A robotic welding method for manually assisted positioning of embedded steel bars in a concrete box girder

Abstract

The invention discloses a robot welding method for manually assisted positioning of embedded steel bars in a concrete box girder, which comprises the following steps: the mobile welding equipment moves to the end of the box girder; input the diameter of the embedded steel bar, and automatically calculate the pixel value b1 occupied by the steel bar in the image corresponding to the calibrated height H0; adjust the pose of the first industrial camera at the end of the industrial robot, and when the steel bar to be welded is located at the center of the image, the industrial robot stops operating; adjust the height Z0 of the first industrial camera in the physical coordinate system so that the pixel value occupied by the steel bar to be welded in the image is b1; identify the welding start point and the welding end point according to the collected image; confirm whether the welding start point and the welding end point are correct. If they are correct, perform welding, otherwise, correct the welding start point and the welding end point; after the current trajectory welding is completed, drive the automatic mobile trolley to the next welding position, and return to the second step. Otherwise, the welding ends; the welding efficiency is high by using the invention.

Description

Technical Field

[0001] The present invention relates to the technical field of steel bar welding, in particular to a robotic welding method for manually assisted positioning of embedded steel bars in concrete box girders. Background Art

[0002] Bridges are important traffic facilities that connect different traffic routes, playing a key role in shortening travel time, improving traffic efficiency, and reducing traffic accidents, and are an important part of the modern traffic network. Common bridges can be divided into beam bridges, arch bridges, suspension bridges, cable-stayed bridges, etc. according to their structural forms; among them, the beam bridges constructed by splicing prestressed box girders are widely used in modern highway and railway bridges due to their advantages such as high structural strength, strong torsion resistance, small deflection, and long service life.

[0003] The box girder bridge mainly includes bridge piers and prestressed box girders. The piers are usually cast on the construction site according to the design requirements, while the prestressed box girders are prefabricated in the beam yard and then transported to the construction site to complete hoisting and splicing. During the manufacturing process of the prestressed box girders, embedded steel bars are pre-buried on both end faces and both sides of the upper surface. After the box girders are hoisted, the embedded steel bars at both ends are welded, and concrete is poured on site to achieve reliable connection of the prestressed box girders. The guardrail steel mesh is spliced on the embedded steel bars on both sides of the upper surface of the box girder, and concrete is poured on site to complete the construction of the bridge guardrail. Finally, the pavement paving, installation of lighting, drainage and other systems are completed.

[0004] However, the welding work of the embedded steel bars in the existing box girder splicing process is generally completed manually. A large number of professional technical workers are required at the construction site. Manual welding has low efficiency and extremely high costs, and the construction site environment is harsh, posing certain safety hazards. Secondly, the technical levels of the welding workers vary, and the welding quality cannot be guaranteed, thus posing a safety hazard for the later service of the bridge. Finally, with the aggravation of the aging population problem and the improvement of the social and economic level, fewer and fewer young people are willing to engage in welding work at the construction site, resulting in an increasingly severe labor shortage at the construction site. Therefore, the traditional manual welding method can no longer meet the needs of modern bridge construction. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0006] In view of the above and / or existing problems in the welding of embedded steel bars, the present invention is proposed.

[0007] Therefore, the object of the present invention is to provide a robotic welding method for the artificial auxiliary positioning of embedded steel bars in concrete box girders. The present invention uses a combination of an industrial robot and a welding gun to replace manual welding of the embedded steel bars in the box girder, reducing the technical level requirements of the operators and improving the welding efficiency.

[0008] To solve the above technical problems, the present invention provides the following technical solution: A robotic welding method for the artificial auxiliary positioning of embedded steel bars in concrete box girders. The mobile welding equipment used includes an automatic mobile vehicle. An operation module, a control module, and a human-machine interface are provided in the cab of the automatic mobile vehicle. The front end of the frame of the automatic mobile vehicle is fixedly connected with an industrial robot. The execution end of the industrial robot is fixedly connected with a base. A first industrial camera for image acquisition and a welding gun for welding steel bars are fixedly connected to the lower end of the base. The first industrial camera, the human-machine interface, and the operation module are respectively electrically connected to the control module. The operation module and the human-machine interface output control signals to the control module, and the industrial robot and the automatic mobile vehicle are controlled by the control module to act. The first industrial camera collects images of the embedded steel bars and sends the collected images to the control module. The control module identifies the welding start point and the welding end point according to the received images of the embedded steel bars, calculates the welding trajectory, and the operator assists in judging whether the welding trajectory is accurate and sends a welding instruction. The control system adjusts the position of the execution end according to the control signal to adjust the poses of the welding gun and the first industrial camera. When welding the embedded steel bars on the upper end surface, the following steps are included.

