Photovoltaic welding robot and photovoltaic welding method

Abstract

The present invention discloses a photovoltaic welding robot and a photovoltaic welding method, wherein the welding robot comprises a point cloud acquisition module for acquiring a point cloud map of a photovoltaic installation site and a point cloud map of welding positions to be welded on each target; a navigation control module for planning the action path of the photovoltaic welding robot according to the point cloud map, and driving the photovoltaic welding robot to move to the target position through a driving motor; a first visual image acquisition module for acquiring a welding visual image and a grabbing visual image; a welding control module for generating a welding control instruction and a grabbing control instruction respectively; and a welding control module for controlling the photovoltaic welding robot to perform welding and grabbing operations according to the welding control instruction and the grabbing control instruction, and then welding and installing the target. The present invention can realize the automatic welding process of the photovoltaic site, reduce the labor cost in the photovoltaic site welding process, and improve the welding quality of the photovoltaic site.

Description

Technical Field

[0001] The invention relates to the technical field of photovoltaic installation, and in particular to a photovoltaic welding robot and a photovoltaic welding method. Background Art

[0002] Since the construction of photovoltaic power stations needs to take into account many factors such as the duration of sunlight, terrain, and climate, photovoltaic power stations with high power generation efficiency are built in relatively remote areas. During the construction of photovoltaic power stations, a large number of PHC pipe piles and photovoltaic brackets need to be welded. These welding works have high requirements for welding quality, and because photovoltaic power stations are located in remote areas, the labor cost has increased significantly. In addition, mistakes are prone to occur during manual welding. If there is a problem with the welding quality at a certain point, it may chain-affect the power generation efficiency of the entire photovoltaic matrix.

[0003] Therefore, there is a need for a photovoltaic welding robot that can intelligently identify the welding position at the photovoltaic site and perform automatic welding, so as to standardize the welding process, welding technology and welding quality at the photovoltaic site, reduce the labor cost during the photovoltaic welding process, and improve the welding quality at the photovoltaic site. Summary of the invention

[0004] In order to solve the technical problems that the welding process in the current photovoltaic installation process is not standardized and requires a lot of manpower costs, the present invention provides a photovoltaic welding robot, and the specific technical solution is as follows:

[0005] The present invention provides a photovoltaic welding robot, comprising a welding mechanical arm for performing welding operations and a grasping mechanical arm for performing grasping operations, comprising:

[0006] A point cloud acquisition module, used to acquire a point cloud map of a photovoltaic installation site and a point cloud map of welding locations to be welded on each target in the photovoltaic installation site;

[0007] a navigation control module, connected to the point cloud acquisition module, for receiving the point cloud map sent by the point cloud acquisition module, planning the action path of the photovoltaic welding robot according to the point cloud map, and driving the photovoltaic welding robot to a target position through a driving motor according to the action path;

[0008] A first visual image acquisition module, used to acquire a welding visual image of the position to be welded at an angle of the welding robot arm and to acquire a grabbing visual image of the position to be welded at an angle of the grabbing robot arm;

[0009] a welding control module, connected to the point cloud acquisition module and the first visual image acquisition module, respectively, for receiving the welding point cloud image sent by the point cloud acquisition module, and the welding visual image and the captured visual image sent by the first visual image acquisition module, and generating a welding control instruction according to the welding visual image and the welding point cloud image, respectively, and generating a capture control instruction according to the captured visual image and the welding point cloud image;

[0010] The welding control module controls the welding robot arm to perform welding operations according to the welding control instructions, and controls the grabbing robot arm to perform grabbing operations according to the grabbing control instructions, and then performs welding installation on the position to be welded.

[0011] The photovoltaic welding robot provided by the present invention generates corresponding control instructions according to the point cloud map obtained by the point cloud acquisition module and the point cloud map of the welding position, and controls the photovoltaic welding robot to automatically navigate to the position to be installed and automatically perform the welding grasping and installation operation during the welding installation process of the position to be welded, thereby realizing the standardized welding and installation operation process of photovoltaic welding, reducing the labor cost in the photovoltaic on-site welding process, reducing the welding and installation errors caused by human errors, and improving the welding quality of the photovoltaic on-site.

[0012] In some embodiments, the first visual image acquisition module acquires a visual image of the welding result of the position to be welded after completing the welding installation;

[0013] The welding control module is also equipped with a welding quality detection model. After receiving the welding result visual image sent by the first visual image acquisition module, the welding control module evaluates the welding quality of the position to be welded according to the welding quality detection model and the welding result visual image. The welding quality detection model generates welding quality parameters according to the welding result visual image.

[0014] The welding quality detection model is generated by deep learning training based on a welding result visual image dataset marked with the welding quality parameters.

[0015] The photovoltaic welding robot provided by the present invention obtains the visual image of the welding result of the position to be welded after the welding installation is completed through the first visual image acquisition module, and combines with the welding quality detection model carried by the welding control module to realize the detection of welding quality after welding is completed, automatically evaluates the welding quality after the welding robot operates, and improves the intelligence of the welding work of the photovoltaic welding robot.

