A spot welding flow processing method

Abstract

The present invention relates to a spot welding assembly line processing method, belonging to the field of spot welding technology. The method comprises the following steps: providing a positioning jig, an assembly line, a welding robot, a handling robot, a laser position finder and a visual inspection mechanism; in the process of handling, the laser position finder scans the parts and the positioning jig to form a scanning path and transmits it to the welding robot, after the scanning is completed, when the positioning jig moves to the welding station through the assembly line, the visual inspection mechanism determines the position of the positioning points on the positioning jig and the parts and forms a new scanning path, and compares the new scanning path with the scanning path accepted by the welding robot. If the comparison results are consistent, the welding robot performs welding according to the set program, and if they are inconsistent, the welding robot corrects the set program according to the comparison error and then performs welding. The present invention solves the problem of poor positioning accuracy when using a two-dimensional camera to position large-size parts in the existing automobile industry production and assembly process.

Description

Technical Field

[0001] The invention belongs to the technical field of spot welding, and in particular relates to a spot welding flow processing method. Background Art

[0002] With the continuous development of science and technology, in order to improve production efficiency, automatic assembly lines are usually used to assemble vehicle parts in the production and assembly process of the existing automobile industry. During the assembly process, after the car body is delivered to the conveyor line or conveyor trolley, the robot receives the signal to start welding, gluing, marking and other work. However, due to the manufacturing errors caused by the conveyor line and multiple sets of trolleys, and the cumulative positioning errors and positioning wear and other factors that occur during the matching operation, each car body will have a certain position change and posture rotation after arriving, resulting in a position difference between the robot tool coordinate point and the actual working point of the car body. Therefore, in the subsequent welding process, intermittent quality accidents will occur due to the cumulative errors, affecting the overall safety of the vehicle. Therefore, in order to complete the precise positioning and guidance of parts on the assembly line and subsequent operations such as grabbing and installation, robots can be combined with visual technology to solve the problem of positioning errors in the robot during welding.

[0003] However, the existing visual system combined with the robot arm positioning guidance method mostly uses a three-dimensional camera, which directly scans the object to obtain the surface space point cloud information to achieve spatial positioning perception. However, the equipment cost and subsequent maintenance cost of the three-dimensional camera itself are high, and it needs to be calibrated when it is used to ensure that the required positioning points can be obtained, which leads to a complicated early calibration process and difficulty in large-scale adaptation and maintenance in the industrial production process. If the currently commonly used, low-cost CCD two-dimensional camera in the industry is used with the PnP model and the robot to complete the positioning, the two-dimensional camera itself has poor accuracy and the system errors at all levels are accumulated. When positioning large parts, the accuracy is difficult to meet the requirements, resulting in the failure of positioning guidance, and only small parts can be positioned. In order to improve the positioning accuracy of the two-dimensional camera positioning method, multi-stage guidance is usually required to achieve step-by-step error elimination and multiple guided positioning, but the positioning time is long and the guidance efficiency is low. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a spot welding assembly process method, which solves the problem of poor positioning accuracy when using a two-dimensional camera to position large-sized parts in the existing automobile industry production and assembly process.

[0005] The purpose of the present invention can be achieved by the following technical solution: A spot welding flow processing method comprises the following steps:

[0006] Provide positioning jigs, assembly lines, welding robots, handling robots, laser position finders and visual inspection mechanisms;

[0007] The positioning jig is located on the assembly line and two positioning points are respectively provided on the positioning jig and the parts. The laser position finder and the visual inspection mechanism are respectively installed on the handling robot and the welding robot;

[0008] During the transportation process, the laser positioning instrument scans the parts and the positioning fixture according to the position of the positioning point to form a scanning path and transmit it to the welding robot. After the scanning is completed, when the positioning fixture moves to the welding station through the assembly line, the visual inspection mechanism determines the position of the positioning points on the positioning fixture and the parts and forms a new scanning path. The new scanning path is compared with the scanning path accepted by the welding robot. If the comparison results are consistent, the welding robot performs welding according to the set program. If not, the welding robot corrects the set program according to the comparison error before welding.