[0009] S1 The operator controls the mobile welding equipment to move to the end of the box girder, so that the forward direction of the chassis of the automatic mobile vehicle is parallel to the end surface of the box girder.

[0010] S2 The operator inputs the diameter B1 of the embedded steel bar to be welded in the human-machine interface, and the control module automatically calculates the pixel value b1 occupied by the steel bar in the image corresponding to the calibration height H0 and the spacing D of the embedded steel bars.

[0011] S3 The operator adjusts the pose of the first industrial camera at the end of the industrial robot through the operation module, observes the image of the embedded steel bars on the bridge displayed on the human-machine interface, and when the steel bar to be welded is in the center of the image, the industrial robot stops moving.

[0012] S4 The operator controls the industrial robot through the operation module to adjust the height Z0 of the first industrial camera in the physical coordinate system, so that the pixel value occupied by the steel bar to be welded in the image is b1, and the control module calculates the height value Z of the position of the steel bar to be welded in the industrial robot coordinate system.

[0013] S5 The control module identifies the welding start point and the welding end point according to the collected images and displays the welding points on the human-machine interface.

[0014] The operator of S6 confirms whether the welding start point and the welding end point are correct. If correct, the operator sends a start welding instruction signal to the control module through the operation module. Otherwise, after correcting the welding start point and the welding end point through the operation module, the start welding instruction is sent again. After confirmation, the control module controls the industrial robot to move so that the welding gun completes the welding according to the preset welding trajectory;

[0015] After the welding of the current trajectory is completed, the industrial robot automatically lifts so that the coordinate Z of the welding gun is 0, and the automatic moving trolley automatically advances a distance D of the embedded steel bar to the next welding position, and returns to step S2 until the trolley moves to the other end of the bridge, the welding operation ends, the operator stops the welding task, and drives the mobile welding equipment away from the construction site;

[0016] When welding the embedded steel bars on the lower end face, the following steps are included,

[0017] Install the welding gun extension arm. A welding gun is fixedly connected to the end of the extension arm far away from the industrial robot. A second industrial camera is connected to the extension arm above the welding gun. The image of the embedded steel bars on the lower end face of the bridge is obtained through the second industrial camera, the welding points are calibrated and identified, the first industrial camera obtains the image of the embedded steel bars on the upper end face, and the operator controls the industrial robot through the operation module to insert the welding gun along the gap between adjacent groups of embedded steel bars on the upper end face;

[0018] After each welding is completed, the operator controls the industrial robot through the operation module to withdraw the welding gun extension arm from the current gap between the embedded steel bars. The operator drives the automatic moving trolley to the next welding position, observes the gap between the embedded steel bars through the image collected by the first industrial camera, and inserts it into the gap between adjacent embedded steel bars, and sequentially completes the welding of all the embedded steel bars on the lower end face.

[0019] As a preferred scheme of the robot welding method for the artificial auxiliary positioning of the embedded steel bars in the concrete box girder of the present invention, wherein: the welding points automatically identified by the system need to be confirmed by the operator whether they are correct. When the identified points are incorrect, the operator needs to modify the welding points through the operation module. When confirming, a confirmation signal is input through the welding operation module, and the operation module sends the confirmation signal to the control module. After receiving the confirmation signal sent by the operation module, the control module controls the industrial robot to move so that the welding gun completes the welding according to the preset welding program.

[0020] As a preferred scheme of the robot welding method for the artificial auxiliary positioning of the embedded steel bars in the concrete box girder of the present invention, wherein: before identifying the welding start point and the welding end point, the image collected by the first industrial camera is processed first, including the following steps,

[0021] The acquired image is an RGB color image. To simplify the model, reduce the computational load, and improve the computational speed, the image is processed using Equation (1) to obtain the grayscale image corresponding to the original image.

[0022] Gray(x,y) = 0.299R(x,y) + 0.587G(x,y) + 0.144B(x,y) (1);

[0023] Where (x,y) represents the coordinates of the pixel point in the image, Gray(x,y) represents the grayscale value at (x,y) in the transformed image, and R(x,y), G(x,y), B(x,y) represent the three primary color values at (x,y) in the original image respectively;

[0024] The original image is processed using the filtering function shown in Equation (2).