[0016] In some embodiments, the point cloud acquisition module includes a laser radar installed on the navigation chassis of the photovoltaic welding robot; the laser radar is used to scan the photovoltaic installation site before the photovoltaic welding robot performs navigation operations, and perform SLAM mapping based on the scanning results to obtain the point cloud map;

[0017] The laser radar is also used to scan the position to be welded before the photovoltaic welding robot performs the welding operation, and perform SLAM mapping based on the scanning results to obtain a point cloud map of the welding position.

[0018] The photovoltaic welding robot provided by the present invention scans and maps the photovoltaic installation site through a laser radar arranged in a point cloud acquisition module, so as to facilitate the control of the photovoltaic welding robot to navigate and move according to the actual situation of the photovoltaic installation site and the target installation position of the photovoltaic site.

[0019] In some embodiments, the photovoltaic welding robot provided by the present invention further includes:

[0020] A second visual image acquisition module, connected to the navigation control module, wherein the second visual image acquisition module has a built-in environmental visual camera for acquiring an environmental visual image of the photovoltaic installation site;

[0021] The navigation control module is further used to receive the environmental visual image sent by the second visual image acquisition module, and identify the to-be-welded position and to-be-welded category of each of the targets in the environmental visual image through the target detection model carried by the navigation control module; the target detection model is generated by deep learning training based on the visual image data set in which the to-be-welded position and the to-be-welded category are pre-labeled;

[0022] The navigation control module divides the positions to be welded into several welding groups according to the categories to be welded, plans the corresponding sub-action paths of the photovoltaic welding robot when operating in each welding group according to the positions to be welded, and composes the action path with each sub-action path.

[0023] The photovoltaic welding robot provided by the present invention acquires the visual image of the photovoltaic installation site through the environmental vision camera, and intelligently identifies the position to be welded and the category to be welded in the visual image in combination with the target detection model, so as to facilitate the control of the photovoltaic welding robot to navigate and move according to the visual image.

[0024] In some embodiments, the navigation control module is further used to mark the location to be welded and the category to be welded on the point cloud map in combination with the environmental visual image.

[0025] The photovoltaic welding robot provided by the present invention combines the point cloud map and the environmental visual image to comprehensively plan the action path of the photovoltaic welding robot, thereby improving the accuracy of the photovoltaic welding robot in navigating to the target installation position on the photovoltaic site during the photovoltaic welding process.

[0026] In some embodiments, the welding robot arm is used to weld the target according to the position to be welded, and a 360-degree steering axis is provided at the connection portion between the welding robot arm and the photovoltaic welding robot, and at least two steering portions are provided on the welding robot arm to realize at least two rotational degrees of freedom of the welding robot arm;

[0027] The working part of the welding robot arm is provided with welding equipment, and the welding equipment includes fusion welding equipment, pressure welding equipment and brazing equipment.

[0028] The first visual image acquisition module includes a welding portion visual camera disposed on the working portion of the welding robot arm, and is used to acquire the welding visual image and the welding result visual image;

[0029] The welding control module is equipped with a welding position detection model, the welding position detection model generates welding position spatial coordinates according to the welding visual image and the welding point cloud image, and the welding control module generates the welding control instruction according to the welding position spatial coordinates;

[0030] The welding position detection model is generated by deep learning training based on a data set including the corresponding relationship between the welding position spatial coordinates, the welding point cloud image and the welding visual image.

[0031] In some embodiments, the grabbing robot arm is used to grab the target and fix the target at the position to be welded according to the grabbing control instruction during the welding installation process of the position to be welded;

[0032] The first visual image acquisition module further includes a grasping portion visual camera disposed on the working portion of the grasping mechanical arm, for acquiring the grasping visual image;

[0033] The welding control module is also equipped with a grasping position detection model, the grasping position detection model generates grasping position spatial coordinates according to the grasping visual image and the welding point cloud image, and the welding control module generates the grasping control instruction according to the grasping position spatial coordinates;

[0034] The grasping position detection model is generated by deep learning training based on a data set including the grasping control instructions, the grasping position spatial coordinates and the corresponding relationship between the grasping visual image.

[0035] The photovoltaic welding robot provided by the present invention obtains a grasping visual image of the position to be welded at the photovoltaic site by arranging a grasping part visual camera on the grasping mechanical arm, and intelligently generates control instructions for the grasping mechanical arm in combination with a grasping position detection model carried in the control module, thereby achieving the effect of intelligently controlling the grasping mechanical arm to fix the target at the position to be welded during the welding process of the target, thereby improving the accuracy of the automatic welding installation process at the photovoltaic site.

[0036] In some embodiments, the navigation control module is further used to receive an externally input navigation remote control command, and generate the action path of the photovoltaic welding robot according to the navigation remote control command;

[0037] The welding control module is also used to receive externally input welding remote control instructions and grabbing remote control instructions, and control the welding robot arm to perform corresponding welding operations according to the welding remote control instructions and the grabbing remote control instructions.

