Welding seam positioning robot and welding seam positioning method

Abstract

The present invention discloses a weld positioning robot and weld positioning method, comprising: a robot body, a weld detection device, a welding gun, a base, and a guide rail device. The robot body has a working end and a supporting end. The weld detection device is rotatably connected to the working end of the robot body and is used to extract a grayscale image of the weld area. The welding gun is fixedly connected to the working end of the robot body, and the welding gun and the detection device are spaced apart. The supporting end of the robot body is rotatably connected to the base. The side of the base away from the robot body is slidably connected to the guide rail device. The present invention solves the technical problem in the prior art of difficulty in achieving rapid positioning of the welding start and end positions and integrated detection and welding functions for welding in complex environments.

Description

Technical Field

[0001] The present invention relates to the technical field of positioning detection, and in particular to a weld positioning robot and a weld positioning method. Background Art

[0002] Welding robots, as the "tailors" of industry, are a crucial processing method in industrial production and a key link in intelligent manufacturing technology. Currently, the deployment of welding robots on industrial assembly lines is highly mature, and the field of welding robots is gradually expanding to non-structural fields such as construction, shipbuilding, and aviation, where the environments are more complex and changeable. Welding robots are gradually being used not only for factory welding but also for on-site welding of medium and thick plates on steel structures. Due to the low machining precision of domestic steel structures, the diverse groove forms, and other harsh factors, as well as the complex and harsh background of on-site welding grooves and the diverse influencing factors, welding robots face welding positioning issues for complex background grooves. The starting and ending positions of welding affect the automated welding process, and a mature solution is urgently needed to guide and promote market development.

[0003] Existing welding robots, as the key to the field of intelligent manufacturing, have been maturely applied in industrial production and manufacturing. However, most of the welding robots that have been successfully applied and entered the product stage in the market can only serve as a link in a complex welding process. Their welding intelligence is low, the working mode is fixed, and the standardization requirements of the welding environment are high. Most of the welding objects are thin plate spot welding, and they are unable to complete multi-layer and multi-pass medium and thick plate welding with complex path planning. For welding in complex environments, it is difficult to achieve rapid positioning of the welding start and end positions and integrated detection and welding functions. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above technical deficiencies, provide a weld positioning robot and a weld positioning method, and solve the technical problem in the prior art that it is difficult to achieve weld detection and welding integration functions for welding in complex environments.

[0005] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a weld positioning robot, comprising: a robot body, a weld detection device, a welding gun, a base, and a guide rail device; wherein:

[0007] The robot body has a working end and a supporting end; the weld detection device is rotatably connected to the working end of the robot body, and the weld detection device is used to extract a grayscale image of the weld area; the welding gun is fixedly connected to the working end of the robot body, and the welding gun and the weld detection device are spaced apart; the supporting end of the robot is rotatably connected to the base; the base is slidably connected to the guide rail device on a side away from the robot body.

[0008] In some embodiments, the weld detection device includes a detection head fixture and a line structured light sensor, the detection head fixture includes two clamping plates, the two clamping plates are arranged opposite to each other and are fixedly connected to the working end of the robot body, and threaded holes are provided on the two clamping plates, and the line structured light sensor is threadedly connected to the clamping plates through the threaded holes.

[0009] In some embodiments, the weld positioning robot also includes a welding gun clamp, which includes a first half ring, a second half ring and a locking piece. One end of the first half ring and the second half ring are fixedly connected to the working end. The locking piece is used to connect the first half ring and the second half ring and make the first half ring and the second half ring embrace each other to form a clamping space. The welding gun is located in the clamping space and abuts against the inner walls of the first half ring and the second half ring.

[0010] In some embodiments, a surface of the base close to the supporting end of the robot body has a groove, the groove fits with the supporting end, and the supporting end is inserted into the groove.

[0011] In some embodiments, the guide rail device includes at least two slide rails and at least two sliders, the at least two slide rails are spaced apart and distributed in parallel, the number of the sliders is consistent with the number of the guide rails, one side of the at least two slide rails is installed on the side of the base away from the support end, and the other side is inserted into the guide rail, and the slider can slide along the guide rail.

