Weld seam tracking device based on visual-haptic fusion and method thereof

Abstract

The application relates to the field of curved welds, and provides a welding seam tracking device and method based on visual-haptic fusion to solve the problem that the existing welding seam tracking technology is mainly based on visual sensors to acquire the position of the welding seam, and in the actual point cloud scanning process, part of the welding seam is easy to be shielded, and the low scanning precision makes it difficult to guarantee the welding quality and safety of the welding seam tracking guided by pure vision. The device comprises a first collecting part arranged above a workpiece to be welded and used for collecting coarse positioning coordinate information of the welding seam; a mechanical arm control module arranged above the workpiece to be welded and located on one side of the first collecting part, the mechanical arm control module being electrically connected with the first collecting part; a welding gun model exploration part connected with the mechanical arm control module; and an exploration haptic feedback assembly comprising a flexible connecting part, a probe, a two-dimensional code and a second collecting part, the flexible connecting part being arranged as a hollow middle part and connected with one end of the welding gun model exploration part away from the mechanical arm control module.

Description

Technical Field

[0001] This application relates to the field of curved welds, and more specifically, to a weld tracking device and method based on visual-tactile fusion. Background Technology

[0002] The application of industrial robots in industrial production has driven the rapid development of intelligent and automated welding, significantly improving welding efficiency and quality. Traditional industrial robot welding operations primarily rely on teach-path models, requiring manual pre-setting of trajectories for repetitive operations. However, this method is inefficient in industrial production involving a large amount of repetitive work, especially when welding complex weld shapes. Simply relying on the machine's teach-path model is insufficient to guarantee welding quality and cannot meet the requirements of modern intelligent welding.

[0003] With the rapid advancement of sensor technology, automated weld seam tracking using sensors has become a new trend. However, existing weld seam tracking technologies mainly rely on vision sensors to obtain weld seam positions. In actual point cloud scanning processes, some weld seams are easily obscured, and the scanning accuracy is relatively low, making it difficult to guarantee welding quality and safety with purely vision-guided weld seam tracking.

[0004] Therefore, existing technologies still need improvement. Summary of the Invention

[0005] The purpose of this application is to provide a weld seam tracking device and method based on visual-tactile fusion, in order to solve the problem that existing weld seam tracking technologies mainly rely on visual sensors to obtain the weld seam position. However, in the actual point cloud scanning process, some weld seams are easily obscured and the scanning accuracy is low, making it difficult to ensure welding quality and safety in purely visual-guided weld seam tracking.

[0006] To achieve the above objectives, the technical solution adopted in the first aspect of this application is as follows:

[0007] A weld seam tracking device based on visual-tactile fusion is used to be positioned above a workpiece to be welded, the weld seam tracking device comprising:

[0008] The first acquisition element is positioned above the workpiece to be welded and is used to acquire the coarse positioning coordinate information of the weld.

[0009] A robotic arm control module is disposed above the workpiece to be welded and located on one side of the first acquisition component. The robotic arm control module is electrically connected to the first acquisition component.

[0010] The welding torch model exploration unit is connected to the robotic arm control module;

[0011] The haptic feedback component includes a flexible connector, a probe, a QR code, and a second acquisition device. The flexible connector is hollowed out in the middle and is connected to the end of the welding torch model exploration section opposite to the robotic arm control module. The probe is connected to the flexible connector, and the end of the probe opposite to the flexible connector is located inside the weld. The QR code is set on the end face of the probe near the flexible connector. The second acquisition device is set inside the welding torch model exploration section and is located directly above the QR code through the hollowed-out flexible connector.

[0012] The welding torch application unit is connected to the robotic arm control module and is located on one side of the welding torch model exploration unit. A welding wire is provided at the end of the welding torch application unit that is away from the robotic arm control module, and the welding wire is placed inside the weld.

[0013] According to the above-described weld seam tracking device based on visual-tactile fusion, the end face of the welding wire away from the welding torch application part and the end face of the probe away from the flexible connector are on the same horizontal line.

[0014] According to the weld seam tracking device based on visual-tactile fusion described above, the welding torch model exploration unit and the exploration tactile feedback component form a combined body, and the shape and size of the combined body are the same as the shape and size of the welding torch application unit.

[0015] According to the weld seam tracking device based on visual-tactile fusion described above, the weld seam tracking device further includes:

[0016] Several lighting modules are embedded around the second acquisition element.

