Multi-layer welding track planning method based on combination of 3D laser vision and robot

Abstract

The invention relates to the technical field of industrial welding robots, and discloses a multi-layer welding track planning method based on combination of 3D laser vision and a robot, which is applied to industrial welding equipment and specifically comprises the following steps: S101, receiving a welding request sent by a controller, the welding request carries welding process parameters of a welding part and a welding material of the welding part; and S102, welding process parameters of the welding part are obtained, and the welding process parameters are input into a preset content text recognition model. According to the method, the 3D laser vision module is used for rapidly scanning to obtain information of the bottom wall and the slope surface of the weld joint opening, the attribute characteristics of the weld joint opening are judged, a three-dimensional model of the weld joint opening is established, and a welding track path of a single-layer weld joint filling surface is obtained, so that industrial welding equipment is controlled to realize multi-layer welding track automatic welding; the defects of tedious teaching, large workload, expensive auxiliary equipment and the like in the prior art are overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial welding robots, and in particular to a multi-layer welding trajectory planning method based on the combination of 3D laser vision and robots. Background Art

[0002] With the development of society, machines have replaced manual labor in various automated operations and have developed rapidly. In the field of medium and thick plates, multi-layer and multi-pass welding is a commonly used welding method. Compared with traditional manual welding, which has high labor intensity and harsh working environment, using robots to achieve automatic welding has become a trend.

[0003] In recent years, with the development of machine vision, image processing, pattern recognition, intelligent control and other technologies, a variety of sensors have been proposed for robot welding. Among them, laser structured light sensors have become an ideal choice for the field of intelligent control of thick plate welding due to their simple structure, large amount of information, high precision and good stability. The basic principle of laser sensors is to project a strip laser onto the workpiece, use CCD to collect the laser stripe image reflected on the workpiece, and then through a series of signal processing, finally obtain the required groove feature information. However, due to the complex surface morphology of multi-layer and multi-pass welds and unclear features, it is often difficult to correctly identify the groove morphology characteristics. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-layer welding trajectory planning method based on the combination of 3D laser vision and robots, to obtain the welding track path of the single-layer weld filling surface, thereby controlling industrial welding equipment to realize multi-layer welding trajectory automated welding, overcoming the shortcomings of the prior art such as cumbersome teaching, heavy workload, and expensive auxiliary equipment. It aims to solve the problem that the prior art uses CCD to collect the laser stripe image reflected on the workpiece, and then obtains the required groove feature information through a series of signal processing, but due to the complex surface morphology of the multi-layer and multi-pass welds, the features are not obvious.

[0005] The present invention is implemented in this way: a multi-layer welding trajectory planning method based on 3D laser vision combined with a robot is applied to industrial welding equipment, specifically comprising the following steps:

[0006] S101: receiving a welding request sent by a controller, wherein the welding request carries welding process parameters of a welding component and welding materials of the welding component;

[0007] S102: Acquire welding process parameters of the welding component, and input the welding process parameters into a preset content text recognition model;

[0008] S103: Obtaining a text recognition result output by the text recognition model based on the welding process parameters, and obtaining the number of welding layers and welding paths of the welding process parameters from the text recognition result;

[0009] S104: using a 3D laser vision module to quickly scan and obtain information about the bottom wall and slope of the weld opening, determine the property characteristics of the weld opening, and establish a three-dimensional model of the weld opening;

[0010] S105: Acquire a three-dimensional model of the weld opening, introduce the number of welding layers and welding paths of the welding process parameters, perform multi-faceted cutting along the bottom wall of the weld opening, acquire the multi-faceted weld filling surface, acquire the welding track path of the single-faceted weld filling surface, collect all acquired welding track paths of the single-faceted weld filling surface, complete the coding arrangement according to the multi-faceted cutting sequence, that is, complete the welding track planning of the multi-layer weld opening, and the welding module completes the execution of the welding track planning.

[0011] Further, in S101, receiving a welding request sent by a controller includes:

[0012] Receiving a connection request sent by a controller through a preset network, wherein the connection request is used to request to establish a connection with the industrial welding equipment;

[0013] Detecting whether the current account of the controller is the target account;

[0014] If the current account of the controller is the target account, the controller is connected according to the connection request, and after the connection is completed, the welding request transmitted by the controller is received.

