CN114523475B - Automatic calibration and compensation device and method for errors of robot assembly system - Google Patents
Automatic calibration and compensation device and method for errors of robot assembly system Download PDFInfo
- Publication number
- CN114523475B CN114523475B CN202210195451.XA CN202210195451A CN114523475B CN 114523475 B CN114523475 B CN 114523475B CN 202210195451 A CN202210195451 A CN 202210195451A CN 114523475 B CN114523475 B CN 114523475B
- Authority
- CN
- China
- Prior art keywords
- calibration
- robot
- assembly
- positioning
- calibration block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 79
- 239000011159 matrix material Substances 0.000 claims abstract description 19
- 230000000007 visual effect Effects 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims description 31
- 230000006870 function Effects 0.000 claims description 26
- 230000033001 locomotion Effects 0.000 claims description 11
- 239000013598 vector Substances 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 238000013519 translation Methods 0.000 claims description 2
- 230000006854 communication Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 5
- 238000009434 installation Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1679—Programme controls characterised by the tasks executed
- B25J9/1687—Assembly, peg and hole, palletising, straight line, weaving pattern movement
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/1605—Simulation of manipulator lay-out, design, modelling of manipulator
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/1633—Programme controls characterised by the control loop compliant, force, torque control, e.g. combined with position control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Numerical Control (AREA)
- Manipulator (AREA)
Abstract
The invention discloses an automatic calibration and compensation device and method for errors of a robot assembly system. Firstly, determining the number and layout scheme of the calibration blocks according to the position and precision requirements of different assembly stations on a robot assembly platform by using a designed universal calibration block; then, rough positioning is carried out on the positions of the calibration blocks based on a visual method, and further, the actual coordinate values of the characteristic points of the positioning holes on each calibration block and the error values of the characteristic points and the theoretical coordinates are obtained based on a force sense method; and then, selecting m calibration block characteristic points closest to the target point according to the position of the target point, and obtaining the error and the actual coordinate value of the target point through error matrix fitting of the calibration blocks, so that the automatic error compensation can be completed. Compared with the prior art, the method is convenient to implement, high in precision, low in cost and easy to realize flexible calibration of assembly system errors.
Description
Technical Field
The invention relates to the technical field of assembly error calibration, in particular to an automatic calibration and compensation device and method for errors of a robot assembly system.
Background
The assembly is an important link in the product manufacturing process, and takes more than half of the time and labor of the manufacturing life cycle of the enterprise product. The assembly robot can greatly improve the manufacturing efficiency, reduce the production cost and maximize the economic benefit of enterprises. However, in the robot assembly system, the robot cannot successfully complete the assembly task due to errors of the robot body and installation errors of each assembly station on the assembly platform.
Reviewing the prior art document, patent document 1 (CN 109048876B) discloses a robot calibration method based on a laser tracker, which can comprehensively calibrate the D-H parameter of the robot, greatly improves the calibration precision of the robot, but has very high calibration cost and poor popularization due to high price of the laser tracker; patent document 2 (CN 113084798 a) discloses a robot calibration device based on multi-station measurement, and the measurement of the calibration device at a plurality of stations is realized through a calibration block, but the calibration process is complex because the structure of the calibration device is complex, and the calibration device needs to be disassembled and assembled for a plurality of times in the calibration process; patent document 3 (CN 113256708 a) discloses a method for calibrating a theoretical 3D space and an actual robot space, where a calibration block is placed in an actual working area of a robot, position coordinates of 4 feature points on the calibration block are obtained through vision, and are substituted into a calibration operation tool to obtain a calibration error, but since the relative positions of the 4 feature points on the calibration block are fixed, adjustment cannot be performed according to the assembly precision requirements of different assembly stations, and the flexibility is low; patent document 4 (CN 111791231B) discloses a robot calibration system and a two-dimensional plane and three-dimensional space motion calibration method, by means of real-time dynamic measurement of the center position of a target, an error response equation of the nominal coordinate and the measured coordinate of the current geometrical parameter of the robot is constructed, the current geometrical parameter of the robot corresponding to the minimum error is solved, and the robot is calibrated on line, but the motion of the robot can interfere with the position of the target in the calibration process due to the mode of 'eyes outside hands', and the calibration process involves coordinate conversion of a plurality of groups of prisms, a camera coordinate system, a world coordinate system and a robot coordinate system, so that the calibration efficiency is low. In summary, the existing calibration method has high equipment price, the used calibration block is difficult to be suitable for assembly occasions with different assembly precision requirements, and only aims at the error of the robot body to calibrate, and the influence of the assembly station error cannot be considered, so that a certain limitation exists, and therefore, the calibration method which is convenient to implement, high in calibration precision, low in calibration assembly cost and easy to realize flexible calibration of the assembly system error is needed.
