Automatic welding feeding and discharging method and device based on flexible control

Abstract

The invention relates to the technical field of welding control, in particular to an automatic welding feeding and discharging method and device based on flexible control. According to the method, a welding seam edge combination is screened out based on the distance uniformity between adjacent edge lines, a welding over-rotation angle is further determined by analyzing the local trend of the edge lines in the welding seam edge combination, the welding over-rotation angle is used for representing the structural stability of a welding seam, and finally the welding stability is obtained in combination with the welding length. And the optimal welding path can be optimized based on the initial weight of the welding stability adjusting algorithm and the final path pheromone weight, and then welding feeding and discharging control is carried out based on the parameter information of the optimal welding path. According to the method, the pheromone weight of the ant colony algorithm in the welding path optimization process is optimized, the optimization efficiency and the result accuracy are improved, and automatic welding feeding and discharging can be conveniently and effectively controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of welding control, and particularly relates to an automatic welding loading and unloading method and device based on flexible control. Background Art

[0002] Flexible control refers to the ability in an automatic welding system to flexibly adjust the control strategies for the welding and loading / unloading processes according to changes in the specifications, shapes, and sizes of different workpieces. This flexibility can reduce equipment switching and setup times and improve the system's ability to adapt to diverse production requirements. Through flexible control, the equipment can automatically adjust welding paths, parameters, etc. to adapt to workpieces of different types and sizes, thereby achieving efficient and stable automated production. During the automatic welding process, the loading / unloading process refers to automatically feeding the workpiece to be welded into the welding equipment and then removing the workpiece from the equipment after welding is completed. This process is usually completed by robotic arms, conveyor belts, or other automated equipment. The purpose of automating the loading / unloading is to improve production efficiency, reduce manual operations, and enhance the accuracy and consistency of welding. The control of loading / unloading needs to be synchronized with the welding process to ensure a continuous and smooth production flow.

[0003] Among them, during the automatic welding process, the identification and positioning of workpieces in the prior art are mainly carried out through two-dimensional image recognition. Image processing algorithms are used to identify several parameters of the workpiece, and then to guide the robotic arm to perform precise loading and unloading processing. To achieve precise control of loading and unloading and smoothly switch the workpieces to be welded, it is necessary to add pre-preparation for the welding process, reasonably arrange the preparation time of the next workpiece, reduce waiting time, achieve seamless connection, and improve welding efficiency. Among them, when identifying welding parameters, there are multiple welding paths on the material surface. The prior art uses the ant colony algorithm to optimize the welding paths, and then determines the optimal welding path for automatic welding and performs loading and unloading control. In the optimization process, the initial weight of each welding path can be set based on the ant colony algorithm. However, there are deviation, rust, and other noise information in the weld seam information, resulting in a waste of a lot of time in these noise information during the optimization process, affecting the efficiency of the optimization algorithm and also the accuracy of the finally obtained optimal welding path, and thus unable to perform accurate loading and unloading control. Summary of the Invention

[0004] In order to solve the technical problem that in the prior art, when using the ant colony algorithm to optimize the weld seam path, it is affected by noise, resulting in low algorithm efficiency and accuracy and unable to perform effective loading and unloading control, the purpose of the present invention is to provide an automatic welding loading and unloading method and device based on flexible control, and the specific technical solutions adopted are as follows: The present invention proposes an automatic welding loading and unloading method based on flexible control, and the method includes: Obtain the surface image of the workpiece to be welded, extract the edge lines in the surface image, and screen out the weld edge combinations according to the distance uniformity between adjacent edge lines; According to the local trend of the edge lines in the weld edge combination, obtain the set of included angles within the seam for each weld edge combination; screen out the welding over-rotation angles from the included angles within the seam; based on the distribution of the welding over-rotation angles in the weld edge combination and the welding length of the weld edge combination, obtain the welding stability of each weld edge combination; Perform welding path planning on the surface image according to the ant colony algorithm. In the ant colony algorithm, for each weld edge combination, adjust the initial weight of the algorithm according to the welding stability to obtain the final path pheromone weight, and plan the optimal welding path based on the final path pheromone weight; Control the loading and unloading of welding according to the path length, path direction, and welding time of the optimal welding path.

