to the long-term operation of its own motors, resulting in changes in its structural parameters, rendering the previous robot error calibration without considering temperature changes invalid and unable to ensure the accuracy requirements for current robot use.

[0010] Based on the above deficiencies existing in the prior art, the present invention is proposed. Summary of the Invention

[0011] The object of the present invention is to provide a measurement method for robot error calibration considering real-time dynamic compensation of thermal errors to solve the problem that the thermal errors of existing robots affect the positioning accuracy of robots.

[0012] To achieve the above object of the invention, the technical solution of the present invention is: a measurement method for robot error calibration considering real-time dynamic compensation of thermal errors, which simultaneously measures a standard part unit and a workpiece to be measured in the same space through a composite measurement system unit set on the robot, and processes the point cloud data of the standard part unit in the measured point cloud to calibrate the kinematic parameters of the robot unit; and dynamically compensates the parameter errors into the kinematic model of the robot unit, and processes the data of the workpiece to be measured in the same space according to the kinematic parameters after compensation of the robot unit, so as to realize real-time dynamic compensation of the kinematic parameters of the robot unit and measurement of the workpiece to be measured unit; the specific steps are as follows:

[0013] Step S1, build a measurement system, and complete the calibration of the probe system and the hand-eye calibration of the robot.

[0014] The measurement system includes a robot unit, a composite measurement system unit installed on the robot unit, a standard part unit and a workpiece to be measured unit; among them: the composite measurement system unit includes a vision measurement system composed of a binocular camera and a cross-structured light laser; the standard part unit includes a rod, a base for fixing the rod, and a plurality of standard balls arranged on the rod.

[0015] The composite measurement system unit is fixed at the end of the robot unit, and the standard part unit and the workpiece to be measured unit are fixed within the scanning range of the vision system unit of the composite measurement system.

[0016] The radius value of the standard ball and the distance between any two ball centers have been calibrated in advance with a coordinate measuring machine as the theoretical value.

[0017] The standard part unit is fixed in the placement space of the workpiece to be measured unit, and only the workpiece to be measured unit is replaced during measurement.

[0018] After the composite measurement system is installed, calibrate the probe system and perform hand-eye calibration.

[0019] Step S2, establish a robot unit model and an error calibration model based on distance;

[0020] The objective function f1 of the error calibration model based on distance is expressed as:

[0021]

[0022] where L Tmk represents the spatial distance between the theoretical positions of the m-th and k-th standard ball centers, and L Amk represents the actual spatial distance between the positions of the m-th and k-th standard ball centers measured by the composite measurement system unit;

[0023] θ represents the joint rotation angle of the robot unit's rotating joints during actual measurement of two standard balls, and Δε represents the kinematic parameter errors of the robot unit, including Δθ i , Δd i , Δa i , Δα i and Δβ i ;

[0024] θ i represents the joint variable; d i represents the link offset; a i represents the link length; α i represents the link twist angle; β i is the rotation variable around the Y-axis;

[0025] Step S3, the composite measurement system measures the standard part and the workpiece to be measured simultaneously;

[0026] The composite measurement system unit measures the standard part unit and the workpiece to be measured simultaneously each time. During the measurement process, the light strip emitted by the cross-structured light laser irradiates on the standard part unit or the workpiece to be measured, and the camera takes pictures of the light strip to extract the center of the light strip, obtaining the coordinates of the light strip center in the camera coordinate system; the visual measurement system performs hand-eye calibration on the robot unit and the visual measurement system, and converts the measured coordinates in the camera coordinate system to the base coordinate system of the robot unit; the robot unit drives the visual measurement system to realize the measurement of the standard part unit and the workpiece to be measured;

[0027] Step S4, calibrate the kinematic parameters of the robot according to the measured point cloud of the standard part;

[0028] Process the measured point cloud data, use the point cloud data of the measured standard part unit to fit the ball center and radius, and calculate the radius value of the standard ball and the distance between any two ball centers;

[0029] Compare with the radius value of the standard sphere measured by the coordinate measuring machine and the theoretical value of the distance between any two sphere centers, and use the LM algorithm to minimize this error function f1 to obtain the kinematic parameter error Δε of the robot, and complete the error calibration;

[0030] Step S5, dynamically compensate and update the kinematic parameter error of the robot, and calculate the compensated point cloud of the workpiece to be measured to achieve the purpose of measuring the workpiece;

[0031] Substitute the calibrated kinematic parameter error Δε of the robot into the corresponding parameters of the robot kinematic model to dynamically update the kinematic parameters of the robot;

[0032] Process the data of the workpiece unit to be measured in this space according to the compensated kinematic parameters of the robot unit, so as to achieve the purpose of real-time dynamic compensation of the kinematic parameters and measurement of the workpiece unit to be measured.

