A cabinet foot automatic welding device and method

Abstract

The present invention relates to the field of automated welding technology, and specifically to an automated welding device and method for cabinet legs. By installing a micro welding gun and an image acquisition device on a welding arm, the image acquisition device moves together with the welding gun, and the welding gun has at least two different servos to control the moving direction, welding is achieved. The method includes acquiring an image of the welding area, determining the detection movement vector of the welding gun at two adjacent moments; determining the angle change stability index according to the servo angle change; and then determining the movement stability coefficient of the welding gun at the current moment; adjusting the preset time range according to the movement stability coefficient at all moments within the preset time range to obtain a target time period; determining the wire feeding speed and welding current of the welding gun at the current moment for the detection movement vector of the welding gun at all moments within the target time period and the width of the weld to be welded. The present invention can effectively improve welding consistency and enhance welding quality.

Description

Technical Field

[0001] The invention relates to the technical field of automated welding, and in particular to an automated welding device and method for cabinet legs. Background Art

[0002] In modern manufacturing, automated welding technology is widely used in the manufacturing and assembly of various metal components, greatly improving production efficiency and welding quality. Especially in the furniture manufacturing industry, cabinet feet are an important supporting structure of the cabinet body, and their welding quality directly affects the overall stability and service life of the cabinet body.

[0003] In the related art, the welding of the cabinet legs is achieved by an automated fixed line. In this method, the welding state of the welding gun changes and the weld width will vary, resulting in poor overall welding consistency, affecting the final welding quality. Summary of the invention

[0004] In order to solve the technical problem in the related art that the welding state of the welding gun changes and the weld width will produce differences, thereby resulting in poor overall welding consistency and affecting the final welding quality, the present invention provides an automatic welding device and method for cabinet legs, and the technical solutions adopted are as follows:

[0005] The present invention proposes an automatic welding method for cabinet legs, wherein a micro welding gun and an image acquisition device are installed on a welding arm, the image acquisition device moves together with the welding gun, and the welding gun has at least two different steering gears to control the moving direction. The method comprises:

[0006] Based on the image acquisition device, the welding area images at different times during the welding process are acquired, and the detection movement vectors of the welding gun at two adjacent times are determined according to the movement of the edges in the welding area images at two adjacent times;

[0007] Determine the angle value of each servo at different times, and determine the angle change stability index of the corresponding servo at the current moment according to the angle values ​​of the same servo at all times within a preset time range before the current moment; determine the movement stability coefficient of the welding gun at the current moment by combining the angle change stability indexes of all servos and the angle change correlation of any two servos;

[0008] The preset time range is adjusted according to the movement stability coefficient at all times within the preset time range to obtain a target time period;

[0009] According to the detected movement vector of the welding gun at all times within the target time period and the width of the weld to be welded, the wire feeding speed and welding current of the welding gun at the current moment are determined.

[0010] Further, according to the movement of the edge in the welding area image at two adjacent moments, the detection movement vector of the welding gun at two adjacent moments is determined, including:

[0011] Perform edge detection on each frame of the welding area image to obtain the edge connected domain of each frame of the welding area image;

[0012] Performing chain code analysis on the edge connected domain to determine the chain code sequence of all edge connected domains in two adjacent frames of welding area images;

[0013] Perform similarity analysis on the chain code sequences of edge connected domains in two adjacent frames of welding area images to determine matching connected domains representing the same region in two adjacent frames of welding area images;

[0014] According to the position change of the center point of each pair of the matching connected domains in the welding area image, the detection movement vector of the welding gun corresponding to the two moments is determined.

[0015] Further, according to the position change of the center point of each pair of the matching connected domains in the welding area image, the detection movement vector of the welding gun corresponding to the two moments is determined, including:

[0016] In any pair of matching connected domains, the image position change of the corresponding center point from the previous welding area image to the next welding area image is used as the connected domain movement vector;

[0017] The connected domain motion vectors of all pairs of matched connected domains are averaged to obtain the detection motion vector.