[0015] Furthermore, the preset network includes one or a combination of a 3G network, a 4G network, a 5G network, and a WIFI network.

[0016] Furthermore, the welding process parameters of the welding parts include:

[0017] The weld opening shape information of the welded parts is used to determine the welding path or welding profile during welding;

[0018] The slope information of the weld opening of the welded parts is used for statistical measurement of the internal space when building the 3D model later;

[0019] Multi-layer segmentation information of the weld opening, which is used to determine the layers of the three-dimensional model established for the weld opening based on the required cutting of the welding material;

[0020] The welding process parameters of the welding filler layer are used to determine the thickness of the welding filler layer and the surface bonding strength.

[0021] Furthermore, in S102, the welding process parameters are input into a preset content text recognition model, including:

[0022] Acquire multiple text recognition models, and sort the categories of the multiple text recognition models in order;

[0023] The content text of the welding process parameters is obtained, and the welding process parameters are input into the text recognition model ranked first.

[0024] Further, in S103, the number of welding layers and welding paths of the welding process parameters are obtained, including:

[0025] Obtaining the number of welding layers includes: after obtaining welding materials based on welding parts, obtaining weld opening shape information of the welding parts and slope information of the weld opening of the welding parts, and obtaining the number of welding layers by combining characteristic attributes of the welding materials, weld opening shape and slope information of the weld opening;

[0026] Obtaining the welding path includes: after obtaining the welding material based on the welding component, obtaining the multi-faceted segmentation information of the weld opening and the welding process parameters of the welding filling layer, and obtaining the welding path by combining the characteristic attributes of the welding material, the multi-faceted segmentation information and the welding process parameters of the welding filling layer.

[0027] Further, in S104, the 3D laser vision module is used to quickly scan and obtain the bottom wall and slope information of the weld opening, including:

[0028] The 3D laser vision module performs laser scanning on the weld opening to obtain the bottom wall of the current weld opening, and scans along the bottom wall of the weld opening along the surrounding contour to gradually form the bottom wall contour of the weld opening. The scanning is stopped when the transmission efficiency of the emitted laser of the 3D laser vision module changes, that is, the bottom wall information of the current weld opening is determined;

[0029] Scan outward along the bottom wall contour edge of the current weld opening to obtain the slope contour points of multiple weld openings, and record the laser emission and receiving time of each slope contour point until all slope contour points are identified and stop scanning.

[0030] Further, determining the property characteristics of the weld opening and establishing a three-dimensional model of the weld opening includes:

[0031] Obtain the multi-contour laser emission and receiving time of the bottom wall information of the current weld opening, and establish bottom wall data points on the coordinate axis, wherein the x-axis of the coordinate axis is the laser emission and receiving time t1, and the y-axis of the coordinate axis is the bottom wall multi-contour number s1 of the weld opening, and connect all the bottom wall data points to determine the slope k1 of the generated curve;

[0032] Obtain the laser emission and reception time of the slope contour point of the current weld opening, and establish the slope data point on the coordinate axis, where the x-axis of the coordinate axis is the laser emission and reception time t2, and the y-axis of the coordinate axis is the slope contour point s2 of the weld opening, and connect the slope contour points of all weld openings to determine the slope k2 of the generated curve;

[0033] It is determined whether the values ​​of k1 and k2 are constant. If they are constant, it is determined that the bottom wall and the slope surface of the weld opening are both planes, and then the three-dimensional model is established using the acquired information.

[0034] Further, 9. In S105, a three-dimensional model of the weld opening is obtained, the number of welding layers and the welding path of the welding process parameters are introduced, and multi-faceted cutting is performed along the bottom wall of the weld opening to obtain a multi-faceted weld filling surface, including:

[0035] The obtained three-dimensional model of the weld opening is placed in the welding layer number and welding path of the welding process parameters to determine the multi-pass welding trajectory planning features matched with the current weld opening;

[0036] Acquire the multi-layer weld filling surface of the matched multi-layer welding trajectory planning feature, and perform spatial calculation on the multi-layer weld filling surface;

[0037] The spatial value of the multi-layer weld filling surface is obtained, and a signal is sent to the welding module, and the welding module fills the welding filling layer according to the spatial value of the multi-layer weld filling surface.