[0038] The photovoltaic welding robot provided by the present invention receives remote control instructions from the user during the photovoltaic welding installation process, and controls the photovoltaic welding robot in real time to perform corresponding operations, so that the user can adjust the operation status of the photovoltaic welding robot in time according to the actual situation at the installation site.

[0039] In some embodiments, the welding control module generates a folding control instruction after the welding installation is completed, and controls the welding robot arm and the grasping robot arm to fold to a preset state after the welding installation is completed according to the folding control instruction.

[0040] After the photovoltaic welding robot provided by the present invention completes the welding installation operation, the welding mechanical arm and the grasping mechanical arm are folded and retracted, thereby reducing the probability of collision and friction between the mechanical arm and environmental facilities when the photovoltaic welding robot moves to the next target installation position to be welded, reducing the resistance of the photovoltaic welding robot during movement, and avoiding accidents such as rollover or scratching of the photovoltaic welding robot during movement, which affects the overall photovoltaic installation efficiency.

[0041] In some embodiments, according to another aspect of the present invention, the present invention further provides a photovoltaic welding method, which is applied to a photovoltaic welding robot including a welding mechanical arm for performing a welding operation and a grasping mechanical arm for performing a grasping operation, comprising the steps of:

[0042] Obtain point cloud maps of photovoltaic installation sites;

[0043] Planning the action path of the photovoltaic welding robot according to the point cloud map;

[0044] According to the action path, the photovoltaic welding robot is driven to move to a target position by a driving motor;

[0045] Acquire a point cloud image of a welding position of a position to be welded, and respectively acquire a welding visual image of the position to be welded at the angle of the welding robot arm, and acquire a grabbing visual image of the position to be welded at the angle of the grabbing robot arm;

[0046] Generate welding control instructions according to the welding visual image and the welding point cloud image respectively, and generate grabbing control instructions according to the grabbing visual image and the welding point cloud image;

[0047] After the welding robot arm is controlled to perform welding operations according to the welding control instructions and the grabbing robot arm is controlled to perform grabbing operations according to the grabbing control instructions, welding and installation are performed on the position to be welded.

[0048] The present invention provides a photovoltaic welding robot and a photovoltaic welding method, which have at least one of the following technical effects:

[0049] (1) Generate corresponding control instructions based on the point cloud map and welding point cloud map obtained by the point cloud acquisition module to control the photovoltaic welding robot to automatically navigate to the installation location and automatically perform welding grabbing and installation operations during the photovoltaic welding installation process, thereby realizing the standardized welding and installation operation process at the photovoltaic site, reducing the labor cost during the photovoltaic site welding process, reducing the welding installation errors caused by human errors, and improving the welding quality at the photovoltaic site;

[0050] (2) The first visual image acquisition module is used to obtain the visual image of the welding result of the position to be welded after the welding installation is completed, and the welding quality detection model carried by the welding control module is combined to realize the detection of welding quality after welding is completed, and the welding quality after the welding robot is operated is automatically evaluated, thereby improving the intelligent welding work of the photovoltaic welding robot;

[0051] (3) The laser radar set in the point cloud acquisition module is used to scan and map the photovoltaic installation site, so as to control the photovoltaic welding robot to navigate and move according to the actual situation of the photovoltaic installation site and the target installation location of the photovoltaic site;

[0052] (4) The visual image of the photovoltaic installation site is obtained through the environmental vision camera, and the target detection model is combined to intelligently identify the welding position and welding category in the visual image, so as to control the photovoltaic welding robot to navigate and move according to the visual image;

[0053] (5) Combine the point cloud map and the environmental visual image to comprehensively plan the action path of the photovoltaic welding robot, and improve the accuracy of the photovoltaic welding robot navigating to the photovoltaic target installation position during the photovoltaic welding process;

[0054] (6) By setting a grasping part visual camera on the grasping robot arm, the grasping visual image of the target to be welded at the photovoltaic site is obtained, and combined with the grasping position detection model carried by the control module, the control instructions of the grasping robot arm are intelligently generated, so as to achieve the effect of intelligently controlling the grasping robot arm to fix the target at the position to be welded during the welding process at the photovoltaic site, thereby improving the accuracy of the photovoltaic automatic welding installation process;

[0055] (7) Receive remote control instructions from the user during the photovoltaic welding installation process, and control the photovoltaic welding robot in real time to perform corresponding operations, so that the user can adjust the operation of the photovoltaic welding robot in time according to the actual situation at the installation site;

[0056] (8) After completing the welding and installation work, the welding robot arm and the grasping robot arm are folded and retracted to reduce the probability of collision and friction between the robot arm and environmental facilities when the photovoltaic welding robot moves to the next target installation position to be welded, reduce the resistance of the photovoltaic welding robot during movement, and avoid accidents such as rollover or scratching of the photovoltaic welding robot during movement, which may affect the efficiency of photovoltaic installation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0058] Figure 1 This is an example diagram of a photovoltaic welding robot according to the present invention;

[0059] Figure 2 This is another exemplary diagram of a photovoltaic welding robot according to the present invention;

[0060] Figure 3 The present invention is a flow chart of a photovoltaic welding method.