[0033] A further preferred technical solution provided by the present invention is:

[0034] It further includes: Step S6, replace the workpiece to be measured for measurement;

[0035] Step S7, determine whether all workpieces have been measured. If so, all completed, then proceed to the next step S8; if not, the measurement is not completed, then update the workpiece to be measured and return to step S3 for measurement;

[0036] Step S8, complete the measurement to obtain the measurement information of all workpieces to be measured of the batch of workpieces to be measured.

[0037] An even further preferred technical solution provided by the present invention is:

[0038] A trigger measurement system is also installed on the robot. The robot is calibrated by measuring the standard part through the trigger measurement system, which is used to supplement the visual measurement system to measure the workpieces to be measured with deep holes or other shapes that are difficult to measure after the robot calibration is completed.

[0039] Another preferred technical solution provided by the present invention is:

[0040] In the step S2, the specific steps of building the error calibration model based on distance are as follows:

[0041] According to the MDH model (MDH is the abbreviation of Modified Denavit-Hartenberg, adding a rotation variable β around the Y axis to the basic robot DH model i to overcome the singularity problem when adjacent joints of the robot are parallel.), the transformation relationship between the (i - 1)-th joint and the i-th joint of the robot unit is expressed as:

[0042]

[0043] Among them, θ i represents the joint variable; d i represents the link offset; a i represents the link length; α i represents the link twist angle; β i is the rotation variable about the Y-axis, cθ i represents the cosine of θ i cosθ i , cα i represents the cosine of α i cosα i , sθ i represents the sine of θ i sinθ i , sα i represents the sine of α i sinα i , Rot represents rotational motion, and Trans represents translational motion;

[0044] The end position P(θ) of the robot unit can be expressed as:

[0045]

[0046] Among them, j is the total number of robot joints, and i is the i-th joint; [p x , p y , p z represents the three-dimensional coordinates of the end position P(θ);

[0047] Performing total differentiation on , the transformation matrix error between each joint is expressed by Equation (3) as:

[0048]

[0049] Among them, Δθ i , Δd i , Δa i , Δα i and Δβ i are the corresponding errors of the kinematic parameters θ i , d i , a i , α i and β i ;

[0050] The position P1 of the vision measurement system in the coordinate system of the robot unit is expressed as:

[0051]

[0052] Among them, The matrix for hand-eye calibration of the robot unit; j is the total number of robot joints, and i is the i-th joint; [x1, y1, z1] is the point cloud position coordinates of the standard part or the workpiece to be measured measured by the vision measurement system.

[0053] The theoretical position coordinates of the m-th and k-th standard ball centers based on distance are respectively denoted as P Tm (x Tm , y Tm , z Tm ) and P Tk (x Tk , y Tk , z Tk ), then the spatial distance L between the theoretical positions of these two standard ball centers Tmk is expressed as:

[0054]

[0055] The position errors of the m-th and k-th standard ball centers are respectively ΔP m and ΔP k . After fitting the measured data, the actual position coordinates of the m-th and k-th standard ball centers measured by the composite measurement system unit are respectively P Am (x Am , y Am , z Am ) and P Ak (x Ak , y Ak , z Ak ), then the spatial distance L between the actual positions of these two standard ball centers Amk is expressed as:

[0056]

[0057] The objective function f1 of the error calibration model based on distance is expressed as:

[0058]

[0059] where N represents the number of squares of the total distance differences, q represents the q-th square of the distance difference; θ represents the joint angles of the six rotational joints of the robot unit 1 at two positions; Δε represents the kinematic parameter errors of the robot unit Δθ i , Δd i , Δa i , Δα i and Δβ i .