[0012] In a second aspect, the present invention further provides a weld positioning method, which is applied to any of the weld positioning robots described above, and the method comprises:

[0013] Acquiring a depth image of the weld area collected by a weld detection device;

[0014] Using a preset weld edge contour extraction method, contour feature extraction is performed on the grayscale image to obtain weld edge contour information;

[0015] Using a preset minimum circumscribed rectangle algorithm to search for the minimum area of ​​the weld edge contour information to obtain a minimum enclosing rectangle;

[0016] According to the coordinates of the corner points of the minimum enclosing rectangle, the starting position and the end position of the weld to be welded by the welding positioning robot are determined. In some embodiments, the preset weld edge contour extraction method is used to extract contour features from the depth image, including:

[0017] An opening operation is performed on the depth image using a preset morphology to adjust the grayscale value of the depth image to obtain an enhanced image; a preset mean filtering algorithm is used to suppress noise on the enhanced image to obtain denoised weld edge contour information;

[0018] Using a preset binary threshold segmentation algorithm, pixel-level segmentation is performed on the denoised weld edge contour information to obtain an enhanced feature area image;

[0019] Based on the feature area image, a preset Canny edge detection algorithm is used to extract weld edge features to obtain a feature separation image.

[0020] In some embodiments, the pixel-level segmentation of the grayscale image using a preset binary threshold segmentation algorithm includes:

[0021] Determine the pixel segmentation threshold according to the grayscale features of the grayscale image;

[0022] According to the pixel segmentation threshold, the grayscale image is segmented to obtain a foreground image and a background image.

[0023] In some embodiments, the method of using a preset minimum bounding rectangle algorithm to perform a minimum area search on the weld edge contour information to obtain a minimum enclosing rectangle includes:

[0024] Using a preset minimum circumscribed rectangle algorithm to search for the minimum area of ​​the weld edge contour information to determine the minimum enclosing polygon;

[0025] Based on the minimum enclosing polygon, all adjacent points are traversed and a preset minimum enclosing rectangle algorithm is used to determine the enclosing rectangle with the smallest area.

[0026] In some embodiments, determining the starting position and the ending position of the welding robot according to the coordinates of the corner points of the minimum enclosing rectangle includes:

[0027] Convert the corner coordinates of the minimum enclosing rectangle into target coordinates corresponding to the robot body coordinate system;

[0028] The starting position and the ending position of the welding robot are determined according to the target coordinates corresponding to the minimum wrapping rectangle.

[0029] Compared with the existing technology, the weld positioning robot and weld positioning method provided by the present invention carry the weld detection device and the welding gun on the robot body. The robot body can rotate relative to the base and move relative to the guide rail device, so that the position and working angle of the weld detection device and the welding gun can be changed. At the same time, the weld position can be quickly identified through the weld detection device, thereby guiding the welding positioning robot to move to the weld position to weld the weld, realizing integrated welding.

[0030] Furthermore, the weld positioning method provided by the present invention first obtains a grayscale image of the weld area using a weld positioning robot; then, a preset weld edge contour extraction method is used to extract contour features from the grayscale image to obtain weld edge contour information; then, a preset minimum enclosing rectangle algorithm is used to search for a minimum area on the weld edge contour information to obtain a minimum enclosing rectangle; and finally, based on the corner coordinates of the minimum enclosing rectangle, the starting and ending positions of the weld to be welded by the welding positioning robot are determined. This achieves accurate positioning of the weld. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a structural diagram of an embodiment of the weld positioning robot provided by the present invention;

[0032] Figure 2 This is a structural diagram of an embodiment of a weld fixture in the weld positioning robot provided by the present invention;

[0033] Figure 3 This is a structural diagram of an embodiment of a base and a guide rail in the weld positioning robot provided by the present invention;

[0034] Figure 4 This is a flow chart of an embodiment of the weld positioning method provided by the present invention;

[0035] Figure 5 This is a flow chart of an embodiment of step S402 in the weld locating method provided by the present invention;

[0036] Figure 6 This is a flow chart of an embodiment of step S501 in the weld locating method provided by the present invention;

[0037] Figure 7 This is a flow chart of an embodiment of step S403 in the weld locating method provided by the present invention; DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0039] In order to solve the technical problem in the prior art that it is difficult to quickly locate the starting and ending positions of welding and integrate detection and welding functions for welding in complex environments, the present invention proposes a line structured light sensor welding robot device for thick plate welding in complex environments. Based on the depth image obtained by scanning the groove of medium and thick plates in complex environments with a line structured light camera, the starting and ending positions of the welding robot are obtained by detection and segmentation of the depth image, thereby solving the problem of the welding robot identifying and locating the starting and ending positions of the groove of medium and thick plates, realizing rapid identification of welds, and expanding the welding working range based on the base moving guide device, thereby greatly improving the intelligence of the welding robot from detection to welding integration process.