[0061] Numbers in the figure: point cloud acquisition module-10, laser radar-11, navigation control module-20, first visual image acquisition module-30, welding part visual camera-31, grabbing part visual camera-32, welding control module-40, second visual image acquisition module-50 and environmental visual camera-51. DETAILED DESCRIPTION

[0062] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0063] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections.

[0064] In order to simplify the drawings, only the parts related to the present invention are schematically shown in each figure, and they do not represent the actual structure of the product. In addition, in order to simplify the drawings and facilitate understanding, in some figures, only one of the parts with the same structure or function is schematically drawn or marked. In this article, "one" not only means "only one", but also means "more than one".

[0065] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0066] In addition, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0067] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings and other implementation methods can be obtained based on these drawings without creative work.

[0068] One embodiment of the present invention, as Figure 1 As shown, the present invention provides a photovoltaic welding robot, including a welding robot arm for performing welding operations and a grasping robot arm for performing grasping operations, and also includes a point cloud acquisition module 10, a navigation control module 20, a first visual image acquisition module 30 and a welding control module 40.

[0069] Among them, after arriving at the installation site, the point cloud acquisition module 10 obtains the point cloud map of the photovoltaic installation site and the point cloud map of the welding positions to be welded on each target in the photovoltaic installation site; the navigation control module 20 is connected to the point cloud acquisition module 10, and is used to receive the point cloud map sent by the point cloud acquisition module 10, plan the action path of the photovoltaic welding robot according to the point cloud map, and drive the photovoltaic welding robot to the target position through the drive motor according to the action path.

[0070] Specifically, the navigation control module 20 is installed on the navigation chassis of the photovoltaic welding robot. The navigation chassis also has a built-in direction drive motor for controlling the movement of the photovoltaic welding robot, and a built-in height drive motor for controlling the photovoltaic welding robot to rise or fall in the vertical direction, so that the photovoltaic welding robot can perform photovoltaic welding installation according to different installation conditions.

[0071] The first visual image acquisition module 30 is used to acquire the welding visual image of the position to be welded at the angle of the welding robot arm and to acquire the grasping visual image of the position to be welded at the angle of the grasping robot arm; the welding control module 40 is connected to the point cloud acquisition module 10 and the first visual image acquisition module 30 respectively, and is used to receive the point cloud image of the welding point sent by the point cloud acquisition module 10, and the welding visual image and the grasping visual image sent by the first visual image acquisition module 30, and respectively generate welding control instructions according to the welding visual image and the point cloud image of the welding point, and generate grasping control instructions according to the grasping visual image and the point cloud image of the welding point; the welding control module 40 controls the welding robot arm to perform the welding operation according to the welding control instruction, and controls the grasping robot arm to perform the grasping operation according to the grasping control instruction, and then performs welding and installation on the position to be welded.

[0072] For example, during the welding installation process of the position to be welded, the welding control module 40 is required to weld the column top seat of the photovoltaic support to the PHC pile, so as to realize the fixed support function of the PHC pile on the photovoltaic support.

[0073] Specifically, in the process of the welding control module 40 generating welding control instructions, the welding visual image of the position to be welded taken by the visual camera in the first visual image acquisition module 30 is combined to identify the spatial coordinates of the position to be welded in the x and y directions in the spatial coordinate system, and the welding point cloud image of the position to be welded collected by the laser radar in the point cloud acquisition module 10 is combined to identify the spatial coordinates of the position to be welded in the z direction in the spatial coordinate system, and the spatial position coordinates of the position to be welded in the spatial coordinate system are comprehensively obtained, and corresponding welding control instructions and grasping control instructions are generated according to the spatial position coordinates.

[0074] The photovoltaic welding robot provided in this embodiment generates corresponding control instructions based on the point cloud map obtained by the point cloud acquisition module and the point cloud map of the welding point, and controls the photovoltaic welding robot to automatically navigate to the installation position and automatically perform welding grasping and installation operations during the photovoltaic welding installation process, thereby realizing the standardized photovoltaic welding and installation operation process, reducing the labor cost in the photovoltaic welding process, reducing the welding installation errors caused by human errors, and improving the photovoltaic welding quality.

[0075] In one embodiment, the first visual image acquisition module 30 acquires a visual image of the welding result of the position to be welded after completing the welding installation; the welding control module 40 is also equipped with a welding quality detection model. After the welding control module 40 receives the visual image of the welding result sent by the first visual image acquisition module 30, the welding quality of the position to be welded is evaluated according to the welding quality detection model and the visual image of the welding result, and the welding quality detection model generates welding quality parameters according to the visual image of the welding result; the welding quality detection model is generated by deep learning training based on a welding result visual image data set marked with welding quality parameters.

[0076] The photovoltaic welding robot provided in this embodiment obtains the visual image of the welding result of the position to be welded after the welding installation is completed through the first visual image acquisition module, and realizes the detection of welding quality after welding is completed in combination with the welding quality detection model carried by the welding control module, automatically evaluates the welding quality after the welding robot operates, and improves the intelligence of the welding work of the photovoltaic welding robot.