[0038] Furthermore, the spatial calculation of the multi-faceted weld fill surface includes:

[0039] The bottom surface value a of the current multi-faceted weld filling surface is obtained by using the three-dimensional model, and then the top surface value b of the previous multi-faceted weld filling surface is obtained;

[0040] Get the height difference between the current bottom value a and the top value b, and get the contour shape of the current multi-faceted weld filling surface for volume calculation. The calculation formula is:

[0041] ∫f′(x)dx=f(x)+C

[0042] Where f(x) is the slope volume and C is the middle volume.

[0043] Compared with the prior art, the multi-layer welding trajectory planning method based on the combination of 3D laser vision and robot provided by the present invention has the following beneficial effects:

[0044] 1. Use the 3D laser vision module to quickly scan and obtain the bottom wall and slope information of the weld opening, determine the attribute characteristics of the weld opening and establish a three-dimensional model of the weld opening, then introduce the welding layer number and welding path of the welding process parameters into the three-dimensional model, perform multi-faceted cutting along the bottom wall of the weld opening, obtain the multi-faceted weld filling surface, and obtain the welding track path of the single-faceted weld filling surface, so as to control the industrial welding equipment to realize multi-layer welding track automatic welding, which overcomes the shortcomings of the prior art such as cumbersome teaching, heavy workload, and expensive auxiliary equipment;

[0045] 2. The present application realizes the early identification of welding process parameters and welding materials of welding parts, so that the system obtains the processing process parameters of multiple weld openings. By combining with the final three-dimensional model of the weld opening, the probability of defects is greatly reduced through parameter optimization. Compared with manual welding, the welding production efficiency is greatly improved, and the weld consistency is improved. At the same time, the three-dimensional model of the weld opening is cut into multiple planes to obtain the multi-plane weld filling surface, and the welding track path of the single-plane weld filling surface is obtained, so that the welding of the weld opening has no intersection and achieves a collision-free effect. This processing method facilitates the welding of different welding trajectories. Combined with the volume calculation of the single-plane weld filling surface, the volume of the Korean solder that needs to be filled is obtained, which ensures the fullness of the internal welding surface, enhances the stability of the welding, and avoids the defects of cold welding and empty welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a schematic flow chart of the multi-layer welding trajectory planning method based on the combination of 3D laser vision and robot proposed in the present invention;

[0047] Figure 2 A schematic block diagram of the process of receiving a welding request sent by a controller in a multi-layer welding trajectory planning method based on a combination of 3D laser vision and a robot proposed by the present invention;

[0048] Figure 3 This is a system structure diagram of industrial welding equipment in the multi-layer welding trajectory planning method based on the combination of 3D laser vision and robots proposed in the present invention;

[0049] Figure 4 This is a statistical principle diagram of the calculation of the 3D laser vision module in the multi-layer welding trajectory planning method based on the combination of 3D laser vision and robots proposed in the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0051] The implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0052] The same or similar numbers in the drawings of this embodiment correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right" and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, it is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limitations on the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0053] Reference Figure 1-2 As shown, a multi-layer welding trajectory planning method based on the combination of 3D laser vision and robot is applied to industrial welding equipment, which specifically includes the following steps:

[0054] S101: receiving a welding request sent by a controller, the welding request carrying welding process parameters of the welding parts and welding materials of the welding parts;

[0055] The receiving controller sends a welding request, including:

[0056] receiving a connection request sent by a controller through a preset network, the connection request being used to request to establish a connection with the industrial welding equipment;

[0057] Check whether the current account of the controller is the target account;

[0058] If the current account of the controller is the target account, the controller is connected according to the connection request, and after the connection is completed, the welding request transmitted by the controller is received.