[0060] Preferably, the lengths of the rods on the standard part and the distribution of the standard balls are designed according to the shape and size of the workpiece to be measured.

[0061] The present invention also provides a technical solution for the application of a measurement method for robot error calibration considering real-time dynamic compensation of thermal error in machine tool calibration.

[0062] Compared with the prior art, the present invention has the following advantages:

[0063] 1) The materials and manufacturing device costs of the robot error calibration method considering real-time dynamic compensation of thermal error according to the present invention are much lower than those of instruments and equipment such as thermal imaging cameras; at the same time, the composite measurement system is easy to install and can be directly installed on the robot, avoiding the cumbersome installation of sensors and the difficulty of continuous measurement on site; by simultaneously measuring the standard part and the workpiece to be measured, real-time dynamic compensation of the robot thermal error can be achieved.

[0064] 2) Compared with the problems of complex modeling and difficult measurement in the traditional complex temperature model, the robot error calibration method considering real-time dynamic compensation of thermal error according to the present invention ensures that the temperature and other factors of the robot are the same when measuring the standard part and the workpiece to be measured through the simultaneous scanning of the standard part and the workpiece to be measured by the composite measurement system. At the same time, the standard part and the workpiece to be measured are in the same space, and the direct compensation effect is good. Using the data of the standard part for calibration and compensating the calibrated robot kinematic model data into the corresponding parameters, the purpose of real-time dynamic compensation of the robot kinematic parameters during the measurement of each workpiece to be measured is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The drawings herein are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present invention and, together with the specification, are used to explain the principles of the present invention.

[0066] Figure 1 It is a schematic structural diagram of the measurement system of the present invention;

[0067] Figure 2 It is a schematic measurement flow diagram of the present invention;

[0068] Figure 3 It is a schematic diagram of an error calibration model based on distance. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings.

[0070] Figure 1 It is a schematic structural diagram of the measurement system of the present invention, Figure 2 It is a schematic measurement flow diagram of the present invention, Figure 3 It is a schematic diagram of an error calibration model based on distance. In combination with the attached Figure 1 、 Figure 2 、 Figure 3As shown in [figure], an embodiment of the present invention provides a measurement method for robot error calibration considering real-time dynamic compensation of thermal error. Through a composite measurement system unit set on the robot, the standard part unit and the workpiece to be measured in the same space are measured simultaneously. The point cloud data of the standard part unit in the measured point cloud is processed to realize the kinematic parameter calibration of the robot unit; and the parameter error is dynamically compensated into the kinematic model of the robot unit, and the data of the workpiece to be measured in the same space is processed according to the compensated kinematic parameters of the robot unit, so as to realize the real-time dynamic compensation of the kinematic parameters of the robot unit and the measurement of the workpiece to be measured unit.

[0071] The specific steps are introduced as follows:

[0072] The schematic diagram of the measurement process of the present invention is as shown in Figure 2 shown.

[0073] Step S1, measurement system setup.

[0074] Install the composite measurement system and complete the calibration of the probe system and the robot hand-eye calibration

[0075] The structure of the measurement system includes Figure 1 the robot unit 1 shown in [figure], the composite measurement system unit 2 installed on the robot unit 1, the standard part unit 3 and the workpiece to be measured unit 4; among them: the composite measurement system unit 2 includes a vision measurement system composed of a binocular camera 201 and a cross-structured light laser 202, and a trigger measurement system 203; the standard part unit 3 includes a rod, a base 301 for fixing the rod 302, and a plurality of standard balls 303 arranged on the rod 302;

[0076] The composite measurement system unit 2 is fixed at the end of the robot unit 1, and the standard part unit 3 and the workpiece to be measured unit 4 are fixed within the scanning range of the vision measurement system unit of the composite measurement system 2;

[0077] The radius value of the standard ball 303 and the distance between any two ball centers have been calibrated in advance with a coordinate measuring machine as the theoretical value;

[0078] The length of the rod 302 on the standard part unit 3 and the distribution of the standard balls 303 can be designed according to the shape and size of the workpiece to be measured unit 4, so that it is placed in the placement space of the workpiece to be measured unit 4, ensuring that it does not move, and only the workpiece to be measured unit 4 is replaced during measurement.