[0077] In one embodiment, Figure 2 As shown, the point cloud acquisition module 10 includes a laser radar 11, which is installed on the navigation chassis of the photovoltaic welding robot; the laser radar 11 is used to scan the photovoltaic installation site before the photovoltaic welding robot performs navigation operations, and perform SLAM (Simultaneous Localization And Mapping) mapping based on the scanning results to obtain a point cloud map.

[0078] Furthermore, the navigation control module 20 also includes a GPS device. After the position of the PHC pipe pile is determined according to the point cloud map each time, the position of the PHC pipe pile is imported into the GPS map. When the laser radar 11 is interfered by environmental signals and the point cloud quality in the acquired point cloud map is low, it is switched to the GPS device. The navigation control module 20 assists the photovoltaic welding robot in navigating the current welding work based on the historical PHC pipe pile position information stored in the GPS map.

[0079] The laser radar 11 is also used to scan the position to be welded before the photovoltaic welding robot performs the welding operation, and perform SLAM mapping based on the scanning results to obtain a point cloud map of the welding position.

[0080] The photovoltaic welding robot provided in this embodiment scans and maps the photovoltaic installation site through a laser radar set in a point cloud acquisition module, which facilitates the control of the photovoltaic welding robot to navigate and move according to the actual situation of the photovoltaic installation site and the target installation position to be welded.

[0081] In one embodiment, Figure 2 As shown, the photovoltaic welding robot provided by the present invention also includes a second visual image acquisition module 50, which is connected to the navigation control module 20. The second visual image acquisition module 50 has a built-in environmental visual camera 51 for acquiring environmental visual images of the photovoltaic installation site.

[0082] The navigation control module 20 is also used to receive the environmental visual image sent by the second visual image acquisition module 50, and identify the to-be-welded position and to-be-welded category of each target in the environmental visual image through the target detection model carried by the navigation control module 20; the target detection model is generated by deep learning training based on the visual image data set that pre-marks the to-be-welded position and to-be-welded category.

[0083] The navigation control module 20 divides the positions to be welded into several welding groups according to the types of welding to be welded, plans the corresponding sub-action paths of the photovoltaic welding robot when operating in each welding group according to the positions to be welded, and composes the sub-action paths into an action path.

[0084] Exemplarily, in the process of marking the positions to be welded and the categories to be welded in the visual image, the positions of the welding points between the PHC pipe piles and the top seats of the photovoltaic bracket columns can be marked, and the categories to be welded that need to be welded can also be marked, for example, welding the top seats of the photovoltaic bracket columns and the PHC, welding the internal structure of the top seats of the photovoltaic bracket columns, or welding the sensors and the top seats of the photovoltaic bracket columns, etc. The positions to be welded can also be divided into several welding groups according to the categories to be welded (Class A PHC pipe piles, Class B PHC pipe piles, and Class C PHC pipe piles), and the welding of Class A PHC pipe piles, Class B PHC pipe piles, and Class C PHC pipe piles can be completed in turn, sub-action paths can be generated in each type of PHC pipe pile welding group, and each sub-action path can be integrated into an action path according to the operation execution order of each welding group.

[0085] The photovoltaic welding robot provided by the present invention acquires the visual image of the photovoltaic installation site through the environmental vision camera, and intelligently identifies the position to be welded and the category to be welded in the visual image in combination with the target detection model, so as to facilitate the control of the photovoltaic welding robot to navigate and move according to the visual image.

[0086] In one embodiment, the navigation control module 20 is further used to mark the location to be welded and the category to be welded on the point cloud map in combination with the environmental visual image.

[0087] The photovoltaic welding robot provided by the present invention combines the point cloud map and the environmental visual image to comprehensively plan the action path of the photovoltaic welding robot, thereby improving the accuracy of the photovoltaic welding robot in navigating to the photovoltaic target installation position during the photovoltaic welding process.

[0088] In one embodiment, Figure 2 As shown, the welding robot arm is used to weld the target according to the position to be welded. The connection part between the welding robot arm and the photovoltaic welding robot is provided with a 360-degree steering axis, and at least two steering parts are provided on the welding robot arm to realize that the welding robot arm has at least two rotational degrees of freedom; the working part of the welding robot arm is provided with welding equipment, and the welding equipment includes fusion welding equipment, pressure welding equipment and brazing equipment.

[0089] The first visual image acquisition module 30 includes a welding portion visual camera 31 disposed on the working portion of the welding robot arm, and is used to acquire a welding visual image and a welding result visual image;

[0090] The welding control module 40 is equipped with a welding position detection model. The welding position detection model generates welding position spatial coordinates based on the welding visual image and the welding point cloud map. The welding control module 40 then generates welding control instructions based on the welding position spatial coordinates. The welding position detection model is generated by deep learning training based on a data set including the corresponding relationship between the welding position spatial coordinates, the welding point cloud map and the welding visual image.

