WO2023202031A1 - Welding method and apparatus, and electronic device and computer-readable storage medium - Google Patents

Abstract

A welding method and apparatus, and an electronic device (61) and a computer-readable storage medium. The welding method is applied to an electronic device (61), which is included in a welding system. The welding system further comprises a robot (62) and a visual sensor (1, 63) fixed onto the robot (62). The method comprises: controlling a visual sensor (1, 63) to acquire a target image corresponding to a position to be welded; according to the target image, acquiring target starting point coordinates and target end point coordinates which correspond to the position to be welded; and controlling the tail end of a robot (62) to perform welding from the target starting point coordinates to the target end point coordinates. The visual sensor (1, 63) is controlled to acquire the target image at the position to be welded, such that the target starting point coordinates and the target end point coordinates can be accurately determined according to the target image, thereby accurately identifying the position to be welded. The tail end of the robot (62) is then controlled to perform welding from the target starting point coordinates to the target end point coordinates, such that the position to be welded can be welded, thereby ensuring the welding precision and the welding quality.

Description

Welding methods, devices, electronic equipment and computer-readable storage media

This application claims priority to the Chinese patent application with application number 202210439273.0 and the application name "Aluminum Alloy Automatic Arc Welding Method and Device" submitted on April 22, 2022, the entire content of which is incorporated into this application by reference.

Technical field

This application belongs to the field of automatic welding technology, and particularly relates to a welding method, device, electronic equipment and computer-readable storage medium.

Background technique

With the continuous development of automatic welding technology, there are more and more ways to automatically weld metal materials, and automatic welding through robotic arms is one of them. How to achieve precise welding through robotic arms has become an urgent problem to be solved.

Contents of the invention

This application provides a welding method, device, electronic equipment and computer-readable storage medium for precise automatic welding of metal materials.

On the one hand, a welding method is provided. The method is applied to electronic equipment included in the welding system. The welding system also includes a robotic arm and a visual sensor fixed to the robotic arm. The method includes:

Control the vision sensor to obtain the target image corresponding to the welding joint;

Obtain the target starting point coordinates and target end point coordinates corresponding to the welding joint according to the target image;

Control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates.

In an exemplary embodiment, the target image is acquired by the visual sensor when the end of the robotic arm is located at the initial coordinates, and the initial coordinates include the initial starting point coordinates and the initial end point coordinates; the method further includes: determining the distance between the initial starting point coordinates and the target starting point coordinates. the first offset, determine the second offset between the initial end point coordinates and the target end point coordinates; determine the movement mode based on the first offset and the second offset, and control the end of the robotic arm from the initial The starting point coordinates move to the target starting point coordinates.

In an exemplary embodiment, the target starting point coordinates and the target end point coordinates are coordinates in the end coordinate system corresponding to the end of the robotic arm; obtaining the target starting point coordinates and the target end point coordinates corresponding to the welding joint according to the target image includes: obtaining according to the target image The reference starting point coordinates and the reference end point coordinates in the sensor coordinate system corresponding to the visual sensor; convert the reference starting point coordinates into the target starting point coordinates according to the first transformation matrix between the end coordinate system and the sensor coordinate system, and convert the reference end point coordinates according to the first transformation matrix The coordinates are converted into target end point coordinates.

In an exemplary embodiment, the method further includes: obtaining a first coordinate and at least two second coordinates. The first coordinate is a coordinate of the center of the calibration element in a base coordinate system corresponding to the base of the robotic arm. At least two second coordinates are obtained. Any second coordinate among the second coordinates corresponds to a pose, and any second coordinate is the coordinate of the center of the calibration element in the sensor coordinate system when the end of the robotic arm is in the corresponding pose; according to the first coordinate, at least two The second coordinates, the pose corresponding to any second coordinate and the second transformation matrix between the terminal coordinate system and the base coordinate system are solved to obtain the first transformation matrix.

In an exemplary embodiment, obtaining the target starting point coordinates and the target end point coordinates corresponding to the welding point based on the target image includes: determining the area where the welding point is located from the target image; based on the area where the welding point is located including reflective pixels, based on the surrounding areas of the reflective pixels The other pixels update the reflective pixels to non-reflective pixels to obtain the updated target image. The reflective pixels are the pixels corresponding to the reflective part of the welding joint; the target starting point coordinates and the target end point coordinates are obtained according to the updated target image.

In an exemplary embodiment, the method further includes: obtaining the characteristic information of the welding place according to the target image, and determining the welding method according to the characteristic information, where the characteristic information includes at least one of connection form, connection situation and connection gap; controlling the mechanical arm. Welding the end from the target starting point coordinates to the target end point coordinates includes: controlling the end of the robotic arm to weld from the target starting point coordinates to the target end point coordinates in a welding manner.

In an exemplary embodiment, the characteristic information includes a connection gap, and determining the welding mode according to the characteristic information includes: determining that the welding mode is pulse welding based on the connection gap being less than a first value; or, determining that the welding mode is pulse welding based on the connection gap being greater than or equal to the first value and If it is less than the second value, it is determined that the welding method is swing welding, and the second value is greater than the first value.

On the one hand, a welding device is provided. The device is applied to electronic equipment included in the welding system. The welding system also includes a robotic arm and a visual sensor fixed to the robotic arm. The device includes:

The first control module is used to control the visual sensor to obtain the target image corresponding to the welding joint;

The acquisition module is used to obtain the target starting point coordinates and target end point coordinates corresponding to the welding joint based on the target image;

The second control module is used to control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates.

In an exemplary embodiment, the target image is acquired by the vision sensor when the end of the robotic arm is located at initial coordinates, and the initial coordinates include initial starting point coordinates and initial end point coordinates;

The acquisition module is also used to determine the first offset between the initial starting point coordinates and the target starting point coordinates, and determine the second offset between the initial end point coordinates and the target end point coordinates; according to the first offset and the second offset The displacement determines the movement mode, and the end of the robotic arm is controlled to move from the initial starting point coordinates to the target starting point coordinates according to the movement mode.

In an exemplary embodiment, the target starting point coordinates and the target end point coordinates are coordinates in an end coordinate system corresponding to the end of the robotic arm;

The acquisition module is configured to convert the reference starting point coordinates into the target starting point coordinates according to the first transformation matrix between the terminal coordinate system and the sensor coordinate system, and convert the reference end point coordinates into the target end point coordinates according to the first transformation matrix.

In an exemplary embodiment, the acquisition module is also configured to acquire a first coordinate and at least two second coordinates. The first coordinate is the coordinate of the center of the calibration element in a base coordinate system corresponding to the base of the robotic arm, at least Any one of the two second coordinates corresponds to a pose, and any second coordinate is the coordinate of the center of the calibration element in the sensor coordinate system when the end of the robotic arm is in the corresponding pose; according to the first coordinate, The first transformation matrix is obtained by solving at least two second coordinates, the posture corresponding to any one of the second coordinates, and the second transformation matrix between the terminal coordinate system and the base coordinate system.

In an exemplary embodiment, the acquisition module is used to determine the area where the welding place is located from the target image; based on the area where the welding place is located includes reflective pixels, and updates the reflective pixels to non-reflective pixels based on other pixels around the reflective pixel, obtaining In the updated target image, the reflective pixels are the pixels corresponding to the reflective part of the welding joint; the target starting point coordinates and the target end point coordinates are obtained according to the updated target image.

In an exemplary embodiment, the acquisition module is further configured to obtain the characteristic information of the welding place according to the target image, and determine the welding method according to the characteristic information, where the characteristic information includes at least one of connection form, connection situation and connection gap;

The second control module is used to control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates in a welding manner.

In an exemplary embodiment, the characteristic information includes a connection gap, and the acquisition module is configured to determine that the welding method is pulse welding based on the connection gap being less than a first value; or, based on the connection gap being greater than or equal to the first value and less than a second value , it is determined that the welding method is swing welding, and the second value is greater than the first value.

In the embodiment of the present application, since the visual sensor is controlled to obtain the target image of the welding point, the target starting point coordinates and the target end point coordinates can be accurately determined based on the target image, thereby achieving accurate identification of the position of the welding point. After that, the end of the robotic arm is controlled to be welded from the target starting point coordinates to the target end point coordinates, so that the welding point can be welded, ensuring the welding accuracy and improving the welding quality.

On the one hand, an aluminum alloy automatic arc welding method is provided, including the following steps: collecting visual data during the aluminum alloy automatic arc welding process; identifying the actual welding seam position of the actual welding workpiece based on the visual data, and calculating the actual welding seam position of the current welding seam. and detect whether the actual weld gap is consistent with the initial weld gap before welding; and when it is detected that the actual weld gap is inconsistent with the initial weld gap, re-match the target welding parameters corresponding to the actual gap and guide the arc welding After the robot moves to the working position corresponding to the actual welding seam position, the arc welding robot is controlled to perform welding actions on the actual welding workpiece based on the target welding parameters.

Optionally, in one embodiment of the present application, before collecting the visual data during the automatic arc welding process of the aluminum alloy, it also includes: converting the manipulator base coordinate system of the arc welding robot, the manipulator end tool coordinate system and the coordinate system set in the arc welding robot. The line scanner coordinate system of the line structured light vision sensor on the robot performs coordinate conversion to obtain the hand-eye calibration conversion relationship.

Optionally, in one embodiment of the present application, identifying the actual weld position of the actually welded workpiece according to the visual data and calculating the actual weld gap of the current weld includes: obtaining the weld in the visual data The starting point image and the welding seam end point image; the entire welding seam length and position are simulated according to the welding seam starting point image and the welding seam end point image, and the actual welding seam position and the actual welding seam gap are obtained.