[0009] As a preferred technical solution of the present invention, a photoelectric sensor is also provided. The photoelectric sensor is installed on the welding robot and is used to detect the welding quality of the weld.

[0010] As a preferred technical solution of the present invention, the detection signals of the photoelectric sensor are visible light-plasma spectrum, near infrared-laser reflection spectrum and far infrared-molten pool spectrum.

[0011] As a preferred technical solution of the present invention, the laser position finder is composed of a CCD camera, a semiconductor laser and a laser protection lens.

[0012] As a preferred technical solution of the present invention, the positioning points on the positioning fixture are located on the side, and the positioning points on the same batch of parts are at the same position.

[0013] As a preferred technical solution of the present invention, it also includes a comparison module, which is communicatively connected with the laser locator and the visual detection mechanism, and compares the position of the locating point in the laser locator with the position of the locating point detected by the visual detection mechanism.

[0014] As a preferred technical solution of the present invention, the visual detection mechanism includes an ARC system, a visual sensor and an ultrasonic sensor, and the welding robot performs automatic welding through the ARC system.

[0015] As a preferred technical solution of the present invention, the visual detection mechanism further includes a fill light, and the irradiation area of ​​the fill light coincides with the shooting area of ​​the visual sensor.

[0016] As a preferred technical solution of the present invention, the welding robot body is a robot with six degrees of freedom joints.

[0017] The beneficial effects of the present invention are:

[0018] 1. The laser position finder is used to simulate the position where welding is required to obtain the welding path of the part on the positioning fixture, and the position between the two is determined by the positioning points on the positioning fixture and the part. Then the visual inspection mechanism determines the position of the part through the positioning points on the positioning fixture and the part, and the welding robot performs welding according to the simulated path of the laser position finder, which solves the problem of poor positioning accuracy when using two-dimensional cameras to locate large-size parts in the existing automobile industry production and assembly process;

[0019] 2. The ARC system in the visual inspection mechanism allows the welding robot to perform automatic welding, which reduces the need to teach the welding robot during the production process and greatly improves production efficiency;

[0020] 3. The simulation results of the laser positioning instrument are compared with the detection results of the visual inspection mechanism through the comparison module to determine whether there is a position deviation between the part and the positioning fixture when the positioning fixture moves to the welding station through the assembly line. The welding path of the welding robot is corrected by detecting the positioning points twice, reducing the problem of deviation in the welding position. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0022] Figure 1 It is a flow chart of the spot welding flow processing method of the present invention;

[0023] Figure 2 This is a distribution diagram of the photoelectric characteristic signal of the present invention; DETAILED DESCRIPTION

[0024] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0025] See also Figure 1-2 ,This embodiment provides a spot welding assembly line processing method, comprising the following steps: providing a positioning fixture, an assembly line, a welding robot, a handling robot, a laser position finder and a visual inspection mechanism;

[0026] In the process of spot welding of parts, after the parts are welded from the previous welding point, the handling robot moves the parts to the positioning fixture, and the positioning fixture moves to the next welding point through the assembly line for welding. During the handling process of the handling robot, the position of the part is determined by the two positioning points set on the part. At the same time, when the part is moved to the positioning fixture, the position of the positioning fixture is also determined by the two positioning points set on the positioning fixture, so that the part is placed on the positioning fixture for fixing. In the process of handling by the handling robot, the positioning points on the positioning fixture and the part are scanned and recorded by the laser positioning instrument, so that the handling robot can accurately carry the parts through the scanning results of the laser positioning instrument. At the same time, the positioning points on the positioning fixture and the part are scanned during the handling to form a scanning path. The path takes a certain position on the positioning fixture as the starting point and another positioning point as the end point. It passes through the two positioning points of the part in turn, and the scanning path is transmitted to the welding robot after the scanning is completed.

[0027] When the positioning jig moves to the welding point along the assembly line, the welding robot detects the position of the positioning jig and the positioning points on the parts through the visual inspection mechanism, and then forms a scanning path based on the position of the positioning points and compares it with the scanning path of the handling robot. If the data of the two scanning paths are the same, welding is performed according to the set welding procedure. If there is an error in the comparison result, the set welding procedure is corrected according to the error data to complete the welding.