[0091] Specifically, the welding robot arm can select a commonly used industrial robot arm to realize the above functions. The welding position detection model generates the welding position spatial coordinates according to the welding visual image and the welding point cloud map. The welding control module 40 then generates the welding control instructions according to the welding position spatial coordinates. The welding position detection model is generated by deep learning training based on a data set including the correspondence between the welding position spatial coordinates, the welding point cloud map and the welding visual image. The welding quality detection model generates welding quality parameters according to the welding result visual image. The welding quality detection model is generated by deep learning training based on the welding result visual image data set marked with welding quality parameters.

[0092] Furthermore, the welding control module 40 evaluates the welding quality of the welding position to be welded according to the welding quality parameter and the preset welding quality threshold, and judges the welding quality to be qualified when the welding quality parameter is greater than the welding quality threshold, and judges the welding quality to be unqualified when the welding quality parameter is not greater than the welding quality threshold.

[0093] The photovoltaic welding robot provided in this embodiment arranges a welding part visual camera on the working part of the welding robot arm to obtain the welding visual image of the position to be welded during the welding process and the welding result visual image of the position to be welded after the welding installation is completed, and combines the preset algorithm in the control module to realize the precise control of the welding robot arm during the welding process and the detection of the welding quality after the welding is completed, thereby improving the intelligence of the welding work of the photovoltaic welding robot.

[0094] In one embodiment, Figure 2 As shown, the grabbing robot arm is used to grab the target and fix the target at the welding position according to the grabbing control instruction during the welding and installation process of the target; the first visual image acquisition module 30 also includes a grabbing part visual camera 32 arranged on the working part of the grabbing robot arm, and the grabbing part visual camera 32 is used to acquire the grabbing visual image; the welding control module 40 is also equipped with a grabbing position detection model, and the grabbing position detection model generates the grabbing position spatial coordinates according to the grabbing visual image and the point cloud map of the welding point, and the welding control module 40 then generates the grabbing control instruction according to the grabbing position spatial coordinates; the grabbing position detection model is generated by deep learning training based on a data set including the corresponding relationship between the grabbing control instruction, the grabbing position spatial coordinates and the grabbing visual image.

[0095] Exemplarily, during the welding and installation operation of the photovoltaic welding robot, the welding robot arm and the grasping robot arm are first used to obtain a grasping visual image by a grasping part visual camera, and a grasping position spatial coordinates are generated according to the grasping visual image and the welding point cloud map through a grasping position detection model. The welding control module 40 then generates a grasping control instruction according to the grasping position spatial coordinates, and controls the grasping robot arm to fix the column top seat of the photovoltaic bracket at the welding position of the PHC pipe pile according to the grasping control instruction. Then, the welding part visual camera 31 obtains the welding visual image of the welding position, and obtains the welding control instruction according to the welding position detection model. The welding robot arm welds the PHC pipe pile and the column top seat of the photovoltaic bracket according to the welding control instruction. After the welding is completed, the welding part visual camera 31 obtains the welding result visual image of the welding position to be welded, and the welding control module 40 determines whether the welding result is qualified according to the welding quality detection model, wherein the grasping position detection model is generated by deep learning training based on a data set including the correspondence between the grasping control instruction, the grasping position spatial coordinates and the grasping visual image.

[0096] The photovoltaic welding robot provided in this embodiment obtains a grasping visual image of the target's position to be welded by setting a grasping part visual camera on the grasping robot arm, and intelligently generates control instructions for the grasping robot arm in combination with the grasping position detection model carried in the welding control module 40, thereby achieving the effect of intelligently controlling the grasping robot arm to fix the target at the position to be welded during the welding process at the photovoltaic site, thereby improving the accuracy of the automatic welding installation process at the photovoltaic site.

[0097] In one embodiment, the navigation control module 20 is also used to receive externally input navigation remote control instructions, and generate the action path of the photovoltaic welding robot according to the navigation remote control instructions; the welding control module 20 is also used to receive externally input welding remote control instructions and grasping remote control instructions, and control the welding robot arm to perform corresponding welding operations according to the welding remote control instructions and the grasping remote control instructions.

[0098] The photovoltaic welding robot provided in this embodiment receives remote control commands from the user during the photovoltaic on-site welding installation process, and controls the photovoltaic welding robot in real time to perform corresponding operations, so that the user can adjust the operating conditions of the photovoltaic welding robot in time according to the actual conditions of the installation site.

[0099] In one embodiment, the welding control module 40 generates a folding control instruction after the welding installation is completed, and controls the welding robot arm and the grabbing robot arm to fold to a preset state according to the folding control instruction after the welding installation is completed.

[0100] After the photovoltaic welding robot provided in this embodiment completes the welding installation work, the welding robot arm and the grasping robot arm are folded and retracted, thereby reducing the probability of collision and friction between the robot arm and environmental facilities when the photovoltaic welding robot moves to the next target installation position to be welded, reducing the resistance of the photovoltaic welding robot during movement, and avoiding accidents such as rollover or scratching of the photovoltaic welding robot during movement, which affects the overall photovoltaic installation efficiency.