Disclosure of Invention
In order to solve the problems, the invention provides an automatic calibration and compensation device and method for errors of a robot assembly system, which utilize a designed calibration block and a layout scheme thereof to acquire actual coordinate values of characteristic points based on a device and a method combining vision and force sense, further fit and obtain errors and actual coordinate values of a target point, and realize flexible calibration of the errors of the assembly system.
The technical scheme adopted by the invention is as follows:
the automatic calibration and compensation device and method for the errors of the robot assembly system comprise the following steps:
Step 1, determining all stations to be assembled on a robot assembly platform, namely N A target points A, wherein the theoretical position coordinates of an ith target point A i are A iT(XiT,YiT,ZiT, simultaneously determining the number N B of calibration blocks B to be laid out on the whole assembly platform and the positions thereof according to the positions of all target points and the assembly precision requirement, installing the calibration blocks at the corresponding positions by utilizing the positioning reference surfaces of the calibration blocks, and calculating the theoretical position coordinates of a characteristic point C j on a jth calibration block B j to be C jT(XjT,YjT,ZjT);
Step 2, referring to theoretical position coordinates C jT of feature points of each calibration block B j, firstly, roughly positioning the actual positions of the feature points based on a visual method, and further, accurately positioning the actual positions of the feature points based on a force sense method to obtain actual position coordinates C jR(XjR,YjR,ZjR of the feature points C j on the calibration blocks;
And 3, selecting m calibration block feature points closest to the theoretical position coordinates of the target point A i according to the theoretical position coordinates of the target point A i, establishing an error matrix model C T＝Di·CR between the theoretical position coordinates and the actual position coordinates of the feature points, fitting and predicting the error of the target point A i through the error matrix model, further obtaining an actual coordinate value A iR(XiR,YiR,ZiR of the target point A i, and executing the operation on all the target points on the assembly platform to realize automatic error compensation.
The device comprises a vision system, a force sense system, an industrial robot and a calibration block, wherein the vision system comprises a switching component, an industrial camera, a camera fixing frame, a height adjusting plate, a light source fixing frame and a light source; the force sense system comprises a locating pin, a locating pin fixing frame, a tool supporting plate, a quick change, a force sensor fixing plate, a six-dimensional force sensor and a tail end adapter plate; the combination of the vision system and the force sense system is that the adapting part is connected with the tail end adapting plate through bolts, the optical axis of the industrial camera is parallel to the axis of the six-dimensional force sensor, the combined vision and force sense system is connected with the flange plate of the industrial robot through the tail end adapting plate, and the calibration blocks are arranged on the assembly station according to actual requirements.
In the vision system, a light source is connected with a light source fixing frame through a bolt, an industrial camera is connected with a camera fixing frame through a bolt, a height adjusting plate is provided with two slender holes, the slender holes are connected with the camera fixing frame through first bolt nuts, and further the height of the camera is adjusted by adjusting the positions of the first bolt nuts on the two slender holes; the switching part is provided with double-row array threaded holes, the slender holes are connected with the threaded holes on the switching part through second bolt nuts, and the unified adjustment of the heights of the light source and the camera is realized by adjusting the positions of the second bolt nuts on the double-row array threaded holes; the industrial camera is positioned right above the light source, the optical axis of the camera is coaxial with the axis of the light source, and the light source does not shade the imaging view of the camera.
In the force sense system, the connection sequence of all the components is as follows: the device comprises a locating pin, a locating pin fixing frame, a tool supporting plate, quick-change connection, a force sensor fixing plate, a six-dimensional force sensor and a tail end adapter plate; the bottom of the locating pin fixing frame is provided with a cylindrical counter bore, the locating pin is matched with the counter bore through a pin hole, a threaded hole is formed in the top of the locating pin along the axis direction, and the locating pin is rigidly connected with the locating pin fixing frame through the assembly of a bolt and the threaded hole.
Further, the calibration block B in the step 1 comprises a positioning hole and two positioning reference surfaces, wherein the positioning hole is a cylindrical counter bore formed in the calibration block along the height direction, and the circle center O of the bottom of the positioning hole is the characteristic point C of the calibration block; the positioning reference surface is two adjacent and mutually perpendicular side surfaces alpha and beta on the calibration block, the position tolerance delta of the axis of the positioning hole relative to the two reference surfaces is set to be the same, and the positioning hole is matched with the positioning pin by adopting a base hole system.
Furthermore, the number and positions of the calibration blocks B in the step 1 should be comprehensively determined according to the positions and assembly accuracy requirements of all the target points a, at least 1 calibration block should be set up near the target point with high assembly accuracy requirements, the total number of the calibration blocks B in the assembly platform should satisfy N B be greater than or equal to 3, the characteristic points of all the calibration blocks cannot all be on the same straight line, each characteristic point should be close to the positions of different target points, and the arranged calibration blocks are required not to interfere with the workpiece, the fixture and the robot motion trail.
