SCARA robot automatic calibration method and device based on vision and storage medium

Abstract

The invention relates to the technical field of robot calibration, in particular to an SCARA robot automatic calibration method and device based on vision and a storage medium. The calibration method comprises the steps that firstly, a robot moves five different calibration plate point positions, robot coordinate positions and visual coordinate positions are recorded in the corresponding point positions, and a rotation translation matrix of a robot coordinate system and a visual coordinate system is worked out according to the five-point calibration method principle; automatically capturing an edge image of a Z-axis lead screw of the robot by using an image processing technology, determining the center position of the image, selecting two different point positions A and B in the visual field range of the camera, transmitting the point positions to the robot, automatically moving the robot to the point A and the point B by adopting a left hand system and a right hand system, and respectively recording robot joint angles A1, A2, B1 and B2 corresponding to the point; the large arm, the small arm and the small arm zero compensation value of the robot are obtained through calculation of a kinematics equation, so that robot parameter calibration is automatically and rapidly achieved, and the calibration precision and consistency are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of robot calibration, and in particular to a vision-based SCARA robot automatic calibration method, device and storage medium. Background Art

[0002] The SCARA robot is a common industrial robot. It is widely used in the electronics, automobile manufacturing and other industries due to its simple structure, high control accuracy and ability to adapt to higher speed requirements. However, in actual use, the absolute positioning accuracy of the SCARA robot may deviate due to processing errors between the actual rod length parameters and the design parameters, thus affecting the product quality of the entire production process.

[0003] In order to enable the robot to complete the task accurately, accurate calibration of robot rod length parameters has become a technical problem that needs to be solved urgently. At present, some robot manufacturers use laser trackers or wire encoders to calibrate SCARA robot rod parameters. Although this method has high calibration accuracy and small measurement error, it is not suitable for small and medium-sized enterprises due to its high cost. Another part of the manufacturers use traditional manual calibration methods for calibration. Although this method is simple to operate and low in cost, it is affected by human factors and has limitations such as large calibration errors, low efficiency, and time-consuming and labor-intensive calibration process. Summary of the invention

[0004] In order to overcome the above-mentioned shortcomings, the purpose of the present invention is to provide a vision-based SCARA robot automatic calibration method, which utilizes a vision system containing a 2D automatic zoom camera to obtain the visual data of the robot's Z-axis screw edge features and the calibration plate, and combines image processing technology and a robot control system to achieve efficient and accurate automatic calibration without human intervention; it also relates to a vision-based SCARA robot automatic calibration device and a storage medium.

[0005] The technical solution of the present invention to solve the technical problem is: The vision-based SCARA robot automatic calibration method includes the following steps: Step S1, connecting the visual system to the robot; Step S2, adjusting the calibration plate and the end surface of the Z axis of the robot to the same horizontal plane, adjusting the position of the calibration plate, and the visual system automatically focusing on the calibration plate; Step S3, move the Z-axis lead screw of the robot to above the center of the visual system's field of view, capture an initial calibration image, and set it as the zero position of the visual system. At the same time, obtain the world coordinates of the robot at the zero position and send it to the visual system; Step S4, moving the robot within the field of view of the visual system, obtaining the world coordinate system and the visual coordinate system of the robot, and then calculating the rotation and translation matrix of the robot coordinate system and the visual coordinate system; Step S5: The robot automatically sends a trigger command to the visual system. The visual system automatically returns a position coordinate after receiving the command. The robot automatically moves to the position coordinate after receiving the information returned by the visual system. The visual system confirms whether the robot has reached the sent position coordinate. According to the left and right hand systems, the robot obtains the joint coordinates of the left and right hand systems respectively based on the position coordinates sent by the visual system and sends them to the visual system. Step S6, selecting another position coordinate different from the position coordinate in step S5, and repeating step S5; Step S7, calculate and obtain the parameters of the robot's upper arm and lower arm and the lower arm zero point compensation value, thereby completing the calibration.

[0006] As an improvement of the present invention, in step S4, the robot moves to any four points and records the world coordinates of the robot. At the same time, the visual system automatically takes pictures of the visual coordinates of the four points. According to the zero point position and the world coordinates and visual coordinates of the four points, the rotation and translation matrices of the robot coordinate system and the visual coordinate system are obtained through coordinate rotation and translation calculation.