Optionally, in one embodiment of the present application, the rematching of the target welding parameters corresponding to the actual gap includes: finding the position of the joint feature in the laser profile information; calculating the The three-dimensional coordinate position of the joint characteristic position in the robot hand-based coordinate system; calculate the distance between the teaching position and the identification position according to the three-dimensional coordinate position, and match the target welding process according to the joint form and the gap between the plates to obtain the target Welding parameters.

Optionally, in one embodiment of the present application, the guiding the arc welding robot to move to the working position corresponding to the actual welding seam position includes: scanning the actual welding seam position to obtain the starting point and end point of the welding seam respectively. Coordinate value information; calculate the TCP (Tool Center Position, Transmission Control Protocol) position coordinates of the arc welding robot based on the coordinate value information of the welding seam starting point and the welding seam end point; guide the arc welding robot based on the TCP position coordinates Perform welding.

On the one hand, an aluminum alloy automatic arc welding device is provided, including: a collection module for collecting visual data during the aluminum alloy automatic arc welding process; and a calculation module for identifying actual weld seams of actual welded workpieces based on the visual data. position, and calculate the actual weld gap of the current weld, and detect whether the actual weld gap is consistent with the initial weld gap before welding; and a welding module for detecting that the actual weld gap is consistent with the When the initial weld gap is inconsistent, the target welding parameters corresponding to the actual gap are re-matched, and after the arc welding robot is guided to move to the working position corresponding to the actual weld position, the arc welding robot is controlled based on the target welding parameters to Describe the welding action performed on the actual welding workpiece.

Optionally, in one embodiment of the present application, it also includes: a conversion module for converting the manipulator base coordinate system of the arc welding robot, the manipulator end tool coordinate system and the line structure provided on the arc welding robot. The line scanner coordinate system of the light vision sensor performs coordinate conversion to obtain the hand-eye calibration conversion relationship.

Optionally, in one embodiment of the present application, the calculation module includes: a first acquisition unit, used to acquire the welding seam starting point image and the welding seam end point image in the visual data; a second acquiring unit, used to obtain the welding seam starting point image and the welding seam end point image in the visual data; The entire weld length and position are simulated based on the weld starting point image and the welding end point image to obtain the actual welding seam position and the actual welding seam gap.

Optionally, in one embodiment of the present application, the welding module includes: a search unit, used to search for the position of the joint feature in the laser profile information; a first calculation unit, used to calculate based on the hand-eye calibration conversion relationship The three-dimensional coordinate position of the joint characteristic position in the robot hand base coordinate system; a matching unit used to calculate the distance between the teaching position and the identification position according to the three-dimensional coordinate position, and match the target according to the joint form and the gap between the plates welding process to obtain the target welding parameters.

Optionally, in one embodiment of the present application, the welding module further includes: a scanning unit, used to scan the actual weld position to obtain the coordinate value information of the weld starting point and the weld end point respectively; a second calculation unit, and a welding unit configured to guide the arc welding robot to perform welding based on the TCP position coordinates.

On the one hand, a vehicle is provided, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the program to implement the above embodiments. The aluminum alloy automatic arc welding method or welding method described above.

On the other hand, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions. The computer instructions are used to cause the computer to execute the aluminum alloy automatic arc welding method or welding method as described in the above embodiment. method.

Embodiments of this application can collect visual data during the automatic arc welding process of aluminum alloys, and then identify the actual weld position, calculate the actual weld gap, and then match the target welding parameters, and automatically call the matching target according to the weld changes during the welding process. Welding parameters, thereby guiding the arc welding robot to perform welding actions at the corresponding position, so that the arc welding robot can still accurately and quickly find the welding position despite uneven weld gaps caused by welding thermal deformation and accumulated tolerances, which can effectively improve welding The automation level of the process improves the welding quality. There is no need for manual assistance to detect the weld position and set parameters, thereby reducing labor costs and eliminating accident losses caused by human judgment errors.

This solves the problem in related technologies that due to the limited scanning of the weld position, the inability to control the weld changes in the welding process in real time makes the automatic arc welding unable to accurately and quickly find the changed weld position during the welding process, thus Technical issues that affect welding quality and increase labor costs.

Description of the drawings

Figure 1 is a flow chart of an aluminum alloy automatic arc welding method provided according to an embodiment of the present application;

Figure 2 is a schematic diagram of the principle of an aluminum alloy automatic arc welding method according to an embodiment of the present application;

Figure 3 is a scanning schematic diagram of an aluminum alloy automatic arc welding method according to an embodiment of the present application;

Figure 4 is a logic flow chart of the adaptive selection of the welding method of the aluminum alloy automatic arc welding method according to one embodiment of the present application;

Figure 5 is a flow chart of an aluminum alloy automatic arc welding method according to an embodiment of the present application;

Figure 6 is a schematic structural diagram of a welding system provided by an embodiment of the present application;

Figure 7 is a schematic structural diagram of another welding system provided by an embodiment of the present application;

Figure 8 is a flow chart of a welding method provided by an embodiment of the present application;

Figure 9 is a schematic diagram of acquiring a target image provided by an embodiment of the present application;

Figure 10 is a flow chart for determining a welding method provided by an embodiment of the present application;

Figure 11 is a schematic structural diagram of an aluminum alloy automatic arc welding device provided according to an embodiment of the present application;

Figure 12 is a schematic structural diagram of a welding device provided by an embodiment of the present application;

Figure 13 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed ways

The embodiments of the present application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present application, but should not be construed as limiting the present application.

The aluminum alloy automatic arc welding method and device according to the embodiment of the present application will be described below with reference to the accompanying drawings.

The welding process of metal materials is extremely complex, and there are many factors that affect the welding quality. Not only are there external uncertain interference factors, but there are also strict requirements for the stability of the incoming workpiece. Most of the welding robots used for metal materials are teaching and reproducing types. Arc welding robots have stable performance but cannot effectively handle changes in welding conditions. Therefore, it is particularly important to control the welding quality during automatic arc welding of metal materials.

Related Technologies With the widespread application of machine vision in the field of industrial automation, such as image detection applications, visual positioning applications, object measurement applications, etc., robots have certain perception capabilities, which are highly automated, efficient, precise and adaptable. Advantages, it can effectively improve the efficiency and quality of automatic arc welding of metal materials.

However, in related technologies, the detection methods for welding robot welding seam visual positioning guidance in metal material welding have roughly the same principle, with uneven performance. The equipment itself is large, and the laser scanning posture of the robot after installation is limited, making it unable to deal with metal materials. The material is highly reflective and interferes with the ability to identify the various forms of weld joints. It cannot handle the uneven weld gaps caused by welding thermal deformation and accumulated tolerances during the welding process, making it impossible for automatic arc welding to accurately and quickly find the welding position, which needs improvement.

Due to the limited scanning of the weld position in related technologies, it is impossible to control the weld changes in the welding process in real time, making it impossible for automatic arc welding to accurately and quickly find the changed weld position during the welding process, thus affecting the welding quality. To solve the technical problem of increasing labor costs, this application provides an aluminum alloy automatic arc welding method.

In this method, visual data during the automatic arc welding process of aluminum alloy can be collected to identify the actual weld position, calculate the actual weld gap, and then match the target welding parameters, and automatically call the matching according to the weld changes during the welding process. The target welding parameters are used to guide the arc welding robot to perform welding actions at the corresponding position, so that the arc welding robot can still accurately and quickly find the welding position despite uneven weld gaps caused by welding thermal deformation and accumulated tolerances, which can effectively improve The automation level of the welding process improves the welding quality. There is no need for manual assistance to detect the weld position and set parameters, thereby reducing labor costs and eliminating accident losses caused by human judgment errors.

This solves the problem in related technologies that due to the limited scanning of the weld position, the inability to control the weld changes in the welding process in real time makes the automatic arc welding unable to accurately and quickly find the changed weld position during the welding process, thus Technical issues that affect welding quality and increase labor costs.

Figure 1 is a schematic flow chart of an aluminum alloy automatic arc welding method provided by an embodiment of the present application.

As shown in Figure 1, the aluminum alloy automatic arc welding method includes the following steps:

In step S101, visual data during the automatic arc welding process of aluminum alloy is collected.

In the actual execution process, as shown in Figure 2, the embodiment of the present application can install the line structured light vision sensor on the end effector of the robot arm such as a welding gun, thereby ensuring that the line structured light is not interfered within the field of view of the welding gun's motion trajectory. That is to ensure that there is no obstruction to the welding seam at any position within 360° on the welding gun.

Furthermore, embodiments of the present application can collect unobstructed visual data during the automatic arc welding process of aluminum alloy through the above-mentioned line structured light vision sensor, which facilitates subsequent judgment of the weld position, thereby realizing arc welding automation and eliminating welding thermal deformation. The weld position is inaccurate.

Optionally, in one embodiment of the present application, before collecting the visual data during the automatic arc welding process of the aluminum alloy, it also includes: converting the manipulator base coordinate system of the arc welding robot, the manipulator end tool coordinate system and the coordinate system set in the arc welding robot. The line scanner coordinate system of the line structured light vision sensor on the robot performs coordinate conversion to obtain the hand-eye calibration conversion relationship.