Further, the visual coarse positioning method in the step 2 refers to selecting an industrial camera with medium precision, controlling the industrial camera to acquire images of corresponding calibration blocks, primarily positioning the positions of the calibration blocks on the assembly platform by using geometric and surface texture information of the calibration blocks, calculating pixel coordinates of the positioning holes in the images by using edge contour information of the positioning holes, converting the pixel coordinates into coordinates under a robot base coordinate system, and controlling the tail end of the robot to move to the position above and near the characteristic points of the positioning holes with the positioning pins.
Further, the force sense method in the step 2 specifically includes:
Step 7.1, taking force and moment data acquired by the six-dimensional force sensor in different directions around the positioning hole [0 degrees, 360 degrees ] as (F x,Fy,Fz,Mx,My,Mz) original training data, training a force sense compliance control model theta=f (F, M) by adopting a machine learning method, and performing parameter optimization on the model through a custom model evaluation index; force and moment data acquired by the six-dimensional force sensor in the actual assembly process are taken as input, and the adjustment direction of the force control process is output through the force sense compliance control model;
Step 7.2, constructing a functional relation s=f (t) of the adjustment step length of the force control process along with the change of the adjustment times, taking the adjustment times as input, and outputting the adjustment step length of the force control process through the function;
Step 7.3, combining the adjustment direction output by the force sense control model and the adjustment step length output by the function relation, controlling the robot to automatically complete the assembly of the positioning hole, further obtaining the actual position coordinates of the characteristic points of the positioning hole,
Further, the custom model evaluation index refers to an average error θ err and a variance s 2 of the model prediction adjustment direction, specifically:
In the method, in the process of the invention, The predicted value of the adjustment direction is represented, and θ i represents the true value of the adjustment direction.
Further, in the step 3, the m calibration block feature points closest to the target point a i are selected, and m=3, that is, the distances between the target point a i and all the calibration block feature points are calculated, and 3 feature points with the smallest distance value and not on the same straight line are selected, so as to establish an error matrix model, so as to obtain the actual coordinate value of the target point a i by fitting.
Further, the error matrix model in the step 3 refers to the motion of regarding the transformation relation C T＝Di·CR of the theoretical position and the actual position of the feature point as a rigid body in space, and is described by a rotation matrix R and a translation vector T, and the calculation formula of the error matrix model D i is as follows:
Wherein C jT＝[XjT,YjT,ZjT]T represents the theoretical coordinates of the jth feature point C j, C jR＝[XjR,YjR,ZjR]T represents the actual coordinates of the jth feature point C j, P cT and P cR represent the theoretical positions and the average center coordinates of the actual positions of the jth to j+m feature points, respectively, U and V are two matrices obtained by SVD singular value decomposition of the matrix H, which are composed of feature vectors of the matrices HH T and H T H, respectively, λ j and ζ j represent feature values, and U j and V j represent feature vectors.
As a preferable technical scheme, the tolerance delta of the positioning hole axis relative to the position of the two reference surfaces is 0.01-0.015 mm.
As an optimal technical scheme, the tolerance level of the positioning hole is IT 10-IT 7, and the tolerance level of the positioning pin is IT 8-IT 5.
As a preferable technical scheme, the distance between the characteristic points and the adjacent target points is not more than 100mm, and the distance interval between the characteristic points is more than 40mm.
The beneficial effects of the invention are as follows:
(1) The low-cost high-precision high-efficiency calibration of the robot assembly system can be realized. Compared with a method for calibrating absolute positioning errors of robots by using expensive laser trackers, the method provided by the invention uses an industrial camera and a force sensor with low price, and a calibration block with simple structure and convenient disassembly and assembly, and the actual positions of characteristic points on each assembly station are rapidly determined by combining visual initial positioning and six-dimensional force accurate positioning, so that the joint calibration of the absolute positioning errors of the robots and the installation errors of the assembly stations is realized, the calibration cost is reduced, and the calibration efficiency is improved. The six-dimensional force sensor can measure force value changes in three directions, can measure moment changes in the calibration process, is beneficial to expanding sample data characteristics, further improves the accuracy of prediction of the force control model, and improves the measurement accuracy of feature point coordinates.
(2) The flexible calibration of the robot assembly system can be realized. The method specifically refers to determining the number and layout scheme of calibration blocks according to the positions and assembly precision requirements of all target points on different assembly stations, and for the target points with higher assembly precision requirements, improving the calibration precision by adjusting the number and positions of the calibration blocks, so that flexible calibration of errors of a robot assembly system is realized, and the method has higher calibration precision.
Drawings
Fig. 1 is a schematic structural diagram of a vision and force sense system device according to an embodiment of the invention.