[0028] The automobile production process is a large-scale assembly line operation, including four major workshops such as stamping, welding, painting and assembly. During the welding process, after the car body is delivered to the conveyor line or conveyor trolley, the robot receives the signal to start welding, gluing, marking and other work. However, due to the manufacturing errors caused by the conveyor line and multiple sets of trolleys, and the cumulative positioning errors and positioning wear during the matching operation, each car body will have a certain position change and posture rotation after arriving, resulting in a position difference between the robot tool coordinate point and the actual working point of the car body. Therefore, during the subsequent welding process, intermittent quality accidents will occur due to the cumulative errors, affecting the overall safety of the vehicle. The traditional method to solve the manufacturing error is to accurately control the physical accuracy of each component, using conventional means such as the spacing of the trolley positioning holes, the accuracy of the conveyor line in place, and the control of the reference hole diameter. The common disadvantage of these means is that they are expensive and laborious, and can only control the tolerance within a certain range. When encountering parts that exceed the tolerance, it is easy to cause the production line to stop, resulting in serious waste of production capacity.

[0029] However, with the continuous development of science and technology, the traditional error control method is inefficient and can no longer meet the existing production needs. Therefore, in some production processes, robots are combined with visual technology to complete the precise positioning and guidance of parts on the assembly line and subsequent grasping and installation operations, thereby replacing the traditional manual handling or robot manual teaching grasping method, greatly improving the level of industrial automation and intelligence. The perception principle of the existing visual system, the visual part of the positioning and guidance system can be divided into: monocular vision positioning, binocular vision positioning, line laser 3D positioning and structured light 3D positioning. Among them, binocular vision or 3D camera realizes positioning by directly acquiring the 3D information of the part space, but the calibration process is complicated and the camera cost is high; while monocular vision indirectly realizes the spatial position perception of the part by combining the 2D feature points extracted from the image collected by the camera with the 3D feature points in space, that is, the 3D positioning based on monocular vision belongs to the PnP (Perspective-n-Point) problem. Specifically, when the 3D digital model information of the part is known, the 3D feature points are selected in advance, and the 2D feature points extracted from the image and the 3D feature points in the digital model are combined through the PnP model, and the external parameters are optimized and solved, so that the part can be solved in the equivalent of the camera space position. Although the monocular camera calibration process is simple and low-cost, the error of this indirect positioning method is significantly higher than that of the direct positioning method. The error mainly comes from the calibration process of the camera's internal parameters, the hand-eye calibration process of the camera and the robot arm, and the error introduced by the non-ideal distortion of the camera lens. Especially when the size of the part to be guided is large, the measurement range and measurement accuracy in visual measurement are often inversely proportional, which ultimately leads to large errors in the positioning guidance link.

[0030] Based on this, the existing vision system cooperates with the positioning and guidance method of the robot arm mostly uses a 3D camera to achieve spatial positioning perception by directly scanning the object to obtain the surface space point cloud information. However, the equipment cost and subsequent maintenance cost of the 3D camera itself are high, and it needs to be calibrated when it is used to ensure that the required positioning points can be obtained, which leads to a complicated early calibration process and difficulty in large-scale adaptation and maintenance in the industrial production process. If the currently commonly used, low-cost CCD two-dimensional camera in industry is used in conjunction with the PnP model and the robot to complete the positioning, the accuracy of the two-dimensional camera itself is poor and the reasons for the accumulation of system errors at all levels are... Therefore, when positioning large parts, it is difficult to meet the accuracy requirements, resulting in failure of positioning guidance, and only the positioning of small parts can be completed; because in the process of positioning large parts, the shooting angle of the two-dimensional camera needs to be expanded first to obtain the overall view of the part, and then the location of the positioning point set on the part is found. At that time, the larger the shooting angle of the camera, the lower the pixel, resulting in poor positioning effect in one time. Therefore, multiple positioning is required to ensure the positioning accuracy. In order to improve the positioning accuracy of the two-dimensional camera positioning method, multi-stage guidance is usually required to achieve step-by-step elimination of errors and perform multiple guided positioning, but the positioning time is long and the guidance efficiency is low.