[0005] Further, the screening of the weld edge combination according to the distance between adjacent edge lines includes: Obtain the edge merging property of adjacent two edge lines according to the distance uniformity at different local positions between adjacent two edge lines; take the two edge lines with the edge merging property greater than the preset merging threshold as the weld edge combination.

[0006] Further, the method for obtaining the edge merging property includes: Take any one of the adjacent two edge lines as the target edge line and the other edge line as the comparison edge line. Uniformly sample multiple sampling points on the target edge line, and take the distance between each sampling point and the comparison edge line as the comparison distance of the sampling point; perform negative correlation mapping on the difference of the comparison distances between adjacent sampling points to obtain the local distance uniformity, and take the average value of all local distance uniformities as the edge merging property.

[0007] Further, the method for obtaining the set of included angles within the seam includes: Take the angle between the connection line of two adjacent sampling points on the target edge line and the horizontal direction as the local angle, calculate the angle difference between adjacent local angles, and obtain the included angle within the seam in the set of included angles within the seam.

[0008] Further, the welding over-rotation angle is the included angle within the seam in the set of included angles within the seam that is greater than the preset angle threshold.

[0009] Further, the method for obtaining the welding stability includes: Perform negative correlation mapping and normalization on the proportion of the welding over-rotation angle in the set of included angles within the seam, and then multiply it by the welding length to obtain the welding stability of the weld edge combination.

[0010] Further, the method for obtaining the weight of the final path pheromone includes: Normalize the welding stability and add it to the positive integer 1 to obtain an adjustment coefficient, and multiply the adjustment coefficient by the initial weight of the algorithm to obtain the weight of the final path pheromone.

[0011] Further, the method for obtaining the welding time of the optimal welding path includes: In the historical database, fit the welding length and welding time in each historical welding process to obtain a welding length-time model; input the path length of the optimal welding path into the welding length-time model to obtain the welding time of the optimal welding path.

[0012] Further, the edge line is obtained by processing the surface image through the canny edge detection algorithm.

[0013] The present invention also provides an automatic welding loading and unloading device based on flexible control, and the device includes: A welding image recognition module, configured to obtain the surface image of the workpiece to be welded, extract the edge line in the surface image, and screen out the weld edge combination according to the distance between adjacent edge lines; A weld information extraction module, configured to obtain the set of included angles within the seam of each weld edge combination according to the local trend of the edge line in the weld edge combination; screen out the welding super-rotation angle from the included angles within the seam; and obtain the welding stability of each weld edge combination according to the distribution of the welding super-rotation angle in the weld edge combination and the welding length of the weld edge combination; An optimal welding path planning module, configured to plan the welding path on the surface image according to the ant colony algorithm. In the ant colony algorithm, for each weld edge combination, adjust the initial weight of the algorithm according to the welding stability to obtain the weight of the final path pheromone, and plan the optimal welding path based on the weight of the final path pheromone; A welding loading and unloading control module, configured to control the welding loading and unloading according to the path length and welding time of the optimal welding path.

[0014] The present invention has the following beneficial effects: The present invention takes into account that a good weld seam should be a weld seam with obvious structural stability and regularity. Therefore, first, a weld seam edge combination is screened based on the distance uniformity between adjacent edge lines, that is, the weld seam edge combination is a combination of two edge lines with a relatively regular distance distribution, and it is more likely to be an excellent weld seam. Further, the welding super-rotation angle is determined by analyzing the local trend of the edge lines in the weld seam edge combination, and the welding super-rotation angle is used to characterize the structural stability of the weld seam. Finally, the welding stability is obtained by combining the welding length. The welding stability can characterize the welding stability degree of the weld seam during the welding process. Therefore, the initial weights of the algorithm can be adjusted based on the welding stability, and the optimal welding path can be found based on the final path pheromone weights. Furthermore, the welding loading and unloading control can be carried out based on the parameter information of the optimal welding path, which is convenient for effectively controlling the automatic welding loading and unloading. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a flowchart of an automatic welding loading and unloading method based on flexible control provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific embodiments, structures, features, and effects of a method and equipment for automatic welding loading and unloading based on flexible control proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0019] The embodiments of the present invention are for the optimization in the process of using the ant colony algorithm to optimize the welding path during the automatic welding process. The ant colony algorithm is an optimization algorithm based on bionics, and it optimizes the welding path by simulating the process of ants looking for food. Its basic principle is: (1) Pheromone mechanism: Ants release pheromones on the path, and the concentration of pheromones on the path reflects the quality of the path. When planning the welding path, the initial path is randomly generated by virtual "ants" according to the weld information in the image.