[0025]

[0026] Where σ is the standard deviation of the filter, and G(x,y) represents the grayscale value at (x,y) in the filtered image;

[0027] The original image is converted into a binary image using Equation (3).

[0028]

[0029] Where T is the threshold value, and g(x,y) is the grayscale value at (x,y) in the segmented image;

[0030] The obtained binary image is processed using Equation (4).

[0031]

[0032] Where A represents the binary image, and B is a 3×3 matrix;

[0033] The image obtained in the previous step is processed using Equation (5) to restore the contour shape of the target in the image.

[0034]

[0035] As a preferred solution of the robotic welding method for the embedded steel bars in the concrete box girder in the present invention, specifically: determining the contours of all features in the target, which specifically includes the following steps.

[0036] Calculate the gradients in two directions of the binary image using Equation (6).

[0037]

[0038] Where G x , G yrespectively represent the gradients in the x and y directions of the binary image;

[0039] and calculate the magnitude and direction of the image gradient, as shown in formula (7):

[0040]

[0041] Determine the contours of all features in the image according to the gradient value and gradient direction.

[0042] As a preferred solution of the robot welding method for the embedded steel bars in the concrete box girder with artificial auxiliary positioning in the present invention, wherein: when identifying the welding points, it includes the following steps,

[0043] Use formula (8) to calculate the matching value between the template image and all the contours in the captured image in the system, and take the contours with r(x, y) < 0.15 (i.e., the similarity is 85%) as the target contours of the welding points. Determine the starting point and ending point of welding according to the template predefined. Among them, the point at the far end of the automatic moving vehicle in the contour is the starting point P1(x1, y1) of welding, and the point closer to the automatic moving vehicle is the ending point P2(x2, y2) of welding;

[0044]

[0045] In the formula, r(x, y) represents the square difference between the image contour area and the template image area, T(i, j) is the pixel value of the template image at (i, j), G(x + i, y + j) represents the pixel value at (x + i, y + j) in the grayscale image, and i, j represent the index variables in the template image T;

[0046] Welding trajectory calculation,

[0047] Since the welding trajectories are all straight lines, use the linear interpolation method to determine the trajectory equation between the welding points P1 and P2,

[0048] AX + BY + C = 0 (9);

[0049] In the formula, X and Y represent the coordinates of points in the industrial robot coordinate system, and A, B, and C respectively represent the coefficients of the linear equation of the welding trajectory, which can be calculated by substituting the coordinates of points P1 and P2.

[0050] As a preferred solution of the robot welding method for the embedded steel bars in the concrete box girder with artificial auxiliary positioning in the present invention, wherein: between step S1 and step S2, it further includes setting the industrial robot coordinate system and parameter setting, specifically,

[0051] Take the installation center position of the industrial robot on the upper surface of the mobile chassis as the coordinate origin of the mobile welding equipment and the industrial robot. The coordinate system is a Cartesian coordinate system, denoted as O(X, Y, Z). Denote the pixel coordinates of the first industrial camera as P(x, y). The absolute position of the first industrial camera in the physical coordinate system O(X, Y, Z) of the robot is determined by the design installation parameters of the first industrial camera and the motion parameters of the industrial robot, denoted as (X0, Y0, Z0).

[0052] There is a linear relationship between the pixel size of the pixel coordinate system of the first industrial camera and the physical size of the physical coordinate system, as shown in formula (10).

[0053] B = K·b (10);

[0054] Where B is the designed size of the calibration block in the physical coordinate system, b is the pixel value occupied by the calibration block in the pixel coordinate system, and K is the mapping coefficient between the pixel coordinate system and the physical coordinate system;

[0055] The specific calibration method is as follows: Fix the first industrial camera at Z = H0, place the calibration block with a width of B on the XOY plane, that is, Z = 0, and read the pixel value b occupied by the calibration block in the image to calculate the coefficient K. In actual use, given the actual width B1 of the object to be measured, the pixel value b1 occupied by the object to be photographed in the image can be obtained from formula (11).

[0056]

[0057] Control the industrial robot to move the absolute position of the first industrial camera in the physical coordinate system so that the pixel occupied by the target object in the image is b1, and record the position (X0, Y0, Z0) of the first industrial camera in the industrial robot coordinate system of the bridge embedded reinforcement. At this time, the coordinate value of the identified target object in the physical coordinate system can be obtained:

[0058]

[0059] At the same time, the spacing D between adjacent embedded reinforcements can be obtained as:

[0060] D = Kd (13);

[0061] d is the pixel distance.