[0079] After the composite measurement system is installed, use the method disclosed in the invention patent with the publication number CN 118559755A and the invention name "A Robot Calibration Method Based on Multi-line Structured Light" applied by the applicant on June 13, 2024 to calibrate the probe system and the hand-eye calibration. Details will not be repeated here.

[0080] Step S2: Establish a robot model and an error model based on distance.

[0081] According to the MDH model (MDH is the abbreviation of Modified Denavit-Hartenberg, and a rotational variable β about the Y-axis is added to the basic robot DH model i to overcome the singularity problem when adjacent joints of the robot are parallel. ), the transformation relationship between the (i - 1)-th joint and the i-th joint of the robot unit 1 is expressed as:

[0082]

[0083] where, θ i represents the joint variable; d i represents the link offset; a i represents the link length; α i represents the link twist angle; when the axes of adjacent joints in the DH model are parallel or approximately parallel, a small angle will cause significant changes in other kinematic parameters. To overcome this problem, a rotational variable β about the Y-axis is added when the axes of adjacent joints are parallel or approximately parallel i . cθ i represents the cosine cosθ of θ i , cα i represents the cosine cosα of α i ; sθ i represents the sine sinθ of θ i ; sα i represents the sine sinα of α i . Rot represents rotational motion, and Trans represents translational motion. i i i i x y z In this embodiment, a six-degree-of-freedom robot unit 1 is selected, and its end position P(θ) can be expressed as:

[0084]

[0085]

[0086] where, [p x , p y , p z represents the three-dimensional coordinates of the end position P(θ).

[0087] Due to the influence of errors such as machining and assembly on the robot unit 1, there is a certain error between the actual value and the theoretical value of the kinematic parameters of the robot unit 1. Performing total differentiation on , the transformation matrix error between each joint is represented by Equation (3).​

[0088]

[0089] Among them, Δθ i , Δd i , Δa i , Δα i and Δβ i are the corresponding errors of the kinematic parameters θ i , d i , a i , α i and β i respectively. Therefore, the position P1 of the vision measurement system and the position P2 measured by the trigger measurement system 203 in the coordinate system of the robot unit 1 can be further expressed as:

[0090]

[0091] Among them, is the matrix of the hand-eye calibration of the robot unit 1. [x1, y1, z1] is the point cloud position coordinates of the standard part or the workpiece to be measured measured by the vision measurement system.

[0092] As Figure 3 shown, the theoretical position coordinates of the m-th and k-th standard ball centers based on the distance are respectively denoted as P Tm (x Tm , y Tm , z Tm ) and P Tk (x Tk , y Tk , z Tk ). Then, the spatial distance L Tmk between the theoretical positions of these two standard ball centers is expressed as:

[0093]

[0094] Due to the influence of the kinematic parameter errors of the robot unit 1, there is an error between the actual position and the theoretical specified position. The position errors of the m-th and k-th standard ball centers are ΔP m and ΔP k respectively. After fitting the measured data, the spatial distance L Amk between the above two positions measured by the composite measurement system unit 2 is expressed as:

[0095]

[0096] The objective function f1 of the error calibration model based on the distance is expressed as:

[0097]

[0098] Wherein, N represents the number of squares of the total distance difference, q represents the square of the q-th distance difference; θ represents the joint angles of the six rotating joints of the robot unit 1 at two positions. Δε represents the kinematic parameter error Δθ of the robot unit 1 i , Δd i , Δa i , Δα i and Δβ i .

[0099] Step S3, the composite measurement system measures the standard part and the workpiece to be measured simultaneously.

[0100] The composite measurement system unit 2 is used to measure the standard part unit 3 and the workpiece to be measured unit 4 simultaneously each time. During the measurement process, the light strip emitted by the cross-structured light laser 202 irradiates on the standard part unit 3 or the workpiece to be measured unit 4, and the camera takes pictures of the light strip to extract the center of the light strip, obtaining the coordinates of the center of the light strip in the coordinate system of the camera 201. The method proposed in the patent with the publication number CN 118559755 A applied by the applicant of the vision measurement system performs hand-eye calibration on the robot unit 1 and the vision measurement system, and can convert the measured coordinates in the coordinate system of the camera 201 to the base coordinate system of the robot unit 201. By driving the vision measurement system with the robot unit 201, the measurement of the standard part unit 3 and the workpiece to be measured unit 4 can be realized.