[0017] According to the above-described weld seam tracking device based on visual-tactile fusion, the weld seam tracking device further includes a light-blocking part, which is disposed on the welding torch model exploration part and located on the outside of the flexible connector. The light-blocking part is used to block the arc light during the welding process of the welding torch application part.

[0018] According to the weld seam tracking device based on visual-tactile fusion described above, the first acquisition element is set as a line laser camera.

[0019] According to the weld seam tracking device based on visual-tactile fusion described above, the robotic arm control module includes:

[0020] The robotic arm, the end of which is connected to the welding torch model exploration section and the welding torch application section via a flange connector;

[0021] An industrial control computer is connected to the first data acquisition device and the second data acquisition device, respectively.

[0022] According to the weld seam tracking device based on visual-tactile fusion described above, the flexible connector is configured as a silicone connector or a nylon connector.

[0023] According to the weld seam tracking device based on visual-tactile fusion described above, the QR code can be set as an Aruco QR code, and the second acquisition device can be set as an endoscopic camera.

[0024] The technical solution adopted in the second aspect of this application is:

[0025] A weld seam tracking method based on visual-tactile fusion, applied to the visual-tactile fusion-based weld seam tracking device as described above, the weld seam tracking method comprising:

[0026] The probe of the tactile feedback component is driven by the robotic arm control module to move along the coarse positioning coordinate information of the weld seam.

[0027] When the probe deflects upon touching the edge of the weld, the deflection information of the QR code on the probe is collected by the second acquisition device under the action of the flexible connector, thereby obtaining the precise positioning coordinate information of the weld.

[0028] The welding torch application unit is driven by the robotic arm control module to move along the precise positioning coordinates of the weld seam to perform weld filling work.

[0029] The beneficial effects of the weld seam tracking device and method based on visual-tactile fusion provided in this application are at least as follows:

[0030] This application uses a robotic arm control module to drive the probe of the tactile feedback component to move along the coarse positioning coordinates of the weld seam. When the probe deflects upon touching the edge of the weld seam, the deflection information of the QR code on the probe is collected by a second acquisition component under the action of a flexible connector to obtain the fine positioning coordinates of the weld seam. The robotic arm control module then drives the welding torch application unit to move along the fine positioning coordinates of the weld seam to perform precise weld seam filling. It is foreseeable that this weld seam tracking device based on visual-tactile fusion can effectively avoid collisions between the welding torch application unit and the workpiece to be welded, ensuring the operational safety of the equipment. At the same time, this application sets up an exploration end (exploration tactile feedback component) and an application end (welding torch application unit), and the robotic arm control module simultaneously starts the operation of the exploration end and the application end. The exploration end explores in front, and the application end applies the corrections based on the exploration end's exploration. Only one operation is required, and this operation is a useful operation (one operation is enough to enable the application end to perform welding torch work), which greatly improves the efficiency and quality of weld seam tracking. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of a weld seam tracking device based on visual-tactile fusion, provided in an embodiment of this application, which is installed above a workpiece being welded.

[0033] Figure 2 This is a schematic diagram of the structure of a weld seam tracking device based on visual-tactile fusion, in which the welding torch model exploration part and the welding torch application part are connected to a flange connector.

[0034] Figure 3 An exploded view of the tactile feedback component of a weld seam tracking device based on visual-tactile fusion, provided in an embodiment of this application, showing the component connected to the exploration section of a welding torch model.

[0035] Figure 4 A flowchart of a weld seam tracking method based on visual-tactile fusion provided in an embodiment of this application.

[0036] Figure 5 This is a schematic diagram of the weld trajectory in a weld tracking method based on visual-tactile fusion provided in an embodiment of this application.

[0037] The following are the labeling elements in the figure:

[0038] 1. First acquisition component; 21. Robotic arm; 22. Industrial control computer; 3. Welding torch model exploration unit; 4. Exploration tactile feedback component; 41. Flexible connector; 42. Probe; 43. QR code; 44. Second acquisition component; 5. Welding torch application unit; 51. Welding wire; 6. Workpiece to be welded; 61. Weld seam; 7. Light blocking part; 8. Flange connector. Detailed Implementation

[0039] To make the technical problems, technical solutions, and beneficial effects to be solved by this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and are not intended to limit the scope of this application.

