Vision-guided six-degree-of-freedom parallel platform docking method and system

Abstract

The invention relates to the technical field of robots, in particular to a visual guidance six-degree-of-freedom parallel platform docking method and system. The device comprises a parallel robot platform, a butt joint piece clamp, a workpiece lifting mechanism, a butt joint piece, a butt joint piece and a butt joint piece support. The system further comprises a camera frame, a binocular camera, a fixed end two-dimensional code adapter, a mobile end two-dimensional code adapter, a platform end two-dimensional code adapter, a fixed end two-dimensional code, a mobile end two-dimensional code and a platform end two-dimensional code. The two-dimensional code pose of the fixed end / the mobile end is detected through the visual system, the pre-calibrated coordinate system conversion relation is combined, the target pose of the parallel robot relative to the butt joint face is solved in real time, and then six-dimensional control parameters are generated to drive the platform to complete high-precision pose adjustment and butt joint.

Description

Technical Field

[0001] The present invention relates to the technical field of robotics, and particularly to a vision-guided six-degree-of-freedom parallel platform docking method and system. Background Art

[0002] With the continuous improvement of industrial automation level, robot technology is more and more widely used in fields such as manufacturing and assembly. In many application scenarios, precise docking is an important technical challenge. Especially in a parallel platform system with multi-degree-of-freedom motion, docking accuracy and efficiency are often the key factors affecting operation quality and production efficiency. Existing six-degree-of-freedom parallel platforms usually use mechanical sensors, force sensors, or encoders and other means for position and attitude detection and control. However, these traditional methods often rely on precise calibration of hardware and complex control algorithms, are difficult to adapt to changing working environments, and have certain limitations for real-time docking effect control in dynamic environments. At the same time, the manual docking method is not only inefficient but also difficult to ensure high precision.

[0003] In order to solve the above problems, a vision-guided six-degree-of-freedom parallel platform docking method is designed. By combining vision measurement technology and parallel platform motion control, it aims to solve the problems existing in the prior art and further improve the accuracy and efficiency of the automatic docking process. Summary of the Invention

[0004] The object of the present invention is to propose a vision-guided six-degree-of-freedom parallel platform docking method and system for the problems existing in the background art. This method can effectively reduce the error caused by the offset of the rotation center due to the installation height of the docked part, and can also reduce the docking error caused by the depth data error of vision, thereby effectively improving the docking accuracy between the docking part and the docked part.

[0005] In the technical solution of the present invention, in the first aspect of the present invention, a vision-guided six-degree-of-freedom parallel platform docking system is provided, including a parallel robot platform, a docking part fixture, a workpiece lifting mechanism, a docking part, a docked part, a docked part support, and further including a camera frame, a binocular camera, a fixed-end two-dimensional code adapter, a mobile-end two-dimensional code adapter, a platform-end two-dimensional code adapter, a fixed-end two-dimensional code, a mobile-end two-dimensional code, and a platform-end two-dimensional code;

[0006] The workpiece lifting mechanism is fixedly installed on the parallel robot platform, and the workpiece fixture is fixedly installed on the workpiece lifting mechanism; the fixed-end two-dimensional code adapter is fixedly installed on the parallel robot platform, the mobile-end two-dimensional code adapter is fixedly installed on the docking part, and the platform-end two-dimensional code adapter is fixedly installed on the docked part; the fixed-end two-dimensional code, the mobile-end two-dimensional code, and the platform-end two-dimensional code are respectively fixed on the fixed-end two-dimensional code adapter, the mobile-end two-dimensional code adapter, and the platform-end two-dimensional code adapter;

[0007] Among them, the central axis of the docking part is made to always be coplanar with the upper plane of the parallel robot platform through the workpiece lifting mechanism.

[0008] Preferably, the docking system includes several coordinate systems, namely: the fixed part QR code coordinate system, the fixed part coordinate system, the moving part coordinate system, the vision system coordinate system, the moving part QR code coordinate system, the platform end QR code coordinate system, the parallel robot control coordinate system, and the robot target pose coordinate system.

[0009] Preferably, the docking system establishes a vision system coordinate system through a binocular camera, determines the conversion relationship between the vision system and the QR code at the platform end, and determines the geometric relationship between the platform end QR code and the robot control system coordinate system; determines the target pose according to the geometric relationship, controls according to the installed target pose, and completes the docking of the parallel robot.

[0010] The second aspect of the present invention provides a vision-guided six-degree-of-freedom parallel platform docking method, which is applied to the above docking system and includes the following specific steps:

[0011] S1. Establish the conversion relationship between the vision system coordinate system and the docking surface coordinate system;

[0012] S2. Determine the origin and direction of each coordinate system;

[0013] S3. Determine the conversion relationship between the fixed part, the moving part, the QR code conversion part, the QR code, and the parallel robot;

[0014] S4. Determine the conversion relationship between each coordinate system;

[0015] S5. Use the vision system to measure the pose of the fixed part QR code, the moving part QR code, and the platform end QR code;

[0016] S6. Through the detected pose data, obtain the required robot target pose H through the conversion relationship between each coordinate system;

[0017] S7. Use the obtained robot target pose H to send a control command to the parallel robot to complete the docking work of the parallel robot.

[0018] Preferably, in step S1, the conversion relationship between the vision system coordinate system and the docking surface coordinate system is expressed as follows:

[0019]

[0020] In the formula, T d , T j are respectively the pose representations of the docking surfaces of the moving part and the fixed part obtained by vision system measurement, H dv , H jvThey are the pose representations of the fixed-end QR code and the mobile-end QR code obtained by the vision system measurement, H d 、H j That is the pose conversion relationship between the QR code and the docking surface.

[0021] Preferably, in step S2, the origin of the fixed part coordinate system is located at the center of the end face of the fixed part. When the parallel robot is in the completed docking state, the x, y, and z axes of this coordinate system are in the same directions as the x, y, and z axes of the parallel robot;

[0022] The origin of the moving part coordinate system is located at the center of the end face of the moving part. When the parallel robot is in the completed docking state, the x, y, and z axes of this coordinate system are in the same directions as the x, y, and z axes of the parallel robot;

[0023] Preferably, in step S2, the origin of the fixed part QR code coordinate system is located at the center of the fixed part QR code. When the parallel robot is in the completed docking state, the x, y, and z axes of this coordinate system are in the same directions as the x, y, and z axes of the parallel robot;

[0024] The origin of the moving part QR code coordinate system is located at the center of the moving part QR code. When the parallel robot is in the completed docking state, the x, y, and z axes of this coordinate system are in the same directions as the x, y, and z axes of the parallel robot;

[0025] The origin of the platform-end QR code coordinate system is located at the center of the platform-end QR code. When the parallel robot is in the completed docking state, the x, y, and z axes of this coordinate system are in the same directions as the x, y, and z axes of the parallel robot;

[0026] Preferably, in step S2, the origin of the robot control coordinate system is located at the center of the circle formed by the centers of the six articulated axes of the parallel robot, and the x, y, and z axes are in the same directions as the x, y, and z axes of the parallel robot.

[0027] Preferably, in step S3, the conversion relationship H j between the fixed part docking surface pose coordinate system and the fixed part QR code pose coordinate system is obtained through the mechanical structure, d the conversion relationship between the moving part docking surface pose coordinate system and the moving part QR code coordinate system, p and the conversion relationship between the robot control system coordinate system and the platform-end QR code pose coordinate system.

[0028] Preferably, in step S4, through the conversion relationship T0 between the vision system and the platform-end QR code and the geometric relationship H p between the platform-end QR code and the robot control system coordinate system,

[0029]

[0030] The H is obtained by measuring the target with a visual system jv , H dv , obtain the coordinate system transformation relationship T of the mobile component and the fixed component under the robot control system d , T j ;

[0031]

[0032] The control matrix H of the robot to reach the target posture and the robot's six-dimensional control parameters corresponding to the matrix H can be obtained from the conversion relationship:

[0033]

[0034] Combining the above formulas, we can get:

[0035]

[0036] Compared with the prior art, the present invention has the following beneficial technical effects:

[0037] 1. The present invention can effectively reduce the excessive control error caused by the control center offset due to installation through the sinking design of the parallel mechanism;

[0038] 2. The visually guided six-degree-of-freedom parallel platform docking system designed by the present invention has low cost, simple operation and high cost performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of a system diagram of the present invention.

[0040] Figure 2 It is a schematic diagram of the system and the positions of each coordinate system of the present invention.

[0041] Figure 3 is the conversion relationship of each coordinate system of the present invention;

[0042] Figure 4 It is a schematic diagram of docking error analysis in the prior art.

[0043] Reference Numerals: 1, camera frame; 2, binocular camera; 3, workpiece to be docked; 4, support for workpiece to be docked; 5, fixed-end QR code; 6, fixed-end QR code adapter; 7, docking part; 8, mobile-end QR code adapter; 9, mobile-end QR code; 10, platform-end QR code adapter; 11, platform-end QR code; 12, workpiece fixture; 13, workpiece lifting mechanism; 14, parallel robot platform; 15, fixed-piece QR code coordinate system; 16, fixed-piece coordinate system; 17, moving-piece coordinate system; 18, vision system coordinate system; 19, moving-piece QR code coordinate system; 20, platform-end QR code coordinate system; 21, parallel robot control coordinate system; 22, robot target pose coordinate system. Detailed Embodiment

[0044] Embodiment 1

[0045] Most current docking systems adjust components through independent five-degree-of-freedom or six-degree-of-freedom columns, and multiple columns are used to adjust the docking pose, resulting in a relatively complex control system. Secondly, when a parallel robot is used to achieve component docking, it usually adopts the form of a six-degree-of-freedom platform combined with a tooling fixture, and the tooling fixture is installed on the upper surface of the six-degree-of-freedom platform. However, since the installation height will affect the adjustment accuracy of the parallel robot, especially the influence of the angle error, the higher the installation height, the greater the influence of the angle error. As Figure 4 shown, when the docking section of the workpiece and the tabletop installation height is H, and when there is an angle error δ in the vision system, according to the position operation of the parallel robot, the resulting position error is:

[0046] Δp = H·sin(δ)

[0047] It can be seen from the above formula that when the angle measurement error δ of the vision system is 0.01 degree, with the increase of H, a large positioning error will be caused. Therefore, it is necessary to minimize the influence on the docking accuracy. This patent redesigned a six-degree-of-freedom parallel robot, combined the tooling fixture with the parallel robot, and reduced the installation height.

[0048] As Figure 1-2 shown, a vision-guided six-degree-of-freedom parallel platform docking system proposed by the present invention includes a parallel robot platform 14, a docking part fixture 12, a workpiece lifting mechanism 13, a docking part 7, a workpiece to be docked 3, a support for workpiece to be docked 4, and also includes a camera frame 1, a binocular camera 2, a fixed-end QR code adapter 6, a mobile-end QR code adapter 8, a platform-end QR code adapter 10, a fixed-end QR code 5, a mobile-end QR code 9, and a platform-end QR code 11;

[0049] The workpiece lifting mechanism 13 is fixedly installed on the parallel robot platform 14, and the workpiece fixture 12 is fixedly installed on the workpiece lifting mechanism 13; the fixed-end QR code adapter 6 is fixedly installed on the parallel robot platform 14, the mobile-end QR code adapter 8 is fixedly installed on the docking part 7, and the platform-end QR code adapter 10 is fixedly installed on the part to be docked 3; the fixed-end QR code 5, the mobile-end QR code 9, and the platform-end QR code 11 are respectively fixed on the fixed-end QR code adapter 6, the mobile-end QR code adapter 8, and the platform-end QR code adapter 10;

[0050] Among them, the central axis of the docking part 7 passes through the workpiece lifting mechanism 13 so that it always maintains a coplanar relationship with the upper plane of the parallel robot platform 14.

[0051] The docking system includes several coordinate systems, namely: the fixed-piece QR code coordinate system 15, the fixed-piece coordinate system 16, the moving-piece coordinate system 17, the vision system coordinate system 18, the moving-piece QR code coordinate system 19, the platform-end QR code coordinate system 20, the parallel robot control coordinate system 21, and the robot target pose coordinate system 22.

[0052] The docking system establishes a vision system coordinate system through the binocular camera 2, determines the conversion relationship between the vision system and the QR code on the platform end, and determines the geometric relationship between the platform-end QR code and the robot control system coordinate system; determines the target pose according to the geometric relationship, and controls according to the installed target pose to complete the docking of the parallel robot.

[0053] In this embodiment, through the sinking design of the parallel mechanism, the excessive control error caused by the offset of the control center due to the installation can be effectively reduced; the designed vision-guided six-degree-of-freedom parallel platform docking system has a low cost, is easy to operate, and has a high cost performance.

[0054] Embodiment 2

[0055] As Figure 3 shown, a vision-guided six-degree-of-freedom parallel platform docking method proposed by the present invention includes the following specific steps in this embodiment

[0056] Step 1: Establish the conversion relationship between the vision system coordinate system and the docking interface coordinate system. The pose conversion relationship between the QR code and the docking surface can be obtained as follows:

[0057]

[0058] In the formula, Td and Tj are the pose representations of the moving part and the fixed part docking surface measured by the vision system respectively, Hdv and Hjv are the pose representations of the fixed-end QR code and the mobile-end QR code measured by the vision system respectively, and Hd and Hj are the pose conversion relationships between the QR code and the docking surface;

[0059] Step 2: Determine the origin and direction of each coordinate system. The origin of the fixed part coordinate system is located at the center of the end face of the fixed part. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as those of the parallel robot's x, y, and z axes; the origin of the moving part coordinate system is located at the center of the end face of the moving part. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as those of the parallel robot's x, y, and z axes; the origin of the fixed part QR code coordinate system is located at the center of the fixed part QR code. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as those of the parallel robot's x, y, and z axes; the origin of the moving part QR code coordinate system is located at the center of the moving part QR code. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as those of the parallel robot's x, y, and z axes; the origin of the platform end QR code coordinate system is located at the center of the platform end QR code. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as those of the parallel robot's x, y, and z axes. The origin of the robot control coordinate system is located at the center of the circle formed by connecting the centers of the six articulated axes of the parallel robot, and the x, y, and z axis directions are the same as those of the parallel robot's x, y, and z axes; the robot target pose coordinate system is the same as the robot control coordinate system.

[0060] Step 3: Determine the conversion relationships between the fixed part, the moving part, the three QR code conversion parts, the three QR codes, and the parallel robot. Since all the structural data are known, the conversion relationship Hj between the fixed part docking surface pose coordinate system and the fixed part QR code pose coordinate system, the conversion relationship Hd between the moving part docking surface pose coordinate system and the moving part QR code coordinate system, and the conversion relationship Hp between the robot control system coordinate system and the platform end QR code pose coordinate system can be obtained from the mechanical structure.

[0061] Step 4: Determine the conversion relationships between the coordinate systems. Through the conversion relationship T0 between the vision system and the platform end QR code and the geometric relationship Hp between the platform end QR code and the robot control system coordinate system, the conversion relationship between the vision system and the robot control system can be obtained

[0062]

[0063] By measuring the target with the vision system to obtain Hjv and Hdv, the coordinate system conversion relationships Td and Tj of the moving component and the fixed component under the robot control system can be further obtained from the Figure 3 conversion relationship

[0064]

[0065] The control matrix H for the robot to reach the target pose and the six-dimensional control parameters of the robot corresponding to the matrix H can be obtained from the conversion relationship:

[0066]

[0067] Substituting Eqs. (4) and (5) into Eq. (6), we can obtain:

[0068]

[0069] Step 5: Use the vision system to measure the pose of the QR code on the fixture, the pose of the QR code on the moving part, and the pose of the QR code on the platform end.

[0070] Step 6: Based on the detected pose data, obtain the required target pose H of the robot through the conversion relationship between the coordinate systems.

[0071] Step 7: Use the obtained target pose H of the robot to send control instructions to the parallel robot, thereby completing the docking work of the parallel robot.

[0072] In this embodiment, the method is applied to fields such as robot automation, precision docking, and assembly, and is particularly suitable for industrial scenarios that require high-precision and high-efficiency docking operations. By combining vision guidance technology with the motion control of the parallel platform, the present invention can achieve the automation, intelligence, and precision of the docking process, and is widely used in industries such as aerospace, automated manufacturing, and precision assembly.

[0073] In the present invention, the tooling fixture is integrated with the parallel robot, and the installation height is reduced through a sinking design, significantly reducing the positioning deviation Δp = H·sin(δ) caused by the visual angle error δ, and improving the adjustment accuracy; a vision system coordinate system is constructed using a binocular camera, and a multi-coordinate system mapping relationship is established through three groups of QR codes (and their adapters) at the fixed end, the mobile end, and the platform end, combined with mechanical calibration parameters (H j 、H d 、H p ), to achieve the spatial closed-loop conversion of pose data; further improving the docking accuracy;

[0074] The vision-guided six-degree-of-freedom parallel platform docking method proposed by the present invention detects the pose of the QR code at the fixed / mobile end through the vision system, combines the pre-calibrated coordinate system conversion relationship, and calculates in real time the target pose of the parallel robot relative to the docking surface, and then generates six-dimensional control parameters to drive the platform to complete high-precision pose adjustment and docking.

[0075] The embodiments of the present invention have been described in detail above with reference to the drawings, but the present invention is not limited thereto. Various changes can be made without departing from the spirit of the present invention within the scope of knowledge possessed by those skilled in the art to which the present invention pertains.

Claims

1. A vision-guided six-degree-of-freedom parallel platform docking system, comprising a parallel robot platform, a docking part fixture, a workpiece lifting mechanism, a docking part, a part to be docked, and a support for the part to be docked, characterized in that, It also includes a camera frame, a binocular camera, a fixed-end QR code adapter, a mobile-end QR code adapter, a platform-end QR code adapter, a fixed-end QR code, a mobile-end QR code, and a platform-end QR code; The workpiece lifting mechanism is fixedly installed on the parallel robot platform, and the workpiece fixture is fixedly installed on the workpiece lifting mechanism; the fixed-end QR code adapter is fixedly installed on the parallel robot platform, the mobile-end QR code adapter is fixedly installed on the docking part, and the platform-end QR code adapter is fixedly installed on the part to be docked; the fixed-end QR code, the mobile-end QR code, and the platform-end QR code are respectively fixed on the fixed-end QR code adapter, the mobile-end QR code adapter, and the platform-end QR code adapter; wherein the central axis of the docking part passes through the workpiece lifting mechanism so that it always maintains a coplanar relationship with the upper plane of the parallel robot platform.

2. The vision-guided six-degree-of-freedom parallel platform docking system according to claim 1, wherein The docking system includes several coordinate systems; namely: the fixed part QR code coordinate system, the fixed part coordinate system, the moving part coordinate system, the vision system coordinate system, the moving part QR code coordinate system, the platform-end QR code coordinate system, the parallel robot control coordinate system, and the robot target pose coordinate system.

3. The vision-guided six-degree-of-freedom parallel platform docking system according to claim 1 or 2, characterized in that, The docking system establishes a vision system coordinate system through the binocular camera, determines the conversion relationship between the vision system and the QR code on the platform end, and determines the geometric relationship between the platform-end QR code and the robot control system coordinate system; determines the target pose according to the geometric relationship, and controls according to the installed target pose to complete the docking of the parallel robot.

4. A vision-guided six-degree-of-freedom parallel platform docking method, applied to the docking system described in any one of claims 1-3, characterized in that, It includes the following specific steps: S1. Establish the conversion relationship between the vision system coordinate system and the docking surface coordinate system; S2. Determine the origin and direction of each coordinate system; S3. Determine the conversion relationship between the fixed part, the moving part, the QR code adapter, the QR code, and the parallel robot; S4. Determine the conversion relationship between each coordinate system; S5. Use the vision system to measure the pose of the fixed part QR code, the moving part QR code, and the platform-end QR code; S6. Through the detected pose data, obtain the required robot target pose H through the conversion relationship between each coordinate system; S7. Use the obtained robot target pose H to send a control instruction to the parallel robot to complete the docking work of the parallel robot.

5. The vision-guided six-degree-of-freedom parallel platform docking method according to claim 4, wherein In step S1, the conversion relationship between the vision system coordinate system and the docking surface coordinate system is expressed by the following formula: where, T d and T j are respectively the pose representations of the mating surfaces of the moving part and the fixed part obtained by visual system measurement. H dv and H jv are respectively the pose representations of the fixed-end QR code and the mobile-end QR code obtained by visual system measurement. H d and H j are the pose conversion relationships between the QR code and the mating surface.

6. The visual guidance six-degree-of-freedom parallel platform docking method according to claim 4, characterized in that In step S2, the origin of the fixed part coordinate system is located at the center of the end face of the fixed part. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as the x, y, and z axis directions of the parallel robot; The origin of the moving part coordinate system is located at the center of the end face of the moving part. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as the x, y, and z axis directions of the parallel robot.

7. The vision-guided six-degree-of-freedom parallel platform docking method according to claim 4, characterized in that, In step S2, the origin of the fixed part QR code coordinate system is located at the center of the fixed part QR code. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as the x, y, and z axis directions of the parallel robot; The origin of the moving part QR code coordinate system is located at the center of the moving part QR code. When the parallel robot is in the completed docking state, the x, y, and z axis directions of this coordinate system are the same as the x, y, and z axis directions of the parallel robot; The origin of the QR code coordinate system on the platform is located at the center of the QR code on the platform. When the parallel robot is in the process of docking, the x, y, and z axis directions of the coordinate system are consistent with the x, y, and z axis directions of the parallel robot.

8. The vision-guided six-degree-of-freedom parallel platform docking method according to claim 4, characterized in that In step S2, the origin of the robot control coordinate system is located at the center of the circle formed by the centers of the six articulated axes of the parallel robot, and the directions of the x, y, and z axes are consistent with those of the parallel robot.

9. The vision-guided six-degree-of-freedom parallel platform docking method according to claim 4, wherein In step S3, the conversion relationship H between the pose coordinate system of the fixed part docking surface and the pose coordinate system of the fixed part QR code is obtained through the mechanical structure j and the conversion relationship H between the pose coordinate system of the moving part docking surface and the moving part QR code coordinate system d and the conversion relationship H between the robot control system coordinate system and the pose coordinate system of the platform-side QR code p .

10. The vision-guided six-degree-of-freedom parallel platform docking method according to claim 4, wherein In step S4, based on the conversion relationship T0 between the vision system and the platform-side QR code and the geometric relationship H between the platform-side QR code and the coordinate system of the robot control system p , the conversion relationship between the vision system and the robot control system can be obtained: Obtain H by measuring the target through the vision system jv and H dv to obtain the coordinate transformation relationships T d and T j ; The control matrix H of the robot to reach the target posture and the robot's six-dimensional control parameters corresponding to the matrix H can be obtained from the conversion relationship: Combining the above formulas, we can get:

Images