[0062] Compared with the prior art, the present invention has the following technical effects: The combination of the industrial robot and the welding gun is used to replace manual labor to complete the welding of the embedded steel bars of the box girder. At the same time, through the welding method of the present invention, the steel bars are automatically identified, and the automatic welding of the upper-end embedded steel bars and the lower-end embedded steel bars is completed, reducing the technical level requirements of the operators, improving the welding efficiency, and reducing the welding labor cost; The operator completes the operation of the industrial robot in the cab, avoiding the potential safety hazards of manual welding and reducing the working intensity of the workers. Brief Description of the Drawings

[0063] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0064] Figure 1 It is a side view of the mobile welding equipment when welding the upper-end embedded steel bars in the present invention.

[0065] Figure 2 It is a side view of the mobile welding equipment when welding the lower-end embedded steel bars in the present invention.

[0066] Figure 3 It is a structural diagram when using the mobile welding equipment to weld the embedded welding at the upper end face.

[0067] Figure 4 It is a structural diagram when using the mobile welding equipment to weld the embedded welding at the lower end face.

[0068] Figure 5 It is a top view when welding the embedded steel bars.

[0069] Figure 6 It is a schematic diagram when the first industrial camera in the present invention collects the image of the embedded steel bars.

[0070] Figure 7 It is a schematic diagram of the image template in the welding area of the present invention.

[0071] Figure 8 It is a schematic diagram of the image parameter calibration in the present invention.

[0072] In the figure, 100 is the automatic moving vehicle, 101 is the cab, 102 is the vehicle frame, 200 is the automatic wire feeder, 300 is the industrial robot, 400 is the welding gun, 500 is the first industrial camera, 600 is the base, 700 is the second industrial camera, 800 is the mounting bracket, 900 is the extension arm, 1000 is the pier, 2000 is the embedded steel bar, and 3000 is the box girder. Detailed Embodiments

[0073] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will give a detailed description of the specific embodiments of the present invention in conjunction with the drawings in the specification.

[0074] In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Persons skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0075] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive of other embodiments.

[0076] Embodiment 1

[0077] Referring to Figures 1 to 3 , this embodiment provides a robotic welding method for manually assisted positioning of embedded steel bars in a concrete box girder, which can further realize the welding of embedded steel bars at the upper and lower end faces of the box girder, with high welding efficiency, low labor cost, and low labor intensity.

[0078] A robotic welding method for manually assisted positioning of embedded steel bars in a concrete box girder, the mobile welding equipment used includes an automatic mobile vehicle. An operation module, a control module, and a human-machine interface are provided in the cab of the automatic mobile vehicle. The front end of the frame of the automatic mobile vehicle is fixedly connected with an industrial robot. The execution end of the industrial robot is fixedly connected with a base. A first industrial camera for image acquisition and a welding gun for welding steel bars are fixedly connected to the lower end of the base. The first industrial camera, the human-machine interface, and the operation module are respectively electrically connected to the control module. The operation module and the human-machine interface output control signals to the control module. The industrial robot is controlled to act through the control module. The first industrial camera acquires the image of the embedded steel bar and sends the acquired image to the control module. The control module identifies the welding start point and the welding end point based on the received image of the embedded steel bar, calculates the welding trajectory, and displays it through the human-machine interface. The operator assists in judging whether the welding trajectory is accurate and sends a welding instruction. The control system adjusts the position of the execution end according to the control signal to adjust the poses of the welding gun and the first industrial camera; the automatic mobile vehicle, the industrial robot, and the wire feeder are all prior arts. Adjusting the pose of the execution end of the industrial robot by using the control module and controlling the automatic wire feeder to automatically feed wire to the welding gun through the control module are both conventional techniques and not the improvement points of this application; referring to Figure 1 , a number of box girders are fixed on the upper ends of a number of bridge piers. A number of groups of embedded steel bars are respectively provided on the upper and lower end faces between adjacent two box girders. The embedded steel bars at the upper end face and the embedded steel bars at the lower end face are arranged staggeredly in the width direction. When welding the embedded steel bars at the upper end face, the following steps are included,

[0079] The operator controls the mobile welding equipment to move to the end of the box girder, so that the advancing direction of the chassis of the automatic mobile vehicle is parallel to the end face of the box girder;

[0080] The operator inputs the diameter of the embedded steel bar to be welded on the human-machine interface, and the control module automatically calculates the pixel value b1 of the steel bar in the corresponding image of the calibration height H0 and the spacing D of the embedded steel bars;