[0020] (2) Path selection: When ants select a path at each step, they tend to choose the path with a higher pheromone concentration and randomly explore according to a certain probability to avoid falling into a local optimum.

[0021] (3) Pheromone update: After each "ant" completes a path, the pheromone is updated according to the quality of the path. The shortest path will retain more pheromones, while the pheromones of the poorer paths will gradually evaporate. Through multiple iterations, the algorithm gradually finds the best welding path.

[0022] The specific algorithm for the ant colony algorithm to optimize the welding path is a well-known technical means for those skilled in the art and will not be elaborated here. The embodiments of the present invention mainly focus on the update of the initial weight in the algorithm.

[0023] The scenario in which the embodiments of the present invention are implemented is an automatic welding scenario. When performing automatic welding, a pre-preparation workbench needs to be equipped to prevent the workpiece to be welded, and a fixing device is used to ensure the stability of the workpiece during image capture and pose adjustment; a manipulator for pose adjustment and a depth camera are equipped to capture the workpiece image.

[0024] (1) Among them, the workbench parameters include: made of aluminum material to ensure high strength and durability; the size can be specifically set based on the size of the workpiece to be welded; the maximum load-bearing weight is set to 500 kg.

[0025] (2) Among them, the parameters of the mechanical part for pose adjustment include: having six degrees of freedom, which can flexibly adjust the position and angle of the workpiece; the repeat positioning accuracy is ±0.05 mm; the maximum load is 50 kg; the radius of the movement range is 1.5 m.

[0026] (3) Among them, the parameters of the depth camera include: the resolution is 1280×720; the field of view angle is 70 degrees; the depth range is from 0.3 m to 2 m; the frame rate is 30 frames.

[0027] The following specifically describes the specific solution of an automatic welding loading and unloading method and device based on flexible control provided by the present invention with reference to the accompanying drawings.

[0028] Please refer to Figure 1 , which shows a flowchart of an automatic welding loading and unloading method based on flexible control provided by an embodiment of the present invention. The method includes: Step S1: Obtain the surface image of the workpiece to be welded, extract the edge lines in the surface image, and screen out the weld edge combinations according to the distance uniformity between adjacent edge lines.

[0029] As described in the above scenario introduction content, in the embodiment of the present invention, the surface image of the welding workpiece is obtained based on the configured depth camera. The size of the pixel value in the surface image represents the distance between the corresponding position of the workpiece and the camera. Since the weld seam has obvious depth information compared with other surface positions, the pixel value in the surface image is also different from that of other surface areas. Therefore, the edge line in the surface image can be extracted based on the pixel value for further analysis.

[0030] In the embodiment of the present invention, after obtaining the surface image, image preprocessing is performed on it to improve the image quality. The image preprocessing operations include operations such as grayscale conversion and mean filtering, which are specific technical means well-known to those skilled in the art and will not be elaborated here.

[0031] In the embodiment of the present invention, the canny edge detection method is used to identify the edge line in the surface image. The canny edge detection is a technical means well-known to those skilled in the art and will not be elaborated here.

[0032] Since the weld seam has a certain width, it can be considered that a weld seam is composed of two edges, and the distance between these two edges is relatively uniform, forming a weld seam with a width. Therefore, the weld seam edge combination can be screened out according to the distance uniformity between adjacent edges. In the subsequent analysis process, only the weld seam edge combination is analyzed, avoiding the interference of information such as patterns on the workpiece surface.

[0033] Preferably, in an embodiment of the present invention, screening out the weld seam edge combination according to the distance between adjacent edge lines includes: Obtaining the edge mergibility of adjacent two edges according to the distance uniformity of different local positions between adjacent two edge lines; taking the two edge lines with the edge mergibility greater than the preset mergibility threshold as the weld seam edge combination. In the embodiment of the present invention, after normalizing the edge mergibility, the mergibility threshold is set to 0.9.