[0018] Further, according to the angle values ​​of the same steering gear at all times within a preset time range before the current time, the angle change stability index of the corresponding steering gear at the current time is determined, including:

[0019] Arrange the angle values ​​of the same servo at all times within a preset time range before the current time in time sequence to obtain a servo angle change sequence;

[0020] Performing first-order difference processing on the steering gear angle change sequence to obtain a first-order difference value;

[0021] Determine the standard deviation of the first-order difference values ​​of the same servo at all corresponding times;

[0022] The inverse of the standard deviation is normalized as a stability index of the angle change of the corresponding steering gear at the current moment.

[0023] Furthermore, the movement stability coefficient of the welding gun at the current moment is determined by combining the angle change stability indexes of all servos and the angle change correlation of any two servos, including:

[0024] According to the servo angle change sequence of any two servos, determine the angle change related indicators of the two servos;

[0025] The average of the angle change related indicators of all the servos in pairs is used as the first stability coefficient of the welding gun at the current moment;

[0026] The average value of the angle change stability index of all servos is used as the second stability coefficient of the welding gun at the current moment;

[0027] The product of the first stability coefficient and the second stability coefficient is calculated and normalized to obtain the movement stability coefficient of the welding gun at the current moment.

[0028] Furthermore, according to the steering gear angle change sequence of any two steering gears, the angle change related indicators of the corresponding two steering gears are determined, including:

[0029] The absolute value of the Pearson correlation coefficient of the two servo angle change sequences is used as the angle change correlation index corresponding to the two servo angles.

[0030] Further, the preset time range is adjusted according to the movement stability coefficient at all times within the preset time range to obtain a target time period, including:

[0031] The average value of the mobile stability coefficient at all times is used as the time adjustment weight;

[0032] The product of the time adjustment weight and the length of the preset time range is calculated to obtain a target time length, and the time range of the target time length before the current moment is used as the target time period.

[0033] Furthermore, the detecting movement vector of the welding gun at all times within the target time period and the width of the weld to be welded, and determining the wire feeding speed and welding current of the welding gun at the current time, include:

[0034] In the target time period, the detection movement vectors of the welding gun at all times are averaged, and the module lengths thereof are normalized as the vector influence index;

[0035] The width of the weld to be welded is normalized as a weld influence index;

[0036] Determining the solder requirement according to the vector influence index and the weld influence index;

[0037] The product of the solder demand degree and the preset maximum wire feeding speed is used as the wire feeding speed of the welding gun at the current moment, and the product of the solder demand degree and the preset maximum welding current is used as the welding current of the welding gun at the current moment.

[0038] Further, determining the solder requirement according to the vector influence index and the weld influence index includes:

[0039] The product of the vector influence index and the weld influence index is calculated, and the maximum and minimum values ​​are normalized as the solder requirement degree.

[0040] On the other hand, an automatic welding device for cabinet legs is proposed, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the automatic welding method for cabinet legs as described above are implemented.

[0041] The present invention has the following beneficial effects:

[0042] The present invention binds an image acquisition device to a welding gun so that they can move together, thereby analyzing image information, and further determining the moving state of the welding gun, and obtaining the width information of the weld from the image, so that the movement of the welding gun can be determined through image analysis; wherein, the detection movement vector is mainly determined by image changes at adjacent moments; and then the angle change stability index is analyzed according to the steering gear rotation angle, and the movement stability coefficient of the welding gun is determined, and the length of the time period for analysis is limited according to the movement stability coefficient to obtain the target time period. The acquisition of the target time period can avoid the influence of movement changes during the welding process and improve the reliability of the overall analysis; within the target time period, the detection movement vector and the weld width are combined to determine the wire feeding speed and welding current of the welding gun at the current moment, so as to facilitate cabinet foot welding based on adaptive wire feeding speed and welding current, improve the consistency of the overall welding process, and enhance the welding quality of the cabinet feet. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A flow chart of an automatic welding method for cabinet legs provided by one embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a cabinet foot provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of the structure of a welding device provided by an embodiment of the present invention;