[0059] S102: Obtain welding process parameters of the welding component, and input the welding process parameters into a preset content text recognition model;

[0060] Among them, the welding process parameters are input into the preset content text recognition model, including:

[0061] Acquire multiple text recognition models, and sort the categories of the multiple text recognition models in order;

[0062] Obtaining the content text of the welding process parameters, and inputting the welding process parameters into the first-ranked text recognition model;

[0063] S103: Obtaining a text recognition result output by a text recognition model based on welding process parameters, and obtaining the number of welding layers and welding paths of the welding process parameters from the text recognition result;

[0064] The number of welding layers and welding paths for obtaining welding process parameters include:

[0065] Obtaining the number of welding layers includes: after obtaining welding materials based on welding parts, obtaining weld opening shape information of the welding parts and slope information of the weld opening of the welding parts, and obtaining the number of welding layers by combining characteristic attributes of the welding materials, weld opening shape and slope information of the weld opening;

[0066] Acquiring the welding path includes: acquiring the welding material based on the welding component, acquiring the multi-face segmentation information of the weld opening and the welding process parameters of the welding filling layer, and acquiring the welding path by combining the characteristic attributes of the welding material, the multi-face segmentation information and the welding process parameters of the welding filling layer;

[0067] S104: using a 3D laser vision module to quickly scan and obtain information about the bottom wall and slope of the weld opening, determine the property characteristics of the weld opening, and establish a three-dimensional model of the weld opening;

[0068] Among them, the 3D laser vision module is used to quickly scan and obtain the bottom wall and slope information of the weld opening, including:

[0069] The 3D laser vision module performs laser scanning on the weld opening to obtain the bottom wall of the current weld opening, and scans along the bottom wall of the weld opening along the surrounding contour to gradually form the bottom wall contour of the weld opening. The scanning is stopped when the transmission efficiency of the emitted laser of the 3D laser vision module changes, that is, the bottom wall information of the current weld opening is determined;

[0070] Scan outward along the bottom wall contour edge of the current weld opening to obtain multiple slope contour points of the weld opening, and record the laser emission and reception time of each slope contour point until all slope contour points are identified and stop scanning;

[0071] S105: obtaining a three-dimensional model of the weld opening, introducing the number of welding layers and welding paths of the welding process parameters, performing multi-faceted cutting along the bottom wall of the weld opening, obtaining the multi-faceted weld filling surface, obtaining the welding track path of the single-faceted weld filling surface, collecting all the obtained welding track paths of the single-faceted weld filling surfaces, completing the coding arrangement according to the multi-faceted cutting sequence, that is, completing the welding track planning of the multi-faceted weld opening, and the welding module completes the execution of the welding track planning;

[0072] Among them, a three-dimensional model of the weld opening is obtained, the number of welding layers and welding paths of the welding process parameters are introduced, multi-faceted cutting is performed along the bottom wall of the weld opening, and a multi-faceted weld filling surface is obtained, including:

[0073] The obtained three-dimensional model of the weld opening is placed in the welding layer number and welding path of the welding process parameters to determine the multi-pass welding trajectory planning features matched with the current weld opening;

[0074] Acquire the multi-layer weld filling surface of the matched multi-layer welding trajectory planning feature, and perform spatial calculation on the multi-layer weld filling surface;

[0075] The spatial values ​​of the multi-layer weld filling surface are obtained and a signal is sent to the welding module. The welding module fills the welding filling layer according to the spatial values ​​of the multi-layer weld filling surface. The technical solution uses a 3D laser vision module to quickly scan and obtain the bottom wall of the weld opening and the slope information of the weld opening, judge the attribute characteristics of the weld opening and establish a three-dimensional model of the weld opening, and then introduce the number of welding layers and welding paths of the welding process parameters into the three-dimensional model, perform multi-layer cutting along the bottom wall of the weld opening, obtain the multi-layer weld filling surface, and obtain the welding track path of the single-layer weld filling surface, thereby controlling the industrial welding equipment to realize automatic welding of multi-layer welding trajectories.

[0076] In this example, the preset network includes one or a combination of 3G network, 4G network, 5G network, and WIFI network.