[0031] Therefore, in order to improve the positioning efficiency of the two-dimensional camera and reduce the positioning error when positioning large parts, the present application uses a handling robot to carry the parts to a positioning jig to complete the fixation and positioning of the parts, and then the handling robot carries the positioning jig to the assembly line for welding. During the transportation process, the laser positioning instrument uses two points on the positioning jig as the starting point and the end point, and then passes through the two positioning points on the part in sequence when simulating the path of the welding position. During the simulation process, the laser positioning instrument determines the relative position between the part and the positioning jig through the positioning jig and the positioning points on the part to form a scanning path. When the laser position finder simulation is completed, the result of the path simulation will be transmitted to the welding robot. When the positioning fixture is located at the welding station, the visual inspection mechanism determines the relative position of the positioning fixture and the part according to the position of the positioning point on the positioning fixture and the position of the positioning point on the part, that is, a new scanning path is formed to be compared with the scanning path of the handling robot. If the comparison result shows consistency, the welding robot will perform welding according to the set program. If it is inconsistent, the set welding parameters will be modified according to the comparison error, so as to complete the welding by correcting the welding parameters according to the actual change of the part position, thereby improving the welding quality.

[0032] The laser position finder is used to simulate the path of the welding position to improve the positioning efficiency of the welding robot with the cooperation of the visual inspection mechanism, and the visual inspection mechanism is used to determine the position between the part and the positioning fixture to improve the positioning accuracy. When the position of the part deviates, the welding parameters can be corrected according to the error, thereby reducing the welding position deviation of the welding robot during the welding process and improving the welding quality.

[0033] In the process of welding, the welding quality of the weld is not only determined by the welding position of the welding robot, but also by factors such as the current during the welding process of the welding robot. Therefore, in order to improve the welding quality and avoid the welding robot causing quality problems in the weld due to other influences during the welding process but unable to be adjusted in time, in this embodiment, a photoelectric sensor is also provided. The photoelectric sensor is installed on the welding robot. The photoelectric sensor is used to detect the welding quality of the weld. In the welding process, dazzling visible light will be generated. The intensity of the visible light is closely related to the plasma. The plasma and the penetration depth and the laser power are strongly linearly related. Therefore, the change in the intensity of the visible light signal can reflect the amount of heat input to a certain extent and indirectly judge the change in penetration depth. At the same time, the plasma is also affected by the flow rate of the protective gas, so the visible light signal can also be used to monitor the fluctuation of the protective gas flow. Therefore, the heat generated by the welding robot during the welding process is detected by the photoelectric sensor, and the heat generated by the weld during the welding process is obtained by detecting the intensity of the visible light of the photoelectric signal to determine whether there are quality problems such as cold welding and spatter.

[0034] The visible light generated during the welding process also contains other light signals, which can be divided into three categories. The first category is the ultraviolet and visible light band (200-750); the second category is the laser reflection band (1030-1080); the third category is the infrared radiation band (900-1700). As long as the correspondence between the light intensity signal and the welding state is clear, welding defect detection can be performed. Therefore, in order to better detect the welding situation, in one embodiment, the detection signal of the photoelectric sensor is the visible light-plasma spectrum, the near-infrared-laser reflection spectrum and the far-infrared-molten pool spectrum. The intensity of the visible light signal is closely related to the plasma. The plasma and the melting depth and laser power are strongly linearly correlated. Therefore, the change in the intensity of the visible light signal can reflect the amount of heat input to a certain extent and indirectly judge the change in melting depth; at the same time, the plasma is also affected by the flow rate of the guard gas, so the visible light signal can also be used to monitor the fluctuation of the guard gas flow.

[0035] The signal intensity in the near-infrared band is mainly related to the reflected light of the laser (1060-1080nm). The laser absorption rate can reach 30-40% during deep fusion welding. At this time, the laser reflection amount is small and the near-infrared signal is weak. However, once spatter, weld penetration, light leakage, etc. occur, the reflected laser energy will change. Therefore, the intensity of the near-infrared laser can also be used to monitor defects such as cold welding, weld penetration, spatter, etc. to a certain extent.