[0040] The embodiment of the present invention provides a welding seam positioning robot, please refer to Figure 1-3 , including: a robot body 1, a weld detection device 2, a welding gun 3, a base 4 and a guide rail device 5; wherein:

[0041] The robot 1 body has a working end and a supporting end; the weld detection device 2 is rotatably connected to the working end of the robot body 1, and the weld detection device 2 is used to extract a grayscale image of the weld area; the welding gun 3 is fixedly connected to the working end of the robot body 1, and the welding gun 3 is spaced apart from the weld detection device 2; the supporting end of the robot body is rotatably connected to the base 4; the base 4 is slidably connected to the guide rail device 5 on the side away from the robot body 1.

[0042] In this embodiment, the weld detection device and the welding gun are mounted on the robot body, integrating weld image information acquisition and welding gun welding. The robot body can rotate relative to the base and move relative to the guide rail device, so that the position and working angle of the weld detection device and the welding gun can be changed. The working range of the welding gun can be expanded, and the image information of the weld can be obtained through the weld detection device. The starting position and end position of the weld can be identified by the weld positioning method, thereby guiding the robot to drive the welding gun to weld the weld, thereby achieving the purpose of integrated detection and welding.

[0043] It should be noted that, in this embodiment, the robot body is a six-axis robot structure, including six joint motor modules, a joint connection converter, a module end cover flange, a robot spindle, a clamping base and an end flange. An end flange is provided on the rear end cover of each joint module motor. There are six symmetrical threaded holes on the end flange, which are connected to the robot spindle through bolts. The corner between the 3rd and 4th axes is connected with a joint intermediate converter bolt. The six joint module motors are connected between the joint modules through the robot spindle and the flange, realizing the six-free motion trajectory planning and welding action of the welding robot.

[0044] In a specific embodiment, the weld detection device is mainly used to extract image information of the weld. By obtaining the image information of the weld and processing the image information of the weld, the starting position and end position of the weld are determined, and the welding object is positioned, thereby guiding the robot to carry the welding gun to weld the weld.

[0045] In some embodiments, the weld detection device 2 includes a detection head fixture 21 and a line structured light sensor 22. The detection head fixture 21 includes two clamping plates, which are arranged opposite to each other and are fixedly connected to the working end of the robot body 1. Threaded holes are provided on the two clamping plates, and the line structured light sensor 22 is threadedly connected to the clamping plates through the threaded holes.

[0046] In this embodiment, the weld seam within the image area of ​​the target detection area is extracted by a line structured light sensor, so that the position of the weld seam can be calibrated while extracting the image; the line structured light sensor is fixed to the working end of the robot body by a clamping plate. When the robot moves, the line structured light sensor moves with it to extract the weld seam image.

[0047] It should be noted that the clamping plate structure can clamp and secure the linear structured light sensor, providing protection for the linear structured light sensor. The clamping space between the two clamping plates can be adjusted according to the size of the linear structured light sensor. Specifically, the clamping plate is fixed to the rear end of the actuator flange at the end of the robot body via bolts.

[0048] In some embodiments, the weld positioning robot also includes a welding gun clamp 6, which includes a first half ring 61, a second half ring 62 and a locking piece 63. One end of the first half ring 61 and the second half ring 62 are fixedly connected to the working end. The locking piece 63 is used to connect the first half ring 61 and the second half ring 62 and make the first half ring 61 and the second half ring 62 embrace to form a clamping space. The welding gun 3 is located in the clamping space and abuts against the inner walls of the first half ring and the second half ring.

[0049] In this embodiment, the welding gun fixture securely secures the welding gun to the robot body, preventing vibration during operation that could affect welding performance. Specifically, the welding gun fixture comprises a first half ring and a second half ring. The ring diameters of the first and second half rings can be adjusted to accommodate welding guns of varying sizes, depending on the size and model of the welding gun.