As a possible implementation method, the embodiment of the present application can use a high-precision standard ball with a known diameter as the hand-eye calibration target between the arc welding robot and the line structured light vision sensor, such as a 25mm standard ball, with the center of the ball As the calibration point, coordinate transformation is performed between the manipulator base coordinate system of the arc welding robot, the manipulator end tool coordinate system, and the line scanner coordinate system of the line structured light vision sensor installed on the arc welding robot, thereby calculating the hand-eye calibration conversion relationship.

For example, in the embodiment of the present application, a high-precision standard ball with a known diameter can be used as a hand-eye calibration target, and the center of the ball is the calibration point. The end of the arc welding robot can move in any posture, and the standard ball is cut by a laser plane, so that the cut surface Circle into an arc-shaped outline point.

Further, through the arc, the embodiment of the present application can fit the center and radius of the circle, and then according to the collinearity of the fitting center and the center of the sphere and the Pythagorean theorem, the line scanner (x of the current line structured light vision sensor) can be obtained , y, z) The y value of the center of the sphere at the position.

It can be understood that during the hand-eye calibration process, there are three coordinate systems: the manipulator base coordinate system {0 _b }, the manipulator end tool coordinate system {0 _t }, and the line scanner coordinate system {0 _s of the line structured light vision sensor }. Among them, the coordinates of any point under the online scanner to the coordinates under the robot base coordinate system can be converted into:

Among them, Ps is a certain point of the current scanner, and Pb is a point under Ps in the base coordinate system. With the above conversion relationship, the arc welding robot can directly obtain the conversion matrix from {0 _t } to {0 _b }

That is the transformation matrix that needs to be calibrated, and the process of solving X is hand-eye calibration.

Embodiments of the present application can control the robot to make the line scanner of the line structured light vision sensor obtain the position coordinates of the fixed point within the target measurement range in the coordinate system of the line scanner of the line structured light vision sensor in different postures, and solve the problem through a certain solution method. Calculating the positional relationship between the laser and the robot can achieve hand-eye calibration of the positional relationship between the line scanner of the line structured light vision sensor and the arc welding robot.

In practical applications, for example, the calibration process of hand-eye calibration can include the following steps:

1. Place the standard diameter ball used for calibration on a flat surface below the manipulator in as safe an area as possible to prevent damage from falling;

2. Manually control the manipulator so that the laser line hits the ball to ensure that the Y coordinate of the ball center in the coordinate system of the measuring instrument is positive;

3. After opening the software and connecting to the device, enter the calibration interface and record the contour and the corresponding manipulator coordinates at this time. At this time, the calibration calculation list will display the sphere center and manipulator UVWXYZ coordinate values obtained from the spherical section profile in the current state;

4. Randomly change the attitude and position of the manipulator (UVWXYZ value), repeat steps 2-3 several times to obtain the corresponding contour and manipulator coordinates;

5. After obtaining enough data, add or delete the data according to the precautions, click the Calculate Hand-Eye Calibration Matrix button on the software, and use the collected data to calculate the coordinate transformation matrix. You can repeatedly import the calculation results and calculate.

In step S102, the actual welding seam position of the actually welded workpiece is identified based on the visual data, the actual welding seam gap of the current welding seam is calculated, and whether the actual welding seam gap is consistent with the initial welding seam gap before welding is detected.

During the actual execution process, the embodiments of the present application can identify the actual welding seam position of the actual welded workpiece based on the collected visual data, and then calculate the actual welding seam gap of the current welding seam, and detect the actual welding seam in real time during the welding operation. seam gap to avoid welding seam deformation caused by welding thermal deformation, causing the actual welding seam position to deviate from the initial welding seam position, thus affecting the actual welding effect. It can improve the automation level of aluminum alloy arc welding and avoid welding thermal deformation on the welding. The impact on quality can reduce manual participation, thereby reducing labor costs and avoiding accident losses caused by human factors, which is beneficial to industrial production.

Optionally, in one embodiment of the present application, identifying the actual weld position of the actually welded workpiece according to the visual data and calculating the actual weld gap of the current weld includes: obtaining the weld starting point image and the welding seam in the visual data. seam end point image; simulate the entire weld length and position based on the weld start point image and weld end point image to obtain the actual weld seam position and actual weld seam gap.

Illustratively, the embodiment of the present application can teach the arc welding robot to reach the welding seam approach point position and the welding seam arc closing position respectively based on the collected visual data, and adjust the appropriate scanning angle, and use the laser vision sensor to detect the welding seam. Scan the starting point and end point images, and calculate the actual weld position and actual weld gap based on the scan results.

For example, as shown in Figure 3, the embodiment of the present application can obtain the arc starting position of the weld seam by scanning the starting point and end point images of the weld seam, and then obtain the line scanner coordinates of the line structured light vision sensor at that point. Three-dimensional coordinates (X ₁ , Y ₁ , Z ₁ ), move the arc welding robot to the closing arc position of the welding seam, adjust the appropriate scanning angle again to scan the closing arc point, and obtain the line structure of the point. Line scanner of light vision sensor The three-dimensional coordinates (X ₂ , Y ₂ , Z ₂ ) under the coordinate system are converted into three-dimensional coordinates (x ₁ , y ₁ , z ₁ ) and (x _{2 )} under the robot coordinate system through the results of hand-eye calibration. ,y ₂ ,z ₂ ), by calculating the coordinate difference D ₁ , D ₂ between the arc starting point and the arc ending point, that is, the offset:

Obtain the starting arc coordinate position (X ₃ , Y ₃ , Z ₃ ) and arc closing coordinate position (X ₅ , Y ₄ , Z ₄ ) that will be guided.

In step S103, when it is detected that the actual weld gap is inconsistent with the initial weld gap, the target welding parameters corresponding to the actual gap are re-matched, and the arc welding robot is guided to move to the working position corresponding to the actual weld position, based on the target welding Parameters control the arc welding robot to perform welding actions on the actual welded workpiece.

As a possible implementation method, the embodiment of the present application can match the corresponding target weld parameters according to the actual weld gap, guide the arc welding robot to move to the working position corresponding to the actual weld position, and control the control according to the target weld parameters. The arc welding robot performs welding actions on the actual welding workpiece, and when it detects that the actual weld gap is inconsistent with the initial gap, it re-matches the target parameters corresponding to the actual gap, which can effectively improve the automation level of the welding process, improve the welding quality, and avoid welding defects. If the welding quality is reduced due to thermal deformation, there is no need for manual assistance to detect the weld position and weld gap and set parameters, thus reducing labor costs and eliminating accident losses caused by human judgment errors.

Optionally, in one embodiment of the present application, re-matching the target welding parameters corresponding to the actual gap includes: finding the joint characteristic position in the laser profile information; calculating the joint characteristic position in the robot hand-based coordinates based on the hand-eye calibration conversion relationship The three-dimensional coordinate position under the system is calculated; the distance between the teaching position and the identification position is calculated based on the three-dimensional coordinate position, and the target welding process is matched according to the joint form and the gap between the plates to obtain the target welding parameters.

Illustratively, the embodiment of this application can pre-set the relevant parameters of the joint so that the relevant database can be subsequently established, and then use the improved robust least squares straight line fitting algorithm, combined with the differential detection method, to automatically find the location in the laser profile information. The characteristic position XYZ of the joint, and based on the hand-eye calibration relationship, calculate the three-dimensional coordinate position (X ₀ , Y ₀ , Z ₀ ) of the joint's characteristic position XYZ in the robot's base coordinate system.

Further, the embodiment of the present application can calculate the distance between the teaching position and the identification position according to the joint characteristic position (X ₀ , Y ₀ , Z ₀ ), obtain the target welding parameters according to the joint form and the gap between the plates, and select the appropriate welding Craftsmanship.

Among them, the joint feature identification parameters may include: joint number, position name, joint form, joint direction, joint endpoint, joint range dividing line, offset, teaching point and image shielding.

Optionally, in one embodiment of the present application, guiding the arc welding robot to move to the working position corresponding to the actual welding seam position includes: scanning the actual welding seam position to obtain the coordinate value information of the welding seam starting point and the welding seam end point respectively; Calculate the TCP position coordinates of the arc welding robot based on the coordinate value information of the welding seam starting point and the welding seam end point; guide the arc welding robot to weld based on the TCP position coordinates.

During the actual execution process, the embodiment of the present application can scan the actual welding seam position in real time, obtain the coordinate value information of the welding seam starting point and the welding seam end point respectively, and convert it into the robot TCP position coordinates through calculation, as shown in Figure 4. Embodiments can use TCP position coordinates and transmit them to the arc welding robot through the TCP/IP network, and then use the private motor of the arc welding robot to drive the welding gun to the guide welding seam position to perform the welding operation.

As shown in FIGS. 2 to 5 , an exemplary embodiment is used to explain in detail the working principle of the aluminum alloy automatic arc welding method according to the embodiment of the present application.

As shown in Figure 2, the embodiment of the present application may include: line structured light vision sensor 1, arc welding robot 2, welding gun 3, TCP/IP communication network cable 4, visual control host 5, arc welding robot welding power supply control cabinet 6 and workpiece 7.

As shown in Figure 5, taking a butt weld and a 2mm uniform gap weld in the production process as an example, the embodiment of the present application may include the following steps:

Step S501: Collect video data. In the actual execution process, as shown in Figure 2, the embodiment of the present application can install the line structured light vision sensor 1 on the end effector of the mechanical arm of the arc welding robot 2, that is, the welding gun 3, thereby ensuring that the line structured light is visible on the welding gun 3 There is no interference within the field of view of the motion trajectory, that is, the welding seam at any position within the 360° range of the welding gun 3 is guaranteed to be unobstructed.