[0101] In one embodiment, according to another aspect of the present invention, the present invention further provides a photovoltaic welding method, which is applied to a photovoltaic welding robot including a welding mechanical arm for performing a welding operation and a grasping mechanical arm for performing a grasping operation, comprising the steps of:

[0102] S100 obtains point cloud maps of photovoltaic installation sites.

[0103] S200 plans the action path of the photovoltaic welding robot based on the point cloud map.

[0104] S300 drives the photovoltaic welding robot to move to the target position through the drive motor according to the action path.

[0105] S400 obtains a point cloud image of the welding position to be welded, and obtains a welding visual image of the welding position to be welded at an angle of a welding robot arm, and obtains a grabbing visual image of the welding position to be welded at an angle of a grabbing robot arm.

[0106] S500 generates welding control instructions according to the welding visual image and the welding point cloud image, and generates grabbing control instructions according to the grabbing visual image and the welding point cloud image.

[0107] S600 controls the welding robot arm to perform welding operations according to the welding control instructions, and controls the grasping robot arm to perform grasping operations according to the grasping control instructions, and then welds and installs the target.

[0108] Specifically, the navigation chassis of the photovoltaic welding robot has a built-in direction drive motor to control the movement of the photovoltaic welding robot, and also has a built-in height drive motor to control the photovoltaic welding robot to rise or lower in the vertical direction, so that the photovoltaic welding robot can weld and install the target according to different installation conditions.

[0109] The photovoltaic welding robot obtains a welding visual image of the position to be welded at the angle of the welding robot arm and a grasping visual image of the position to be welded at the angle of the grasping robot arm, and generates welding control instructions according to the welding visual image and the point cloud image of the welding position, and generates grasping control instructions according to the grasping visual image and the point cloud image of the welding position.

[0110] The photovoltaic welding robot controls the welding robot arm to perform welding operations according to the welding control instructions, and controls the grasping robot arm to perform grasping operations according to the grasping control instructions, and then welds and installs the target.

[0111] Exemplarily, during the welding and installation process of the target, it is necessary to weld the top seat of the column of the photovoltaic bracket to the PHC pipe pile to realize the fixed support effect of the PHC pipe pile on the photovoltaic bracket. In the process of generating welding control instructions, the welding visual image of the position to be welded taken by the visual camera is combined to identify the spatial coordinates of the position to be welded in the x and y directions in the spatial coordinate system, and the welding point cloud image of the position to be welded collected by the laser radar is combined to identify the spatial coordinates of the position to be welded in the z direction in the spatial coordinate system, and the spatial position coordinates of the position to be welded in the spatial coordinate system are obtained comprehensively, and corresponding welding control instructions and grasping control instructions are generated according to the spatial position coordinates.

[0112] The photovoltaic welding method provided in this embodiment generates corresponding control instructions based on the point cloud map obtained by the point cloud acquisition module and the point cloud map of the welding point, and controls the photovoltaic welding robot to automatically navigate to the installation position and automatically perform the welding grasping and installation operation during the photovoltaic welding installation process, thereby realizing the standardized photovoltaic welding and installation operation process, reducing the labor cost in the photovoltaic welding process, reducing the welding installation errors caused by human errors, and improving the photovoltaic welding quality.

[0113] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0114] Those of ordinary skill in the art will appreciate that the units and steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed with hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0115] In the embodiments provided in the present application, it should be understood that the disclosed photovoltaic welding robot can be implemented in other ways. For example, the photovoltaic welding robot embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the communication connection between each other shown or discussed can be through some interfaces, communication connections of devices or units or integrated circuits, which can be electrical, mechanical or other forms.