[0077] In this example, the welding process parameters for welding components include:

[0078] The weld opening shape information of the welding component is used to determine the welding path or welding contour during welding; the slope information of the weld opening of the welding component is used for statistical measurement of the internal space when the three-dimensional model is established later; the multi-faceted segmentation information of the weld opening is used to determine the face of the three-dimensional model established by the weld opening based on the required cutting face of the welding material; the welding process parameters of the welding filler layer are used to determine the thickness of the welding filler layer and the surface bonding strength. The present application realizes the early identification of the welding process parameters and the welding materials of the welding component, so that the system obtains the processing process parameters of multiple weld openings, and through the three-dimensional segmentation of the final weld opening The combination of the two models greatly reduces the probability of defects through parameter optimization, greatly improves welding production efficiency and weld consistency compared to manual welding, and performs multi-faceted cutting on the three-dimensional model of the weld opening to obtain multi-faceted weld filling surfaces and the welding track path of the single-faceted weld filling surface, so that the welding of the weld opening has no intersections and achieves a collision-free effect. This processing method facilitates welding of different welding trajectories, and combines the volume calculation of the single-faceted weld filling surface to obtain the volume of the solder that needs to be filled, thereby ensuring the fullness of the internal welding surface, enhancing the stability of the welding, and avoiding the defects of cold welding and empty welding.

[0079] In this example, the property characteristics of the weld opening are determined and a three-dimensional model of the weld opening is established, including:

[0080] Obtain the multi-contour laser emission and receiving time of the bottom wall information of the current weld opening, and establish bottom wall data points on the coordinate axis, where the x-axis of the coordinate axis is the laser emission and receiving time t1, and the y-axis of the coordinate axis is the bottom wall multi-contour number s1 of the weld opening, and connect all the bottom wall data points to determine the slope k1 of the generated curve;

[0081] Obtain the laser emission and reception time of the slope contour point of the current weld opening, and establish the slope data point on the coordinate axis, where the x-axis of the coordinate axis is the laser emission and reception time t2, and the y-axis of the coordinate axis is the slope contour point s2 of the weld opening, and connect the slope contour points of all weld openings to determine the slope k2 of the generated curve;

[0082] Determine whether the values ​​of k1 and k2 are constant. If they are constant, that is, determine that the bottom wall and slope of the weld opening are both planes, and then use the acquired information to complete the establishment of the three-dimensional model. The technical solution determines the attribute characteristics of the weld opening and establishes a three-dimensional model of the weld opening, and then introduces the number of welding layers and welding paths of the welding process parameters into the three-dimensional model, performs multi-faceted cutting along the bottom wall of the weld opening, obtains the multi-faceted weld filling surface, and obtains the welding track path of the single-faceted weld filling surface, thereby controlling the industrial welding equipment to realize multi-layer welding trajectory automated welding, overcoming the shortcomings of the prior art such as cumbersome teaching, heavy workload, and expensive auxiliary equipment.

[0083] Reference Figure 3 As shown, the system diagram of industrial welding equipment includes:

[0084] A startup module is used to connect with the controller to realize command transmission;

[0085] An identification module, used for identifying welding process parameters of welding parts;

[0086] 3D laser vision module, used to quickly scan and obtain the bottom wall and slope information of the weld opening, and determine the attribute characteristics of the weld opening;

[0087] A three-dimensional generation module is used to establish a three-dimensional model of the weld opening based on the information obtained by the 3D laser vision module;

[0088] A three-dimensional processing module performs multi-faceted cutting along the bottom wall of the weld opening on the established three-dimensional model to obtain a multi-faceted weld filling surface and a welding track path of a single-faceted weld filling surface;

[0089] The welding module is used to execute the welding trajectory planning of multi-layer weld openings. This technical solution uses a 3D laser vision module to quickly scan and obtain the bottom wall and slope information of the weld opening, determine the attribute characteristics of the weld opening and establish a three-dimensional model of the weld opening, and then introduce the number of welding layers and welding paths of the welding process parameters into the three-dimensional model, perform multi-faceted cutting along the bottom wall of the weld opening, obtain the multi-faceted weld filling surface, and obtain the welding track path of the single-faceted weld filling surface, thereby controlling the industrial welding equipment to realize multi-layer welding trajectory automated welding, overcoming the shortcomings of the prior art such as cumbersome teaching, heavy workload, and expensive auxiliary equipment.