Furthermore, the embodiment of the present application can collect unobstructed visual data during the automatic arc welding process of aluminum alloy through the above-mentioned line structured light vision sensor 1, so as to facilitate the subsequent judgment of the welding seam position, thereby realizing arc welding automation.

Step S502: Standard ball hand-eye calibration. As a possible implementation method, the embodiment of the present application can use a high-precision standard ball with a known diameter as the hand-eye calibration target between the arc welding robot 2 and the line structured light vision sensor 1, such as a 25mm standard ball. The center of the sphere is the calibration point. The manipulator base coordinate system of the arc welding robot 2, the manipulator end tool coordinate system and the line scanner coordinate system of the line structured light vision sensor 1 installed on the arc welding robot 2 are coordinate converted to calculate Hand-eye calibration conversion relationship.

For example, in the embodiment of the present application, a high-precision standard ball with a known diameter can be used as a hand-eye calibration target, and the center of the ball is the calibration point. The end of the arc welding robot 2 can move in any posture, and the standard ball is cut by a laser plane, so that The tangent circle forms an arc-shaped contour point.

Further, through the arc, the embodiment of the present application can fit the center and radius of the circle, and then according to the collinearity of the fitting center and the center of the sphere and the Pythagorean theorem, the line scanner of the current line structured light vision sensor 1 can be obtained ( The y value of the center of the sphere at the x, y, z) position.

It can be understood that during the hand-eye calibration process, there are three coordinate systems: the manipulator base coordinate system {0 _b }, the manipulator end tool coordinate system {0 _t }, and the line scanner coordinate system {0 of the line structured light vision sensor 1 _s }. Among them, the coordinates of any point under the line scanner of the online structured light vision sensor 1 to the coordinates under the manipulator base coordinate system can be converted into:

Among them, Ps is a certain point of the current scanner, and Pb is the point under Ps in the base coordinate system. With the above conversion relationship, the arc welding robot 2 can directly obtain the conversion matrix from {0 _t } to {0 _b }

That is the transformation matrix that needs to be calibrated, and the process of solving X is hand-eye calibration.

In the embodiment of the present application, the robot can be controlled to make the line scanner of the line structured light vision sensor 1 obtain the position coordinates of the fixed point within the target measurement range in the line scanner coordinate system of the line structured light vision sensor 1 in different postures. The method solves the positional relationship between the line scanner of the line structured light vision sensor 1 and the arc welding robot 2, and can realize the hand-eye calibration of the positional relationship between the line scanner of the line structured light vision sensor 1 and the arc welding robot 2.

In practical applications, for example, the calibration process of hand-eye calibration can include the following steps:

1. Place the standard diameter ball used for calibration on a flat surface below the manipulator in as safe an area as possible to prevent damage from falling;

2. Manually control the manipulator so that the laser line hits the ball to ensure that the Y coordinate of the ball center in the coordinate system of the measuring instrument is positive;

3. After opening the software and connecting to the device, enter the calibration interface and record the contour and the corresponding robot hand coordinates at this time. At this time, the calibration calculation list will display the center of the ball and the robot hand UVXYZ coordinate values obtained from the spherical section profile in the current state;

4. Randomly change the attitude and position of the manipulator (UVWXYZ value), repeat steps 2-3 several times to obtain the corresponding contour and manipulator coordinates;

5. After obtaining enough data, add or delete the data according to the precautions, click the Calculate Hand-Eye Calibration Matrix button on the software, and use the collected data to calculate the coordinate transformation matrix. You can repeatedly import the calculation results and calculate.

Step S503: Weld seam position scanning. Illustratively, the embodiment of the present application can teach the arc welding robot 2 to reach the welding seam approach point position and the welding seam arc closing position respectively based on the collected visual data, and adjust the appropriate scanning angle, using the line structured light vision sensor 1 , scan the image of the starting point and end point of the weld, and calculate the actual weld position and the actual weld gap based on the scanning results, and detect the actual weld gap in real time during the welding operation to avoid welding thermal deformation caused by The deformation of the weld seam causes the actual weld seam position to deviate from the initial weld seam position, thus affecting the actual welding effect.

For example, as shown in Figure 3, the embodiment of the present application can obtain the arc starting position of the weld seam by scanning the starting point and end point images of the weld seam, and then obtain the line scanner coordinates of the line structured light vision sensor 1 at that point. Lower the three-dimensional coordinates (X ₁ , Y ₁ , Z ₁ ), move the arc welding robot to the arc closing position of the welding seam, adjust the appropriate scanning angle again to scan the arc closing point, and obtain the line of the structured light vision sensor 1 at that point The three-dimensional coordinates (X ₂ , Y ₂ , Z ₂ ) under the scanner coordinate system are converted into three-dimensional coordinates (x ₁ , y 1 , z 1 ) and (x 1 , y ₁ , z ₁ ) under the robot coordinate system through the results of hand-eye calibration. x ₂ , y ₂ , z ₂ ), by calculating the coordinate differences D ₁ and D ₂ between the arc starting point and the arc ending point, that is, the offset:

Obtain the starting arc coordinate position (X ₃ , Y ₃ , Z ₃ ) and arc closing coordinate position (X ₄ , Y ₄ , Z ₄ ) that will be guided.

Step S504: Set weld parameter information. In this embodiment of the present application, joint-related parameters can be set for later establishment of a database.

Step S505: Calculate the welding seam position. Embodiments of the present application can use the improved robust least squares straight line fitting algorithm, combined with the differential detection method, to automatically find the characteristic position XYZ of the positioning joint in the laser profile information, and calculate the characteristic position XYZ of the joint on the robot based on the hand-eye calibration relationship. The three-dimensional coordinate position (X ₀ , Y ₀ , Z ₀ ) in the base coordinate system.

Step S506: Adaptively call appropriate welding parameters. Further, the embodiment of the present application can calculate the distance between the teaching position and the identification position according to the joint characteristic position (X ₀ , Y ₀ , Z ₀ ), obtain the target welding parameters according to the joint form and the gap between the plates, and select the appropriate welding process, and when it is detected that the actual weld gap is inconsistent with the initial gap, re-match the target parameters corresponding to the actual gap, which can effectively improve the automation level of the welding process, improve the welding quality, and avoid the degradation of welding quality caused by welding thermal deformation, without Manually assist in detecting the welding seam position and welding seam gap and setting parameters, thereby reducing labor costs and eliminating accident losses caused by human judgment errors.

For example, when the vision device enters the state of waiting for the manipulator signal, after the embodiment of the present application recognizes the positioning signal, the feature point positions of the current contour data of the line scanner of the line structured light vision sensor 1 can be calculated based on the matrix, and then the feature points are The offset of the position is written into the robot and the joint feature identification parameters are set, including:

1. Joint number: JOB number, indicating which welding joint;

2. Position name: the joint name indicating the JOB number;

3. Joint form: butt joint, corner joint, lap joint;

4. Joint direction: divided into rising edge and falling edge, determined according to the outline of the left and right sides of the inflection point (joint);

5. Joint end point: left point or right point of the joint position;

6. Connector range dividing line: Set the shielding function for the connector characteristic position area;

7. Offset: XYZ compensation added when there is a fixed deviation in calibration;

8. Teaching point: XYZ value at the end of the robot when teaching;

9. Image masking: Use a masking algorithm to image reflective areas.

Furthermore, in the embodiment of this application, after setting two positions for a certain joint, it waits for the robot signal to guide. After the positioning is successful, the line scanner coordinates of the line structured light vision sensor 1 of the joint position will be displayed on the interface. (x′,0,z′), manipulator coordinates (X′,Y′,Z′) and offset (ΔX,ΔY,ΔZ).

Step S507: TCP/IP signal coordinate conversion. During the actual execution process, the embodiment of the present application can scan the actual welding seam position, obtain the coordinate value information of the welding seam starting point and the welding seam end point respectively, and convert it into the robot TCP position coordinates through calculation, as shown in Figure 4, the implementation of this application For example, TCP position coordinates can be used and transmitted to the arc welding robot 2 through the TCP/IP network.

Step S508: The arc welding robot 2 guides the welding. The memory unit transmits the coordinate value to the control unit, and the control unit outputs a signal value to the robot's private motor to drive the welding gun 3 to the guided weld position to perform the welding operation.

According to the aluminum alloy automatic arc welding method proposed in the embodiment of the present application, visual data during the aluminum alloy automatic arc welding process can be collected, and then the actual weld position can be identified, the actual weld gap can be calculated, and the target welding parameters can be matched. According to the changes in the weld seam, the arc welding robot can automatically call the matching target welding parameters to guide the arc welding robot to perform welding actions at the corresponding position, so that the arc welding robot can still achieve uneven weld seam gaps caused by welding thermal deformation and accumulated tolerances. Finding the welding position accurately and quickly can effectively improve the automation level of the welding process and improve the welding quality. There is no need for manual assistance to detect the welding seam position and set parameters, thereby reducing labor costs and eliminating accident losses caused by human judgment errors. This solves the problem in related technologies that due to the limited scanning of the weld position, the inability to control the weld changes in the welding process in real time makes the automatic arc welding unable to accurately and quickly find the changed weld position during the welding process, thus Technical issues that affect welding quality and increase labor costs.