[0050] In some embodiments, a surface of the base 4 close to the supporting end of the robot body has a groove, the groove fits with the supporting end, and the supporting end is inserted into the groove.

[0051] In this embodiment, a groove is provided on the base, the support end of the robot body is inserted into the groove, and the diameter of the groove is basically adapted to the size of the support end of the robot body, which can prevent the robot body from shifting and deviating when rotating relative to the base.

[0052] In some embodiments, the guide rail device 5 includes at least two slide rails and at least two sliders, the at least two slide rails are spaced apart and distributed in parallel, the number of the sliders is consistent with the number of the guide rails, one side of the at least two slide rails is installed on the side of the base away from the support end, and the other side is inserted into the guide rail, and the slider can slide along the guide rail.

[0053] In this embodiment, by providing a guide rail device, the robot body can move a set distance along the guide rail device, so that the robot drives the welding gun to increase its working range.

[0054] It should be noted that the length of the guide rail can be set according to actual needs, and the number of guide rails can also be set according to actual needs. It can be understood that the more guide rails there are, the higher the maintenance cost. On the basis of ensuring stable movement, the number of guide rails in this embodiment is 2, and correspondingly, the number of sliders is also 2. The two sliders are installed at the bottom of the support end of the robot body and inserted into the slide rail. The movement of the slider in the slide rail drives the movement of the robot body, thereby increasing the working range of the welding gun and the scanning range of the weld detection device.

[0055] Based on the above-mentioned weld positioning robot, the present invention also proposes a weld positioning method, see Figure 4 , methods include:

[0056] S401, acquiring a depth image of the weld area;

[0057] S402, using a preset weld edge contour extraction method to extract contour features from the depth image to obtain weld edge contour information;

[0058] S403, using a preset minimum circumscribed rectangle algorithm to perform a minimum area search on the weld edge contour information to obtain a minimum enclosing rectangle;

[0059] S404: Determine the starting position and the ending position of the weld to be welded by the welding positioning robot according to the coordinates of the corner points of the minimum enclosing rectangle.

[0060] In this embodiment, the robot body is first moved to the weld area, and a grayscale image of the weld area is collected by an acquisition device. Then, a preset weld edge contour extraction method is used to extract contour features from the grayscale image to obtain weld edge contour information. Then, a preset minimum circumscribed rectangle algorithm is used to perform a minimum area search on the weld edge contour information to obtain a minimum enclosing rectangle. Finally, the starting and ending positions of the welding robot are determined based on the coordinates of the corner points of the minimum enclosing rectangle. The present invention achieves the purpose of rapid positioning of robot welding and integration of detection and welding functions.

[0061] It should be noted that the grayscale image obtained by scanning with a line structured light sensor can reflect the statistical distribution of different grayscale levels of the welding object, which can be represented by a one-dimensional discrete function:

[0062] h(k)=n k k=0,1,…,L-1

[0063] n k is the number of grayscale pixels in the image f(x,y), and each column of the histogram corresponds to a height n k The histogram reflects the distribution of the grayscale image of the welding object. The mean and variance of the histogram are also the mean and variance of the grayscale image. The relative frequency P of the grayscale level of the normalized histogram is r (k):

[0064] P r (k) = n k / N

[0065] Where N is the total number of pixels of the image f(x,y), the discrete form of the transformation function of the grayscale image T(r k ) can be expressed as:

[0066]

[0067] r k Denotes the normalized grayscale level, and k denotes the grayscale level before normalization. Histogram averaging is used to process the depth image. This histogram averaging correction increases the grayscale intervals of the weld groove depth image, enhances the weld groove features, and increases the dynamic range of grayscale differences between pixels in the groove image, thereby enhancing the contrast between the groove features and the background. This improves the overall visibility of the weld groove. When scanning the depth image of the weld groove, it often contains many small noise points. Because the groove features contain a large number of edges, a median filter algorithm is used to filter out salt and pepper noise while preserving the weld groove edges and preventing blur.

[0068] Furthermore, contour feature extraction is performed on the grayscale image, including extracting edge features of the weld using a preset weld edge contour extraction method, so as to better separate local and small welds from the image background.