[0047] Figure 4Schematic diagram of the welding arm structure provided by an embodiment of the present invention;

[0048] Figure 5 Schematic diagram of the structure of an automatic welding device for cabinet feet provided by an embodiment of the present invention;

[0049] The reference numerals in the figure are: 1 - base, 2 - positioner, 3 - shielding gas cylinder, 4 - smoke exhaust system, 5 - control cabinet, 6 - welding arm, 7 - servo motor, 8 - welding torch, 9 - image acquisition device. Detailed implementation manners

[0050] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following combines the accompanying drawings and preferred embodiments to detail the specific implementation manners, structures, features and effects of an automatic welding device and method for cabinet feet 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.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0052] The following specifically describes the specific solutions of an automatic welding device and method for cabinet feet provided by the present invention with reference to the accompanying drawings.

[0053] Please refer to Figure 1 , which shows a flowchart of an automatic welding method for cabinet feet provided by an embodiment of the present invention. The method includes:

[0054] S101: Obtain welding area images at different moments during the welding process based on the image acquisition device 9, and determine the detected movement vector of the welding torch 8 for two adjacent moments according to the movement of the edges in the welding area images of two adjacent moments.

[0055] It should be noted that the cabinet feet in the embodiments of the present invention are corresponding metal cabinet feet, and the shape is "I"-shaped. For details, please refer to Figure 2 , Figure 2 Schematic diagram of the cabinet feet provided by an embodiment of the present invention.

[0056] In the embodiments of the present invention, the cabinet feet are welded by an automatic welding device. Refer to Figure 3 , Figure 3 Schematic diagram of the structure of the welding device provided by an embodiment of the present invention. In Figure 3 , the base 1, positioner 2, shielding gas cylinder 3, smoke exhaust system 4 and control cabinet 5 are all common structures during the welding process, and will not be elaborated herein.

[0057] In the embodiment of the present invention, a micro welding gun 8 and an image acquisition device 9 are installed on the welding arm 6. The image acquisition device 9 moves together with the welding gun 8. The welding gun 8 has at least two different steering gears 7 to control the moving direction. The specific structural diagram of the welding arm 6 is shown in FIG. Figure 4 As shown, in the welding arm 6, the image acquisition device 9 moves together with the welding gun 8, so that the cabinet leg area to be welded can be more accurately and effectively acquired during the welding process of the welding gun 8.

[0058] The robot arm includes steering gears 7 in different directions. For example, three steering gears 7 with different directions can be provided to control the welding gun 8 to move in multiple angles and in all directions.

[0059] In the embodiment of the present invention, during the welding process, the welding area images at different times during the welding process are collected by the image acquisition device 9. Since the image acquisition device 9 moves with the welding gun 8, the welding area image also represents the image of the welding state of the welding gun 8 at the corresponding time.

[0060] The two adjacent moments represent the two sampling moments that are closest to each other. In the embodiment of the present invention, it can be set to sample the welding area image every 0.1 second.

[0061] It can be understood that the change in the welding area image at two adjacent moments represents the movement of the corresponding welding gun 8. The longer the welding gun 8 moves, the greater the change in the corresponding welding area image. Therefore, in the embodiment of the present invention, the movement characteristics of the welding gun 8 can be analyzed based on the change in the welding area image at two adjacent moments.

[0062] Therefore, the movement of the welding gun 8 can be accurately determined by analyzing the changes in the image, so there is no need to collect data in the movement dimension of the welding gun 8. Data analysis from another dimension requires multi-dimensional data matching, which will make the overall accuracy poor. Therefore, only performing welding gun 8 movement analysis based on image analysis can improve the accuracy of the overall analysis.