[0007] As a further improvement of the present invention, in step S5, the robot automatically sends a trigger command to the visual system, and the visual system automatically returns a position coordinate upon receiving the command. After the robot receives the information returned by the visual system, the center of the robot's Z-axis screw moves to the position coordinate under the left-hand system, and the visual system takes a photo to confirm whether the robot has reached the position coordinate. If not, the visual system sends a position compensation value to the robot, so that the robot moves until it reaches the position coordinate.

[0008] As a further improvement of the present invention, in step S5, if the robot reaches the position coordinate, the robot obtains the joint coordinate information of the current left hand system at the position coordinate and sends it to the visual system.

[0009] As a further improvement of the present invention, in step S5, the system automatically switches to the right-hand system, and the visual system takes a photo again to confirm whether the robot has reached the position coordinate. If not, the visual system sends the position compensation value to the robot, so that the robot moves until it reaches the position coordinate.

[0010] As a further improvement of the present invention, if the robot reaches the position coordinate, the robot obtains the joint coordinate information of the current right hand system at the position coordinate and sends it to the visual system, and triggers a command to the visual system at the same time.

[0011] As a further improvement of the present invention, in step S6, the visual system will automatically send another position coordinate different from the position coordinate in step S5 to the robot after receiving the command. After receiving the coordinate information, the robot will automatically move to the position coordinate and repeat step S5.

[0012] A vision-based automatic calibration device for a SCARA robot, comprising a calibration fixture fixed on the Z-axis screw of the robot and a vision system for automatic focus photography for calibration, wherein the calibration fixture comprises a calibration plate limit block, a calibration plate and a fixed plate, the upper surface of the calibration plate limit block is symmetrically connected with a first clamp and a second clamp, a positioning groove is arranged on the fixed plate, the calibration plate is fixed in the positioning groove, the fixed plate is provided with a fixing plate through hole for exposing the calibration plate, the bottom surface of the calibration plate limit block is connected to the fixed plate to cover the calibration plate, and the Z-axis screw of the robot passes through the calibration fixture.

[0013] As a further improvement of the present invention, the visual system includes a bracket connected to the robot, a ring-shaped light source and a camera connected to the bracket, and an external industrial computer, the ring-shaped light source is located above the camera, the camera is electrically connected to the industrial computer, and the industrial computer is provided with a receiving module for receiving information, a processing module for analysis and processing, and a storage module for storing data running the processing module and recording data.

[0014] A vision-based storage medium, wherein the storage medium stores data information for robot automatic calibration for implementing the automatic calibration method when executed by a processing module.

[0015] In the present invention, by combining a visual system including a 2D automatic zoom camera, image processing technology is used to automatically capture the edge image of the robot's Z-axis lead screw and determine the center position of the image, and the robot's rod length parameters and zero point compensation value are calculated in combination with the robot's position and posture, thereby quickly completing the calibration process and performing self-correction, greatly improving the calibration accuracy and consistency; the present invention automatically completes operations such as recording the joint angles of each axis of the robot and calculating the displacement parameters through the coordination of the visual system and the control method, further reducing human operation errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] For ease of explanation, the present invention is described in detail with reference to the following preferred embodiments and the accompanying drawings.

[0017] Figure 1 A flowchart of the vision-based automatic calibration method for a SCARA robot according to the present invention; Figure 2 A schematic diagram of a flow chart of a vision-based automatic calibration method for a SCARA robot according to the present invention; Figure 3The structure diagram of the vision-based automatic calibration device for SCARA robots of the present invention is shown in FIG. Figure 1 ; Figure 4 The structure diagram of the vision-based automatic calibration device for SCARA robots of the present invention is shown in FIG. Figure 2 ; Figure 5 It is a front view of the vision-based automatic calibration device for SCARA robots of the present invention; Figure 6 It is a left view of the vision-based automatic calibration device for SCARA robots of the present invention; Figure 7 A top view of the vision-based automatic calibration device for SCARA robots of the present invention; Figure 8 The structure of the calibration tool of the present invention is shown in FIG. Figure 1 ; Fig. 9 The structure of the calibration tool of the present invention is shown in FIG. Figure 2 ; Fig.10 It is a schematic diagram of the calibration plate of the present invention under visual conditions; Fig.11 It is a front view of the calibration tool of the present invention; Fig.12 It is a rear view of the calibration tool of the present invention; Fig.13 It is the forward and inverse solution diagram of robot kinematics; Figure markings: 1-calibration tooling, 11-calibration plate limit block, 12-calibration plate, 13-fixed plate, 14-first clamp, 15-second clamp, 2-visual system, 21-bracket, 22-ring light source, 23-camera, 24-industrial computer, 3-robot, 31-screw, 4-Z-axis screw center, 5-Z-axis end face of the robot. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] like Figures 1 to 12 As shown, the vision-based SCARA robot automatic calibration device of the present invention includes a calibration tool 1 fixed on the Z-axis screw 31 of the robot 3 and a visual system 2 for automatically focusing and photographing for calibration. The calibration tool 1 is fixed on the Z-axis screw 31 of the robot, and the visual system includes a 2D autofocus camera, which is located below the calibration tool 1 and can automatically focus on the position.