Referring to FIG. 6 , an embodiment of the present application provides a welding system, which includes an electronic device 61 , a robotic arm 62 , and a visual sensor 63 fixed to the robotic arm 62 . The electronic device 61 is connected to the visual sensor 63 and the mechanical arm 62 through communication respectively. The electronic device 61 can send signals through the communication connection to control the visual sensor 63 and the mechanical arm 62 . Thereby, the electronic device 61 can control the visual sensor 63 to obtain an image of the welding place to be welded. The electronic device 61 can also control the movement of the end of the robotic arm 62 so that the end of the robotic arm 62 welds the welding point.

For example, as shown in FIG. 7 , one end of the robotic arm 62 is installed on the base 64 , and the other end of the robotic arm 62 , that is, the end of the robotic arm 62 , is the end used for welding. For example, if the robotic arm 62 includes a welding element 621, such as a welding gun, and the welding element 621 includes a nozzle for welding, then the nozzle is the end of the robotic arm 62. The movement of the end of the robotic arm 62 is to adjust the posture of the welding element 621, so that the welding place can be welded through the nozzle of the welding element 621, that is, the end of the robotic arm 62. The pose includes at least one of a position and an attitude, and the robotic arm 62 including the welding element 621 and the base 64 form a welding robot.

In addition, a visual sensor 63 is fixed on the robotic arm 62. The visual sensor 63 can be a camera, a video recorder, a scanner, a line structured light visual sensor, etc. There is no limitation here. The error of the visual sensor can be limited to within 0.1 mm. . The visual sensor 63 can be fixed at any position within a 360-degree range of the welding element 621 axially. It should be understood that the embodiment of the present application does not limit the fixed position of the visual sensor 63, as long as it ensures that the visual sensor 63 is not blocked during the movement of the robotic arm 62, and does not affect the visual sensor 63 in acquiring the image of the welding joint.

For example, the electronic device 61 may be connected to the visual sensor 63 and the robotic arm 62 through wired or wireless communication. For example, the method shown in FIG. 7 is a wired method, that is, the electronic device 61 communicates with the visual sensor 63 and the robotic arm 62 through the network cable 65 respectively. The network cable 65 includes but is not limited to: TCP (Transmission Control Protocol, Transmission Control Protocol)/IP (Internet Protocol, Internet Protocol) communication network cable. Continuing to refer to FIG. 7 , the electronic device 61 may include a vision control host for controlling the vision sensor 63 , and a mechanical control host for controlling the robotic arm 62 .

In addition, FIG. 7 also shows the object platform 66 and the calibration component 67 for hand-eye calibration. The hand refers to the welding component 621 on the robot arm 62 and the eye refers to the visual sensor 63 . The following method embodiments will describe the process of hand-eye calibration in detail, which will not be described in detail here.

The embodiment of the present application provides a welding method, which can be applied to the electronic equipment included in the welding system shown in Figure 6 or Figure 7. As shown in Figure 8, the method includes the following steps 801 to 803.

Step 801: Control the visual sensor to obtain the target image corresponding to the welding point.

Among them, welding is a method of connecting metal materials. Wherever metal materials need to be connected, there is the welding point that needs to be welded. The welding point is also called a joint. A welding point can be uniquely indicated by an identifier, which can be a name or a string. The string is, for example, JOB (Job Number). In addition, the metal material can be an alloy, such as aluminum alloy, etc., or it can be a pure metal, which is not limited here. Welding includes but is not limited to arc welding, gas welding, etc., which are also not limited here.

The embodiment of the present application is not limited to where the visual sensor obtains the target image. For example, the electronic device can obtain the initial coordinates, and after controlling the end of the robotic arm to reach the initial coordinates, control the visual sensor to obtain the target image corresponding to the welding point. That is, the target image is acquired by the vision sensor with the end of the robotic arm located at the initial coordinates.

The initial coordinates are coordinates near the welding point. It can be considered that the initial coordinates are estimated and not accurate enough. The initial coordinates can be set manually based on experience or actual needs, or can be automatically calculated based on the volume of the robotic arm, the volume of the metal material, the area of the welding joint, etc. The method of obtaining the initial coordinates is not limited here.

After obtaining the initial coordinates, the initial coordinates are reached by controlling the end of the robotic arm. Then, when the end of the robotic arm is located at the initial coordinates, the vision sensor is controlled to obtain the target image corresponding to the welding point at an appropriate angle. The target image is used to determine the target starting point coordinates and the target end point coordinates. Since the embodiment of the present application ensures that the visual sensor will not be blocked during the movement of the robotic arm, the movement of the robotic arm will not be restricted. No matter how the robotic arm moves, the visual sensor can obtain an accurate target image. .

The target starting point coordinates and the target end point coordinates are relatively accurate coordinates, more accurate than the above-mentioned initial coordinates. The target starting point coordinates and the target end point coordinates are no longer the coordinates near the welding point, but the coordinates of the welding point itself that can represent the welding point. For example, see Figure 9, the target starting point coordinates are the coordinates of the starting point shown in Figure 9, the starting point is the point at the end of the robotic arm where welding begins, and the target end point coordinates are the coordinates of the end point shown in Figure 9, and the end point is the end point of the robotic arm. The end ends at a point where welding is done.

In the embodiment of this application, the reason why it is necessary to determine the target starting point coordinates and the target end point coordinates after acquiring the target image is because: the welding joint may be uneven, and this unevenness is difficult to predict, and it is necessary to obtain the target image before Only then can the actual situation of the welding joint be determined, thereby determining the accurate target starting point coordinates and target end point coordinates. For example, the causes of uneven welding may include but are not limited to: the welding has been welded once before, and the previous welding process caused deformation and accumulated tolerances in the welding, thus making the welding uneven. No matter what causes the welding joint to be uneven, the embodiment of the present application can determine the accurate target starting point coordinates and target end point coordinates when the welding joint is uneven, so as to achieve accurate welding for the welding joint.

For example, since the end of the robot arm needs to reach the target starting point coordinates and the target end point coordinates, the target starting point coordinates and the target end point coordinates are coordinates that can be recognized by the end of the robot arm. For example, the target starting point coordinates and the target end point coordinates are The coordinates in the end coordinate system corresponding to the end of the robotic arm. This coordinate belongs to TCP (Tool Center Point) coordinates. Of course, the initial coordinates can also be coordinates in the end coordinate system.

Illustratively, the number of initial coordinates is one or more. For example, the initial coordinates may include an initial starting point coordinate and an initial end point coordinate. The initial starting point coordinate is a coordinate located near the starting point of the welding point, and the initial end point coordinate is a coordinate located near the end point of the welding point. Correspondingly, referring to Figure 9, after controlling the end of the robotic arm to reach the initial coordinates, controlling the visual sensor to obtain the target image corresponding to the welding point includes: controlling the end of the robotic arm to reach the initial starting point coordinates, controlling the visual sensor according to the appropriate first angle Obtain the first image corresponding to the starting point of the welding point, and also control the end of the robotic arm to reach the initial end point coordinates, adjust the first angle to the second angle, and control the visual sensor to obtain the second image corresponding to the end point of the welding point according to the second angle. . Wherein, the first angle is the same as or different from the second angle, and the above-mentioned target image includes the first image and the second image.

It should be understood that the end of the robotic arm can first reach the initial starting point coordinates and then reach the initial end point coordinates, or it can first reach the initial end point coordinates and then reach the initial starting point coordinates, which is not limited here.

Step 802: Obtain the target starting point coordinates and the target end point coordinates corresponding to the welding point according to the target image.

Among them, the embodiment of the present application can obtain the target starting point coordinates and the target end point coordinates based on the same target image. Alternatively, embodiments of the present application may also obtain the target starting point coordinates and the target end point coordinates based on different target images. For example, when the target image includes a first image and a second image, the target starting point coordinates can be obtained according to the first image, and the target end point coordinates can be obtained according to the second image.

For example, the embodiment of the present application can identify the target image to obtain the outline of the welding point, and the outline reflects the length and position of the welding point. Among them, if the target image records all the welding joints, the entire contour of the welding joint can be obtained directly based on the target image. If the target image only records part of the welding joint, then part of the contour of the welding joint can be obtained based on the target image. , and the unobtained partial contours are completed through the simulation process.

The contour of the welding joint is often composed of multiple points. A certain algorithm can be used to select two points from multiple points, and the coordinates of these two points are used as the target starting point coordinates and the target end point coordinates respectively. Among them, the algorithms used include but are not limited to the least squares straight line fitting algorithm to improve robustness, the differential detection algorithm, etc., which are not limited here.

In an exemplary embodiment, obtaining the target starting point coordinates and the target end point coordinates corresponding to the welding point based on the target image includes: determining the area where the welding point is located from the target image; based on the area where the welding point is located including reflective pixels, based on the surrounding areas of the reflective pixels The other pixels update the reflective pixels to non-reflective pixels to obtain the updated target image. The reflective pixels are the pixels corresponding to the reflective part of the welding joint; the target starting point coordinates and the target end point coordinates are obtained according to the updated target image.

Among them, the embodiment of the present application can determine the area where the welding place is located from the target image through image recognition. For example, you can input the target image into the image recognition model and obtain the image output by the image recognition model. Compared with the target image, the image output by the image recognition model has more area edge lines. The area surrounded by the area edge lines is where the welding place is. Area.

Since the welding joint is a metal material, there may be a reflective part at the welding joint, and the reflective part may reflect light, causing the area where the welding joint is located in the target image to include reflective pixels. Since such reflective pixels may affect the determination of the target starting point coordinates and the target end point coordinates, the reflection needs to be eliminated, that is, the reflective pixels are updated to non-reflective pixels to obtain an updated target image. Thus, the accuracy of the determined target starting point coordinates and target end point coordinates can be ensured. For example, when updating reflective pixels to non-reflective pixels, a masking algorithm may be used, and the masking algorithm used is not limited here.