[0063] Furthermore, in some embodiments of the present invention, the detection movement vector of the welding gun 8 at two adjacent moments is determined according to the movement of the edge in the welding area image at two adjacent moments, including: performing edge detection on each frame of the welding area image to obtain the edge connected domain of each frame of the welding area image; performing chain code analysis on the edge connected domain to determine the chain code sequence of all edge connected domains in the two adjacent frames of the welding area image; performing similarity analysis on the chain code sequence of the edge connected domain in the two adjacent frames of the welding area image to determine the matching connected domain representing the same area in the two adjacent frames of the welding area image; and determining the detection movement vector of the welding gun 8 at the corresponding two moments according to the position change of the center point of each pair of matching connected domains in the welding area image.

[0064] The edge detection may specifically be, for example, Canny edge detection. Of course, in the embodiment of the present invention, various other edge detection methods may also be used, such as Sobel edge detection, without limitation thereto.

[0065] During welding, the area where the welding gun 8 moves will be welded, which will produce a large image change, while other areas only represent the movement characteristics of the welding gun 8 itself. Therefore, edge detection determines the edge connected domain and performs similarity analysis on the edge connected domain to determine the matching connected domain with higher similarity that represents the same object.

[0066] The similarity analysis in the embodiment of the present invention is specifically a chain code analysis, wherein the chain code may be specifically, for example, an eight-chain code, or, in other embodiments, may also be a four-chain code or a sixteen-chain code, without limitation thereto.

[0067] Since the welding gun 8 can be rotated to different angles for welding, and the image will also produce corresponding angle changes during rotation, conventional image matching will cause inaccurate matching due to the angle change, while chain code matching can avoid this influence and accurately obtain the matching connected domain representing the same area.

[0068] Furthermore, in some embodiments of the present invention, the detection movement vector of the welding gun 8 corresponding to the two moments is determined according to the position change of the center point of each pair of matching connected domains in the welding area image, including: in any pair of matching connected domains, the image position change of the corresponding center point from the previous welding area image to the next welding area image is used as the connected domain movement vector; and the connected domain movement vectors of all pairs of matching connected domains are averaged to obtain the detection movement vector.

[0069] Among them, since a pair of matching connected domains generated by chain code matching represent the same object, the time series changes of the center points in the same pair of matching connected domains are determined as the corresponding connected domain movement vectors, and the mean of all connected domain movement vectors is calculated as the detection movement vector. The detection movement vector can characterize the overall movement characteristics, that is, the movement characteristics of the welding gun 8.

[0070] S102: Determine the angle value of each servo 7 at different moments, and determine the angle change stability index of the corresponding servo 7 at the current moment according to the angle values ​​of the same servo 7 at all moments within a preset time range before the current moment; determine the movement stability coefficient of the welding gun 8 at the current moment by combining the angle change stability indexes of all servos 7 and the angle change correlation between any two servos 7.

[0071] Since the steering gear 7 rotates in different directions and there is also a stability requirement during the rotation process, the faster the rotation, the worse the corresponding stability. In the embodiment of the present invention, stability analysis is implemented for the change of the angle value of each steering gear 7.

[0072] The preset time range may be, for example, a time range of 2 seconds before the current moment. Of course, in other embodiments of the present invention, the preset time range may be adjusted according to actual detection requirements, and there is no limitation to this.

[0073] Further, in some embodiments of the present invention, an angle change stability index of the corresponding steering gear 7 at the current moment is determined based on the angle values ​​of the same steering gear 7 at all moments within a preset time range before the current moment, including: arranging the angle values ​​of the same steering gear 7 at all moments within the preset time range before the current moment in time sequence to obtain a steering gear angle change sequence; performing first-order difference processing on the steering gear angle change sequence to obtain a first-order difference value; determining the standard deviation of the first-order difference values ​​of the same steering gear 7 at all corresponding moments; and normalizing the inverse of the standard deviation as the angle change stability index of the corresponding steering gear 7 at the current moment.