[0020] In the present invention, the calibration tool 1 includes a calibration plate limit block 11, a calibration plate 12 and a fixed plate 13, the upper surface of the calibration plate limit block 11 is symmetrically connected with a first clamp 14 and a second clamp 15, a positioning groove is provided on the fixed plate 13, the calibration plate 12 is fixed in the positioning groove, the fixed plate 13 is provided with a fixed plate through hole for exposing the calibration plate 12, the bottom surface of the calibration plate limit block 11 is connected to the fixed plate 13 so as to cover the calibration plate 12, and the Z-axis screw 31 of the robot passes through the calibration tool 1; specifically, the calibration tool 1 includes a first clamp 14, a second clamp 15, a calibration plate limit block 11, a calibration plate 12, and a fixed block 13, the calibration plate 12 is marked with a checkerboard, the fixed block 13 has 8 screw through holes on the edges around it, and its inner edge has a positioning groove, and the fixed The depth of the positioning groove is consistent with the thickness of the calibration plate 12. The calibration plate 12 is fixed on the positioning groove. The center of the calibration plate 12 is circularly hollowed out, and its diameter is large enough to allow the robot's screw 31 to pass through. There is a circular through hole at the center of the calibration plate limit block 11, and its diameter is consistent with the diameter of the circular hollow in the center of the calibration plate 12. The fixed block 13 is connected to the calibration plate 12 on the calibration plate limit block 11 by 8 screws. The calibration plate limit block 11 is fixed to the second clamp 15 by four screws, and the second clamp 15 is connected to the first clamp 14 by 2 screws on the side. The calibration tool 1 can be fixed to the end of the Z-axis screw 31 of the robot. The sides of the first clamp 14 and the second clamp 15 have arc-shaped through holes, and the through hole radius is consistent with the radius of the circular through hole at the calibration plate limit block 11.

[0021] In the present invention, the visual system 2 includes a bracket 21 connected to the robot, a ring light source 22 and a camera 23 connected to the bracket 21, and an external industrial computer 24. The ring light source 22 is located above the camera 23. The ring light source 22 provides lighting for the calibration tool to facilitate the automatic focusing of the camera 23. The camera 23 adopts a 2D autofocus camera. The camera 23 is electrically connected to the industrial computer 24. The industrial computer 24 is fixedly connected to the frame of the robot. The industrial computer 24 is provided with a receiving module for receiving information, a processing module for analysis and processing, and a storage module for storing data of the processing module operation and recording data.

[0022] The vision-based storage medium of the present invention stores data information for robot automatic calibration that is used to implement the automatic calibration method when executed by a processing module.

[0023] The vision-based automatic calibration method of the SCARA robot of the present invention comprises the following steps: Step S1, connecting the visual system to the robot; Step S2, adjusting the calibration plate and the end surface of the Z axis of the robot to the same horizontal plane, adjusting the position of the calibration plate, and the visual system automatically focusing on the calibration plate; Step S3, move the Z-axis lead screw of the robot to above the center of the visual system's field of view, capture an initial calibration image, and set it as the zero position of the visual system. At the same time, obtain the world coordinates of the robot at the zero position and send it to the visual system; Step S4, moving the robot within the field of view of the visual system, acquiring the world coordinate system and the visual coordinate system of the robot, and obtaining the rotation and translation matrix of the robot coordinate system and the visual coordinate system; Step S5: The robot automatically sends a trigger command to the visual system. The visual system automatically returns a position coordinate after receiving the command. The robot automatically moves to the position coordinate after receiving the information returned by the visual system. The visual system confirms whether the robot has reached the sent position coordinate. According to the left and right hand systems, the robot obtains the joint coordinates of the left and right hand systems respectively based on the position coordinates sent by the visual system and sends them to the visual system. Step S6, selecting another position coordinate different from the position coordinate in step S5, and repeating step S5; Step S7, calculate and obtain the parameters of the robot's upper arm and lower arm and the lower arm zero point compensation value, thereby completing the calibration.