[0069] In some embodiments, see Figure 5 , the preset weld edge contour extraction method is used to extract contour features from the depth image, including:

[0070] S501, performing an opening operation on the depth image using a preset morphology to adjust the grayscale value of the depth image to obtain an enhanced image;

[0071] S502, using a preset mean filter algorithm to suppress noise on the enhanced image to obtain denoised weld edge contour information;

[0072] S503, using a preset binary threshold segmentation algorithm to perform pixel-level segmentation on the denoised weld edge contour information to obtain an enhanced feature region image;

[0073] S504: Based on the feature region image, a preset Canny edge detection algorithm is used to extract weld edge features to obtain a feature separation image. In this embodiment, since the weld groove occupies a small area in the depth image, in order to accurately locate the weld position from the majority of invalid welding background and obtain the feature region, an adaptive binary threshold segmentation algorithm based on the least squares method is used to enhance the feature region of the weld image. Subsequently, in order to obtain the edge features of the depth image, i.e., the location where the attributes of the weld groove and the background region suddenly change, the preset Canny edge detection algorithm is used to extract the edge features of the weld groove, further narrow the ROI area, separate the groove features from the background, and facilitate the search for the start and end positions of the weld. A depth morphological opening operation is then performed on the extracted edge feature region, and the feature region is first corroded and then expanded to filter out noise blocks that are smaller than the structural elements of the weld groove, eliminate burrs, and connect and fill small holes to improve the groove information missing due to noise.

[0074] In some embodiments, see Figure 6 , the grayscale image is segmented at the pixel level using a preset binary threshold segmentation algorithm, including:

[0075] S601, determining a pixel segmentation threshold according to the grayscale features of the grayscale image;

[0076] S602: Perform value segmentation on the grayscale image according to a pixel segmentation threshold to obtain a foreground image and a background image.

[0077] In this embodiment, when the grayscale value of a pixel is lower than the threshold, it is assigned a value of 0, and when it is higher than the threshold, it is assigned a value of 255. The segmented image satisfies the following formula:

[0078]

[0079] Traverse all the pixels of the image and divide the image into foreground and background according to the grayscale characteristics of the image

[0080] w0, its average gray value is u0, the background w1 its average gray value is u1, the total average gray value of the image is u, and the variance of the foreground and background is g:

[0081]

[0082] The threshold at this time is calculated based on the corresponding threshold of each small area on the image. Therefore, different thresholds are used for different areas on the same image. The grayscale T when the variance g is the largest is the optimal threshold. Pixel-level segmentation is performed based on T between the welding groove and the welding groove background to enhance its feature area.

[0083] In some embodiments, see Figure 7 The method of using a preset minimum bounding rectangle algorithm to search the minimum area of ​​the weld edge contour information to obtain the minimum enclosing rectangle includes:

[0084] S701, using a preset minimum circumscribed rectangle algorithm to perform a minimum area search on the weld edge contour information to determine a minimum enclosing polygon;

[0085] S702: Based on the minimum enclosing polygon, traverse all adjacent points and use a preset minimum enclosing rectangle algorithm to determine the enclosing rectangle with the smallest area.

[0086] In this embodiment, the minimum enclosing rectangle is searched for by the minimum enclosing rectangle algorithm to determine the starting and ending positions of the weld seam. The minimum enclosing rectangle algorithm is performed on the weld groove contour to obtain the minimum area polygon of the weld groove area. The line connecting two adjacent points of the minimum area polygon is used as the rectangle edge L1. The parallel line of the polygon from the edge L1 is found and regarded as the second edge L2 of the rectangle; the points on the convex hull are projected onto the edge L1, and the two points farthest apart are found among the projected points. A perpendicular line to L1 is drawn through these two points and regarded as the other two sides L3 and L4 of the rectangle; all adjacent points of the polygon are traversed, and the above minimum enclosing rectangle algorithm steps are repeated to obtain the rectangle with the smallest area. Since the obtained rectangle corner points are located in the image coordinate system, the coordinate system needs to be converted to be consistent with the coordinate system of the detection data to obtain the starting and ending positions of the weld groove rectangle. The coordinate system is converted to find the four corner point information of the groove to obtain the starting and ending positions of the welding robot.

[0087] Furthermore, since there is no correlation between the grayscale image collected by the line structured light sensor and the robot's motion trajectory, the corner point information needs to be converted into the robot's motion coordinate system. During the trajectory planning stage, the robot automatically selects two of the four corner points as the starting and ending positions.