[0081] The operator adjusts the pose of the first industrial camera at the end of the industrial robot through the operation module, observes the image of the bridge embedded steel bar displayed on the human-machine interface, and when the steel bar to be welded is at the center of the image, the industrial robot stops moving;

[0082] The operator controls the industrial robot through the control operation module to adjust the height Z0 of the first industrial camera in the physical coordinate system, so that the pixel value of the steel bar to be welded in the image is b1, and the control module calculates the height value Z of the position of the steel bar to be welded in the industrial robot coordinate system;

[0083] The control module identifies the welding start point and the welding end point according to the collected image and displays the welding points on the human-machine interface;

[0084] The operator confirms whether the welding start point and the welding end point are correct. If correct, the operator sends the start welding command signal to the control module through the operation module. Otherwise, after correcting the welding start point and the welding end point through the operation module, the start welding command is sent again. After confirmation, the control module controls the industrial robot to move so that the welding gun completes the welding according to the preset welding trajectory;

[0085] After the welding of the current trajectory is completed, the industrial robot automatically lifts so that the coordinate Z of the welding gun is 0, and the automatic mobile trolley automatically advances a distance D of the embedded steel bar to the next welding position, and returns to step S2 until the trolley moves to the other end of the bridge, the welding operation ends, the operator stops the welding task, and drives the mobile welding equipment away from the construction site;

[0086] When welding the embedded steel bars on the upper end face, the following steps are included,

[0087] Install the welding gun extension arm. The extension arm is fixedly connected to the end of the base. A welding gun is fixedly connected to the end of the extension arm away from the base, so that the welding gun can be inserted along the gap between adjacent groups of embedded steel bars on the upper end face. An installation bracket is fixedly connected to the end of the extension arm, and a second industrial camera is fixedly connected to the installation bracket. The second industrial camera is electrically connected to the control module. The image of the bridge embedded steel bars on the lower end face is obtained through the second industrial camera, the welding points are calibrated and identified, the image of the embedded steel bars on the upper end face is obtained by the first industrial camera, and the operator controls the industrial robot through the operation module to insert the welding gun along the gap between adjacent groups of embedded steel bars on the upper end face;

[0088] The specific calibration and recognition process is the same as that of the embedded steel bars on the upper welding surface, which will not be elaborated here. After each welding is completed, the operator controls the industrial robot through the operation module to withdraw the welding gun extension arm from the current gap between the embedded steel bars. The operator drives the automatic mobile trolley to the next welding position. The operator observes the gap between the embedded steel bars through the images collected by the first industrial camera and inserts it into the adjacent gap between the embedded steel bars, and sequentially completes the welding of all the embedded steel bars on the lower end surface.

[0089] The welding points automatically recognized by the system need to be confirmed by the operator whether they are correct. When the recognized points are incorrect, the operator needs to modify the welding points through the operation module. When confirming, a confirmation signal is input through the welding operation module, and the operation module sends the confirmation signal to the control module. After receiving the confirmation signal sent by the operation module, the control module controls the industrial robot to act so that the welding gun completes the welding according to the preset welding procedure.

[0090] The recognition result is manually confirmed. The operator adjusts and confirms the recognition result obtained by the template matching method, then calculates the welding trajectory, and executes the subsequent welding procedure. Due to the complex lighting conditions and background colors at the construction site, and many rust points on the embedded steel bars, there are phenomena of missing targets and recognition deviations in the automatic recognition algorithm. The operator's manual confirmation based on the captured images can eliminate the recognition errors and mistakes of the automatic recognition algorithm and ensure that all the embedded steel bars are welded.

[0091] The operation module includes a control joystick, a trackball, a confirmation button, and a mode switching button. The control joystick, trackball, confirmation button, and mode switching button are respectively electrically connected to the control module. In step S6, if it is correct, the operator presses the confirmation button, and the confirmation button transmits the confirmation signal to the control module. If it is incorrect, the operator corrects the welding point through the trackball. After the correction is completed, the operator presses the confirmation button. After receiving the signal indicating the end of the correction from the confirmation button, the control module sends a welding instruction; the control module is connected to the driving system of the automatic mobile trolley (mainly electrically connected to the driving motor that drives the automatic mobile trolley forward and the steering motor that adjusts the steering of the automatic mobile trolley. This is a prior art, and the specific implementation structure will not be elaborated). The operation mode is switched through the mode switching button, and the mode switching button sends the operation mode signal to the control module. When switching to the trolley control mode through the mode switching button, the control joystick is used to adjust the position and transfer process of the automatic mobile trolley, and the control joystick sends the adjustment signal to the control module, and the control module controls the movement and steering of the automatic mobile trolley; when switching to the industrial robot control mode through the mode switching button, the control joystick sends the signal for controlling the actions of the industrial robot to the control module, and the control module adjusts the X and Y coordinates of the end of the industrial robot in the robot coordinate system according to the received signal, that is, controls the pose of the industrial robot through the control joystick to realize the position adjustment of the industrial robot during calibration and lower end surface welding.