[0090] Reference Figure 4 As shown in the figure, the spatial calculation of multi-faceted weld filling surfaces includes:

[0091] The three-dimensional model is used to obtain the bottom surface value a of the current multi-faceted weld filling surface, and then the top surface value b of the previous multi-faceted weld filling surface is obtained; the height difference between the current bottom surface value a and the top surface value b is obtained, and the contour shape of the current multi-faceted weld filling surface is obtained for volume calculation. The calculation formula is:

[0092] ∫f′(x)dx=f(x)+C

[0093] Among them, f(x) is the slope volume, C is the middle volume, and the application realizes the early identification of welding process parameters and welding materials of welding parts, so that the system obtains the processing parameters of multiple weld openings, and combines with the final three-dimensional model of the weld opening. The probability of defects is greatly reduced through parameter optimization, which greatly improves the welding production efficiency and weld consistency compared with manual welding. At the same time, the three-dimensional model of the weld opening is cut into multiple planes to obtain the multi-plane weld filling surface, and the welding track path of the single-plane weld filling surface is obtained, so that the welding of the weld opening has no intersection and achieves a collision-free effect. This processing method is convenient for welding different welding trajectories. Combined with the volume calculation of the single-plane weld filling surface, the volume of the Korean solder that needs to be filled is obtained, which ensures the fullness of the internal welding surface, enhances the stability of the welding, and avoids the defects of cold welding and empty welding.

[0094] In this embodiment, the entire operation process can be controlled by a computer, and with the addition of a PLC, etc., automated operation control can be achieved. In each operation link, sensors can be set up to provide signal feedback to achieve sequential execution of steps. These are all common knowledge of current automated control and will not be described in detail in this embodiment.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-layer welding trajectory planning method based on 3D laser vision and robot, characterized in that: Applied to industrial welding equipment, it includes the following steps: S101: receiving a welding request sent by a controller, wherein the welding request carries welding process parameters of a welding component and welding materials of the welding component: S102: Acquire welding process parameters of the welding component, and input the welding process parameters into a preset content text recognition model; S103: Obtaining a text recognition result output by the text recognition model based on the welding process parameters, and obtaining the number of welding layers and welding paths of the welding process parameters from the text recognition result; S104: using a 3D laser vision module to quickly scan and obtain information about the bottom wall and slope of the weld opening, determine the property characteristics of the weld opening, and establish a three-dimensional model of the weld opening; S105: Acquire a three-dimensional model of the weld opening, introduce the number of welding layers and welding paths of the welding process parameters, perform multi-faceted cutting along the bottom wall of the weld opening, acquire the multi-faceted weld filling surface, acquire the welding track path of the single-faceted weld filling surface, collect all acquired welding track paths of the single-faceted weld filling surface, complete the coding arrangement according to the multi-faceted cutting sequence, that is, complete the welding track planning of the multi-layer weld opening, and the welding module completes the execution of the welding track planning.

2. The multi-layer welding trajectory planning method based on 3D laser vision and robot combination as claimed in claim 1 is characterized in that: In S101, receiving a welding request sent by a controller includes: Receiving a connection request sent by a controller through a preset network, wherein the connection request is used to request to establish a connection with the industrial welding equipment; Detecting whether the current account of the controller is the target account; If the current account of the controller is the target account, the controller is connected according to the connection request, and after the connection is completed, the welding request transmitted by the controller is received.

3. The multi-layer welding trajectory planning method based on 3D laser vision and robot combination as claimed in claim 2 is characterized in that: The preset network includes one or a combination of a 3G network, a 4G network, a 5G network, and a WIFI network.

4. The multi-layer welding trajectory planning method based on 3D laser vision and robot as claimed in claim 3 is characterized in that: The welding process parameters of the welding parts include: The weld opening shape information of the welded parts is used to determine the welding path or welding profile during welding; The slope information of the weld opening of the welded parts is used for statistical measurement of the internal space when building the 3D model later; Multi-layer segmentation information of the weld opening, which is used to determine the layers of the three-dimensional model established for the weld opening based on the required cutting of the welding material; The welding process parameters of the welding filler layer are used to determine the thickness of the welding filler layer and the surface bonding strength.

5. The multi-layer welding trajectory planning method based on 3D laser vision and robot as claimed in claim 4 is characterized in that: In S102, the welding process parameters are input into a preset content text recognition model, including: Acquire multiple text recognition models, and sort the categories of the multiple text recognition models in order; The content text of the welding process parameters is obtained, and the welding process parameters are input into the text recognition model ranked first.

