Humanoid welding robot based on machine vision
By integrating the left arm welding gun and right arm vision device on the welding robot, the problem of limiting the degree of freedom and reachable range of the welding robot is solved, and multi-angle image data acquisition and welding accuracy are achieved to adapt to complex welding tasks.
Patent Information
- Application Number
- CN202421309434.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-11
- Publication Date
- 2025-07-04
- Estimated Expiration
- 2034-06-11
AI Technical Summary
Existing welding robots are unable to perform complex welding tasks due to their limitations in degrees of freedom and reachability.
A humanoid welding robot based on machine vision is designed. The left arm is equipped with a welding gun and the right arm is equipped with a visual device. The welding gun arm and the visual arm are integrated on the same device, which move independently, improve freedom and reachability, and obtain multi-angle image data through the visual arm to improve welding accuracy.
It realizes precise identification and welding of welding robots at different angles, and can weld weld parts of various structural forms, improving welding quality and flexibility.
Smart Images
Figure CN223057026U_ABST
Abstract
Description
Technical Field
[0001] The utility model relates to the technical field of humanoid robots, in particular to a humanoid welding robot based on machine vision. Background Art
[0002] Welding robots are one of the key areas for R & D, manufacturing and competition among enterprises in the industrial robot industry in China. Currently, most welding robots are single-arm collaborative robots, and their welding quality and application scenarios are restricted by factors such as their own degrees of freedom and reachable ranges, and they are unable to perform complex welding tasks. Content of the Utility Model
[0003] The purpose of the utility model is to provide a humanoid welding robot based on machine vision.
[0004] In order to achieve the above purpose, the utility model provides the following technical solutions:
[0005] 1. A humanoid welding robot based on machine vision, comprising:
[0006] A humanoid robot body, including a left arm equipped with a welding torch and a right arm equipped with a first vision device; the left arm is configured to drive the welding torch to move so that the welding torch performs welding; the right arm is configured to drive the first vision device to move to obtain image data, and the image data includes at least one of the following: welding part image data, welding torch image data, and weld image data;
[0007] A welding vision processing device, connected to the first vision device, and configured to perform data processing on the image data to obtain welding information, where the welding information includes at least one of the following: welding part information obtained based on the welding part image data; welding torch information obtained based on the welding torch image data; and weld information obtained based on the weld image data;
[0008] A robot controller, connected to the welding vision processing device, the left arm, and the right arm, and configured to control the movement of the left arm and the right arm based on the welding information.
[0009] 2. The humanoid welding robot based on machine vision according to Technical Solution 1, where the types of the welding information include at least one of the following: two-dimensional grayscale image, two-dimensional color image, depth image, and three-dimensional point cloud data.
[0010] 3. The humanoid welding robot based on machine vision according to any one of Technical Solutions 1 to 2, where the welding part information includes the shape, size, and pose of the welding part;
[0011] The welding torch information includes the contour and pose of the welding torch.
[0012] 4. The humanoid welding robot based on machine vision according to any one of Technical Solutions 1 to 3, wherein the first vision device includes a 2D camera and / or a 3D camera.
[0013] 5. The humanoid welding robot based on machine vision according to Technical Solution 4, wherein the 3D camera is any one or more of a passive binocular camera, a passive multi-camera, an active binocular camera, and an active multi-camera.
[0014] 6. The humanoid welding robot based on machine vision according to any one of Technical Solutions 1 to 5, wherein the humanoid robot body further includes a head, a neck, a torso, and legs; wherein, the head and / or the neck and / or the torso is provided with a second vision device configured to obtain welding environment image data; the welding vision processing device is connected to the second vision device and is configured to obtain welding environment information based on the welding environment image data.
[0015] 7. The humanoid welding robot based on machine vision according to Technical Solution 6, wherein the robot controller is further configured to control the movement of the humanoid robot body based on the welding environment information and / or the welding part information.
[0016] 8. The humanoid welding robot based on machine vision according to Technical Solution 6, wherein the legs are tracked mechanical legs, wheeled mechanical legs, or bipedal mechanical legs.
[0017] 9. The humanoid welding robot based on machine vision according to any one of Technical Solutions 1 to 8, wherein the left arm is a humanoid structure, including a first shoulder, a first upper arm, a first forearm, and a first hand connected in sequence; the left arm is connected to the humanoid robot body through the first shoulder.
[0018] 10. The humanoid welding robot based on machine vision according to Technical Solution 9, wherein the welding torch is disposed on the first hand or the first forearm, preferably on the first hand.