[0040] It should be noted that when a component is referred to as "fixed to" or "set on" another component, it may be directly or indirectly located on that other component. When a component is referred to as "connected to" another component, it may be directly or indirectly connected to that other component. The terms "upper," "lower," "left," "right," "front," "rear," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate orientations or positions based on the accompanying drawings, and are for ease of description only, and should not be construed as limiting the technical solution. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features. "A plurality" means two or more, unless otherwise explicitly defined.

[0041] The application of industrial robots in industrial production has driven the rapid development of intelligent and automated welding, significantly improving welding efficiency and quality. Traditional industrial robot welding operations primarily rely on teach-path models, requiring manual pre-setting of trajectories for repetitive operations. However, this method is inefficient in industrial production involving a large amount of repetitive work, especially when welding complex weld shapes. Simply relying on the machine's teach-path model is insufficient to guarantee welding quality and cannot meet the requirements of modern intelligent welding.

[0042] With the rapid advancement of sensor technology, automated weld seam tracking using sensors has become a new trend. However, existing weld seam tracking technologies mainly rely on vision sensors to obtain weld seam positions. In actual point cloud scanning processes, some weld seams are easily obscured, and the scanning accuracy is relatively low, making it difficult to guarantee welding quality and safety with purely vision-guided weld seam tracking.

[0043] For this purpose, please refer to Figure 1 , Figure 2 and Figure 3 The first aspect of this application provides a weld seam tracking device based on visual-tactile fusion, which is used to be set above a workpiece 6 to be welded with a weld seam 61. The weld seam tracking device includes a first acquisition unit 1, a robotic arm control module, a welding torch model exploration unit 3, an exploration tactile feedback component 4, and a welding torch application unit 5.

[0044] The first acquisition component 1 is positioned above the workpiece 6 to be welded and is used to acquire the coarse positioning coordinate information of the weld 61. The robotic arm control module is positioned above the workpiece 6 to be welded and located to one side of the first acquisition component 1. The robotic arm control module is electrically connected to the first acquisition component 1. The welding torch model exploration unit 3 is connected to the robotic arm control module. The exploration tactile feedback component 4 includes a flexible connector 41, a probe 42, a QR code 43, and a second acquisition component 44. The flexible connector 41 is hollowed out in the middle and is connected to the end of the welding torch model exploration unit 3 that is away from the robotic arm control module. The probe 42 is connected to the flexible connector 41. One end of the probe 42 away from the flexible connector 41 is located inside the weld 61. The QR code 43 is disposed on the end face of the probe 42 near the flexible connector 41. The second acquisition element 44 is disposed inside the welding torch model exploration part 3 and is located directly above the QR code 43 through the hollowed-out flexible connector 41. The welding torch application part 5 is connected to the robotic arm control module and is located on one side of the welding torch model exploration part 3. The welding torch application part 5 away from the robotic arm control module is provided with a welding wire 51, which is disposed inside the weld 61.

[0045] In this embodiment, the probe 42 of the tactile feedback component 4 is driven by the robotic arm control module to move along the coarse positioning coordinates of the weld 61. When the probe 42 deflects upon touching the edge of the weld 61, the deflection information of the QR code 43 on the probe 42 is collected by the second acquisition component 44 under the action of the flexible connector 41 to obtain the fine positioning coordinates of the weld 61. The welding torch application unit 5 is then driven by the robotic arm control module to move along the fine positioning coordinates of the weld 61 to perform precise weld filling work. It is foreseeable that this embodiment... For example, the weld seam tracking device based on visual-tactile fusion can effectively avoid collisions between the welding torch application unit 5 and the workpiece 6 to be welded, ensuring the safe operation of the equipment. At the same time, this embodiment sets up an exploration end (exploration tactile feedback component 4) and an application end (welding torch application unit 5), and the work of the exploration end and the application end are started simultaneously through the robotic arm control module. The exploration end explores in front, and the application end applies the corrections based on the exploration end's exploration. Only one operation is required, and this operation is a useful operation (one operation is enough to enable the application end to work with the welding torch), which greatly improves the efficiency and quality of weld seam tracking.

[0046] Optional, see below Figure 1 In one embodiment, the robotic arm control module includes a robotic arm 21 and an industrial computer 22. The end of the robotic arm 21 is connected to the welding torch model exploration unit 3 and the welding torch application unit 5 via a flange connector 8. The industrial computer 22 is connected to the first acquisition unit 1 and the second acquisition unit 44 via data cables.

[0047] Optionally, in one embodiment, the first acquisition device 1 may be configured as a line laser camera.