FIG. 2 is a flow chart of an error automatic calibration and compensation method in an embodiment of the invention.
Fig. 3 is a schematic diagram of the structure of the labeled block and three views according to the embodiment of the present invention.
FIG. 4 is a schematic diagram of the installation layout of the labeled block according to an embodiment of the present invention.
FIG. 5 is a schematic diagram of the error solving and compensating process according to the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail with reference to the following examples and the accompanying drawings. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Referring to fig. 1, a flowchart of an automatic calibration and compensation device and method for errors in a robot assembly system according to a first embodiment of the present application is shown. An automatic calibration and compensation device and method for errors of a robot assembly system according to a first embodiment of the present application include the following steps.
S1, designing a general calibration block, wherein the specific process is as follows:
S1.1. The designed calibration block comprises a positioning hole and two positioning reference surfaces, wherein the positioning hole is a cylindrical counter bore formed in the height direction of the calibration block, and the depth is The circle center O of the hole bottom is the characteristic point of the calibration block; the positioning reference surface is two adjacent and mutually perpendicular side surfaces alpha and beta on the calibration block, and the position tolerance delta of the axis of the positioning hole relative to the two reference surfaces is set to be the same, specifically 0.01mm; the positioning hole and the positioning pin are matched by adopting a base hole, and the pin hole matching precision is H8/g6, so that the diameter of the positioning hole is/>Locating pin diameter is/>
S1.2, comprehensively determining the number and positions of calibration blocks according to the positions of all target points and the assembly precision requirement, wherein for the calibration blocks which are at least 1 near the target points with high assembly precision requirement, the total number of the calibration blocks in an assembly platform is more than or equal to N B and is not less than 3, all characteristic points of all the calibration blocks cannot be on the same straight line, all the characteristic points are respectively close to the positions of different target points, the distance between the set characteristic points and the adjacent target points is not more than 100mm, and the arranged calibration blocks are not interfered with a workpiece, a clamp and a robot motion track; in addition, the distance interval between the characteristic points is recommended to be larger than 40mm, so that the calibration cost is reduced, the calibration efficiency is improved, and the installation and arrangement diagram of the calibration blocks is shown in fig. 3.
S2, utilizing offline programming software to compile a robot calibration program, wherein the specific process is as follows:
s2.1, importing the three-dimensional models of the calibration block and the robot assembly platform into offline programming software, and arranging the calibration block models according to the layout scheme described in S1.2, so that the layout of the calibration block models on the assembly platform model corresponds to the actual layout of the calibration block one by one.
S2.2, simulating the action track of the robot in the calibration process by using off-line programming software, wherein the calibration process comprises the functions of controlling an industrial camera to collect calibration block images, processing images and converting coordinates, collecting six-dimensional force sensor data, controlling an air pressure switch and the like, and setting the three-dimensional coordinates, the maximum movement speed, the initial acceleration and the like of each point position in the movement process of the robot in the software.
S2.3, establishing a corresponding relation between the robot output signals and the function instructions, distributing a unique output signal for each function instruction in the calibration program, transmitting the function instructions by utilizing the output signals, and enabling the corresponding relation between the output signals and the function instructions to be as follows:
S2.4, adding the corresponding robot output signal to the position needing to trigger the function in the calibration procedure according to the function needing to be executed in the calibration procedure, and adjusting the state of the signal according to the requirement. If a function needs to be turned on, the output signal corresponding to the function is set to $OUT=TRUE, and if a function needs to be turned off, the output signal corresponding to the function is set to $OUT=FALSE, wherein the output signal of the robot is represented by the output signal.
S2.5, after the setting of the motion trail and the output signal of the robot is completed, the motion trail of the robot is checked, optimized and corrected by utilizing the simulation function of the off-line programming software, and after the programming of the calibration program is completed, the corresponding post-processing plug-in is selected according to the brand of the actual robot, and the calibration program is converted into a program which can be recognized by the actual robot.
S3, adding a communication function for the calibration program, wherein the specific process is as follows:
S3.1, creating a communication configuration file in an actual robot controller, wherein the communication configuration file comprises a definition communication server and a client, an IP address and a port number for setting communication connection and a storage format for configuring communication data, the data sent by the robot comprise current position coordinates and function instructions of the robot, and the received data are new position coordinates for driving the robot to continue to move.
S3.2, defining a communication function in the calibration program obtained in the step S2, firstly packaging the current position coordinate of the robot into XML tag data according to a data storage format in a function body, writing the XML tag data into a transmission channel, then entering a blocking state waiting for receiving, automatically analyzing effective data according to the data storage format after receiving the XML tag data sent by an industrial personal computer, storing the effective data in corresponding variables, and then writing the current position coordinate of the robot into the transmission channel, thereby circulating. In the process, the method comprises the step of sending the function instruction in the calibration program besides the current position coordinate of the robot. The state of the output signal of the robot is scanned, and corresponding function instructions are written into the transmitting channel according to the state of the output signal and transmitted to the industrial personal computer together with the position coordinates of the robot.