[0034] Further, the method for obtaining the edge mergibility includes: Take any one of the adjacent two edge lines as the target edge line and the other as the comparison edge line. Uniformly sample multiple sampling points on the target edge line, and take the distance between each sampling point and the comparison edge line as the comparison distance of the sampling point. It should be noted that the distance between the sampling point and the comparison edge line is the minimum value of the distances between the sampling point and each edge point on the comparison edge line. If these two edge lines are the edge lines of a weld, the comparison distances should exhibit a uniform distribution characteristic, and there will be no significant change in the comparison distances between different sampling points. Therefore, further perform a negative correlation mapping on the differences in the comparison distances between adjacent sampling points to obtain the local distance uniformity, that is, the greater the difference in the comparison distances between a sampling point and its adjacent sampling point, the greater the abnormality in the distribution of the comparison distances at this local position, and the smaller the local distance uniformity. Take the average value of all local distance uniformities as the edge combination property.

[0035] As an example, the edge combination property is expressed by the formula: Among them, is the edge combination property between the k-th group of adjacent two edge lines, is the number of sampling points, is the comparison distance of the n-th sampling point, is the comparison distance of the (n + 1)-th sampling point.

[0036] In the edge combination property formula, the differences in the comparison distances are negatively correlated through the reciprocal form. At the same time, to avoid the denominator being 0, a positive integer 1 is added to the denominator. That is, the closer the final edge combination property is to 1, the more likely the two edge lines are the edge combination of the weld.

[0037] In the embodiment of the present invention, the sampling points are 100 points uniformly set on the target edge line.

[0038] Step S2: According to the local trend of the edge lines in the weld edge combination, obtain the set of included angles within the seam for each weld edge combination; screen out the welding over-rotation angle from the included angles within the seam; according to the distribution of the welding over-rotation angle in the weld edge combination and the welding length of the weld edge combination, obtain the welding stability of each weld edge combination.

[0039] The workpiece to be welded has a relatively complete geometric structure, and the shape structure of the area to be welded is relatively regular. For example: straight welds, which are commonly used in plate splicing, connection of support structures, etc., and are commonly found in steel structures, frame structures, etc.; curved welds, which are commonly used in pipelines, vehicle body structures, flange connections, etc., and require welding around a circumference or a complex curved surface. During the automatic welding process, regardless of the structural shape of the area to be welded on the workpiece, the optimized welding method needs to ensure the smoothness of the entire automatic welding process. The smoother the welding, the higher the stability and welding efficiency of the entire welding process; conversely, the lower the smoothness, the more complex the welding process, the higher the requirements for the robotic arm, and the higher the risk of welding errors. Therefore, in the embodiments of the present invention, the set of included angles within the seam of each weld edge combination is obtained based on the local trend of the edge line in the weld edge combination, and then the welding super-angle is screened out. That is, at the position of the welding super-angle, the robotic arm of the automatic welding will perform a welding posture conversion with a higher difficulty. The more welding super-angles there are in the weld edge combination, the worse the smoothness of this path. Further combined with the welding length of the weld edge combination, the longer the welding length, the more margin the robotic arm has for adjustment and control, and the stronger the welding stability, thereby obtaining the welding stability of each weld edge combination.

[0040] Preferably, in the embodiments of the present invention, the method for obtaining the set of included angles within the seam includes: Taking the angle between the line connecting two adjacent sampling points on the target edge line and the horizontal direction as the local angle, calculating the angular difference between adjacent local angles, and obtaining the included angles within the seam in the set of included angles within the seam. That is, the greater the angular difference between the local angles, the greater the amplitude of the welding control performed by the robotic arm at this position. It should be noted that since there are 100 sampling points in the embodiments of the present invention, 99 local angles will be generated, and there are 98 angular differences, that is, the number of elements in the set of included angles within the seam is 98.

[0041] In the embodiments of the present invention, the welding super-angle is the included angle within the set of included angles within the seam that is greater than the preset included angle threshold. The included angle threshold is set to 90 degrees.

[0042] Preferably, in the embodiments of the present invention, the method for obtaining the welding stability includes: Performing a negative correlation mapping and normalization on the proportion of the welding super-angle in the set of included angles within the seam, and then multiplying it by the welding length to obtain the welding stability of the weld edge combination.

[0043] It should be noted that in the embodiments of the present invention, the proportion of non-welded super-rotation angles in the set of in-seam angles can be directly selected as the negatively correlated mapping and normalization result of the proportion of welded super-rotation angles in the set of in-seam angles. That is, the smaller the proportion of welded super-rotation angles, the larger the proportion of non-welded super-rotation angles, indicating that the control of the robotic arm during the automated welding process is smoother and more stable, and the welding stability is stronger.