[0092] The present invention uses an industrial robot to replace manual labor to complete the welding of embedded steel bars in box girders, reducing the technical level requirements of operators, ensuring welding quality, improving welding efficiency, avoiding potential safety hazards of manual welding, and reducing welding labor costs and work intensity.

[0093] Embodiment 2

[0094] Refer to Figures 4 to 8 , this embodiment is based on Embodiment 1. The difference from Embodiment 1 is that this embodiment provides a robotic welding method for manual auxiliary positioning of embedded steel bars in concrete box girders, which can...

[0095] Specifically, before identifying the welding start point and the welding end point, the image collected by the first industrial camera is processed first, including the following steps:

[0096] The collected image is an RGB trichromatic image. To simplify the model, reduce the calculation amount, and improve the calculation rate, the formula (1) is used to process the image to obtain the grayscale image corresponding to the original image.

[0097] Gray(x, y) = 0.299R(x, y) + 0.587G(x, y) + 0.144B(x, y) (1);

[0098] In the formula, (x, y) represents the coordinates of the pixel point in the image, Gray(x, y) represents the grayscale value at (x, y) in the transformed image, and R(x, y), G(x, y), B(x, y) respectively represent the trichromatic values at (x, y) in the original image.

[0099] To reduce the noise and details of the original image and improve the quality and clarity of the image, the filtering function shown in formula (2) is used to process the original image.

[0100]

[0101] In the formula, σ is the standard deviation of the filter, and G(x, y) represents the grayscale value at (x, y) in the filtered image.

[0102] The preprocessed image is subjected to threshold segmentation processing to separate the target from the background in the image, mainly including the following steps: The original image is converted into a binary image using formula (3).

[0103]

[0104] In the formula, T is the set grayscale threshold, and g(x, y) is the grayscale value at (x, y) in the segmented image.

[0105] The obtained binary image is processed using formula (4).

[0106]

[0107] In the formula, A represents a binary image, B is a 3-order matrix with all its elements being 1, s is the increment of the image pixel in the x direction, t is the increment in the y direction, and min (x，y)∈B {A(x + s, y + t)} represents the minimum value of each pixel point after the matrix B performs convolution calculation on the image A;

[0108] Use formula (5) to process the image obtained in the previous step to restore the contour shape of the target in the image.

[0109]

[0110] In the formula, max (x，y)∈B {A(x - s, y - t)} represents the maximum value of each pixel point after the matrix B performs convolution calculation on the image A;

[0111] By processing the image, the uneven and discontinuous phenomena of the target contour caused by illumination, steel bar rust, and background color can be eliminated; after the above steps of processing, a relatively regular target contour can be obtained, improving the accuracy of subsequent template matching recognition.

[0112] Determine the contours of all features in the target, specifically including the following steps.

[0113] Use formula (6) to calculate the gradients in two directions of the binary image:

[0114]

[0115] In the formula, G x , G y respectively represent the gradients of the binary image in the x and y directions;

[0116] And calculate the magnitude and direction of the image gradient, as shown in formula (7):

[0117]

[0118] In the formula, M is the gradient value of the image, and θ is the direction angle of the image gradient;

[0119] Determine the contours of all features in the image according to the gradient value and gradient direction;

[0120] Identify the welding points, including the following steps.

[0121] The matching values between the template image in the control module and all the contours in the captured image are calculated using Equation (8). The contours with r(x, y) < 0.15 (i.e., similarity of 85%) are taken as the target contours of the welding points. The starting and ending points of welding are determined according to the template pre - definition. Among them, the point on the contour farthest from the automatic mobile vehicle is the starting point P1(x1, y1) of welding, and the point closer to the automatic mobile vehicle is the ending point P2(x2, y2) of welding;

[0122]

[0123] In the formula, r(x, y) represents the mean square error between the image contour area and the template image area, T(i, j) is the pixel value of the template image at (i, j), G(x + i, y + j) represents the pixel value at (x + i, y + j) in the grayscale image, and i, j represent the index variables in the template image T;

[0124] Welding trajectory calculation,

[0125] Since the welding trajectories are all straight lines, a linear interpolation method is used to determine the trajectory equation between the welding points P1 and P2, and the coefficients of the linear equation are calculated from the coordinates of points P1 and P2,

[0126] Ax + By + C = 0 (9)

[0127] In the formula, A, B, and C respectively represent the coefficients of the welding trajectory in the robot coordinate system. The robot coordinate system is as Figure 1 shown, and the Y - axis is the direction passing through the coordinate origin and perpendicular to the paper surface.