S3.3, taking the robot output signal as a search zone bit, traversing all the output signals in the calibration program, and inserting the call of the communication function after the output signals. When the program is executed to the position, the robot establishes communication with the industrial personal computer, and in the communication process, the robot sends the current position coordinates and the function instruction to the industrial personal computer, and the industrial personal computer executes the corresponding function according to the function instruction.
S4, acquiring theoretical coordinates and actual coordinates of the calibration block, wherein the specific process is as follows:
S4.1, importing the calibration program obtained in the step S3 into an actual robot controller, setting an IP address of an industrial personal computer to enable the industrial personal computer and the robot to be in the same local area network, and running the robot calibration program to enable the robot to be in communication connection with the industrial personal computer.
S4.2, enabling the robot to move to a theoretical position C 1T of a characteristic point C 1 on the calibration block according to an off-line programming preset track, receiving current coordinate data sent by the robot through data transmission between the industrial personal computer and the robot, further obtaining a theoretical coordinate C 1T(X1T,Y1T,Z1T of the locating hole, and similarly executing the operation on other calibration blocks B j to obtain the theoretical coordinate C jT(XjT,YjT,ZjT of the locating hole.
S4.3, selecting an industrial camera with medium precision, controlling the industrial camera to acquire an image of a calibration block, eliminating noise interference in the image through image preprocessing, utilizing geometric and surface texture information of the calibration block, initially positioning the position of the calibration block on an assembly platform based on a template matching method, then calculating pixel coordinates of the center of a positioning hole in the image based on an elliptic contour detection algorithm by utilizing edge contour information of the positioning hole, and then converting the pixel coordinates of the positioning hole into coordinates of a robot base coordinate system through camera calibration and hand-eye calibration, and further controlling the tail end of the robot to move to the position above a characteristic point of the positioning hole with a positioning pin.
S4.4, taking force and moment data (F x,Fy,Fz,Mx,My,Mz) acquired by the six-dimensional force sensor at different positions [0 DEG, 360 DEG ] around the positioning hole as original training data, training a force sense compliance control model theta=f (F, M) by adopting a machine learning method, taking the force and moment data acquired by the six-dimensional force sensor in the actual assembly process as input, and outputting the adjustment direction of the force control process through the force sense compliance control model; the method comprises the steps of performing parameter tuning on a model by using a custom model evaluation index, namely an average error theta err and a variance s 2 of model prediction adjustment direction, wherein the force control model with better performance can be obtained, and the custom model evaluation index comprises the following specific steps:
In the method, in the process of the invention, The predicted value of the adjustment direction is represented, and θ i represents the true value of the adjustment direction.
S4.5, constructing a functional relation s=f (t) of the adjustment step length of the force control process along with the change of the adjustment times, taking the adjustment times as input, and outputting the adjustment step length of the force control process through the function, wherein the functional relation is specifically as follows:
In the formula, n represents the adjustment times, s min represents the minimum adjustment step length, and is taken as 0.01mm, so that the meaning of avoiding the situation that the step length approaches 0 along with the increase of the adjustment times is that a is a undetermined coefficient, and is taken as 1.38, and the size of the initial step length is determined.
S4.6, combining the adjustment direction output by the force sense control model and the adjustment step length output by the functional relation, controlling the robot to automatically complete positioning hole assembly, and further obtaining the actual position coordinates C jR(XjR,YjR,ZjR of the positioning hole feature points C jR.
S5, calculating an error according to the position of the target point A i and completing error compensation, wherein the specific process is as follows:
S5.1, selecting 3 calibration block characteristic points closest to the target point A i, namely respectively calculating the distances between the target point A i and all the calibration block characteristic points, and selecting 3 characteristic points which are the smallest in distance and are not on the same straight line.
S5.2, according to the theoretical coordinates C jT and the actual coordinates C jR of the characteristic points of the 3 selected calibration blocks, an error matrix model C T＝Di -CR of the characteristic points is established, and then the calculation formula of the error matrix model D i is as follows:
Wherein C jT＝[XjT,YjT,ZjT]T represents the theoretical coordinates of the jth feature point C j, C jR＝[XjR,YjR,ZjR]T represents the actual coordinates of the jth feature point C j, P cT and P cR represent the theoretical positions and the average center coordinates of the actual positions of the jth to j+m feature points, respectively, U and V are two matrices obtained by SVD singular value decomposition of the matrix H, which are composed of feature vectors of the matrices HH T and H T H, respectively, λ j and ζ j represent feature values, and U j and V j represent feature vectors.