[0116] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0117] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0118] It should be noted that the above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A photovoltaic welding robot, characterized in that: It includes a welding robot arm for performing welding operations and a grabbing robot arm for performing grabbing operations, including: A point cloud acquisition module, used to acquire a point cloud map of a photovoltaic installation site and a point cloud map of welding locations to be welded on each target in the photovoltaic installation site; a navigation control module, connected to the point cloud acquisition module, for receiving the point cloud map sent by the point cloud acquisition module, planning the action path of the photovoltaic welding robot according to the point cloud map, and driving the photovoltaic welding robot to a target position through a driving motor according to the action path; A first visual image acquisition module, used to acquire a welding visual image of the position to be welded at an angle of the welding robot arm and to acquire a grabbing visual image of the position to be welded at an angle of the grabbing robot arm; a welding control module, connected to the point cloud acquisition module and the first visual image acquisition module, respectively, for receiving the welding point cloud image sent by the point cloud acquisition module, and the welding visual image and the captured visual image sent by the first visual image acquisition module, and generating a welding control instruction according to the welding visual image and the welding point cloud image, respectively, and generating a capture control instruction according to the captured visual image and the welding point cloud image; The welding control module controls the welding robot arm to perform welding operations according to the welding control instructions, and controls the grabbing robot arm to perform grabbing operations according to the grabbing control instructions, and then performs welding installation on the position to be welded; The first visual image acquisition module acquires a visual image of a welding result of the position to be welded after the welding installation is completed; The welding control module is also equipped with a welding quality detection model. After receiving the welding result visual image sent by the first visual image acquisition module, the welding control module evaluates the welding quality of the position to be welded according to the welding quality detection model and the welding result visual image. The welding quality detection model generates welding quality parameters according to the welding result visual image. The welding quality detection model is generated by deep learning training based on a welding result visual image dataset marked with the welding quality parameters; The point cloud acquisition module includes a laser radar installed on the navigation chassis of the photovoltaic welding robot; the laser radar is used to scan the photovoltaic installation site before the photovoltaic welding robot performs navigation operations, and perform SLAM mapping based on the scanning results to obtain the point cloud map; The laser radar is also used to scan the position to be welded before the photovoltaic welding robot performs the welding operation, and perform SLAM mapping according to the scanning results to obtain a point cloud map of the welding position; A second visual image acquisition module, connected to the navigation control module, wherein the second visual image acquisition module has a built-in environmental visual camera for acquiring an environmental visual image of the photovoltaic installation site; The navigation control module is further used to receive the environmental visual image sent by the second visual image acquisition module, and identify the to-be-welded position and to-be-welded category of each of the targets in the environmental visual image through the target detection model carried by the navigation control module; the target detection model is generated by deep learning training based on the visual image data set in which the to-be-welded position and the to-be-welded category are pre-labeled; The navigation control module divides the positions to be welded into a plurality of welding groups according to the categories to be welded, plans the corresponding sub-action paths of the photovoltaic welding robot when operating in each welding group according to the positions to be welded, and composes the action path from each of the sub-action paths; The navigation control module is further used to mark the position to be welded and the category to be welded on the point cloud map in combination with the environmental visual image; The navigation control module is further used to receive an externally input navigation remote control command, and generate the action path of the photovoltaic welding robot according to the navigation remote control command; The welding control module is further used to receive externally input welding remote control instructions and grabbing remote control instructions, and control the welding robot arm to perform corresponding welding operations according to the welding remote control instructions and the grabbing remote control instructions; The welding control module generates a folding control instruction after the welding installation is completed, and controls the welding robot arm and the grasping robot arm to fold to a preset state after the welding installation is completed according to the folding control instruction.

2. A photovoltaic welding robot according to claim 1, characterized in that: The welding robot arm is used to weld the target according to the position to be welded, and a 360-degree steering axis is arranged at the connection part between the welding robot arm and the photovoltaic welding robot, and at least two steering parts are arranged on the welding robot arm to realize that the welding robot arm has at least two rotational degrees of freedom; The working part of the welding robot arm is provided with welding equipment, which includes fusion welding equipment, pressure welding equipment and brazing equipment; The first visual image acquisition module includes a welding portion visual camera disposed on the working portion of the welding robot arm, and is used to acquire the welding visual image and the welding result visual image; The welding control module is equipped with a welding position detection model, the welding position detection model generates welding position spatial coordinates according to the welding visual image and the welding point cloud image, and the welding control module generates the welding control instruction according to the welding position spatial coordinates; The welding position detection model is generated by deep learning training based on a data set including the corresponding relationship between the welding position spatial coordinates, the welding point cloud image and the welding visual image.

3. A photovoltaic welding robot according to claim 1, characterized in that: The grabbing robot arm is used to grab the target and fix the target at the position to be welded according to the grabbing control instruction during the welding installation process of the position to be welded; The first visual image acquisition module further includes a grasping portion visual camera disposed on the working portion of the grasping mechanical arm, for acquiring the grasping visual image; The welding control module is also equipped with a grasping position detection model, the grasping position detection model generates grasping position spatial coordinates according to the grasping visual image and the welding point cloud image, and the welding control module generates the grasping control instruction according to the grasping position spatial coordinates; The grasping position detection model is generated by deep learning training based on a data set including the grasping control instructions, the grasping position spatial coordinates and the corresponding relationship between the grasping visual image.

4. A photovoltaic welding method, characterized in that: The photovoltaic welding robot applied to any one of claims 1 to 3 comprises a welding robot arm for performing welding operations and a grasping robot arm for performing grasping operations, and the photovoltaic welding method comprises the steps of: Obtain point cloud maps of photovoltaic installation sites; Planning the action path of the photovoltaic welding robot according to the point cloud map; According to the action path, the photovoltaic welding robot is driven to move to a target position by a driving motor; Acquire a point cloud image of a welding position of a position to be welded, and respectively acquire a welding visual image of the position to be welded at the angle of the welding robot arm, and acquire a grabbing visual image of the position to be welded at the angle of the grabbing robot arm; Generate welding control instructions according to the welding visual image and the welding point cloud image respectively, and generate grabbing control instructions according to the grabbing visual image and the welding point cloud image; After the welding robot arm is controlled to perform welding operations according to the welding control instructions and the grabbing robot arm is controlled to perform grabbing operations according to the grabbing control instructions, welding and installation are performed on the position to be welded.

Images