[0019] 11. The humanoid welding robot based on machine vision according to Technical Solution 10, wherein the first hand is a humanoid manipulator.
[0020] 12. The humanoid welding robot based on machine vision according to Technical Solution 11, wherein the welding torch is installed in a gripping or clamping manner.
[0021] 13. The humanoid welding robot based on machine vision according to any one of Technical Solutions 1 to 12, wherein the right arm is a humanoid structure, including a second shoulder, a second upper arm, a second forearm, and a second hand connected in sequence; the right arm is connected to the humanoid robot body through the second shoulder.
[0022] 14. The humanoid welding robot based on machine vision according to Technical Solution 13, wherein the welding torch is disposed on the second hand or the second forearm or the second upper arm, preferably on the second hand.
[0023] 15. The humanoid welding robot based on machine vision according to Technical Solution 14, wherein the second hand is a humanoid manipulator.
[0024] 16. The humanoid welding robot based on machine vision according to Technical Solution 15, wherein the mounting form of the first vision device is holding or clamping.
[0025] The above technical solutions of the present utility model have the following advantages:
[0026] The humanoid robot is applied to the welding scenario, and a welding torch is equipped on one robotic arm of the humanoid robot, and a vision device is equipped on the other robotic arm. The welding torch arm and the vision arm are integrated on the same device and are independent moving structures, greatly improving the degrees of freedom and reach range. The vision arm can move at different angles based on requirements to obtain image data at different angles to ensure the accuracy of welding and improve the welding quality. The vision arm can move from different angles to drive the first vision device to obtain weld seam image data at multiple angles, accurately identify the weld seam, and be used to detect the weld seam quality. The welding torch arm can move from different angles to drive the welding torch to weld at various angles, so that the welding robot of the present application is not restricted by the placement position or structure of the welded parts, enabling it to weld welded parts of various structural forms. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present utility model, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present utility model. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0028] Figure 1 is a structural block diagram of a humanoid welding robot provided by some embodiments of the present application;
[0029] Figure 2 is a schematic diagram of a humanoid robot body provided by some embodiments of the present application;
[0030] Figure 3 is a schematic diagram of a humanoid robot body provided by some embodiments of the present application.
[0031] Reference Signs:
[0032] 10: Humanoid robot body; 11: Left arm; 12: Right arm;
[0033] 20: Welding vision processing device;
[0034] 30: Robot controller. Detailed implementation manners
[0035] To make the objectives, technical solutions and advantages of the present utility model clearer, the technical solutions of the present utility model will be clearly and completely described below in conjunction with the embodiments of the present utility model. Obviously, the described embodiments are only a part of the embodiments of the present utility model, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present utility model without creative efforts shall fall within the protection scope of the present utility model.
[0036] In the description of this application, the terms "first", "second", "third", etc. are used to distinguish different objects and do not indicate any order or importance.
[0037] In the description of this application, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.
[0038] In the description of this application, the term "and / or" is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone.
[0039] In the description of this application, words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.
[0040] In the description of this application, "up", "down", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the described object changes, the relative position relationship may also change accordingly.
[0041] It should be noted that, without conflict, the features in the following embodiments and implementation manners can be combined with each other.
[0042] Refer to Figure 1 , Figure 1 which shows a humanoid welding robot based on machine vision, including:
[0043] The humanoid robot body 10 includes a left arm 11 equipped with a welding torch and a right arm 12 equipped with a first vision device; the left arm 11 is configured to drive the welding torch to move so that the welding torch performs welding; the right arm 12 is configured to drive the first vision device to move to obtain image data, and the image data includes at least one of the following: welding part image data, welding torch image data, and weld seam image data;
[0044] The welding vision processing device 20 is connected to the first vision device and is configured to perform data processing on the image data to obtain welding information, and the welding information includes at least one of the following: welding part information obtained based on the welding part image data; welding torch information obtained based on the welding torch image data; and weld seam information obtained based on the weld seam image data;
[0045] The robot controller 30 is connected to the welding vision processing device 20, the left arm 11, and the right arm 12, and is configured to control the movement of the left arm 11 and the right arm 12 based on the welding information.
[0046] Applying the humanoid robot to the welding scenario, and one robotic arm of the humanoid robot is equipped with a welding torch, and one robotic arm is equipped with a vision device. The welding torch arm (i.e., the left arm 11 equipped with the welding torch) and the vision arm (i.e., the right arm 12 equipped with the first vision device) are both integrated on the same device (i.e., the humanoid robot body 10) and are independent moving structures, greatly improving the degrees of freedom and reach range.