S5.3, fitting the target point A i according to the obtained error matrix D i so as to obtain an actual coordinate value A iR(XiR,YiR,ZiR of the target point A i, and finally replacing the theoretical coordinates with the actual coordinates of the obtained target point A i by modifying a coordinate file in a robot assembly program, and executing the operations on all the target points on the assembly platform, so that automatic error compensation can be realized, and the error solving and compensating process is shown in fig. 4.
It is apparent that the above examples of the present invention are merely illustrative of the present invention and are not limiting of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary or impossible to exemplify all embodiments herein. And obvious changes and modifications which come within the spirit of the invention are desired to be protected.
Claims (5)
1. The automatic calibration and compensation method for the errors of the robot assembly system is characterized by comprising the following steps of: the method is applied to an automatic calibration and compensation device for errors of a robot assembly system, and the device comprises the following components: the system comprises a visual system, a force sense system, an industrial robot and a calibration block B, wherein the visual system comprises an adapter component (1), an industrial camera (2), a camera fixing frame (3), a height adjusting plate (4), a light source fixing frame (5) and a light source (6); the force sense system comprises a locating pin (7), a locating pin fixing frame (8), a tool supporting plate (9), a quick change (10), a force sensor fixing plate (11), a six-dimensional force sensor (12) and a tail end adapter plate (13); the combination of the vision system and the force sense system is that the adapting part (1) is connected with the tail end adapting plate (13) through bolts, the optical axis of the industrial camera (2) is parallel to the axis of the six-dimensional force sensor (12), the combined vision and force sense system is connected with the flange plate of the industrial robot through the tail end adapting plate (13), and the calibration block B is arranged on the assembly station; in the vision system, a light source (6) is connected with a light source fixing frame (5) through a bolt, an industrial camera (2) is connected with a camera fixing frame (3) through a bolt, a height adjusting plate (4) is provided with two slender holes (4-1), the slender holes are connected with the camera fixing frame (3) through first bolt nuts, and then the camera height is adjusted by adjusting the positions of the first bolt nuts on the two slender holes; the switching component (1) is provided with double-row array threaded holes (1-1), the slender holes are connected with the threaded holes on the switching component (1) through second bolt nuts, and then the unified adjustment of the heights of the light source and the camera is realized by adjusting the positions of the second bolt nuts on the double-row array threaded holes (1-1); the industrial camera (2) is positioned right above the light source (6), the optical axis of the camera is coaxial with the axis of the light source, and the light source can not shade the imaging visual field of the camera; in the force sense system, the connection sequence of all the components is as follows: the device comprises a locating pin (7), a locating pin fixing frame (8), a tool supporting plate (9), a quick change (10), a force sensor fixing plate (11), a six-dimensional force sensor (12) and a tail end adapter plate (13); the bottom of the locating pin fixing frame (8) is provided with a cylindrical counter bore (8-1), the locating pin (7) is matched with the counter bore through a pin hole, a threaded hole is formed in the top of the locating pin (7) along the axis direction, and the locating pin (7) is rigidly connected with the locating pin fixing frame (8) through the assembly of a bolt and the threaded hole;
The method comprises the following steps: step 1, determining all stations to be assembled on a robot assembly platform, namely N A target points A, wherein the theoretical position coordinates of an ith target point A i are A iT(XiT,YiT,ZiT, simultaneously determining the number N B of calibration blocks B to be laid out on the whole assembly platform and the positions thereof according to the positions of all target points and the assembly precision requirement, installing the calibration blocks B at the corresponding positions by utilizing the positioning reference surfaces of the calibration blocks B, and calculating the theoretical position coordinates of a characteristic point C j on a jth calibration block B j as C jT(XjT,YjT,ZjT);
Step 2, referring to a theoretical position coordinate C jT of a feature point C j of a calibration block B j, firstly, roughly positioning the actual position of the feature point C j based on a visual method, and further, accurately positioning the actual position of the feature point C j on the calibration block B j based on a force sense method to obtain an actual position coordinate C jR(XjR,YjR,ZjR of the feature point C j;
Step 3, according to the theoretical position coordinates of the target point A i, selecting m calibration block B characteristic points closest to the theoretical position coordinates, establishing an error matrix model C T＝Di·CR between the theoretical position coordinates and the actual position coordinates of the characteristic points, fitting and predicting the error of the target point A i through the error matrix model, further obtaining the actual position coordinates A iR(XiR,YiR,ZiR of the target point A i), and executing the operation on all the target points on the assembly platform to realize automatic error compensation; the error matrix model in the step 3 refers to the motion of regarding the transformation relation C T＝Di·CR of the theoretical position and the actual position of the feature point as a rigid body in space, and is described by a rotation matrix R and a translation vector T, and the calculation formula of the error matrix model D i is as follows:
Wherein C jT＝[XjT,YjT,ZjT]T represents the theoretical coordinates of the jth feature point C j, C jR＝[XjR,YjR,ZjR]T represents the actual coordinates of the jth feature point C j, P cT and P cR represent the theoretical positions and the average center coordinates of the actual positions of the jth to j+m feature points, respectively, U and V are two matrices obtained by SVD singular value decomposition of the matrix H, which are composed of feature vectors of the matrices HH T and H T H, respectively, λ j and ζ j represent feature values, and U j and V j represent feature vectors;
The calibration block B in the step 1 comprises a positioning hole (15) and two positioning reference surfaces (16), wherein the positioning hole (15) is a cylindrical counter bore formed in the calibration block B along the height direction, and the circle center O of the bottom of the positioning hole is the characteristic point C of the calibration block B; the positioning reference surface (16) is two adjacent and mutually perpendicular side surfaces alpha and beta on the calibration block B, the position tolerance delta of the axis of the positioning hole relative to the two reference surfaces is set to be the same, and the value range is 0.01-0.015 mm; the positioning holes are matched with the positioning pins by adopting a base hole, the tolerance level of the positioning holes is IT 10-IT 7, and the tolerance level of the positioning pins is IT 8-IT 5;
The number and the positions of the calibration blocks B in the step 1 are comprehensively determined according to the positions of all the target points A and the assembly precision requirement, at least 1 calibration block B is set up near the target point with high assembly precision requirement, the total number of the calibration blocks B in the assembly platform is more than or equal to 3 and N B, the characteristic points of all the calibration blocks B cannot be all on the same straight line, and all the characteristic points are respectively close to the positions of different target points; the distance between the set characteristic point and the adjacent target point is not more than 100mm, and the arranged calibration block B is required not to interfere with the workpiece, the clamp and the robot motion track; further, the distance interval between the feature points is greater than 40mm.
2. The automatic calibration and compensation method for errors of a robot assembly system according to claim 1, wherein the method comprises the following steps: the visual method in the step 2 refers to selecting an industrial camera with medium precision, controlling the industrial camera to acquire images of corresponding calibration blocks B, utilizing geometric and surface texture information of the calibration blocks B to initially position the calibration blocks B on an assembly platform, then utilizing edge contour information of a positioning hole to calculate pixel coordinates of the positioning hole in the images, converting the pixel coordinates into coordinates under a robot base coordinate system, and further controlling the tail end of the robot to move above a characteristic point with a positioning pin.
3. The automatic calibration and compensation method for errors of a robot assembly system according to claim 2, wherein the method comprises the following steps: the force sense method in the step 2 specifically comprises the following steps: force and moment data (F x,Fy,Fz,Mx,My,Mz) acquired by the six-dimensional force sensor in different directions of a positioning hole [0 DEG, 360 DEG ] are used as original training data, a machine learning method is adopted to train a force sense compliance control model theta=f (F, M), and parameter optimization is carried out on the model through a custom model evaluation index; force and moment data acquired by the six-dimensional force sensor in the actual assembly process are taken as input, and the adjustment direction of the force control process is output through the force sense compliance control model;
Constructing a functional relation s=f (t) of the adjustment step length of the force control process along with the change of the adjustment times, taking the adjustment times as input, and outputting the adjustment step length of the force control process through the function; the functional relation is specifically:
Wherein n represents the adjustment times, s min represents the minimum adjustment step length, and is taken as 0.01mm, the meaning of the adjustment times is that the step length is prevented from approaching 0 along with the increase of the adjustment times, a is a undetermined coefficient, and is taken as 1.38, and the size of the initial step length is determined;
And the robot is controlled to automatically complete the assembly of the positioning holes so as to obtain the actual position coordinates of the feature points.
4. A method for automatic calibration and compensation of errors in a robotic assembly system according to claim 3, wherein: the custom model evaluation index refers to average error theta err and variance s 2 of model prediction adjustment direction, and specifically comprises the following steps:
In the method, in the process of the invention, The predicted value of the adjustment direction is represented, and θ i represents the true value of the adjustment direction.