[0024] In the present invention, the calibration tool 1 is fixed at the end of the robot (i.e., the action execution end of the robot). It adopts a 2D autofocus camera vision system and is combined with the calibration tool 1 fixed at the end of the Z-axis of the SCARA robot for calibration. The robot 3 establishes a connection with the vision system 2 through SOCKET TCP communication, and the two exchange data to complete the calibration work.

[0025] In step S2 , firstly, the calibration plate 12 marked with checkerboard grids and the Z-axis end face of the robot are adjusted to the same level, and the position of the robot 3 is adjusted so that the visual system 2 can automatically focus on the calibration plate 12 .

[0026] In step S3 , the center of the screw rod 31 is moved roughly to above the center of the field of view of the camera 23 to take a picture, which is recorded as the zero position P0 of the visual system 2 , and the world coordinates of the robot 3 are sent to the visual system 2 .

[0027] In step S4, the robot moves to any four points and records the world coordinates of the robot. At the same time, the visual system automatically takes pictures of the visual coordinates of the four points. According to the zero point position and the world coordinates and visual coordinates of the four points, the rotation and translation matrix of the robot coordinate system and the visual coordinate system is obtained through coordinate rotation and translation calculation; specifically, the robot is moved within the field of view of the calibration plate to any four points (note: the X and Y of the four points are different), namely P1, P2, P3, and P4, and the world coordinates of the robot are recorded. At the same time, the visual coordinates of the four points are automatically recorded by taking pictures. The coordinate relationship between the visual system and the robot is calibrated through 5 points, and the spatial coordinate rotation and translation matrix K of the robot coordinate system and the visual coordinate system can be obtained through coordinate rotation and translation calculation.

[0028] The calibration method of the present invention comprises the following steps: firstly, the robot moves to 5 different calibration plate points (P0, P1, P2, P3, P4), the corresponding points all record the robot coordinate position and the visual coordinate position, and the rotation and translation matrix of the robot coordinate system and the visual coordinate system is obtained according to the principle of the five-point calibration method; then, the image processing technology is used to automatically capture the edge image of the Z-axis screw rod of the robot and determine the center position of the image, two different points A and B are selected within the camera field of view and transmitted to the robot, the robot uses the left and right hand systems to automatically move to the points A and B and respectively record the robot joint angles A1, A2, B1, B2 corresponding to the points, and the robot upper arm, lower arm and lower arm zero point compensation values ​​can be obtained according to the calculation of the kinematic equation, thereby realizing automatic calibration.

[0029] Specifically: Assume that the robot coordinates are , the visual coordinates are , then the robot coordinates can be obtained from the visual coordinates through rotation and translation, as shown in formula (1): (1); in, is the robot coordinate, is the visual coordinate, R is the rotation matrix, and T is the translation matrix.

[0030] Furthermore, (2); Among them, X is the X-axis coordinate of the robot coordinate, Y is the Y-axis coordinate of the robot coordinate, x is the x-axis coordinate of the visual coordinate, and y is the y-axis coordinate of the visual coordinate. is the matrix conversion coefficient.

[0031] The conversion relationship between robot coordinates and visual coordinates is as follows: (3); (4); Among them, X0, X1, X2, Y0, Y1, and Y2 are the XY coordinates of three different points in the robot, and x0, x1, x2, y0, y1, and y2 are the XY coordinates of three different points in the vision system. is the matrix conversion coefficient; according to the above formula, at least 3 points are needed to calculate the matrix conversion coefficient, but in order to improve the accuracy and reduce the error, 5 points are selected for calibration.

[0032] Expressed in matrix form: (5); Similarly, we can get: (6); Among them, X0, X1, X 2、 X3, X 4、 Y0, Y1, Y 2、 Y3 and Y4 are the XY axis coordinates of the robot at the positions of P0, P1, P2, P3, and P4, respectively. 2、 x3 and x4 are the xy axis coordinates of the visual coordinates of the positions P0, P1, P2, P3, and P4 respectively. is the matrix conversion coefficient.

[0033] Combining (5) and (6), we can get the following matrix: (7); in, is the visual coordinates of the points P0, P1, P2, P3, and P4. are the robot coordinates of the points P0, P1, P2, P3, and P4, is the rotation and translation matrix of the robot coordinate system and the visual coordinate system, where is the matrix conversion coefficient.

[0034] but , where K is the rotation and translation matrix of the robot coordinate system and the visual coordinate system.

[0035] In step S5, the robot automatically sends a trigger command to the visual system 2, and the visual system 2 automatically returns a position coordinate after receiving the command. After the robot receives the information returned by the visual system 2, the center of the Z-axis screw 31 of the robot moves to the position coordinate in the left-hand system (right-hand coordinate system: in the robot system, the right-hand coordinate system is defined as the index finger of the right hand pointing to the positive direction of the X-axis, the middle finger pointing to the positive direction of the Y-axis, and the thumb pointing to the positive direction of the Z-axis; left-hand coordinate system: the left-hand coordinate system is defined as the index finger of the left hand pointing to the positive direction of the X-axis, the middle finger pointing to the positive direction of the Y-axis, and the thumb pointing to the positive direction of the -Z axis), and the visual system takes a photo to confirm whether the robot has reached the position coordinate. If not, the visual system sends the position compensation value to the robot, so that the robot moves until it reaches the position coordinate; if the robot reaches the position At the position coordinate, the robot obtains the joint coordinate information of the current left-hand system at the position coordinate and sends it to the visual system; specifically, the robot 3 automatically sends a trigger command to the visual system 2, and at this time the visual system 2 receives the command and automatically returns to send the position of point A (any center point of the chessboard within the camera's field of view) to the robot, and then according to the received coordinate information, the robot's lead screw center moves to the point A position under the left-hand system, and the camera takes a picture again to confirm whether the robot has reached the point A position. If not, the visual system sends the position compensation value to the robot again, and the robot moves again until it reaches point A, where the position compensation value is the deviation compensation value between the robot's arrival position and the target position given by the visual system; if it has arrived, the robot obtains the joint coordinate information A1 of the current hand system and sends it to the visual system.

[0036] After that, it automatically switches to the right-hand system, and the visual system takes a picture again to confirm whether the robot has reached the position coordinate, wherein the left-hand system and the right-hand system are calibrated in order to solve the zero-point compensation value of the forearm; if not, the visual system sends the position compensation value to the robot, so that the robot moves until it reaches the position coordinate; if the robot reaches the position coordinate, the robot obtains the joint coordinate information of the current right-hand system at the position coordinate and sends it to the visual system, and triggers a command to the visual system; specifically, it automatically switches to the right-hand system, and the camera 23 takes a picture again to confirm whether the robot has reached point A. If not, the visual system 2 sends the position compensation value to the robot again, and the robot moves again until it reaches point A, wherein the position compensation value is the deviation compensation value between the robot's arrival position and the target position given by the visual system; if it has arrived, the robot obtains the joint coordinate information A2 of the current hand system and sends it to the vision, and sends a trigger command to the visual system 2.

[0037] In step S6, the visual system will automatically send another position coordinate different from the position coordinate in step S5 to the robot after receiving the command. After receiving the coordinate information, the robot will automatically move to the position coordinate and repeat step S5. Specifically, the visual system 2 will automatically return and send the position of point B (any chessboard center point far from point A within the camera's field of view) to the robot after receiving the command. Then, according to the received coordinate information, the robot moves to point B under the left-hand system, and then repeats the above steps to send the joint coordinate information of B1 and B2 to the visual system 2. Points A and B are used to solve the parameters of the upper arm L1 and the forearm L2, and a set of equations to be solved are listed. If only one point is taken, the solution conditions are insufficient.

[0038] In the present invention, the coordinate relationship between the visual system and the robot is calibrated by recording the above-mentioned five points (0, A1, A2, B1, B2), the visual system sends the position information of points A and B, the robot automatically moves to the two points in turn, and uses the left and right hand systems to obtain the joint coordinate information of A1, A2, B1, B2 respectively, and then sends it to the vision, and the vision automatically records the joint coordinate information of the four points of the robot; the coordinate relationship between the visual system and the robot is calibrated by recording the five points, the vision sends the position information of points A and B, the robot automatically moves to the two points in turn, and uses the left and right hand systems to obtain the joint coordinate information of A1, A2, B1, B2 (the points of different hand systems of A and B respectively), and then sends it to the vision, and the vision automatically records the position information of points AB.

[0039] like Fig.13 As shown, the calibration steps are as follows: (1) The calibration plate is fixed on the end of the robot and connected to the vision through socket TCP. The automatic calibration program is compiled for calibration. First, the left-hand system is used to make point A1 coincide with the center point of the camera, and the angles of the two joints at this time are recorded. ; (2) The robot points to the camera from the right hand so that point A2 coincides with the center point of the camera and records the angles of the two joints. ; (3) Using the same method, make points B1 and B2 coincide with the center point of the camera and record the joint angles to obtain ; The kinematic equations associated with the calibration steps are: (8); Where: L1 is the length of the robot's upper arm, L2 is the length of the robot's lower arm, is the angle between the x-axis and OA; is the angle between OA and AB; It is the zero-point angle deviation between the theoretical value and the actual value of the robot's upper and lower arms.

[0040] According to the joint angles obtained at point A in the above steps and equation (8), the following set of equations can be obtained: (9); Among them, L1 is the rod length of the robot's upper arm, and L2 is the rod length of the robot's lower arm. is the zero-point angle deviation between the theoretical value and the actual value of the robot's upper and lower arms, are the joint angles of J1 and J2 at point A1, They are the joint angles of J1 and J2 at point A2 respectively.

[0041] Similarly, the equations at points B1 and B2 are: (10); Among them, L1 is the rod length of the robot's upper arm, and L2 is the rod length of the robot's lower arm. is the zero-point angle deviation between the theoretical value and the actual value of the robot's upper and lower arms, They are the joint angles of the upper arm J1 and the lower arm J2 at point B1 and the joint angles of the upper arm J1 and the lower arm J2 at point B2 respectively.

[0042] Solving the above equations, we can obtain the zero point compensation values ​​of L1, L2 and J2. .

[0043] The present invention realizes automatic calibration of SCARA robots through a high-precision visual system, reduces manual intervention, and improves calibration accuracy and efficiency, which is reflected in the following points: 1. Improve calibration accuracy: The existing SCARA robot calibration usually relies on manual measurement or simple calibration tools. The calibration results are easily affected by human factors, and the accuracy and repeatability are poor. The present invention can more accurately obtain the position, compensation value and other parameters of each joint of the robot, significantly improving the calibration accuracy.

[0044] 2. Reduce human intervention: Traditional calibration methods require professionals to perform precise mechanical measurements and manual adjustments, which is not only time-consuming and labor-intensive, but also prone to human errors. The present invention greatly reduces the need for human intervention and the risk of errors in manual operations through automated visual inspection and calculation.

[0045] 3. Improve calibration efficiency: Traditional calibration methods often require repeated debugging and manual correction, which is time-consuming. The present invention utilizes the automated photography and feedback calculation mechanism of the 2D automatic zoom vision system, which can significantly shorten the calibration cycle and improve calibration efficiency.

[0046] 4. Reduce costs: Traditional calibration methods rely on a lot of manual measurement and adjustment, and each time the machine is replaced or adjusted, it needs to be manually recalibrated, resulting in high labor costs and time loss. In addition, high-precision laser tracker calibration equipment is expensive and has high maintenance costs. The present invention can reduce the reliance on manual measurement and specialized equipment, thereby reducing overall production costs.

[0047] 5. Adapt to complex working environment: The calibration methods in the prior art are often difficult to adapt to the changing production environment and complex working conditions; the present invention can flexibly adapt to different working conditions and environmental conditions, and can still perform efficient and accurate calibration in a more complex production environment.

[0048] The present invention has the following advantages: 1. Automated calibration under vision guidance: Traditional SCARA robot calibration methods mostly rely on mechanical measurement tools (such as laser displacement sensors, standard blocks, MARK point auxiliary tools, etc.). This method is often not very accurate and complicated to operate. The present invention combines a 2D automatic zoom camera vision system and uses image processing technology to automatically capture the edge image of the robot's Z-axis screw rod and determine the center position of the image. The robot's rod length parameters and zero point compensation value are calculated based on the robot's position and posture, thereby quickly completing the calibration process and performing self-correction, greatly improving the calibration accuracy and consistency.

[0049] 2. Fully automatic calibration process: This invention avoids the tedious steps of manual intervention. Through the cooperation of the visual system and the control algorithm, it automatically completes the operations such as recording the joint angles of each axis of the robot and calculating the displacement parameters, further reducing the human operation error. This method does not require complex measurement tools, and can quickly generate a calibration model only through standard visual image data, saving equipment and labor costs.

[0050] 3. Accurate algorithm based on 2D auto-zoom image processing: The present invention uses a high-resolution auto-zoom camera to realize auto-focus photography of the chessboard calibration plate and the edge of the screw, can extract key points with high precision, and use efficient image processing algorithms for precise measurement; it is different from the traditional mechanical calibration method. Visual calibration can avoid the cumulative error caused by error transmission and provide more precise parameter calculation, thereby realizing the precise correction of small errors and improving the accuracy of the calibration process.

[0051] 4. Intelligent feedback and adaptive adjustment mechanism: The present invention adopts an intelligent feedback system, and the visual detection results interact with the robot control system in real time. By automatically feeding back errors, the system can automatically adjust the position of the robot according to real-time data to achieve the optimal calibration requirements. It can not only automatically adjust the algorithm parameters according to the image contour fitting accuracy, but also adaptively optimize the calibration strategy for different types of working environments.

[0052] 5. Modular and universal design: The present invention is highly versatile and can adapt to SCARA robots of different models and configurations. The configuration and control method of the 2D camera vision system can be flexibly adjusted to meet the calibration requirements in different application scenarios. It is not only suitable for SCARA robots, but can also be applied to other types of industrial robots or automation equipment through appropriate adjustments.

[0053] 6. Storage medium and calibration data management: During the calibration process of the present invention, all calibration data (such as joint angle, position error, correction parameters, etc.) can be stored in a database or cloud storage for subsequent retrieval and analysis. By storing and managing historical calibration data, the two-axis zero point compensation value can be automatically backfilled during the next zero point setting.

[0054] The present invention not only improves the calibration accuracy and efficiency of the SCARA robot, but also effectively reduces the production cost, has significant technical advantages and market application prospects, and has the following advantages: 1. Low cost: Traditional calibration methods rely on a lot of manual measurement and adjustment, and each time the machine is replaced or adjusted, it needs to be manually recalibrated, resulting in high labor costs and time loss; in addition, high-precision laser tracker calibration equipment is expensive and has high maintenance costs; the present invention can reduce the reliance on manual measurement and specialized equipment, thereby reducing overall production costs.

[0055] 2. High precision: The existing SCARA robot calibration usually relies on manual measurement or simple calibration tools. The calibration results are easily affected by human factors, and the accuracy and repeatability are poor. The present invention can more accurately obtain the position, compensation value and other parameters of each joint of the robot, thereby significantly improving the calibration accuracy.

[0056] 3. High efficiency: Traditional calibration methods often require repeated debugging and manual correction, which is time-consuming. The present invention utilizes the automated photography and feedback calculation mechanism of the 2D automatic zoom vision system, which can significantly shorten the calibration cycle, improve the calibration efficiency, and continuously realize large-scale robot calibration and debugging.

[0057] 4. Stability: The calibration methods in the prior art are often difficult to adapt to changing production environments and complex working conditions. The 2D automatic zoom camera vision system of the present invention can flexibly adapt to different working conditions and environmental conditions, and can still perform efficient and accurate calibration in a more complex production environment. It has shown good repeatability and stability in multiple experiments and is suitable for long-term application.

[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A vision-based SCARA robot automatic calibration method, characterized in that: The steps include: Step S1, connecting the visual system to the robot; Step S2, adjusting the calibration plate and the end surface of the Z axis of the robot to the same horizontal plane, adjusting the position of the calibration plate, and the visual system automatically focusing on the calibration plate; Step S3, move the Z-axis lead screw of the robot to above the center of the visual system's field of view, capture an initial calibration image, and set it as the zero position P0 of the visual system, and simultaneously obtain the world coordinates of the robot at the zero position and send it to the visual system; Step S4, moving the robot within the field of view of the visual system, acquiring the world coordinate system and the visual coordinate system of the robot, and obtaining the rotation and translation matrix of the robot coordinate system and the visual coordinate system; Step S5: The robot automatically sends a trigger command to the visual system. The visual system automatically returns a position coordinate after receiving the command. The robot automatically moves to the position coordinate after receiving the information returned by the visual system. The visual system confirms whether the robot has reached the sent position coordinate. According to the left and right hand systems, the robot obtains the joint coordinates of the left and right hand systems respectively based on the position coordinates sent by the visual system and sends them to the visual system. Step S6, selecting another position coordinate different from the position coordinate in step S5, and repeating step S5; Step S7, calculate and obtain the parameters of the robot's upper arm and lower arm and the lower arm zero point compensation value, thereby completing the calibration.

2. The vision-based SCARA robot automatic calibration method according to claim 1, characterized in that: In step S4, the robot moves to any four points (P1, P2, P3, P4) and records the world coordinates of the robot. At the same time, the visual system automatically takes pictures of the visual coordinates of the four points. According to the zero point position and the world coordinates and visual coordinates of the four points, the rotation and translation matrices of the robot coordinate system and the visual coordinate system are obtained through coordinate rotation and translation calculation.

3. The vision-based SCARA robot automatic calibration method according to claim 2, characterized in that: In step S5, the robot automatically sends a trigger command to the vision system. After receiving the command, the vision system automatically returns a position coordinate. After the robot receives the information returned by the vision system, the center of the robot's Z-axis screw moves to the position coordinate under the left-hand system. The vision system takes a photo to confirm whether the robot has reached the position coordinate. If not, the vision system sends the position compensation value to the robot, so that the robot moves until it reaches the position coordinate.

4. The vision-based SCARA robot automatic calibration method according to claim 3, characterized in that: In step S5, if the robot reaches the position coordinate, the robot obtains the joint coordinate information of the current left hand system at the position coordinate and sends it to the visual system.

5. The vision-based SCARA robot automatic calibration method according to claim 4, characterized in that: In step S5, the robot automatically switches to the right-hand system, and the visual system takes pictures again to confirm whether the robot has reached the position coordinates. If not, the visual system sends the position compensation value to the robot, causing the robot to move until it reaches the position coordinates.

6. The vision-based automatic calibration method for SCARA robots according to claim 5, characterized in that: If the robot reaches the position coordinate, the robot obtains the joint coordinate information of the current right hand system at the position coordinate and sends it to the visual system, and triggers a command to the visual system at the same time.

7. The vision-based automatic calibration method for SCARA robots according to claim 6, characterized in that: In step S6, the visual system automatically sends another position coordinate different from the position coordinate in step S5 to the robot after receiving the command. After receiving the coordinate information, the robot automatically moves to the position coordinate and repeats step S5.

8. The vision-based SCARA robot automatic calibration device is characterized by: It includes a calibration tool fixed on the Z-axis screw of the robot and a visual system for calibration of automatic focus photography, the calibration tool includes a calibration plate limit block, a calibration plate and a fixed plate, the upper surface of the calibration plate limit block is symmetrically connected with a first clamp and a second clamp, a positioning groove is provided on the fixed plate, the calibration plate is fixed in the positioning groove, the fixed plate is provided with a fixing plate through hole for exposing the calibration plate, the bottom surface of the calibration plate limit block is connected to the fixed plate to cover the calibration plate, and the Z-axis screw of the robot passes through the calibration tool.

9. The vision-based SCARA robot automatic calibration device according to claim 8, characterized in that: The visual system includes a bracket connected to the robot, a ring-shaped light source and a camera connected to the bracket, and an external industrial computer. The ring-shaped light source is located above the camera. The camera is electrically connected to the industrial computer. The industrial computer is provided with a receiving module for receiving information, a processing module for analysis and processing, and a storage module for storing data running the processing module and recording data.

10. A vision-based storage medium, characterized in that It stores data information for robot automatic calibration that is used to implement the automatic calibration method as described in any one of claims 1-7 when executed by the processing module.

Images