[0048] Optionally, in one embodiment, the flexible connector 41 may be configured as a silicone connector or a nylon connector.

[0049] Optionally, in one embodiment, the probe 42 may be configured as a metal probe.

[0050] Optionally, in one embodiment, the QR code 43 can be set as an Aruco QR code.

[0051] Optional, see below Figure 1 and Figure 2 In one embodiment, the second acquisition element 44 can be configured as an endoscopic camera.

[0052] Optionally, in one embodiment, in order to provide good illumination when recognizing QR code 43, the weld seam tracking device further includes several lighting modules (not shown in the figure), which are embedded around the second acquisition element 44, wherein the lighting modules can be configured as LED lights.

[0053] Optional, see below Figure 1 and Figure 2 In one embodiment, the end face of the welding wire 51 facing away from the welding gun application part 5 is on the same horizontal line as the end face of the probe 42 facing away from the flexible connector 41.

[0054] Optional, see below Figure 1 and Figure 2 In one embodiment, the welding torch model exploration unit 3 and the exploration tactile feedback component 4 form a combined body, the shape and size of which are the same as those of the welding torch application unit 5.

[0055] Optional, see below Figure 2 In one embodiment, the weld seam tracking device further includes a light-blocking part 7, which can be configured as a hollow cylindrical structure. The light-blocking part 7 is disposed on the welding torch model exploration part 3 and located on the outside of the flexible connector 41. The light-blocking part 7 is used to block the arc light during the welding process of the welding torch application part 5, so as to prevent the arc light from affecting the second acquisition part 44's acquisition of the deflection information of the QR code 43 on the probe 42.

[0056] See Figure 4 The second aspect of this application provides a weld seam tracking method based on visual-tactile fusion, which is applied to the weld seam tracking device based on visual-tactile fusion as described above. The weld seam tracking method includes:

[0057] S10. Drive the probe of the tactile feedback component to move along the coarse positioning coordinate information of the weld seam through the robotic arm control module.

[0058] S20. When the probe deflects upon touching the edge of the weld, the deflection information of the QR code on the probe is collected by the second acquisition device under the action of the flexible connector to obtain the precise positioning coordinate information of the weld.

[0059] S30. The welding torch application unit is driven by the robotic arm control module to move along the precise positioning coordinate information of the weld seam to perform weld seam filling work.

[0060] Specifically, the coarse positioning coordinate information is acquired through a first acquisition device, which can be a line laser camera. The robotic arm control module includes a robotic arm and an industrial computer. The robotic arm is connected to the welding torch model exploration unit and the welding torch application unit, respectively, and the industrial computer is electrically connected to the robotic arm.

[0061] First, a line laser camera is used to coarsely scan and locate the weld seam, obtaining its coarse positioning coordinates in the line laser camera coordinate system. Coordinate transformation is then performed to obtain its coarse positioning coordinates in the robotic arm coordinate system. Subsequently, the industrial control computer controls the tactile feedback component on the robotic arm, causing the probe on the tactile feedback component to move along the coarse positioning coordinates of the weld seam. Due to the inaccuracy of the coarse positioning coordinates, when the probe deflects upon touching the edge of the weld seam, the QR code on the probe also deflects. Because a flexible connector separates the probe from the welding torch model exploration section, the second acquisition component within the welding torch model exploration section does not deflect. Therefore, the second acquisition component can collect the deflection (pose) information of the QR code on the probe to obtain tactile feedback at the probe end, thus obtaining the fine positioning coordinates of the weld seam. Finally, the robotic arm control module drives the welding torch application section to move along the fine positioning coordinates of the weld seam for precise weld seam filling.

[0062] Optionally, in one embodiment, the specific steps for obtaining the coarse positioning coordinate information in the robotic arm coordinate system through coordinate transformation can be as follows: preprocessing the coarse positioning coordinate information of the weld in the line laser camera coordinate system to remove noise, thereby improving the accuracy and reliability of the information; using the RANSAC (Random Sample Consensus) algorithm to perform linear fitting on the preprocessed data (coarse positioning coordinate information) to identify the surface of the workpiece to be welded; obtaining the position and shape of the weld notch by calculating the difference between the fitted data (fitted data) and the original data (coarse positioning coordinate information obtained by the line laser camera); determining the average position of the weld on each cross section by calculating the average value of the weld notch data of each cross section; stitching together the weld position data of each cross section one by one to form an overall weld position map, and taking the center of the bottom end of the weld as the position of the weld feature point; obtaining the hand-eye transformation relationship matrix between the line laser camera and the robotic arm through hand-eye calibration to obtain the coarse positioning coordinate information of visual positioning in the robotic arm coordinate system.