[0074] Among them, the first-order difference processing is to make a difference between the latter parameter and the previous parameter in the corresponding steering gear angle change sequence, so as to obtain the corresponding first-order difference value. Since the parameters in the steering gear angle change sequence represent the angle values ​​of rotation at two adjacent moments, that is, assuming that at the first moment, the angle of the steering gear 7 is a, the angle of the steering gear 7 is b at the second moment, and the angle of the steering gear 7 is c at the third moment, then the corresponding angle value of the steering gear 7 rotation at the second moment is |ba|, the angle value of the steering gear 7 rotation at the third moment is |cb|, and so on. The corresponding first-order difference value is |cb|-|ba|, wherein || represents the absolute value.

[0075] In combination with the above content, the first-order difference value represents the value of the corresponding angle change speed. Under stable conditions, the values ​​of all first-order difference values ​​are closer to 0 and the overall change is smaller. In the embodiment of the present invention, the discreteness analysis of the first-order difference value is performed through the standard deviation.

[0076] Among them, the larger the standard deviation of the first-order difference values ​​at all times, the larger the change of the first-order difference values, that is, the more unstable the change of the angle of the steering gear 7 is. Therefore, in the embodiment of the present invention, the inverse of the standard deviation is normalized as the angle change stability index of the corresponding steering gear 7 at the current moment.

[0077] Of course, in other embodiments of the present invention, variance may also be used when performing discreteness analysis, and there is no limitation to this.

[0078] Among them, when the welding gun 8 moves along a straight line, the angles of multiple servos 7 change proportionally. At this time, not only is the angle change of a single servo 7 relatively stable, but the angles of multiple servos 7 also show a strong correlation. When presenting a relatively stable moving state, such as moving in a straight line, the corresponding servo 7 angle change will show certain correlation characteristics, and normal welding requires linear push welding, so this feature can be combined for stability analysis.

[0079] Further, in some embodiments of the present invention, the movement stability coefficient of the welding gun 8 at the current moment is determined in combination with the angle change stability indicators of all the servos 7 and the angle change correlation of any two servos 7, including: determining the angle change related indicators corresponding to the two servos 7 according to the servo angle change sequence of any two servos 7; taking the average of the angle change related indicators of all the servos 7 in pairs as the first stability coefficient of the welding gun 8 at the current moment; taking the average of the angle change stability indicators of all the servos 7 as the second stability coefficient of the welding gun 8 at the current moment; calculating the product of the first stability coefficient and the second stability coefficient and normalizing them to obtain the movement stability coefficient of the welding gun 8 at the current moment.

[0080] In the embodiment of the present invention, correlation analysis is performed by analyzing the change sequence of the steering gear angles between different multi-stages. Correlation analysis includes many methods, and the present invention specifically takes the Pearson correlation algorithm as an example.

[0081] Furthermore, in some embodiments of the present invention, according to the servo angle change sequences of any two servos 7, an angle change related index corresponding to the two servos 7 is determined, including: taking the absolute value of the Pearson correlation coefficient of the two servo angle change sequences as the angle change related index corresponding to the two servos 7.

[0082] Since the Pearson correlation coefficient calculated by the Pearson correlation algorithm takes values ​​of [-1, 1] in the value range, and the closer the value is to 0, the less correlation it indicates; the closer the value is to 1 or -1, the stronger the correlation is, therefore, the absolute value is taken, so that the larger the absolute value obtained is, the stronger the correlation between the servo angle change sequences of the two servos 7 is, that is, the larger the value of the angle change related index is, the stronger the corresponding correlation is.

[0083] Then, the angle change related indexes of all the two-by-two combinations of the steering gears 7 are comprehensively analyzed, that is, the average of all the two-by-two combinations of the angle change related indexes is used as the first stability coefficient. For example, if there are three steering gears 7, three groups of angle change related indexes are obtained by two-by-two combinations, and the corresponding average values ​​are used as the first stability coefficient.

[0084] The angle change stability index represents the stability characteristic of a corresponding steering gear 7 , and the average value of the angle change stability indexes of all steering gears 7 is taken as the second stability coefficient.