Exemplarily, obtaining the target starting point coordinates and the target end point coordinates corresponding to the welding joint according to the target image includes: obtaining the reference starting point coordinates and the reference end point coordinates in the sensor coordinate system corresponding to the visual sensor according to the target image; according to the end coordinate system and the sensor coordinates The first conversion matrix between the systems converts the reference starting point coordinates into the target starting point coordinates, and converts the reference end point coordinates into the target end point coordinates according to the first conversion matrix. Of course, this solution can be combined with the above-described solution for eliminating reflections. For example, when the updated target image is obtained through the solution for eliminating reflections, this solution can correspondingly obtain the corresponding image of the welding joint based on the updated target image. Target starting point coordinates and target end point coordinates.

Among them, since the target image is obtained through the visual sensor, the coordinates determined directly based on the target image, that is, the reference starting point coordinates and the reference end point coordinates, are coordinates in the sensor coordinate system. The sensor coordinate system and the end coordinate system are different coordinate systems. Therefore, it is necessary to convert the reference starting point coordinates and the reference end point coordinates in the sensor coordinates to the target starting point coordinates and the target end point coordinates in the end coordinate system according to the first conversion matrix, so that the end of the robotic arm can identify and reach the target starting point coordinates and target end point coordinates to achieve welding at the welding point.

For example, the reference starting point coordinates are (M ₁ , P ₁ , Q ₁ ), and the reference starting point coordinates can be converted into the target starting point coordinates (m ₁ , p ₁ , q ₁ ) according to the first transformation matrix. The reference end point coordinates are (M ₂ , P ₂ , Q ₂ ), and the reference starting point coordinates can be converted into the target starting point coordinates (m ₂ , p ₂ , q ₂ ) according to the first transformation matrix.

For example, the electronic device may store the first conversion matrix, so that when the first conversion matrix needs to be used, the stored first conversion matrix is used. The first conversion matrix can be directly obtained by the electronic device from other devices, or the first conversion matrix can be determined by the electronic device through a hand-eye calibration process.

During the hand-eye calibration process, see FIG. 9 , the calibration element is placed on the bearing platform. The calibration element can be a highly precise sphere with a certain diameter, for example, 25 mm in diameter. The bearing platform can be a flat surface located below the end of the robotic arm, and the flat surface has a certain degree of safety. The calibration element is not easily dropped and damaged after being placed on the bearing platform.

After the calibration element is placed, the first coordinate of the center of the calibration element in a base coordinate system corresponding to the base of the robotic arm can be obtained. The base coordinate system is, for example, a geodetic coordinate system. In addition, at least two second coordinates need to be obtained. The second coordinates are the coordinates of the center of the calibration element in the sensor coordinate system when the end of the robotic arm belongs to a certain orientation. Among them, the following process can be performed at least twice, and each time a second coordinate is obtained, thereby obtaining at least two second coordinates, each second coordinate corresponds to a pose, and the pose can be represented by six degrees of freedom information. .

In this process, the robotic arm is manually controlled so that the end of the robotic arm is in a certain position. In addition, the visual sensor is caused to emit a laser plane. For example, the dotted line area in Figure 9 is an exemplary laser plane. The laser plane is incident on the calibration element, which is equivalent to cutting the calibration element to obtain a section, and the section can be located above the center of the calibration element. , and then, according to each point on the edge of the section, the coordinates of the center of a calibration element in the sensor coordinate system can be obtained by fitting, that is, a second coordinate. For example, when the calibration element is a sphere and the center of the calibration element is the center of the sphere, then the calibration element is cut to obtain a circular section, and the center coordinates and radius of the circular section can be fitted according to the points on the edge of the circular section, and then combined The collinear theorem of the center of a circle and the center of a sphere and the Pythagorean theorem determine the coordinates of the center of the sphere, which is the second coordinate.

In addition, in different processes, the position and posture of the end of the robotic arm are also different. Moreover, the posture of the end of the robotic arm needs to be within a certain range to prevent the end of the robotic arm from shaking when it is in a certain posture, thus affecting the accuracy of the hand-eye calibration process.

After obtaining the first coordinate, at least two second coordinates and the pose corresponding to any second coordinate, the obtained content can be directly used to solve to obtain the first transformation matrix, or the obtained content can be supplemented or filtered, such as adding Or delete some second coordinates and the poses corresponding to these second coordinates to obtain updated content. Afterwards, the obtained content or updated content can be directly used to solve to obtain the first transformation matrix. For example, based on the following formula (1), combined with the pose corresponding to any second coordinate, the first transformation matrix is obtained:

In formula (1), the meanings of each symbol are as follows:

Pb represents the first coordinate, Ps represents the second coordinate;

The second transformation matrix is expressed as

The second transformation matrix is a transformation matrix between the end coordinate system {Ot} and the base coordinate system {Ob}. The second transformation matrix can be directly obtained based on the connection relationship between the base and the end of the robotic arm;

The first transformation matrix that needs to be solved is expressed as

The first conversion matrix is the conversion matrix between the terminal coordinate system {Ot} and the sensor coordinate system {Os}. The process of hand-eye calibration is to solve and obtain the first conversion matrix.

Based on the above description, it can be concluded that for the method of determining the first transformation matrix through the hand-eye calibration process, the method also includes: obtaining the first coordinate and at least two second coordinates, where the first coordinate is the center of the calibration element at the base of the robot arm. The coordinates in the base coordinate system corresponding to the base. Any one of at least two second coordinates corresponds to a pose. Any one of the second coordinates is the center of the calibration element when the end of the robotic arm is in the corresponding pose. Coordinates in the sensor coordinate system; solve to obtain the first transformation based on the first coordinate, at least two second coordinates, the pose corresponding to any second coordinate and the second transformation matrix between the end coordinate system and the base coordinate system matrix.

Step 803: Control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates.

After determining the target starting point coordinates and the target end point coordinates, the end movement of the robotic arm can be controlled. In some embodiments, if the target image is acquired by the vision sensor when the end of the robotic arm is located at the initial coordinate, it is necessary to first control the end of the robotic arm to move from the initial coordinate to the target starting point coordinate, then the end of the robotic arm moves from the target starting point Start welding at the coordinates and move while welding until the end of the robotic arm reaches the target end coordinates and then end the welding, thereby achieving welding from the target starting point coordinates to the target end point coordinates. Or, in other embodiments, there is no limit on where the target image is obtained, as long as it is ensured that the end of the robotic arm can reach the target starting point coordinates and weld from the target starting point coordinates to the target end point coordinates. After the welding is completed, a weld seam will be formed at the welding point. For example, Figure 9 shows an exemplary weld seam.

In an exemplary embodiment, for the case where the initial coordinates include the initial starting point coordinates and the initial end point coordinates, the method further includes: determining a first offset between the initial starting point coordinates and the target starting point coordinates, and determining the initial end point coordinates and the target end point coordinates. the second offset between them; determine the movement mode according to the first offset and the second offset, and control the end of the robotic arm to move from the initial starting point coordinates to the target starting point coordinates according to the movement mode.

For example, if the initial starting point coordinates are (m ₃ , p ₃ , q ₃ ) and the target starting point coordinates are (m ₁ , p ₁ , q ₁ ), then the first offset E ₁ is expressed as the following formula (2):

For another example, if the initial end point coordinates are (m ₄ , p ₄ , q ₄ ) and the target end point coordinates are (m ₂ , p ₂ , q ₂ ), then the second offset W ₂ is expressed as the following formula (3):

The movement mode may include a movement speed. When at least one of the first offset and the second offset is greater than an offset threshold, the movement mode may be determined to be fast movement, so that the end of the robotic arm can move quickly. from the initial starting point coordinates to the target starting point coordinates to avoid affecting the welding efficiency due to large offsets. If both the first offset and the second offset are less than or equal to the offset threshold, the movement mode can be determined to be slow movement to ensure smoothness and stability of the robotic arm during movement.

Alternatively, the movement method may also include a movement path, and an appropriate movement path may be selected based on the first offset and the second offset. The movement path is used for the end of the robotic arm to move from the initial starting point coordinates to the target starting point coordinates. For example, multiple alternative paths can be determined based on the first offset and the second offset, and from the multiple alternative paths, an alternative path can be selected that can make the robotic arm move stably and smoothly based on the performance of the robotic arm. , or select an alternative path that can minimize the movement cost of the robotic arm, etc. The selected alternative path is the movement path. The moving cost of the robotic arm may be the wear and tear generated when the robotic arm moves along the alternative path.

Of course, the embodiment of the present application does not limit the movement method. The movement method may also include both movement speed and movement path, and may also include other contents besides movement speed and movement path, which are not limited here. In addition, after determining the first offset and the second offset, the electronic device can also send the first offset and the second offset to the robotic arm through the communication connection, and the robotic arm determines the first offset and the second offset. The second offset completes the movement on its own. In addition, the electronic device can also display the first offset and the second offset, and can also display the above coordinates according to actual needs, such as initial coordinates, reference starting point coordinates, reference end point coordinates, target starting point coordinates, and target end point coordinates.