6. The multi-layer welding trajectory planning method based on 3D laser vision and robot as claimed in claim 5 is characterized in that: In S103, the number of welding layers and welding paths of the welding process parameters are obtained, including: Obtaining the number of welding layers includes: after obtaining welding materials based on welding parts, obtaining weld opening shape information of the welding parts and slope information of the weld opening of the welding parts, and obtaining the number of welding layers by combining characteristic attributes of the welding materials, weld opening shape and slope information of the weld opening; Obtaining the welding path includes: after obtaining the welding material based on the welding component, obtaining the multi-faceted segmentation information of the weld opening and the welding process parameters of the welding filling layer, and obtaining the welding path by combining the characteristic attributes of the welding material, the multi-faceted segmentation information and the welding process parameters of the welding filling layer.

7. The multi-layer welding trajectory planning method based on 3D laser vision and robot combination as claimed in claim 6 is characterized in that: In S104, the 3D laser vision module is used to quickly scan and obtain the bottom wall and slope information of the weld opening, including: The 3D laser vision module performs laser scanning on the weld opening to obtain the bottom wall of the current weld opening, and scans along the bottom wall of the weld opening along the surrounding contour to gradually form the bottom wall contour of the weld opening. The scanning is stopped when the transmission efficiency of the emitted laser of the 3D laser vision module changes, that is, the bottom wall information of the current weld opening is determined; Scan outward along the bottom wall contour edge of the current weld opening to obtain the slope contour points of multiple weld openings, and record the laser emission and receiving time of each slope contour point until all slope contour points are identified and stop scanning.

8. The multi-layer welding trajectory planning method based on 3D laser vision and robot combination as claimed in claim 7, characterized in that: Determine the property characteristics of the weld opening and establish a 3D model of the weld opening, including: Obtain the multi-contour laser emission and receiving time of the bottom wall information of the current weld opening, and establish bottom wall data points on the coordinate axis, wherein the x-axis of the coordinate axis is the laser emission and receiving time t1, and the y-axis of the coordinate axis is the bottom wall multi-contour number s1 of the weld opening, and connect all the bottom wall data points to determine the slope k1 of the generated curve; Obtain the laser emission and reception time of the slope contour point of the current weld opening, and establish the slope data point on the coordinate axis, where the x-axis of the coordinate axis is the laser emission and reception time t2, and the y-axis of the coordinate axis is the slope contour point s2 of the weld opening, and connect the slope contour points of all weld openings to determine the slope k2 of the generated curve; It is determined whether the values ​​of k1 and k2 are constant. If they are constant, it is determined that the bottom wall and the slope surface of the weld opening are both planes, and then the three-dimensional model is established using the acquired information.

9. The multi-layer welding trajectory planning method based on 3D laser vision and robot as claimed in claim 8, characterized in that: In S105, a three-dimensional model of the weld opening is obtained, the number of welding layers and the welding path of the welding process parameters are introduced, and multi-faceted cutting is performed along the bottom wall of the weld opening to obtain a multi-faceted weld filling surface, including: The obtained three-dimensional model of the weld opening is placed in the welding layer number and welding path of the welding process parameters to determine the multi-pass welding trajectory planning features matched with the current weld opening; Acquire the multi-layer weld filling surface of the matched multi-layer welding trajectory planning feature, and perform spatial calculation on the multi-layer weld filling surface; The spatial value of the multi-layer weld filling surface is obtained, and a signal is sent to the welding module, and the welding module fills the welding filling layer according to the spatial value of the multi-layer weld filling surface.

10. The multi-layer welding trajectory planning method based on 3D laser vision and robot combination as claimed in claim 9, characterized in that: Spatial calculations for multi-faceted weld fill surfaces include: The bottom surface value a of the current multi-faceted weld filling surface is obtained by using the three-dimensional model, and then the top surface value b of the previous multi-faceted weld filling surface is obtained; Get the height difference between the current bottom value a and the top value b, and get the contour shape of the current multi-faceted weld filling surface for volume calculation. The calculation formula is: ∫f′(x)dx=f(x)+C Where f(x) is the slope volume and C is the middle volume.

Images