[0101] Step S4, calibrate the robot kinematic parameters according to the measured point cloud of the standard part.

[0102] Process the measured point cloud data, use the point cloud data of the measured standard part unit 3 to fit the sphere center and radius, and calculate the radius value of the standard sphere 303 and the distance between any two sphere centers.

[0103] And compare it with the theoretical values of the radius value of the standard sphere 303 and the distance between any two sphere centers measured by the coordinate measuring machine, and use the LM algorithm to minimize this error function f1 to obtain the robot kinematic parameter error Δε.

[0104] Step S5, dynamically compensate and update the robot kinematic parameter error, and calculate the compensated point cloud of the workpiece to achieve the purpose of measuring the workpiece.

[0105] Substitute the calibrated robot kinematic parameter error Δε into the corresponding parameters of the robot kinematic model to dynamically update the robot kinematic parameters.

[0106] Finally, process the data of the workpiece to be measured unit 4 in this space according to the compensated kinematic parameters of the robot unit 1, so as to achieve the purpose of real-time dynamic compensation of the kinematic parameters and measurement of the workpiece to be measured unit 4.

[0107] The simultaneous scanning of the standard part unit 3 and the workpiece to be measured unit 4 by the composite measurement system unit 2 ensures that the temperature and other factors of the robot unit 1 are the same when measuring both. At the same time, the standard part unit 3 and the workpiece to be measured unit 4 are in the same space, and the direct compensation effect is good. The data of the standard part unit 3 is used for calibration, and the calibrated kinematic model data of the robot unit 1 is compensated into the corresponding parameters, so as to achieve the purpose of real-time dynamic compensation of the kinematic parameters of the robot unit 1 when measuring each workpiece to be measured unit 4.

[0108] Step S6, replace the workpiece to be measured for measurement.

[0109] Step S7, determine whether all workpieces have been measured? If all are completed, proceed to the next step S8. If the measurement is not completed, update the workpiece to be measured and return to step S3 for measurement.

[0110] Step S8, complete the measurement and obtain the measurement information of this batch of workpieces to be measured.

[0111] In this embodiment, the trigger measurement system 203 is similar to the vision measurement system and can also be used for robot kinematic parameter calibration. The standard part unit 3 is measured by the trigger probe 203 and compared with the theoretical value measured by the coordinate measuring machine. The LM algorithm is used to minimize this error function f1 to achieve the purpose of robot calibration. Although the trigger measurement system 203 cannot perform real-time calibration of robot kinematic parameters like the vision measurement system (that is, it cannot compensate for thermal errors in real time dynamically). However, after it completes the robot kinematic parameter calibration, it supplements some occasions where the vision measurement system is difficult to measure deep holes or other shapes and requires high precision during the measurement process.

[0112] The standard part unit 3 is installed at the spatial position of the workpiece to be measured unit 4. The vision measurement system driven by the robot unit 201 can realize the simultaneous measurement of the standard part unit 3 and the workpiece to be measured unit 4. Since it is a simultaneous measurement, the thermal error of the robot unit 1 when measuring both is the same. The data of the measured standard part unit 3 is used for iterative optimization to obtain the error value Δε of the robot kinematic parameters, which is compensated into the robot model to calculate the coordinates of the light strip part of the workpiece to be measured unit 4 to achieve real-time compensation of the robot thermal error. Since the data of the standard part unit 3 and the workpiece to be measured unit 4 can be obtained simultaneously in each measurement, through the standard part unit 3 measured each time, the robot kinematic parameter error Δε obtained by using the LM algorithm for parameter iterative optimization can be compensated into the measurement of the workpiece to be measured unit 4, so as to achieve the purpose of real-time dynamic update and compensation of the robot kinematic parameter error with the change of the robot thermal error.