[0085] Therefore, the first stability coefficient and the second stability coefficient both characterize the stability characteristics of the movement of the steering gear 7. The product of the first stability coefficient and the second stability coefficient is calculated, and the maximum and minimum values ​​are normalized to obtain the movement stability coefficient of the welding gun 8 at the current moment.

[0086] S103: adjusting the preset time range according to the movement stability coefficient at all times within the preset time range to obtain a target time period.

[0087] Different moments have corresponding movement stability coefficients. By analyzing the movement stability coefficients of all moments, it is used to adjust to obtain a more stable time range, or to narrow the detection time range when it is unstable, so that it is closer to the movement characteristics of the current moment, thereby improving the accuracy of subsequent welding analysis and ensuring overall reliability.

[0088] It should be noted that during the overall welding process, since the welding gun 8 is moving and is controlled by the servo 7 and the corresponding mechanical arm, when the movement state changes, for example, when moving forward and then turning right, the turning point will cause the angle change of the servo 7 to be different from that when moving forward, and the Euclidean distance analysis during the corresponding speed calculation will also change due to the completion of the turn. At this time, if the analysis is still performed according to the time range corresponding to the turning point, the overall analysis will produce a large abnormal effect.

[0089] In the embodiment of the present invention, the preset time range is adjusted by moving the stability coefficient to determine the target time period, and then subsequent wire feeding analysis can be performed according to the target time period to avoid the operation turning point affecting the overall wire feeding effect.

[0090] Furthermore, in some embodiments of the present invention, the preset time range is adjusted according to the mobile stability coefficient at all moments in the preset time range to obtain a target time period, including: taking the average of the mobile stability coefficients at all moments as the time adjustment weight; calculating the product of the time adjustment weight and the length of the preset time range to obtain the target time length, and taking the time range of the target time length before the current moment as the target time period.

[0091] By calculating the time adjustment weight, the corresponding time influence is characterized, and then the preset time range is adjusted to obtain the target time period, which can improve the accuracy of the overall welding analysis.

[0092] S104: Determine the wire feeding speed and welding current of the welding gun 8 at the current moment according to the detected movement vectors of the welding gun 8 at all moments in the target time period and the width of the weld to be welded.

[0093] In the embodiment of the present invention, the welding is controlled so that the welding quality is kept consistent. The control effect of welding can be achieved through the wire feeding speed and the welding current. The wire feeding speed refers to the speed at which the wire feeder of the welding machine pushes the welding wire per minute during the welding process. The current is the magnitude of the output current of the welding machine. The wire feeding speed and the current are two important parameters in the welding process. The relationship between them will directly affect the welding effect and welding strength. Therefore, the wire feeding speed and the welding current are adjusted by the detection movement vector of the welding gun 8 at all times within the target time period and the width of the weld to be welded, so as to ensure the consistency of the overall welding effect.

[0094] Among them, the weld width is obtained by analyzing the welding area image obtained at the current moment, that is, the area of ​​the concave feature before the position of the welding gun 8 is the weld, and the Euclidean distance on both sides of the weld is the weld width.

[0095] Furthermore, in some embodiments of the present invention, the wire feeding speed and welding current of the welding gun 8 at the current moment are determined based on the detected movement vectors of the welding gun 8 at all moments in the target time period and the width of the weld to be welded, including: within the target time period, averaging the detected movement vectors of the welding gun 8 at all moments, and normalizing its module length as a vector influence index; normalizing the width of the weld to be welded as a weld influence index; determining the degree of solder demand based on the vector influence index and the weld influence index; determining the wire feeding speed and welding current of the welding gun 8 at the current moment based on the degree of solder demand.

[0096] Since the target time period represents a time period in which the welding gun 8 operates relatively stably, the detected movement vectors of the welding gun 8 at all moments in the target time period are vector averaged and normalized as a vector influence index.