[0128] Between step S1 and step S2, it also includes setting the robot coordinate system and parameter settings. Specifically,

[0129] Take the industrial robot installation center position on the upper surface of the mobile chassis as the origin of the industrial robot coordinate system. The coordinate system is a Cartesian coordinate system, denoted as O(X, Y, Z). Denote the pixel coordinates of the first industrial camera as P(x, y). The absolute position of the first industrial camera in the physical coordinate system O(X, Y, Z) of the industrial robot is determined by the design and installation parameters of the first industrial camera and the motion parameters of the industrial robot, denoted as (X0, Y0, Z0);

[0130] There is a linear relationship between the pixel size of the pixel coordinate system of the first industrial camera image and the physical size of the physical coordinate system, as shown in Equation (10),

[0131] B = K·b (10);

[0132] Among them, B is the design size of the calibration block (a square iron block placed on the automatic mobile vehicle) in the physical coordinate system, b is the pixel value occupied by the calibration block in the pixel coordinate system, and K is the mapping coefficient between the pixel coordinate system and the physical coordinate system;

[0133] The specific calibration method is to fix the first industrial camera at Z = H0 and place the calibration block with a width of B on the XOY plane, that is, Z = 0. The coefficient K can be calculated by reading the pixel value b occupied by the calibration block in the image. In actual use, if the actual width B1 of the object to be measured is known, the pixel value b1 occupied by the photographed object in the image can be calculated by formula (11),

[0134]

[0135] Control the industrial robot to move the absolute position of the first industrial camera in the physical coordinate system so that the pixel occupied by the target object in the image is b1, and record the position of the first industrial camera in the bridge pre-embedded reinforcement industrial robot coordinate system (X0, Y0, Z0). At this time, the coordinate value of the identified target object in the physical coordinate system can be obtained.

[0136]

[0137] At the same time, the spacing D between adjacent embedded bars can be obtained as:

[0138] D = Kd (13);

[0139] d is the pixel distance.

[0140] The images taken by industrial cameras are two-dimensional images without three-dimensional spatial information, and the specific height of the target in the robot coordinate system cannot be determined. The calibration method proposed in this application uses the actual size of the target object to establish the spatial relationship between the embedded steel bars and the robot by manually calibrating the first welding point, which facilitates the control system to obtain the height information of the target object in the robot coordinate system, and then determine the three-dimensional coordinate information of the welding trajectory, providing a basis for the welding robot to perform welding tasks.