No matter how it is controlled, the end of the robot arm can reach the target starting point coordinates. After that, you need to weld from the target starting point coordinates to the target end point coordinates. In an exemplary embodiment, the method further includes: obtaining the characteristic information of the welding place according to the target image, and determining the welding method according to the characteristic information, where the characteristic information includes at least one of connection form, connection situation and connection gap; controlling the mechanical arm. Welding the end from the target starting point coordinates to the target end point coordinates includes: controlling the end of the robotic arm to weld from the target starting point coordinates to the target end point coordinates in a welding manner.

Among them, by identifying the target image, the characteristic information of the welding joint can be obtained. According to the characteristic information, at least one suitable welding method can be adaptively determined, so that the welding from the target starting point coordinate to the target end point coordinate can be completed according to the welding method, thereby ensuring the welding quality. These characteristic information can be stored in the database corresponding to the identification of the welding point. When the characteristic information needs to be used, the characteristic information of the welding point can be obtained from the database according to the identification of the welding point.

Since the welding point is also called a joint, the connection form included in this feature information is also called the joint form, and the connection situation is also called the joint direction. Since a weld seam will be formed at the welding point, such as the weld seam shown in Figure 9, the connection gap included in the characteristic information is also called a weld seam gap. Figure 9 also shows an exemplary connection gap. For example, the connection forms include but are not limited to butt joints, corner joints, lap joints, etc., and the joint directions include but are not limited to the rising edge direction and the falling edge direction. The joint direction can be based on the outline of the joint in the specified direction shown in Figure 9 Sure.

For example, when the characteristic information includes a connection gap, see FIG. 10 , determining the welding mode according to the characteristic information may include: determining that the welding mode is pulse welding based on the connection gap being smaller than a first value; or, determining the welding mode as pulse welding based on the connection gap being larger than a first value. Or equal to the first value and less than the second value, it is determined that the welding method is swing welding, and the second value is greater than the first value. The embodiments of the present application do not limit the values of the first numerical value and the second numerical value. For example, the first numerical value is 2 mm, and the second numerical value is 4 mm.

Among them, when the connection gap is greater than or equal to the first value and less than the second value, it means that the connection gap is wide, so swing welding can be used, that is, the end of the robot arm swings in the above specified direction while welding. , in order to form a sufficiently wide weld to fill the wider connection gap and ensure the welding quality and welding success rate. Or, when the connection gap is smaller than the first value, it means that the connection gap is narrow, so there is no need to use swing welding, and pulse welding can be used directly. In addition, based on the fact that the connection gap is greater than the second value, it means that the connection gap is too wide and the welding difficulty is too great, so the welding can be stopped, and a manual welding can also be prompted.

Above, through steps 801 to 803, the process of welding a welding point is explained. If there are multiple continuous welding places, the above-mentioned steps 801 to 803 can be performed once for each welding place to complete the welding. In other words, if long, irregular metal materials need to be welded in some complex welding scenarios, the part to be welded can be divided into multiple welding locations, and the above step 801 is performed once for each welding location. to 803. Therefore, the welding method provided by the embodiment of the present application can also be applied to complex welding scenarios, thereby enhancing the applicability of the welding method.

To sum up, in the embodiment of the present application, the visual sensor is controlled to obtain the target image of the welding point, so the target starting point coordinates and the target end point coordinates can be accurately determined based on the target image, thereby achieving accurate identification of the position of the welding point. After that, the end of the robotic arm is controlled to be welded from the target starting point coordinates to the target end point coordinates, so that the welding point can be welded, ensuring the welding accuracy and improving the welding quality.

Next, the aluminum alloy automatic arc welding device proposed according to the embodiment of the present application will be described with reference to the accompanying drawings.

Figure 11 is a block diagram of an aluminum alloy automatic arc welding device according to an embodiment of the present application.

As shown in FIG. 11 , the aluminum alloy automatic arc welding device 10 includes: a collection module 100 , a calculation module 200 and a welding module 300 .

The collection module 100 is used to collect visual data during the automatic arc welding process of aluminum alloy.

The calculation module 200 is used to identify the actual weld position of the actual welded workpiece based on the visual data, calculate the actual weld gap of the current weld, and detect whether the actual weld gap is consistent with the initial weld gap before welding.

The welding module 300 is used to re-match the target welding parameters corresponding to the actual gap when it is detected that the actual weld gap is inconsistent with the initial weld gap, and guide the arc welding robot to move to a working position corresponding to the actual weld position, based on the target The welding parameters control the arc welding robot to perform welding actions on the actual welding workpiece.

Optionally, in one embodiment of the present application, the aluminum alloy automatic arc welding device 10 further includes: a conversion module.

Among them, the conversion module is used to perform coordinate conversion between the manipulator base coordinate system of the arc welding robot, the manipulator end tool coordinate system, and the line scanner coordinate system of the line structured light vision sensor installed on the arc welding robot to obtain the hand-eye calibration conversion relationship. .

Optionally, in one embodiment of the present application, the computing module 200 includes: a first acquisition unit and a second acquisition unit.

Among them, the first acquisition unit is used to acquire the welding seam starting point image and the welding seam end point image in the visual data.

The second acquisition unit is used to simulate the length and position of the entire weld based on the weld starting point image and the welding end point image, and obtain the actual welding seam position and the actual welding seam gap.

Optionally, in one embodiment of the present application, the welding module 300 includes: a search unit, a first calculation unit and a matching unit.

Among them, the search unit is used to find the position of the joint feature in the laser profile information.

The first calculation unit is used to calculate the three-dimensional coordinate position of the joint's characteristic position in the robot hand-based coordinate system based on the hand-eye calibration conversion relationship.

The matching unit is used to calculate the distance between the teaching position and the identification position based on the three-dimensional coordinate position, and match the target welding process based on the joint form and the gap between the plates to obtain the target welding parameters.

Optionally, in one embodiment of the present application, the welding module 300 further includes: a scanning unit, a second calculation unit and a welding unit.

Among them, the scanning unit is used to scan the actual welding seam position and obtain the coordinate value information of the welding seam starting point and the welding seam end point respectively.

The second calculation unit is used to calculate the TCP position coordinates of the arc welding robot based on the coordinate value information of the welding seam starting point and the welding seam end point.

The welding unit is used to guide the arc welding robot for welding based on TCP position coordinates.

It should be noted that the foregoing explanation of the embodiment of the aluminum alloy automatic arc welding method is also applicable to the aluminum alloy automatic arc welding device of this embodiment, and will not be described again here.

According to the aluminum alloy automatic arc welding device proposed in the embodiment of the present application, the visual data during the aluminum alloy automatic arc welding process can be collected, and then the actual weld position can be identified, the actual weld gap can be calculated, and the target welding parameters can be matched. According to the changes in the weld seam, the arc welding robot can automatically call the matching target welding parameters to guide the arc welding robot to perform welding actions at the corresponding position, so that the arc welding robot can still achieve uneven weld seam gaps caused by welding thermal deformation and accumulated tolerances. Finding the welding position accurately and quickly can effectively improve the automation level of the welding process and improve the welding quality. There is no need for manual assistance to detect the welding seam position and set parameters, thus reducing labor costs and eliminating accident losses caused by human judgment errors. This solves the problem in related technologies that due to the limited scanning of the weld position, the inability to control the weld changes in the welding process in real time makes the automatic arc welding unable to accurately and quickly find the changed weld position during the welding process, thus Technical issues that affect welding quality and increase labor costs.

Figure 12 is a schematic structural diagram of a welding device provided by an embodiment of the present application. The device is applied to electronic equipment included in the welding system. The welding system also includes a robotic arm and a visual sensor fixed to the robotic arm. Referring to Figure 12, the welding device includes the following modules.

The first control module 1201 is used to control the visual sensor to obtain the target image corresponding to the welding joint;

The acquisition module 1202 is used to obtain the target starting point coordinates and the target end point coordinates corresponding to the welding joint according to the target image;

The second control module 1203 is used to control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates.

In an exemplary embodiment, the target image is acquired by the vision sensor when the end of the robotic arm is located at initial coordinates, and the initial coordinates include initial starting point coordinates and initial end point coordinates;

The acquisition module 1202 is also used to determine the first offset between the initial starting point coordinates and the target starting point coordinates, and determine the second offset between the initial end point coordinates and the target end point coordinates; according to the first offset and the second The offset determines the movement mode, according to which the end of the robotic arm is controlled to move from the initial starting point coordinates to the target starting point coordinates.

In an exemplary embodiment, the target starting point coordinates and the target end point coordinates are coordinates in an end coordinate system corresponding to the end of the robotic arm;

The acquisition module 1202 is configured to convert the reference starting point coordinates into the target starting point coordinates according to the first transformation matrix between the terminal coordinate system and the sensor coordinate system, and convert the reference end point coordinates into the target end point coordinates according to the first transformation matrix.

In an exemplary embodiment, the acquisition module 1202 is also used to acquire the first coordinate and at least two second coordinates. The first coordinate is the coordinate of the center of the calibration element in the base coordinate system corresponding to the base of the robot arm, Any one of the at least two second coordinates corresponds to a pose, and any second coordinate is the coordinate of the center of the calibration element in the sensor coordinate system when the end of the robotic arm is in the corresponding pose; according to the first coordinate , at least two second coordinates, the pose corresponding to any second coordinate and the second transformation matrix between the end coordinate system and the base coordinate system, and the first transformation matrix is obtained by solving.

In an exemplary embodiment, the acquisition module 1202 is used to determine the area where the welding place is located from the target image; based on the area where the welding place is located includes reflective pixels, update the reflective pixels to non-reflective pixels based on other pixels around the reflective pixels, The updated target image is obtained, and the reflective pixels are the pixels corresponding to the reflective part of the welding joint; the target starting point coordinates and the target end point coordinates are obtained according to the updated target image.