[0113] At the same time, the measurement method for robot error calibration considering real-time dynamic compensation of thermal error provided by the present invention is not only applicable to robots, but also applicable to the calibration of machine tools.

[0114] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiments can be implemented by means of software plus a hardware platform. Based on such an understanding, the technical solutions of the above embodiments can be embodied in the form of a software product, and this software product can be stored in the NVIDIA development board, so that the entire system constitutes a product, and it is more convenient to execute the methods described in various embodiments of the present invention.

[0115] As mentioned above, the above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A measurement method for robot error calibration considering real-time dynamic compensation of thermal error, characterized in that: By means of a composite measuring system unit arranged on the robot, the standard part unit and the workpiece unit to be measured in the same space are measured simultaneously, and the point cloud data of the standard part unit in the measured point cloud is processed to realize the kinematic parameter calibration of the robot unit; and the parameter error is dynamically compensated to the kinematic model of the robot unit, and the data of the workpiece to be measured in the same space are processed according to the kinematic parameters of the robot unit after compensation, so as to realize the real-time dynamic compensation of the kinematic parameters of the robot unit and the measurement of the workpiece unit to be measured; the specific steps include the following: Step S1, measurement system construction, probe system calibration and robot hand-eye calibration; The measuring system comprises a robot unit, a composite measuring system unit mounted on the robot unit, a standard part unit and a workpiece unit to be measured; wherein: the composite measuring system unit comprises a visual measuring system composed of a binocular camera and a cross-structured light laser; the standard part unit comprises a rod, a base for fixing the rod, and a plurality of standard balls arranged on the rod; The compound measuring system unit is fixed at the end of the robot unit, and the standard part unit and the workpiece unit to be measured are fixed within the scanning range of the compound measuring system visual system unit; The radius of the standard sphere and the distance between any two sphere centers have been calibrated in advance by a three-dimensional coordinate measuring machine as theoretical values; The standard part unit is fixed in the placement space of the workpiece unit to be measured, and only the workpiece unit to be measured is replaced during measurement; After the composite measuring system is installed, the probe system and hand-eye calibration are performed; Step S2, establishing a robot unit model and a distance-based error calibration model; The objective function f1 of the distance-based error calibration model is expressed as: Where, N represents the number of the squares of the total distance differences, q represents the square of the q-th distance difference; L Tmk represents the spatial distance between the theoretical positions of the m-th and k-th standard sphere centers; L Amk represents the actual spatial distance between the positions of the m-th and k-th standard sphere centers measured by the composite measurement system unit; θ represents the joint rotation angle of the robot unit at the time of actually measuring two standard spheres; Δε represents the kinematic parameter error of the robot unit, including Δθ i , Δd i , Δa i , Δα i and Δβ i ; θ i represents the joint variable; d i represents the link offset; a i represents the link length; α i represents the link twist angle; β i is the rotation variable about the Y-axis; Step S3, the composite measuring system measures the standard part and the workpiece to be measured simultaneously; The composite measurement system unit is used to measure the standard part unit and the workpiece unit to be measured at the same time. During the measurement process, the light strip emitted by the cross-structured light laser is irradiated on the standard part unit or the workpiece unit to be measured. The camera takes a picture of the light strip to extract the center of the light strip and obtain the coordinates of the center of the light strip in the camera coordinate system. The visual measurement system performs hand-eye calibration on the robot unit and the visual measurement system, and converts the measured coordinates in the camera coordinate system to the robot unit base coordinate system. The robot unit drives the visual measurement system to realize the measurement of the standard part unit and the workpiece unit to be measured. Step S4, calibrating the robot kinematic parameters according to the measured standard part point cloud; Process the measured point cloud data, use the measured point cloud data of the standard part unit to fit the sphere center and radius, and calculate the radius value of the standard sphere and the distance between any two sphere centers; Compared with the radius of the standard ball measured by the three-dimensional coordinate measuring machine and the theoretical value of the distance between any two ball centers, the LM algorithm is used to minimize this error function f1 to obtain the robot kinematic parameter error Δε, completing the error calibration; Step S5, dynamically compensating and updating the robot kinematic parameter errors, and calculating the compensated point cloud of the workpiece to be measured to achieve the purpose of measuring the workpiece; Substitute the calibrated robot kinematic parameter error Δε into the corresponding parameters of the robot kinematic model to dynamically update the robot kinematic parameters; Process the data of the workpiece unit to be measured in this space according to the kinematic parameters after compensation of the robot unit, so as to achieve the purpose of real-time dynamic compensation of kinematic parameters and measurement of the workpiece unit to be measured.