[0047] It can be understood that the left arm 11 of the humanoid robot body can be equipped with the first vision device, and the right arm 12 is equipped with the welding torch. This solution is equivalent to the solution where the left arm 11 is equipped with the welding torch and the right arm 12 is equipped with the first vision device.
[0048] The identification and positioning of the welding part and the welding torch are an indispensable step in welding, and the identification and positioning results of the two directly affect the welding quality. In the technical solution of the present application, the vision arm can move at different angles based on requirements to obtain image data at different angles to ensure the accuracy of welding and improve the welding quality. For example, before welding, the vision arm can move from different angles to drive the first vision device to obtain welding part image data at multiple angles, improving the accuracy of welding part identification and positioning. For example, before welding, the vision arm can move from different angles to drive the first vision device to obtain welding torch image data at multiple angles, improving the accuracy of welding torch identification and positioning.
[0049] Weld seam identification and weld seam quality detection are an indispensable step in welding. In the technical solution of the present application, during welding, the vision arm can move from different angles to drive the first vision device to obtain weld seam image data at multiple angles, accurately identifying the weld seam for detecting the weld seam quality.
[0050] The welding torch arm can be moved at different angles to drive the welding torch to perform welding at various angles, so that the welding robot of the present application is not restricted by the placement position or structure of the welded parts, enabling it to weld welded parts of various structural forms. For example, it can weld straight seams, fillet welds, fillet welds between tube sheets and tube heads, etc.
[0051] The humanoid robot body 10 is used to provide corresponding installation areas or spaces for each component, realizing the integration and concentration of each component to form a humanoid welding robot. For example, each component includes the two arms of the humanoid robot: the left arm 11 and the right arm 12. It can be understood that considering the humanoid structure of the humanoid robot, the positional relationship between the left arm 11 and the right arm 12 on the humanoid robot body 10 is the same as or similar to that of a human's two arms.
[0052] The left arm 11 can be of a humanoid structure, including a first shoulder, a first upper arm, a first forearm, and a first hand connected in sequence. The left arm 11 is connected to the humanoid robot body 10 through the first shoulder. The left arm 11 is equipped with a welding torch, which is preferably arranged on the first hand, and the specific installation form is not limited. In some preferred embodiments, the first hand is a humanoid manipulator, and it can be understood that its structural form is similar to that of a human hand. At this time, the first hand can hold the welding torch. It can be understood that the welding torch can also be arranged on the first hand through other existing installation forms, such as being clamped between any two fingers, or being clamped at the back of the hand position. In some embodiments, the welding torch can be arranged on the forearm, and the specific installation form is not limited.
[0053] The right arm 12 is a vision arm and is equipped with a first vision device. Similar to the left arm 11, the right arm 12 can be of a humanoid structure, including a second shoulder, a second upper arm, a second forearm, and a second hand connected in sequence. The right arm 12 is connected to the humanoid robot body 10 through the second shoulder. The first vision device can be arranged on the second hand, and the specific installation form is not limited. In some preferred embodiments, the second hand is a humanoid manipulator, and it can be understood that its structural form is similar to that of a human hand. At this time, the second hand can hold the first vision device. It can be understood that the first vision device can also be arranged at other parts of the right arm 12 through other existing installation forms, such as the second upper arm or the second forearm.
[0054] The first vision device can be a 2D camera or a 3D camera. The 3D camera can be any one or more of a passive binocular camera, a passive multi-camera, an active binocular camera, and an active multi-camera. For example, the 3D camera can be a binocular line laser camera or an RGB-D camera.
[0055] In the technical solution provided by the embodiment of the present application, the welding vision processing device 20 can be regarded as an image data processing center, which processes the image data collected by the first vision device to obtain corresponding welding information. The welding information includes at least one of the following: welding part information obtained based on the welding part image data; welding torch information obtained based on the welding torch image data; and weld seam information obtained based on the weld seam image data. In some embodiments, the welding part information includes the shape, size, and pose of the welding part; the welding torch information includes the contour and pose of the welding torch. The types of welding information may include at least one of the following: two-dimensional grayscale images, two-dimensional color images, depth images, and three-dimensional point cloud data. It can be understood that the welding vision processing device 20 can be a color image system plus a stereo vision system.
[0056] It can be understood that the welding vision processing device 20 can be a processor, which can be embedded in the main board of the first vision device or the main board of the humanoid robot body 10. It can be understood that the processor has existing vision processing algorithm modules.