[0097] The ratio of the modulus of the mean value of the detection movement vector of the welding gun 8 at all times within the target time period to the maximum modulus of the detection movement vector of the welding gun 8 is used as the corresponding vector influence index.

[0098] Similarly, normalizing the weld width means taking the ratio of the weld width to be welded to the maximum value of all weld widths as the weld influence index.

[0099] It can be understood that the larger the modulus of the detected moving vector is, the faster the corresponding moving speed of the welding gun 8 is, and the faster the welding wire is required to melt in order to achieve the ideal welding effect. The larger the weld width is, the more welding wire is required to melt. Therefore, in the embodiment of the present invention, the product of the vector influence index and the weld influence index is calculated, and the maximum and minimum values ​​are normalized as the solder demand degree.

[0100] The product of the solder demand and the preset maximum wire feeding speed is used as the wire feeding speed of the welding gun 8 at the current moment, and the product of the solder demand and the preset maximum welding current is used as the welding current of the welding gun 8 at the current moment.

[0101] Among them, the preset maximum wire feeding speed is the maximum wire feeding speed corresponding to the welding of the welding gun 8. Specifically, it can be, for example, 1 cm per second, and the maximum welding current is 360A. Of course, it can also be adjusted according to the actual detection situation, and there is no restriction on this.

[0102] By combining the welding conditions within the target time period and the weld width, the wire feeding speed and welding current in the subsequent welding process can be determined, which can make the overall welding process more consistent and improve the welding quality of the cabinet legs.

[0103] The present invention binds the image acquisition device 9 to the welding gun 8 so that they can move together, thereby analyzing the image information, and then determining the movement state of the welding gun 8, and obtaining the width information of the weld from the image, so that the movement of the welding gun 8 can be determined through image analysis; wherein, the image analysis mainly determines the detection movement vector through the image changes at adjacent moments; and then analyzes the angle change stability index according to the rotation angle of the servo 7, determines the movement stability coefficient of the welding gun 8, and limits the length of the analysis time period according to the movement stability coefficient to obtain the target time period. The acquisition of the target time period can avoid the influence of movement changes during the welding process and improve the reliability of the overall analysis; within the target time period, the detection movement vector and the weld width are combined to determine the wire feeding speed and welding current of the welding gun 8 at the current moment, so as to facilitate the cabinet foot welding based on the adaptive wire feeding speed and welding current, improve the consistency of the overall welding process, and enhance the welding quality of the cabinet foot.

[0104] The present invention also provides an automatic welding device for cabinet legs, see Figure 5 , Figure 5 A schematic structural diagram of an automatic cabinet leg welding device provided for one embodiment of the present invention, wherein the device 600 includes a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and executable on the processor 602, and when the processor 602 executes the computer program 603, the steps of the aforementioned automatic cabinet leg welding method are implemented.

[0105] It should be noted that the sequence of the above 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 accompanying 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.

[0106] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A cabinet foot automatic welding method, characterized in that: A micro welding gun and an image acquisition device are mounted on a welding arm, the image acquisition device moves together with the welding gun, and the welding gun has at least two different steering gears to control the moving direction. The method comprises: Based on the image acquisition device, the welding area images at different times during the welding process are acquired, and the detection movement vectors of the welding gun at two adjacent times are determined according to the movement of the edges in the welding area images at two adjacent times; Determine the angle value of each servo at different times, and determine the angle change stability index of the corresponding servo at the current moment according to the angle values ​​of the same servo at all times within a preset time range before the current moment; determine the movement stability coefficient of the welding gun at the current moment by combining the angle change stability indexes of all servos and the angle change correlation of any two servos; The preset time range is adjusted according to the movement stability coefficient at all times within the preset time range to obtain a target time period; According to the detected movement vector of the welding gun at all times within the target time period and the width of the weld to be welded, the wire feeding speed and welding current of the welding gun at the current moment are determined, wherein the concave feature area before the welding gun position in the welding area image at the current moment is the weld, and the Euclidean distance on both sides of the weld is the weld width; According to the angle values ​​of the same servo at all times within the preset time range before the current time, the angle change stability index of the corresponding servo at the current time is determined, including: Arrange the angle values ​​of the same servo at all times within a preset time range before the current time in time sequence to obtain a servo angle change sequence; Performing first-order difference processing on the steering gear angle change sequence to obtain a first-order difference value; Determine the standard deviation of the first-order difference values ​​of the same servo at all corresponding times; The inverse of the standard deviation is normalized as a stability index of the angle change of the corresponding steering gear at the current moment; Combine the angle change stability index of all servos and the angle change correlation of any two servos to determine the movement stability coefficient of the welding gun at the current moment, including: According to the servo angle change sequence of any two servos, determine the angle change related indicators of the two servos; The average of the angle change related indicators of all the servos in pairs is used as the first stability coefficient of the welding gun at the current moment; The average value of the angle change stability index of all servos is used as the second stability coefficient of the welding gun at the current moment; The product of the first stability coefficient and the second stability coefficient is calculated and normalized to obtain the movement stability coefficient of the welding gun at the current moment.