2. The measurement method for robot error calibration considering real-time dynamic compensation of thermal error according to claim 1, characterized in that It further includes: Step S6, replace the workpiece to be measured for measurement; Step S7, determine whether all workpieces have been measured. If yes, all are completed, then proceed to the next step S8; If no, the measurement is not completed, then update the workpiece to be measured and return to step S3 for measurement; Step S8, complete the measurement to obtain the measurement information of all workpieces to be measured in the batch of workpieces to be measured.

3. A measurement method for robot error calibration considering real-time dynamic compensation of thermal error according to claim 2, characterized in that: A trigger-type measurement system is also installed on the robot. The robot is calibrated by measuring a standard part through the trigger-type measurement system, which is used to supplement the visual measurement system to measure workpieces to be measured with deep holes or other difficult-to-measure shapes after the robot calibration is completed.

4. The measurement method for robot error calibration considering real-time dynamic compensation of thermal error according to claim 1, characterized in that: In the step S2, the modeling steps of the error calibration model based on distance are specifically as follows: According to the MDH model, the transformation relationship T between the (i - 1)-th joint and the i-th joint of the robot unit i i-1 is expressed as: Among them, θ i represents the joint variable; d i represents the link offset; a i represents the link length; α i represents the link twist angle; β i is the rotation variable about the Y-axis, cθ i represents the cosine of θ i cosθ i , cα i represents the cosine of α i cosα i , sθ i represents the sine of θ i sinθ i , sα i represents the sine of α i sinα i , Rot represents rotational motion, and Trans represents translational motion; The end position P(θ) of the robot unit can be expressed as: where j is the total number of robot joints, and i is the i-th joint; [p x , p y , p z represents the three-dimensional coordinates of the end position P(θ); For T i i-1 Performing total differentiation, the transformation matrix error between each joint is expressed by Equation (3) as follows: Among them, Δθ i , Δd i , Δa i , Δα i and Δβ i are the corresponding errors of the kinematic parameters θ i , d i , a i , α i and β i ; The position P1 of the visual measurement system in the coordinate system of the robot unit is expressed as: Among them, T c j is the matrix for hand-eye calibration of the robot unit; j is the total number of robot joints, and i is the i-th joint; [x1, y1, z1] is the point cloud position coordinates of the standard part or the workpiece to be measured measured by the vision measurement system. The theoretical position coordinates of the m-th and k-th standard sphere centers based on distance are respectively denoted as P Tm (x Tm , y Tm , z Tm ) and P Tk (x Tk , y Tk , z Tk ). Then the spatial distance L Tmk between the theoretical positions of these two standard sphere centers is expressed as: The position errors of the m-th and k-th standard ball centers are ΔP m and ΔP k , respectively. After fitting the measured data, the actual position coordinates of the m-th and k-th standard ball centers measured by the composite measurement system unit are P Am (x Am , y Am , z Am ) and P Ak (x Ak , y Ak , z Ak ), respectively. Then the spatial distance L Amk between the actual positions of these two standard ball centers is expressed as: The objective function f1 of the error calibration model based on distance is expressed as: Where N represents the number of the squares of the total distance differences, q represents the square of the q-th distance difference; θ represents the joint angles of the six revolute joints of the robotic unit 1 at two positions; Δε represents the kinematic parameter errors Δθ i , Δd i , Δa i , Δα i and Δβ i .

5. A measurement method for robot error calibration considering real-time dynamic compensation of thermal error according to claim 1, characterized in that: The lengths of the rods and the distribution of the standard balls on the standard part are designed according to the shape and size of the workpiece to be measured.

6. Application of a measurement method for robot error calibration considering real-time dynamic compensation of thermal error as described in any one of claims 1-5 in machine tool calibration.

Images