[0057] In the technical solution provided by the embodiment of the present application, the robot controller 30 is connected to the welding vision processing device 20, the left arm 11, and the right arm 12, and is configured to control the movement of the left arm 11 and the right arm 12 based on the welding information. For example, when receiving a path, the robot controller 30 controls the welding arm to move the welding torch along the corresponding path to weld the welding part.
[0058] The robot controller 30 can also be configured to control the movement of the left arm 11 and the right arm 12 based on user operation instructions. For example, before welding, based on the corresponding user operation instructions, the robot controller 30 controls the vision arm to move from different angles to drive the first vision device to obtain welding part image data at multiple angles. For example, before welding, based on the corresponding user operation instructions, the robot controller 30 controls the vision arm to move from different angles to drive the first vision device to obtain welding torch image data at multiple angles. For example, during welding, based on the corresponding user operation instructions, the robot controller 30 controls the vision arm to move from different angles to drive the first vision device to obtain weld seam image data at multiple angles.
[0059] In some embodiments, referring to Figure 2 and Figure 3 , the humanoid robot body 10 further includes a head, a neck, a torso, and legs; wherein, the head and / or the neck and / or the torso are provided with a second vision device, which is configured to obtain welding environment image data; the welding vision processing device 20 is connected to the second vision device and is configured to obtain welding environment information based on the welding environment image data. At this time, the robot controller 30 is further configured to control the movement of the humanoid robot body 10 based on the welding environment information and / or the welding part information.
[0060] In some embodiments, the legs can be tracked robotic legs, wheeled robotic legs or bipedal robotic legs.
[0061] It can be understood that the mechanical designs of the various parts of the humanoid robot body 10 can adopt existing design solutions, which are not limited herein.
[0062] It can be understood that the second vision device can be a 2D camera or a 3D camera. The 3D camera can be any one or more of a passive binocular camera, a passive multi-camera, an active binocular camera, and an active multi-camera. For example, the 3D camera can be a binocular line laser camera or an RGB-D camera.
[0063] The technical solutions provided by the embodiments of the present invention have been introduced in detail above. Specific examples are used in this article to elaborate on the principles and implementation manners of the present invention. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, the technical solutions recorded in the foregoing embodiments can be modified, or some of the technical features can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the ideas and scopes of the technical solutions of the embodiments of the present invention.
Claims
1. A humanoid welding robot based on machine vision, characterized in that Comprising: A humanoid robot body (10), including a left arm (11) equipped with a welding torch and a right arm (12) equipped with a first vision device; the left arm (11) is configured to drive the welding torch to move so that the welding torch performs welding; the right arm (12) is configured to drive the first vision device to move to obtain image data, and the image data includes at least one of the following: welding part image data, welding torch image data, and weld seam image data; A welding vision processing device (20), connected to the first vision device, and configured to perform data processing on the image data to obtain welding information, where the welding information includes at least one of the following: welding part information obtained based on the welding part image data; welding torch information obtained based on the welding torch image data; and weld seam information obtained based on the weld seam image data; A robot controller (30), connected to the welding vision processing device (20), the left arm (11), and the right arm (12), and configured to control the movement of the left arm (11) and the right arm (12) based on the welding information.
2. The humanoid welding robot based on machine vision according to claim 1, wherein, The types of the welding information include at least one of the following: two-dimensional grayscale image, two-dimensional color image, depth image, and three-dimensional point cloud data.
3. The humanoid welding robot based on machine vision according to claim 1, characterized in that, The welding part information includes the shape, size, and pose of the welding part; The welding torch information includes the contour and pose of the welding torch.
4. The humanoid welding robot based on machine vision according to any one of claims 1 to 3, characterized in that, The first vision device includes a 2D camera and / or a 3D camera.
5. The humanoid welding robot based on machine vision according to claim 4, characterized in that, The 3D camera is any one or more of a passive binocular camera, a passive multi-camera, an active binocular camera, and an active multi-camera.
6. The humanoid welding robot based on machine vision according to any one of claims 1 to 3, characterized in that, The humanoid robot body further includes a head, a neck, a torso, and legs; wherein, a second vision device is provided on the head and / or the neck and / or the torso, and is configured to obtain welding environment image data; the welding vision processing device is connected to the second vision device and is configured to obtain welding environment information based on the welding environment image data.
7. The humanoid welding robot based on machine vision according to claim 6, characterized in that, The robot controller (30) is further configured to control the movement of the humanoid robot body based on the welding environment information and / or the welding part information.
8. The humanoid welding robot based on machine vision according to claim 6, characterized in that, The legs are tracked mechanical legs, wheeled mechanical legs, or bipedal mechanical legs.