2. The automatic welding method for cabinet legs according to claim 1, characterized in that: According to the movement of the edge in the welding area image at two adjacent moments, the detection movement vector of the welding gun at two adjacent moments is determined, including: Perform edge detection on each frame of the welding area image to obtain the edge connected domain of each frame of the welding area image; Performing chain code analysis on the edge connected domain to determine the chain code sequence of all edge connected domains in two adjacent frames of welding area images; Perform similarity analysis on the chain code sequences of edge connected domains in two adjacent frames of welding area images to determine matching connected domains representing the same region in two adjacent frames of welding area images; According to the position change of the center point of each pair of the matching connected domains in the welding area image, the detection movement vector of the welding gun corresponding to the two moments is determined.

3. The automatic welding method for cabinet legs according to claim 2, characterized in that: According to the position change of the center point of each pair of the matching connected domains in the welding area image, the detection movement vector of the welding gun corresponding to two moments is determined, including: In any pair of matching connected domains, the image position change of the corresponding center point from the previous welding area image to the next welding area image is used as the connected domain movement vector; The connected domain motion vectors of all pairs of matched connected domains are averaged to obtain the detection motion vector.

4. The automatic welding method for cabinet legs according to claim 1, characterized in that: According to the servo angle change sequence of any two servos, the angle change related indicators of the corresponding two servos are determined, including: The absolute value of the Pearson correlation coefficient of the two servo angle change sequences is used as the angle change correlation index corresponding to the two servo angles.

5. The automatic welding method for cabinet legs according to claim 1, characterized in that: The preset time range is adjusted according to the movement stability coefficient at all moments in the preset time range to obtain a target time period, including: The average value of the mobile stability coefficient at all times is used as the time adjustment weight; The product of the time adjustment weight and the length of the preset time range is calculated to obtain a target time length, and the time range of the target time length before the current moment is used as the target time period.

6. The automatic welding method for cabinet legs according to claim 1, characterized in that: The method of determining the wire feeding speed and welding current of the welding gun at the current moment based on the detected movement vector of the welding gun at all moments in the target time period and the width of the weld to be welded comprises: In the target time period, the detection movement vectors of the welding gun at all times are averaged, and the module lengths thereof are normalized as the vector influence index; The width of the weld to be welded is normalized as a weld influence index; Determining the solder requirement according to the vector influence index and the weld influence index; The product of the solder demand degree and the preset maximum wire feeding speed is used as the wire feeding speed of the welding gun at the current moment, and the product of the solder demand degree and the preset maximum welding current is used as the welding current of the welding gun at the current moment.

7. The automatic welding method for cabinet legs according to claim 6, characterized in that: Determining the solder requirement according to the vector influence index and the weld influence index includes: The product of the vector influence index and the weld influence index is calculated, and the maximum and minimum values ​​are normalized as the solder requirement degree.

8. An automatic welding device for cabinet legs, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images