[0141] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A robotic welding method for manually assisted positioning of embedded steel bars in a concrete box girder, characterized in that: The mobile welding equipment used includes an automatic mobile vehicle. Inside the cab of the automatic mobile vehicle, there is an operation module, a control module, and a human-machine interface. At the front end of the frame of the automatic mobile vehicle, an industrial robot is fixedly connected. At the execution end of the industrial robot, a base is fixedly connected. At the lower end of the base, a first industrial camera for image acquisition and a welding gun for welding steel bars are fixedly connected. The first industrial camera, the human-machine interface, and the operation module are respectively electrically connected to the control module. The operation module and the human-machine interface output control signals to the control module, and the control module controls the actions of the industrial robot. The first industrial camera collects images of the embedded steel bars and sends the collected images to the control module. The control module identifies the welding start point and the welding end point based on the received images of the embedded steel bars, calculates the welding trajectory, and the operator assists in judging whether the welding trajectory is accurate and sends a welding instruction. The control system adjusts the position of the execution end according to the control signal to adjust the poses of the welding gun and the first industrial camera. When welding the upper-end embedded steel bars, the following steps are included, S1 The operator controls the mobile welding equipment to move to the end of the box girder, so that the forward direction of the chassis of the automatic mobile vehicle is parallel to the end face of the box girder; S2 The operator inputs the diameter B1 of the embedded steel bar to be welded in the human-machine interface, and the control module automatically calculates the pixel value b1 occupied by the steel bar in the image corresponding to the calibration height H0; S3 Adjust the pose of the first industrial camera at the end of the industrial robot through the operation module, and observe the image of the bridge embedded steel bar displayed in the human-machine interface. When the steel bar to be welded is at the center of the image, the industrial robot stops moving; S4 The operator controls the industrial robot through the operation module to adjust the height Z0 of the first industrial camera in the physical coordinate system, so that the pixel value occupied by the steel bar to be welded in the image is b1, and at the same time measures the pixel spacing d between adjacent embedded steel bars. The control module calculates the height value Z of the position of the steel bar to be welded in the industrial robot coordinate system and the spacing D between adjacent embedded steel bars; S5 The control module identifies the welding start point and the welding end point based on the collected images and displays the welding points on the human-machine interface; S6 The operator confirms whether the welding start point and the welding end point are correct. If correct, the operator sends a start welding instruction signal to the control module through the operation module. Otherwise, after correcting the welding start point and the welding end point through the operation module, send the start welding instruction again. After confirmation, the control module controls the industrial robot to act so that the welding gun completes the welding according to the preset welding trajectory; S7 After the current trajectory welding is completed, the industrial robot automatically lifts so that the coordinate of the welding gun Z = 0, and the automatic mobile trolley automatically advances a distance D of one embedded steel bar to the next welding position, and returns to step S2 until the trolley moves to the other end of the bridge, the welding operation ends, the operator stops the welding task, and drives the mobile welding equipment away from the construction site; When welding the lower-end embedded steel bars, the following steps are included, Install a welding gun extension arm. A welding gun is fixedly connected to one end of the extension arm away from the industrial robot. A second industrial camera is connected to the extension arm above the welding gun. The image of the embedded steel bars on the lower end face of the bridge is obtained through the second industrial camera to calibrate and identify the welding points. The first industrial camera obtains the image of the embedded steel bars on the upper end face. The operator controls the industrial robot through the operation module to insert the welding gun along the gap between adjacent groups of embedded steel bars on the upper end face; After each welding is completed, the operator controls the industrial robot through the operation module to withdraw the welding gun extension arm from the current gap between the embedded steel bars. The operator drives the automatic mobile trolley to the next welding position. The operator observes the gap between the embedded steel bars through the image collected by the first industrial camera and inserts it into the gap between adjacent embedded steel bars to complete the welding of all the embedded steel bars on the lower end face in sequence.

2. The robotic welding method for manual assisted positioning of embedded steel bars in a concrete box girder according to claim 1, characterized in that: The welding points automatically identified by the system need to be confirmed by the operator whether they are correct. When the identified points are incorrect, the operator needs to modify the welding points through the operation module. When confirming, a confirmation signal is input through the welding operation module. The operation module sends the confirmation signal to the control module. After receiving the confirmation signal sent by the operation module, the control module controls the industrial robot to act so that the welding gun completes the welding according to the preset welding procedure.

3. The robot welding method for artificial auxiliary positioning of embedded steel bars in a concrete box girder as claimed in claim 1, wherein: Between step S1 and step S2, it further includes setting the coordinate system and parameters of the industrial robot. Specifically, Take the industrial robot installation center position on the upper surface of the mobile chassis as the coordinate origin of the mobile welding equipment and the industrial robot. The coordinate system is a Cartesian coordinate system, denoted as , and record the pixel coordinates of the first industrial camera as . The absolute position of the first industrial camera in the physical coordinate system of the robot is determined by the design and installation parameters of the first industrial camera and the motion parameters of the industrial robot, denoted as ; The pixel size of the pixel coordinate system of the first industrial camera image and the physical size of the physical coordinate system are in a linear relationship, as shown in formula (10), （10）； where B is the designed size of the calibration block in the physical coordinate system, b is the number of pixels occupied by the calibration block in the pixel coordinate system, and K is the mapping coefficient between the pixel coordinate system and the physical coordinate system; The specific calibration method is to fix the first industrial camera at Place the calibration block with width B on the XOY plane, that is Read the pixel value b occupied by the calibration block in the image to calculate the coefficient K. In actual use, when the actual width of the object to be measured is known Then the pixel value occupied by the photographed target object in the image can be obtained from formula (11). （11）； Control the industrial robot to move the first industrial camera to the absolute position in the physical coordinate system so that the number of pixels occupied by the target object in the image is , and record the position of the first industrial camera in the industrial robot coordinate system of the bridge embedded bar . At this time, the coordinate value of the recognized target object in the physical coordinate system can be obtained: （12）； At the same time, the distance D between adjacent embedded steel bars can be obtained as: （13）； d is the pixel distance.

Images