[0063] Optionally, in one embodiment, the specific steps of obtaining tactile feedback at the probe end by acquiring the deflection (pose) information of the QR code on the probe through the second acquisition device, thereby obtaining the precise positioning coordinate information of the weld seam, can be as follows: obtaining the intrinsic parameter matrix of the second acquisition device through chessboard calibration, and obtaining the hand-eye conversion relationship matrix between the second acquisition device and the flange connector at the end of the robotic arm through hand-eye calibration. During the movement of the robotic arm, the second data acquisition unit reads the position of the QR code in the pixel coordinate system. Based on the probe size, the transformation matrix M from the QR code to the probe tip is obtained, thus determining the probe tip position in the second acquisition device coordinate system. This allows us to obtain the probe end position in the coordinate system of the robotic arm end flange connector. The robotic arm reads the transformation matrix between its end flange connector and the robotic arm coordinate system. Obtain the position of the probe end in the robot arm coordinate system That is, precise positioning coordinate information in the robotic arm coordinate system based on visual-tactile fusion.

[0064] Optional, see below Figure 5 In one embodiment, the specific steps of the above-mentioned precise weld filling work by driving the welding torch application part to move along the precise positioning coordinate information of the weld seam through the robotic arm control module can be as follows: the rotation of the end flange connector controlled by the robotic arm drives the welding torch application part and the welding wire on it to move along the precise positioning coordinate information at a constant speed of V0 to perform precise weld filling work.

[0065] Optional, see below Figure 1 and Figure 2In one embodiment, the positions of the center of the robotic arm flange connector, the tip of the probe, and the tip of the welding wire on the plane of the workpiece to be welded are respectively P tcp P tip P torch The center of the robotic arm flange connector moves at a speed along the visually coarse weld seam. The magnitude of the tangential velocity normal velocity magnitude Where, k p f fn , These are custom speed constants, custom gain coefficients, and The length along the normal direction of the coarse weld trajectory and the custom base offset can be used to calculate the magnitudes of the velocities along the x and y axes of the center of the robot arm flange connector in the robot arm coordinate system. in These are the unit vectors of the robotic arm's coordinate system.

[0066] When the rotational speed of the flange connector is At that time, the speed of the welding wire tip along the visual precision weld seam trajectory is in The length is determined by the relative position of the end effector flange connector and the end of the welding wire, and the direction is read by the industrial control computer connected to the robotic arm, specifying the tangential speed. The magnitude of the velocity in the normal direction is in These are the unit tangent vector and unit normal vector at the current probe end position on the precision weld trajectory, respectively.

[0067] To ensure that the tip of the welding wire moves along the tactile precision weld seam trajectory, i.e.

[0068] To ensure that the tip of the welding wire maintains a constant speed and magnitude of movement, i.e.

[0069] United:

[0070]

[0071] in The direction can be obtained through visual coarse weld seam trajectory, based on Calculate using known parameters and

[0072] And:

[0073]

[0074] At this point, the end effector speed V of the robotic arm can be determined by the trajectory of the precision weld and the relative positions of each component.x V y , To achieve constant speed trajectory tracking of the weld seam by the welding torch application unit.