[0036] Far infrared signal: The signal detected by the infrared sensor mainly comes from the infrared radiation of the molten pool. The infrared radiation signal of the molten pool can better reflect the weld width, weld bead, excess height and penetration of the weld, and is linearly related to the amount of welding line energy under certain conditions. Generally, the more liquid molten pool, the stronger the far infrared signal, so far infrared can be used to monitor the change of the molten width. Therefore, by detecting and analyzing the visible light signal, near infrared band signal and far infrared signal, it can be judged whether there are defects in the welding process.

[0037] In order to reduce the positioning error of the welding robot during the welding process, when the welding path is simulated by a laser position finder, in order to reduce the error, during the path simulation process, the principle is to adopt the laser triangulation reflection principle, that is, the laser beam is amplified to form a laser line and projected onto the surface of the object to be measured. The reflected light passes through the high-quality optical system and is projected onto the imaging matrix. The distance from the sensor to the surface to be measured (Z axis) and the position information along the laser line (X axis) are calculated. Therefore, in order to achieve this effect, in one embodiment, the laser position finder is composed of a CCD camera, a semiconductor laser and a laser protection lens. The main function of the CCD camera in the weld tracking sensor is to read the image. When the CCD camera is used to shoot the scene, the light reflected by the scene is transmitted to the CCD through the lens of the CCD camera. When the CCD is exposed, the photodiode is excited by the light and releases the charge, and the electrical signal of the photosensitive element is generated.

[0038] The CCD control chip uses the control signal circuit in the photosensitive element to control the current generated by the photodiode, which is output by the current transmission circuit. The CCD camera collects the electrical signals generated by the image and outputs them to the amplifier. The amplified and filtered electrical signals are sent to the A / D, which converts the electrical signals (analog signals at this time) into digital signals. The value is proportional to the strength of the electrical signal, that is, the voltage. These values ​​are actually the image data.

[0039] However, the image data obtained in the previous step cannot be used to generate an image directly. It must be output to a digital signal processor (DSP). In the DSP, the image data is post-processed by color correction, white balance processing (depending on the user's settings in the CCD camera), and encoded into the image format, resolution and other data formats supported by the camera before being stored as an image file. Finally, the image file is written to the memory of the CCD camera (built-in or external memory).

[0040] Semiconductor lasers generate power through a linear generator, with uniform density, high straightness, good stability, and a straight-line output. The lasers are available in red and blue light, and the corresponding wavelength angle and line width can be selected according to the specific requirements of the customer.

[0041] The filter protection lens is used to protect the laser camera on the one hand, and can filter light to reduce the impact during data collection on the other hand.

[0042] The laser positioning device is used to determine the position of the parts that need to be welded during transportation or positioning, so that the subsequent welding robot can reduce the steps of multiple positioning by visual detection laser, thereby improving the efficiency of positioning.

[0043] Since both the laser positioning device and the visual inspection mechanism are positioned through positioning points, in order to improve the positioning accuracy, in one embodiment, the positioning points on the positioning fixture are located on the side, and the positioning points on the same batch of parts are in the same position. After the positioning fixture fixes the parts, two positioning pins are set on the side of the positioning fixture where the parts are not blocked. The positioning pins are used as the positioning points of the fixture, and at the same time, parts with obvious structural features are selected on the parts as the positioning points of the parts. This is to allow the welding robot to accurately weld according to the simulated path after the laser positioning device performs path simulation. The visual inspection mechanism will determine the positioning points on the positioning fixture and the parts, and ensure that the welding robot will not have welding quality problems due to position deviation of the positioning fixture during movement when welding according to the simulated path based on the relative positions between the positioning points.