[0044] Step S3: Perform welding path planning on the surface image according to the ant colony algorithm. In the ant colony algorithm, for each weld edge combination, adjust the initial weight of the algorithm according to the welding stability to obtain the final path pheromone weight, and plan the optimal welding path based on the final path pheromone weight.

[0045] After the above steps, each weld edge combination corresponds to a welding stability, and thus the initial weight of the algorithm can be adjusted in the ant colony algorithm. When using the final path pheromone weight for optimization, compared with the traditional ant colony optimization algorithm, it can improve the convergence speed and the accuracy of the optimal welding path more according to the basic characteristics of the workpiece to be welded.

[0046] Preferably, in an embodiment of the present invention, the method for obtaining the final path pheromone weight includes: Normalize the welding stability and add it to the positive integer 1 to obtain an adjustment coefficient, and multiply the adjustment coefficient by the initial weight of the algorithm to obtain the final path pheromone weight.

[0047] It should be noted that the normalization in the embodiments of the present invention adopts the linear normalization method, and methods such as sigmoid function mapping can also be used, which will not be elaborated and limited here.

[0048] Step S4: Perform pre-preparation work for welding loading and unloading according to the path length, path direction, and welding time of the optimal welding path.

[0049] After obtaining the optimal welding path, its welding length and the main direction corresponding to this path can be intuitively obtained. In the subsequent automatic welding process, the main direction can be used as a reference value for attitude adjustment, greatly improving the turning effectiveness during the welding process.

[0050] In the control of loading and unloading during the automatic welding process, the welding time is also an important welding parameter. By determining the path length, path direction, and welding time of the optimal welding path, the pre-preparation work for automatic welding is completed. In the automatic welding process, automatic welding can be directly carried out through these welding parameters, controlling the loading and unloading parameters and welding direction during the welding process.

[0051] Preferably, in the embodiments of the present invention, the method for obtaining the welding time of the optimal welding path includes: In the historical database, the welding length and welding time in each historical welding process are fitted to obtain a welding length-time model; the path length of the optimal welding path is input into the welding length-time model to obtain the welding time of the optimal welding path. In the embodiment of the present invention, the least squares method is selected for linear fitting to obtain the welding length-time model.

[0052] In summary, the present invention first screens out the weld edge combinations based on the distance uniformity between adjacent edge lines, further determines the welding super-rotation angle by analyzing the local trends of the edge lines in the weld edge combinations, uses the welding super-rotation angle to characterize the structural stability of the weld, and finally obtains the welding stability in combination with the welding length. The initial weights of the algorithm are adjusted based on the welding stability, and the optimal welding path can be optimized based on the final path pheromone weights, and then the welding loading and unloading control is performed based on the parameter information of the optimal welding path. By optimizing the pheromone weights of the ant colony algorithm during the welding path optimization process, the present invention improves the optimization efficiency and the accuracy of the results, and facilitates the effective control of automatic welding loading and unloading.

[0053] Based on the same inventive concept, the present invention also proposes an automatic welding loading and unloading device based on flexible control, and the device includes: A welding image recognition module, configured to acquire a surface image of a workpiece to be welded, extract edge lines in the surface image, and screen out weld edge combinations according to the distance between adjacent edge lines; A weld information extraction module, configured to obtain a set of in-seam angles for each weld edge combination according to the local trends of the edge lines in the weld edge combination; screen out the welding super-rotation angle from the in-seam angles; and obtain the welding stability of each weld edge combination according to the distribution of the welding super-rotation angle in the weld edge combination and the welding length of the weld edge combination; An optimal welding path planning module, configured to perform welding path planning on the surface image according to the ant colony algorithm. For each weld edge combination in the ant colony algorithm, adjust the initial weights of the algorithm according to the welding stability to obtain the final path pheromone weights, and plan the optimal welding path based on the final path pheromone weights; A welding loading and unloading control module, configured to control the welding loading and unloading according to the path length and welding time of the optimal welding path.