[0075] In summary, this application provides a weld seam tracking device and method based on visual-tactile fusion, which is used to be positioned above a workpiece to be welded with a weld seam. The weld seam tracking device includes a first acquisition component, a robotic arm control module, a welding torch model exploration unit, an exploration tactile feedback component, and a welding torch application unit. The first acquisition component is positioned above the workpiece to be welded and is used to acquire coarse positioning coordinate information of the weld. The robotic arm control module is positioned above the workpiece to be welded and located to one side of the acquisition component. The robotic arm control module is electrically connected to the acquisition component. The welding torch model exploration unit is connected to the robotic arm control module. The exploration tactile feedback component includes a flexible connector, a probe, a QR code, and a second acquisition component. The flexible connector is hollowed out in the middle and connected to the end of the welding torch model exploration unit away from the robotic arm control module. The probe is connected to the flexible connector, and the end of the probe away from the flexible connector is located inside the weld. The QR code is positioned on the end face of the probe near the flexible connector. The second acquisition component is positioned inside the welding torch model exploration unit and is located directly above the QR code through the hollowed-out flexible connector. The welding torch application unit is connected to the robotic arm control module and is located to one side of the welding torch model exploration unit. The end of the welding torch application unit away from the robotic arm control module is provided with a welding wire, which is located inside the weld. This application uses a robotic arm control module to drive the probe of the tactile feedback component to move along the coarse positioning coordinates of the weld seam. When the probe deflects upon touching the edge of the weld seam, the deflection information of the QR code on the probe is collected by a second acquisition component under the action of a flexible connector to obtain the fine positioning coordinates of the weld seam. The robotic arm control module then drives the welding torch application unit to move along the fine positioning coordinates of the weld seam to perform precise weld seam filling. It is foreseeable that this weld seam tracking device based on visual-tactile fusion can effectively avoid collisions between the welding torch application unit and the workpiece to be welded, ensuring the operational safety of the equipment. At the same time, this application sets up an exploration end (exploration tactile feedback component) and an application end (welding torch application unit), and the robotic arm control module simultaneously starts the operation of the exploration end and the application end. The exploration end explores in front, and the application end applies the corrections based on the exploration end's exploration. Only one operation is required, and this operation is a useful operation (one operation is enough to enable the application end to perform welding torch work), which greatly improves the efficiency and quality of weld seam tracking.

[0076] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A weld seam tracking device based on visual-tactile fusion, used to be positioned above a workpiece to be welded with a weld seam, characterized in that, The weld seam tracking device includes: The first acquisition element is positioned above the workpiece to be welded and is used to acquire the coarse positioning coordinate information of the weld. A robotic arm control module is disposed above the workpiece to be welded and located on one side of the first acquisition component. The robotic arm control module is electrically connected to the first acquisition component. The welding torch model exploration unit is connected to the robotic arm control module; The haptic feedback component includes a flexible connector, a probe, a QR code, and a second acquisition device. The flexible connector is hollowed out in the middle and is connected to the end of the welding torch model exploration section away from the robotic arm control module. The probe is connected to the flexible connector, and the end of the probe away from the flexible connector is located inside the weld. When the probe touches the edge of the weld and deflects, under the action of the flexible connector, the second acquisition device collects the deflection information of the QR code on the probe to obtain the precise positioning coordinate information of the weld. The QR code is set on the end face of the probe near the flexible connector, and the second acquisition device is set inside the exploration part of the welding gun model and is located directly above the QR code through the hollowed-out flexible connector. The welding torch application unit is connected to the robotic arm control module and is located on one side of the welding torch model exploration unit. A welding wire is provided at the end of the welding torch application unit that is away from the robotic arm control module, and the welding wire is placed inside the weld.

2. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The end face of the welding wire away from the welding gun application part and the end face of the probe away from the flexible connector are on the same horizontal line.

3. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The welding torch model exploration section and the exploration tactile feedback component form a combined body, and the shape and size of the combined body are the same as those of the welding torch application section.

4. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The weld seam tracking device also includes: Several lighting modules are embedded around the second acquisition element.

5. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The weld seam tracking device also includes a light-blocking part, which is disposed on the welding torch model exploration part and located on the outside of the flexible connector. The light-blocking part is used to block the arc light during the welding process of the welding torch application part.

6. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The first acquisition device is set as a line laser camera.

7. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The robotic arm control module includes: The robotic arm, the end of which is connected to the welding torch model exploration section and the welding torch application section via a flange connector; An industrial control computer is connected to the first data acquisition device and the second data acquisition device, respectively.

8. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The flexible connector is configured as a silicone connector or a nylon connector.

9. The weld seam tracking device based on visual-tactile fusion according to claim 1, characterized in that, The QR code can be set to an Aruco QR code, and the second acquisition device can be set to an endoscopic camera.

10. A weld seam tracking method based on visual-tactile fusion, characterized in that, It is applied to the weld seam tracking device based on visual-tactile fusion as described in claim 1, wherein the weld seam tracking method includes: The probe of the tactile feedback component is driven by the robotic arm control module to move along the coarse positioning coordinate information of the weld seam. When the probe deflects upon touching the edge of the weld, the deflection information of the QR code on the probe is collected by the second acquisition device under the action of the flexible connector, thereby obtaining the precise positioning coordinate information of the weld. The welding torch application unit is driven by the robotic arm control module to move along the precise positioning coordinates of the weld seam to perform weld filling work.

Images