5. The automatic calibration and compensation method for errors in a robot assembly system according to claim 4, wherein the method comprises the steps of: in the step 3, the m calibration block B feature points closest to the target point a i are selected, and m=3, that is, the distances between the target point a i and all the calibration block B feature points are calculated, and 3 feature points with the smallest distance value and not on the same straight line are selected, so as to establish an error matrix model, and to obtain the actual coordinate value of the target point a i by fitting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210195451.XA CN114523475B (en) | 2022-03-01 | 2022-03-01 | Automatic calibration and compensation device and method for errors of robot assembly system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210195451.XA CN114523475B (en) | 2022-03-01 | 2022-03-01 | Automatic calibration and compensation device and method for errors of robot assembly system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114523475A CN114523475A (en) | 2022-05-24 |
| CN114523475B true CN114523475B (en) | 2024-06-18 |
Family
ID=81626658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210195451.XA Active CN114523475B (en) | 2022-03-01 | 2022-03-01 | Automatic calibration and compensation device and method for errors of robot assembly system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114523475B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115122333A (en) * | 2022-07-20 | 2022-09-30 | 上海节卡机器人科技有限公司 | Robot calibration method and device, electronic equipment and storage medium |
| CN117021113B (en) * | 2023-09-20 | 2024-03-19 | 苏州诺克智能装备股份有限公司 | Mechanical arm cooperative positioning assembly method, system and medium |
| CN117226853B (en) * | 2023-11-13 | 2024-02-06 | 之江实验室 | Robot kinematics calibration method, device, storage medium and equipment |
| CN118332708B (en) * | 2024-06-17 | 2024-10-18 | 成都飞机工业（集团）有限责任公司 | Large-size component assemblability evaluation method based on characteristic point distance error matrix |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106272416A (en) * | 2016-08-29 | 2017-01-04 | 上海交通大学 | Feel based on power and the robot slender axles Fine Boring system and method for vision |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6141863A (en) * | 1996-10-24 | 2000-11-07 | Fanuc Ltd. | Force-controlled robot system with visual sensor for performing fitting operation |
| CN106853639A (en) * | 2017-01-04 | 2017-06-16 | 河北工业大学 | A kind of battery of mobile phone automatic assembly system and its control method |
| CN109839075A (en) * | 2017-11-29 | 2019-06-04 | 沈阳新松机器人自动化股份有限公司 | A kind of robot automatic measurement system and measurement method |
| CN109543823B (en) * | 2018-11-30 | 2020-09-25 | 山东大学 | Flexible assembly system and method based on multi-mode information description |
| CN111590581B (en) * | 2020-05-26 | 2021-10-22 | 珠海格力智能装备有限公司 | Positioning compensation method and device for robot |
-
2022
- 2022-03-01 CN CN202210195451.XA patent/CN114523475B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106272416A (en) * | 2016-08-29 | 2017-01-04 | 上海交通大学 | Feel based on power and the robot slender axles Fine Boring system and method for vision |
Non-Patent Citations (1)
| Title |
|---|
| robotic fasten assembly using vision and force sensing;WEI QIAN ET AL;<2019 IEEE INTERNATIONAL CONFERENCE ON REAL-TIME COMPUTING AND ROBOTICS>;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114523475A (en) | 2022-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114523475B (en) | Automatic calibration and compensation device and method for errors of robot assembly system | |
| CN103895023B (en) | A kind of tracking measurement method of the mechanical arm tail end tracing measurement system based on coding azimuth device | |
| CN102825602B (en) | PSD (Position Sensitive Detector)-based industrial robot self-calibration method and device | |
| US10974393B2 (en) | Automation apparatus | |
| CN102706277B (en) | Industrial robot online zero position calibration device based on all-dimensional point constraint and method | |
| CN112833786B (en) | Cabin attitude and pose measuring and aligning system, control method and application | |
| CN102183205A (en) | Method for matching optimal assembly poses of large-sized parts | |
| CN109366503B (en) | Large-scale component-oriented processing method based on mobile series-parallel robot | |
| CN109304730A (en) | A kind of robot kinematics' parameter calibration method based on laser range finder | |
| CN109848989B (en) | Robot execution tail end automatic calibration and detection method based on ruby probe | |
| CN102654387A (en) | Online industrial robot calibration device based on spatial curved surface restraint | |
| CN110125455A (en) | A method of for optimizing drill bit pose in robotic drill | |
| CN111515928B (en) | Mechanical arm motion control system | |
| Li et al. | Real-time trajectory position error compensation technology of industrial robot | |
| CN111055289A (en) | Method and device for calibrating hand and eye of robot, robot and storage medium | |
| CN113910239B (en) | Industrial robot absolute positioning error compensation device and method | |
| CN111546334A (en) | Industrial robot online pose error compensation method for reducing contour error | |
| CN110174074A (en) | A kind of measuring device and method for industrial robot thermal deformation error compensation | |
| Hvilshøj et al. | Calibration techniques for industrial mobile manipulators: Theoretical configurations and best practices | |
| Li et al. | A laser-guided solution to manipulate mobile robot arm terminals within a large workspace | |
| CN110940351A (en) | Robot precision compensation method based on parameter dimension reduction identification | |
| Liu et al. | Positioning accuracy improvement for target point tracking of robots based on Extended Kalman Filter with an optical tracking system | |
| CN113733155A (en) | Six-axis industrial robot calibration device and calibration method | |
| Jian et al. | On-line precision calibration of mobile manipulators based on the multi-level measurement strategy | |
| CN114516055B (en) | Real-time calibration method and device for mechanical arm without shutdown based on binocular vision and deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |