Butt-joint assembly method for clamping cabin section A and fixed section B based on visual guidance robot

Abstract

The invention relates to a method for clamping a cabin section A to complete butt-joint assembly with a cabin section B by a visual guidance robot. The mode that six degrees of freedom x, y, z, rx, ry and rz are measured and matched separately is adopted, measurement and matching of a circumferential rolling pin groove angle rx are completed through monocular vision, fixed alignment and matching of a pitching angle ry and a yaw angle rz are completed through a clamping tool and a fixing workpiece, distance adjustment in the y direction and the z direction in a butt joint plane is completed through monocular vision, and matching is completed. Adjustment of the distance in the x direction is completed through teaching to complete matching, finally six-degree-of-freedom adjustment and matching of the butt joint cabin sections A and B are achieved, and efficient and high-precision automatic assembly of the small-caliber and thin-wall cabin sections A and B is achieved.

Description

Technical Field

[0001] The invention relates to the field of robots, vision-guided methods and thin-walled cabin section docking, and in particular to a method for completing docking and assembly of a thin-walled cabin section A with a fixed section B by a robot clamping the thin-walled cabin section based on monocular vision guidance. Background Art

[0002] Cabin docking and assembly is a necessary link and key technology for missile assembly. Currently, manual docking is mostly used, which has problems such as low assembly efficiency, difficulty in ensuring accuracy and quality. Automatic assembly needs to solve the problem of adjusting the 6-DOF posture of cabin A and cabin B to be consistent and matched. Since it is difficult to complete the docking posture measurement of the entire period from the beginning to the end of cabin docking, automatic docking becomes extremely difficult. Researchers dock by scanning three-dimensional models, reverse modeling, indirect targets, etc., but when using the above methods for some thin-walled parts and small-caliber cabins, the docking assembly becomes extremely complicated and becomes inapplicable. Serial robots have the advantages of large moving range and high repeatability, and visual measurement has the advantages of non-contact high precision. If visual guidance robots are used, the DOF split matching is used to gradually solve the 6-DOF adjustment matching problem, which can solve the problem of automatic docking of small-diameter and thin-walled cabins and improve the efficiency of docking assembly and the consistency of product quality. Summary of the invention

[0003] The purpose of the present invention is to provide a docking assembly method for a robot-clamped compartment section A and a fixed section B based on vision guidance, so as to overcome the above-mentioned defects.

[0004] The docking assembly method of the cabin section A and the fixed section B by the robot clamping the cabin section A based on vision guidance is specifically as follows: the cabin section B is fixed on the bracket, the robot clamps the cabin section A and uses vision guidance to the template position, and the robot teaching is used to complete the automatic docking assembly of the cabin section A and the cabin section B, wherein the use of vision guidance to the template position includes the following steps:

[0005] 1) Construct the robot coordinate system;

[0006] 2) Use monocular vision to obtain cross-sectional images of the two compartments and obtain the center position of the axis of the thin-walled circular section and the position of the pin groove;

[0007] 3) Calculate the center position deviation and angle deviation between the current position of cabin A and the template position using the fitting calculation method;

[0008] 4) Based on the center position deviation and angle deviation, use the target instruction iteration method to adjust compartment A to the template position.

[0009] The step 1) is specifically as follows:

[0010] A right-handed coordinate system is established with the horizontal axis of the cabin as the x-axis and the vertical axis within the cabin section as the z-axis.

[0011] The step 3) comprises the following steps:

[0012] 3.1) Set the coordinates of the center of the thin-walled circle in the image coordinate system to (u0, v0), the center of the pin groove to (u1, v1), and the fitting radius to Perform Hough transform at (u0, v0) to obtain the expanded coordinates of the slot as m and the rotation angle as rx = m / (2π*r);

[0013] 3.2) The scale factor μ is obtained through calibration calculation, and then the distance that needs to be adjusted is obtained as du=μ(U0-u0), dv=μ(V0-v0).

[0014] The scale factor μ is specifically:

[0015] The actual measured physical diameter of the thin-walled circular section is D mm, and the diameter measured in the image coordinate system is d pixels, so μ = D / d.

[0016] The step 4) is specifically as follows:

[0017] The robot calculates the angle of the current position by sending instructions, obtains the deviation, adjusts the deviation, sends instructions again, and adjusts the deviation until the current position angle deviation and position deviation are both less than the threshold. The position and angle are matched, the sending is stopped, and the automatic assembly teaching stage is entered.

[0018] The present invention has the following advantages and beneficial effects:

[0019] 1. The present invention describes a method based on monocular vision to guide a robot to clamp the thin-walled cabin section A and complete the docking assembly with the fixed section B. The sensor only uses monocular vision, which has the advantages of few measurement parameters, low cost and easy use.

[0020] 2. The method for assembling the thin-walled cabin section A and the fixed section B by using a monocular vision-guided robot to clamp the thin-walled cabin section A is described in the present invention. It does not require scanning, reverse modeling, etc., which greatly simplifies the process of automated assembly and improves assembly efficiency.

[0021] 3. The method for the present invention of using a monocular vision-guided robot to clamp the thin-walled cabin section A and complete docking and assembly with the fixed section B adopts step-by-step matching of six degrees of freedom, which improves the flexibility of detection and has the advantages of efficient decoupling and high matching accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a characteristic schematic diagram of cabin section A and cabin section B of the present invention;

[0023] Figure 2 This is a schematic diagram of the assembly of cabin section A and cabin section B of the present invention;

[0024] Figure 3 Assemble the flow chart of the present invention;

[0025] Figure 4 This is a flow chart of the target instruction iteration method of the present invention. DETAILED DESCRIPTION

[0026] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0027] like Figure 1 As shown, cabin section A and cabin section B are thin-walled parts with pin groove positioning. The position and posture of A and B need to be adjusted to complete the assembly of A and B. The posture includes that the axes of A and B need to be parallel, the phase angles of the pin grooves need to be adjusted to be consistent, and the position adjustment requires that the cross-sectional center position coordinates of A and B are consistent. After the docking is completed, the two cross sections coincide.

[0028] like Figure 2 As shown, for the convenience of description, a coordinate system is established as shown in the figure, with the axial direction being the x-axis, the vertical direction in the cross section being the z-axis, and a right-handed coordinate system is established; wherein, cabin section A uses a manipulator through a specific fixture to keep the axial direction constant each time the cabin section is clamped; cabin section B uses a specific fixed bracket to keep the axial direction of cabin section B constant on the bracket each time.

[0029] like Figure 3 As shown, the specific operation process is to fix cabin section B on the bracket, the robot clamps cabin section A and adopts visual guidance to the template position, and uses robot teaching to complete the automatic docking and assembly of cabin section A and cabin section B. The present invention adopts a method of separate degree of freedom matching, wherein the pitch angle ry and yaw angle rz are guaranteed by clamping tools, which is superior to a thin-walled tube as a cylinder. After fixing and tightening by clamping hands and fixed brackets, the consistent fixation of the axis can be achieved, that is, ry and rz can be adaptively adjusted by mechanical devices to achieve self-matching of the axis angle, and it has a high repeatability.

[0030] The rolling angle rx and the docking center position (y, z) are detected by monocular vision. The specific detection method is to obtain the cross-sectional image of the product by monocular vision. The cross-sectional image is a thin-walled cylindrical section with pin and groove features. The center position of the axis of the thin-walled circular section and the pin and groove position can be obtained by image processing technology, and then the center position deviation and angle deviation are calculated by fitting calculation method.

[0031] Set the image coordinate system section thin-wall circle center coordinates to (u0, v0), the pin groove center coordinates to (u1, v1), and the fitting radius to Perform Hough transform at (u0, v0), the expanded coordinates of the slot after the transform are m, and the rotation angle is rx = m / (2π*r). Set the template center position coordinates to (U0, V0), and the scale factor μ can be obtained through calibration calculation. The specific method for calibrating the scale factor is that the actual measured physical diameter of the target is Dmm, and the diameter measured in the image system is d pixels, then μ = D / d. The distance to be adjusted is du = μ(U0-u0), dv = μ(V0-v0). The adjustment method is the target instruction iteration method. The robot sends the current position angle calculation instruction to obtain the deviation, then adjusts the deviation, and then sends the instruction again until the current position angle deviation dr<0.05°, the position deviation du<0.05mm, and dv<0.05mm. The position and angle matching is completed, the sending stops, and the automatic assembly teaching stage is entered.

[0032] Through the above method, the matching of position and angle is completed, and then by using robot teaching, the six-degree-of-freedom adjustment and matching of docking sections A and B are finally realized, realizing efficient and high-precision automatic assembly of small-caliber, thin-walled sections A and B.

[0033] like Figure 4 As shown in the figure, the servo adjustment method of the robot and vision is the target instruction iteration method. The robot sends the current position angle calculation instruction to obtain the deviation, then adjusts the deviation, and sends the instruction again until the current position angle deviation dr<0.05°, the position deviation du<0.05mm, and dv<0.05mm. The position and angle matching is completed and the automatic assembly stage is entered.

[0034] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should be regarded as within the scope of protection of the present invention.

Claims

1. A docking assembly method for a robot-held compartment section A and a fixed section B based on vision guidance, characterized in that: Specifically, cabin section B is fixed on the bracket, the robot clamps cabin section A and uses visual guidance to the template position, and uses robot teaching to complete the automatic docking and assembly of cabin section A and cabin section B, wherein the use of visual guidance to the template position includes the following steps: 1) Construct the robot coordinate system; 2) Use monocular vision to obtain cross-sectional images of the two compartments and obtain the center position of the axis of the thin-walled circular section and the position of the pin groove; 3) Calculate the center position deviation and angle deviation between the current position of cabin A and the template position using the fitting calculation method; 4) Based on the center position deviation and angle deviation, use the target instruction iteration method to adjust compartment A to the template position.

2. The method for docking and assembling the cabin section A and the fixed section B based on vision-guided robot according to claim 1 is characterized in that: The step 1) is specifically as follows: A right-handed coordinate system is established with the horizontal axis of the cabin as the x-axis and the vertical axis within the cabin section as the z-axis.

3. The method for docking and assembling the cabin section A and the fixed section B based on vision-guided robot according to claim 1 is characterized in that: The step 3) comprises the following steps: 3.1) Set the coordinates of the center of the thin-walled circle in the image coordinate system to (u0, v0), the center of the pin groove to (u1, v1), and the fitting radius to Perform Hough transform at (u0, v0) to obtain the expanded coordinates of the slot as m and the rotation angle as rx = m / (2π*r); 3.2) The scale factor μ is obtained through calibration calculation, and then the distance that needs to be adjusted is obtained as du=μ(U0-u0), dv=μ(V0-v0).

4. The method for docking and assembling the cabin section A and the fixed section B based on vision-guided robot according to claim 3 is characterized in that: The scale factor μ is specifically: The actual measured physical diameter of the thin-walled circular section is D mm, and the diameter measured in the image coordinate system is d pixels, so μ = D / d.

5. The method for docking and assembling the cabin section A and the fixed section B based on vision-guided robot according to claim 1, characterized in that: The step 4) is specifically as follows: The robot calculates the angle of the current position by sending instructions, obtains the deviation, adjusts the deviation, sends instructions again, and adjusts the deviation until the current position angle deviation and position deviation are both less than the threshold. The position and angle are matched, the sending is stopped, and the automatic assembly teaching stage is entered.

Images