[0054] It should be noted that: the above-mentioned sequence of embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0055] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An automatic welding loading and unloading method based on flexible control, characterized in that, The method includes: Obtain the surface image of the workpiece to be welded, extract the edge lines in the surface image, and screen out the weld edge combinations according to the distance uniformity between adjacent edge lines; According to the local trend of the edge lines in the weld edge combination, obtain the set of included angles within the seam for each weld edge combination; screen out the welding over-rotation angles from the included angles within the seam; according to the distribution of the welding over-rotation angles in the weld edge combination and the welding length of the weld edge combination, obtain the welding stability of each weld edge combination; Perform welding path planning on the surface image according to the ant colony algorithm. In the ant colony algorithm, for each weld edge combination, adjust the initial weight of the algorithm according to the welding stability to obtain the final path pheromone weight, and plan the optimal welding path based on the final path pheromone weight; Control the welding loading and unloading according to the path length, path direction, and welding time of the optimal welding path.

2. The automatic welding loading and unloading method based on flexible control according to claim 1, characterized in that, The screening of the weld edge combination according to the distance between adjacent edge lines includes: Obtain the edge merger of adjacent two edge lines according to the distance uniformity at different local positions between adjacent two edge lines; take the two edge lines with the edge merger greater than the preset merger threshold as the weld edge combination.

3. A method for automatic welding loading and unloading based on flexible control according to claim 2, characterized in that, The method for obtaining the edge merger includes: Take any one of the adjacent two edge lines as the target edge line and the other edge line as the comparison edge line. Uniformly sample multiple sampling points on the target edge line, and take the distance between each sampling point and the comparison edge line as the comparison distance of the sampling point; perform a negative correlation mapping on the difference in the comparison distances between adjacent sampling points to obtain the local distance uniformity, and take the average value of all local distance uniformities as the edge merger.

4. The automatic welding loading and unloading method based on flexible control according to claim 3, characterized in that, The method for obtaining the set of included angles within the seam includes: Take the angle between the connection line of two adjacent sampling points on the target edge line and the horizontal direction as the local angle, calculate the angle difference between adjacent local angles, and obtain the included angle within the seam in the set of included angles within the seam.

5. A method for automatic welding loading and unloading based on flexible control according to claim 1, characterized in that, The welding over-rotation angle is the included angle within the seam in the set of included angles within the seam that is greater than the preset angle threshold.

6. A method for automatic welding loading and unloading based on flexible control according to claim 1, characterized in that, The method for obtaining the welding stability includes: Perform a negative correlation mapping and normalization on the proportion of the welding over-rotation angle in the set of included angles within the seam, and then multiply it by the welding length to obtain the welding stability of the weld edge combination.

7. A method for automatic welding loading and unloading based on flexible control according to claim 1, characterized in that, The method for obtaining the final path pheromone weight includes: Normalize the welding stability and then add the positive integer 1 to obtain the adjustment coefficient, and multiply the adjustment coefficient by the initial weight of the algorithm to obtain the final path pheromone weight.

8. A method for automatic welding loading and unloading based on flexible control according to claim 1, characterized in that, The method for obtaining the welding time of the optimal welding path includes: In the historical database, fit the welding length and welding time in each historical welding process to obtain a welding length-time model; input the path length of the optimal welding path into the welding length-time model to obtain the welding time of the optimal welding path.

9. A method for automatic welding loading and unloading based on flexible control according to claim 1, characterized in that, The edge line is obtained by processing the surface image through the canny edge detection algorithm.

10. An automatic welding loading and unloading device based on flexible control, characterized in that, The device includes: The welding image recognition module is used to obtain the surface image of the workpiece to be welded, extract the edge lines in the surface image, and screen out the weld edge combinations according to the distances between adjacent edge lines; The weld information extraction module is used to obtain the set of included angles within the seam for each weld edge combination according to the local trends of the edge lines in the weld edge combination; screen out the welding over-rotation angles from the included angles within the seam; and obtain the welding stability of each weld edge combination according to the distribution of the welding over-rotation angles in the weld edge combination and the welding length of the weld edge combination; The optimal welding path planning module is used to perform welding path planning on the surface image according to the ant colony algorithm. In the ant colony algorithm, for each weld edge combination, adjust the initial weight of the algorithm according to the welding stability to obtain the final path pheromone weight, and plan the optimal welding path based on the final path pheromone weight; The welding loading and unloading control module is used to control the welding loading and unloading according to the path length of the optimal welding path and the welding time.

Images