[0088] The specific embodiments of the present invention described above do not limit the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A weld positioning method, characterized in that: A weld positioning robot is used, which includes: a robot body, a weld detection device, a welding gun, a base and a guide rail device; wherein: The robot body has a working end and a supporting end; the weld detection device is rotatably connected to the working end of the robot body; the welding gun is fixedly connected to the working end of the robot body, and the welding gun and the weld detection device are spaced apart; the supporting end of the robot body is rotatably connected to the base; the base is slidably connected to the guide rail device on a side away from the robot body; Weld location methods include: Acquiring a depth image of the weld area collected by a weld detection device; Using a preset weld edge contour extraction method, contour feature extraction is performed on the depth image to obtain weld edge contour information; Using a preset minimum circumscribed rectangle algorithm to search for the minimum area of ​​the weld edge contour information to obtain a minimum enclosing rectangle; The starting point and the ending point of the weld to be welded by the welding positioning robot are determined according to the coordinates of the corner points of the minimum enclosing rectangle.

2. The weld positioning method according to claim 1, characterized in that: The weld detection device includes a detection head fixture and a line structured light sensor. The detection head fixture includes two clamping plates. The two clamping plates are arranged opposite to each other and are fixedly connected to the working end of the robot body. Both clamping plates are provided with threaded holes, and the line structured light sensor is threadedly connected to the clamping plates through the threaded holes.

3. The weld positioning method according to claim 1, characterized in that: The weld positioning robot also includes a welding gun clamp, which includes a first half ring, a second half ring and a locking piece. One end of the first half ring and the second half ring are both fixedly connected to the working end. The locking piece is used to connect the first half ring and the second half ring and make the first half ring and the second half ring embrace to form a clamping space. The welding gun is located in the clamping space and abuts against the inner walls of the first half ring and the second half ring.

4. The weld positioning method according to claim 1, characterized in that: A side of the base close to the supporting end of the robot body has a groove, the groove fits with the supporting end, and the supporting end is inserted into the groove.

5. The weld positioning method according to claim 4, characterized in that: The guide rail device includes at least two slide rails and at least two sliders. The at least two slide rails are distributed in parallel at intervals. The number of the sliders is consistent with the number of the guide rails. The at least two slide rails are installed on one side of the base away from the support end at intervals, and the other side is inserted into the guide rail. The slider can slide along the guide rail.

6. The weld positioning method according to claim 1, characterized in that: The method of extracting contour features from the depth image using a preset weld edge contour extraction method includes: An opening operation is performed on the depth image using a preset morphology to adjust the grayscale value of the depth image to obtain an enhanced image; a preset mean filtering algorithm is used to suppress noise on the enhanced image to obtain denoised weld edge contour information; Using a preset binary threshold segmentation algorithm, pixel-level segmentation is performed on the denoised weld edge contour information to obtain an enhanced feature area image; Based on the feature area image, a preset Canny edge detection algorithm is used to extract weld edge features to obtain a feature separation image.

7. The weld positioning method according to claim 6, characterized in that: The pixel-level segmentation of the grayscale image is performed using a preset binary threshold segmentation algorithm, including: Determine the pixel segmentation threshold according to the grayscale features of the grayscale image; According to the pixel segmentation threshold, the grayscale image is segmented to obtain a foreground image and a background image.

8. The weld positioning method according to claim 1, characterized in that: The method of using a preset minimum circumscribed rectangle algorithm to search the minimum area of ​​the weld edge contour information to obtain a minimum enclosing rectangle includes: Using a preset minimum circumscribed rectangle algorithm to search for the minimum area of ​​the weld edge contour information to determine the minimum enclosing polygon; Based on the minimum enclosing polygon, all adjacent points are traversed and a preset minimum enclosing rectangle algorithm is used to determine the enclosing rectangle with the smallest area.

9. The weld positioning method according to claim 1, characterized in that: The step of determining the starting position and the ending position of the welding robot according to the coordinates of the corner points of the minimum enclosing rectangle includes: Convert the corner coordinates of the minimum enclosing rectangle into target coordinates corresponding to the robot body coordinate system; The starting position and the ending position of the welding robot are determined according to the target coordinates corresponding to the minimum wrapping rectangle.

Images