In an exemplary embodiment, the acquisition module 1202 is also configured to obtain the characteristic information of the welding place according to the target image, and determine the welding method according to the characteristic information. The characteristic information includes at least one of connection form, connection situation and connection gap;

The second control module 1203 is used to control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates in a welding manner.

In an exemplary embodiment, the characteristic information includes a connection gap, and the acquisition module 1202 is configured to determine that the welding method is pulse welding based on the connection gap being less than a first value; or, based on the connection gap being greater than or equal to the first value and less than a second value. The value determines that the welding method is swing welding, and the second value is greater than the first value.

In the embodiment of the present application, the visual sensor is controlled to obtain the target image of the welding point, so the target starting point coordinates and the target end point coordinates can be accurately determined based on the target image, thereby achieving accurate identification of the position of the welding point. After that, the end of the robotic arm is controlled to be welded from the target starting point coordinates to the target end point coordinates, so that the welding point can be welded, ensuring the welding accuracy and improving the welding quality.

FIG. 13 is a schematic structural diagram of an example electronic device provided by an embodiment of the present application. The electronic device may include: a memory 701, a processor 702, and a computer program stored on the memory 701 and executable on the processor 702.

When the processor 702 executes the program, the aluminum alloy automatic arc welding method provided in the above embodiment is implemented. For example, when the processor 702 executes the computer program, the electronic device implements the aluminum alloy automatic arc welding method corresponding to Figures 1 and 5. Alternatively, when the processor 702 executes the computer program, the electronic device implements the welding method corresponding to FIG. 8 .

Further, the electronic device also includes: a communication interface 703 for communication between the memory 701 and the processor 702 .

The memory 701 may include high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.

If the memory 701, the processor 702 and the communication interface 703 are implemented independently, the communication interface 703, the memory 701 and the processor 702 can be connected to each other through a bus and complete communication with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of presentation, only one thick line is used in Figure 13, but it does not mean that there is only one bus or one type of bus.

Alternatively, if the memory 701, the processor 702 and the communication interface 703 are integrated and implemented on one chip, the memory 701, the processor 702 and the communication interface 703 can communicate with each other through the internal interface.

The processor 702 may be a central processing unit (Central Processing Unit, CPU for short), or an Application Specific Integrated Circuit (ASIC for short), or one or more processors configured to implement the embodiments of the present application. integrated circuit.

This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by the processor, the above aluminum alloy automatic arc welding method is implemented. For example, when the computer program stored on the computer-readable storage medium is executed by the processor, the electronic device implements the aluminum alloy automatic arc welding method corresponding to Figures 1 and 5, or implements the welding method corresponding to Figure 8.

In the description of this specification, reference to the terms "one embodiment," "some embodiments," "examples," or "some examples" or the like means that features, structures, materials, or characteristics are described in connection with the embodiment or example. Included in at least one embodiment or example of this application. In this specification, the schematic expressions of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the described features, structures, materials or characteristics may be combined in any suitable manner in any one or N embodiments or examples. Furthermore, those skilled in the art may combine and combine different embodiments or examples and features of different embodiments or examples described in this specification unless they are inconsistent with each other.

In addition, the terms “first” and “second” are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Therefore, features defined as "first" and "second" may explicitly or implicitly include at least one of these features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly limited.

Any process or method descriptions in flowcharts or otherwise described herein may be understood to represent modules, segments, or portions of code that include one or more executable instructions for implementing customized logical functions or steps of the process. , and the scope of the embodiments of the present application includes additional implementations in which functions may be performed out of the order shown or discussed, including in a substantially simultaneous manner or in the reverse order depending on the functionality involved, which should be It is understood by those skilled in the art to which the embodiments of the present application belong.

The logic and/or steps represented in the flowchart diagrams or otherwise described herein, for example, may be considered a sequenced list of executable instructions for implementing the logical functions, and may be implemented in any computer-readable medium to For use by, or in conjunction with, instruction execution systems, devices or devices (such as computer-based systems, systems including processors or other systems that can fetch instructions from and execute instructions). equipment for use. For the purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Examples (non-exhaustive list) of computer readable media include the following: electrical connections with one or N wires (electronic device), portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium may even be paper or other suitable medium on which the program may be printed, as the paper or other medium may be optically scanned, for example, and subsequently edited, interpreted, or otherwise suitable as necessary. process to obtain the program electronically and then store it in computer memory.

It should be understood that various parts of the present application can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if it is implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: discrete logic gate circuits with logic functions for implementing data signals; Logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

Those of ordinary skill in the art can understand that all or part of the steps involved in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The program can be stored in a computer-readable storage medium. When executed, one of the steps of the method embodiment or a combination thereof is included.

In addition, each functional unit in various embodiments of the present application can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software function modules. If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

The storage media mentioned above can be read-only memory, magnetic disks or optical disks, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above-mentioned embodiments are illustrative and cannot be understood as limitations of the present application. Those of ordinary skill in the art can make modifications to the above-mentioned embodiments within the scope of the present application. The embodiments are subject to changes, modifications, substitutions and variations.

Claims (10)

1. A welding method, characterized in that the method is applied to the electronic equipment included in the welding system, the welding system also includes a mechanical arm and a visual sensor fixed to the mechanical arm, the method includes:

   Control the vision sensor to obtain the target image corresponding to the welding point;

   Obtain the target starting point coordinates and target end point coordinates corresponding to the welding point according to the target image;

   Control the end of the robotic arm to be welded from the target starting point coordinates to the target end point coordinates.
2. The method according to claim 1, wherein the target image is acquired by the visual sensor when the end of the robotic arm is located at initial coordinates, and the initial coordinates include initial starting point coordinates and initial end point coordinates;

   The method also includes:

   Determine a first offset between the initial starting point coordinates and the target starting point coordinates, and determine a second offset between the initial end point coordinates and the target end point coordinates;

   A movement mode is determined according to the first offset amount and the second offset amount, and the end of the robotic arm is controlled to move from the initial starting point coordinates to the target starting point coordinates according to the movement mode.
3. The method according to claim 1, characterized in that the target starting point coordinates and the target end point coordinates are coordinates in an end coordinate system corresponding to the end of the robotic arm;

   Obtaining the target starting point coordinates and target end point coordinates corresponding to the welding point according to the target image includes:

   Obtain the reference starting point coordinates and the reference end point coordinates in the sensor coordinate system corresponding to the visual sensor according to the target image;

   The reference starting point coordinates are converted into the target starting point coordinates according to the first transformation matrix between the terminal coordinate system and the sensor coordinate system, and the reference end point coordinates are converted into the target end point according to the first transformation matrix. coordinate.
4. The method of claim 3, further comprising:

   Obtain a first coordinate and at least two second coordinates. The first coordinate is the coordinate of the center of the calibration element in the base coordinate system corresponding to the base of the robotic arm. Among the at least two second coordinates, Any second coordinate corresponds to a pose, and the any second coordinate is the coordinate of the center of the calibration element in the sensor coordinate system when the end of the robotic arm is in the corresponding pose;

   According to the first coordinate, the at least two second coordinates, the posture corresponding to any one of the second coordinates, and the second transformation matrix between the end coordinate system and the base coordinate system, the solution is the first transformation matrix.
5. The method according to any one of claims 1 to 4, characterized in that, obtaining the target starting point coordinates and target end point coordinates corresponding to the welding point according to the target image includes:

   Determine the area where the weld is located from the target image;

   Based on the fact that the area where the welding point is located includes reflective pixels, the reflective pixels are updated to non-reflective pixels according to other pixels around the reflective pixel, and an updated target image is obtained. The reflective pixels are the reflection of the welding point. Partially corresponding pixels;

   The target starting point coordinates and the target end point coordinates are obtained according to the updated target image.
6. The method according to any one of claims 1-4, characterized in that,

   The method further includes: obtaining characteristic information of the welding place according to the target image, and determining a welding method according to the characteristic information, where the characteristic information includes at least one of connection form, connection situation and connection gap;

   Controlling the end of the robotic arm to be welded from the target starting point coordinate to the target end point coordinate includes: controlling the end of the robotic arm to be welded from the target starting point coordinate to the target end point coordinate according to the welding method. .
7. The method according to claim 6, wherein the characteristic information includes the connection gap, and determining the welding method according to the characteristic information includes:

   Based on the connection gap being less than the first value, it is determined that the welding method is pulse welding;

   Alternatively, it is determined that the welding method is swing welding based on the fact that the connection gap is greater than or equal to the first value and less than a second value, and the second value is greater than the first value.
8. A welding device, characterized in that the device is applied to electronic equipment included in the welding system. The welding system also includes a robotic arm and a visual sensor fixed to the robotic arm. The device includes:

   The first control module is used to control the visual sensor to obtain the target image corresponding to the welding joint;

   An acquisition module, configured to acquire the target starting point coordinates and the target end point coordinates corresponding to the welding point according to the target image;

   The second control module is used to control the end of the robotic arm to be welded from the target starting point coordinate to the target end point coordinate.
9. An electronic device, characterized in that it includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program so that the electronic device The equipment implements the welding method according to any one of claims 1-7.
10. A computer-readable storage medium with a computer program stored thereon, characterized in that the computer program is executed by a processor, so that the electronic device is used to implement the welding method according to any one of claims 1-7.

Images