[0044] In order to further reduce errors and improve welding quality, in one embodiment, a comparison module is also included. The comparison module is communicated with the laser locator and the visual detection mechanism. The comparison module compares the position of the positioning point in the laser locator with the position of the positioning point detected by the visual detection mechanism. After the positioning fixture arrives at the welding station, the visual detection mechanism will detect the relative position between the positioning fixture and the positioning point on the part. Since the laser locator uses the two positioning points on the positioning fixture as the starting point and the end point when performing path simulation, and passes through the two positioning points on the part in sequence, the visual detection mechanism will detect the relative position between the positioning fixture and the positioning point on the part, and use the comparison module to compare the laser locator with the positioning point on the part. The distance between the positioning points detected by the optical position finder during path simulation is compared with the distance detected by the visual inspection mechanism. When the comparison results of the comparison module are consistent, it means that the welding position of the part during welding is consistent with the saved welding position. Therefore, the welding robot can now run according to the saved simulated path to complete the welding, reducing the time for re-adjustment and ensuring that the quality of the product remains consistent during the welding process. When the comparison results of the comparison module are inconsistent, the visual inspection mechanism will correct the simulation path according to the comparison error, allowing the welding robot to weld according to the corrected path to avoid position deviations of the parts and affect the welding quality.

[0045] When the comparison results of the comparison module are inconsistent, the simulation path needs to be corrected to reduce the error. Therefore, in order to better correct the error to obtain a new welding path, in one embodiment, the visual inspection mechanism includes an ARC system, a visual sensor and an ultrasonic sensor. The welding robot performs automatic welding through the ARC system. The ARC system can use visual devices such as cameras or depth cameras to obtain images or point cloud data of objects. By analyzing and processing the image or point cloud, the features of the product, such as edges, colors, textures, etc., can be extracted to determine the position and distance of the object. Therefore, the position and distance of the parts are obtained through visual sensors and ultrasonic sensors. When the positioning fixture has a position deviation, the visual inspection module can be corrected through the ARC system to ensure that the welding process will not be affected by errors.

[0046] In order to ensure that the positioning points on the parts can be clearly photographed to obtain their positions during the inspection process of the visual inspection mechanism, in one embodiment, the visual inspection mechanism also includes a fill light, and the irradiation area of ​​the fill light coincides with the shooting area of ​​the visual sensor. The fill light can increase the brightness of the inspected object, especially in a darker environment. This helps to provide a clearer image, allowing the inspection algorithm to better identify and analyze the characteristics of the object. The fill light can also reduce the impact of light, such as specific shadows, distortions or color changes. This helps to eliminate unnecessary errors in inspection and improve the accuracy of inspection. In addition, different inspection locations or environments may have different lighting conditions, which will affect the stability of the inspection algorithm and the reliability of the results. Adding a fill light can provide a uniform lighting condition, making the inspection process more controllable and stable.

[0047] In order to ensure that the welding robot can perform welding according to the simulated path, in one embodiment, the welding robot body is a robot with six degrees of freedom joints. The six degrees of freedom joint robot has high motion accuracy and control accuracy, and can achieve a welding effect with higher surface quality. Its welding process has high stability, high welding speed, high welding quality, and high automation level, and is more suitable for high-quality, large-scale, and efficient welding production.

[0048] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A spot welding flow processing method, characterized in that: The following steps are involved: Provide positioning jigs, assembly lines, welding robots, handling robots, laser position finders and visual inspection mechanisms; The positioning jig is located on the assembly line and two positioning points are respectively provided on the positioning jig and the parts. The laser position finder and the visual inspection mechanism are respectively installed on the handling robot and the welding robot; During the handling process, the laser position finder scans the parts and the positioning fixture according to the position of the positioning point to form a scanning path and transmit it to the welding robot. After the scanning is completed, when the positioning fixture moves to the welding station through the assembly line, the visual inspection mechanism determines the position of the positioning points on the positioning fixture and the parts and forms a new scanning path, and compares the new scanning path with the scanning path accepted by the welding robot. If the comparison results are consistent, the welding robot performs welding according to the set program. If not, the welding robot corrects the set program according to the comparison error and then performs welding; A photoelectric sensor is also provided, which is installed on the welding robot and is used to detect the welding quality of the weld; The positioning points on the positioning fixture are located on the side, and the positioning points on the same batch of parts are located at the same position; The two positioning points set on the positioning fixture determine the position of the positioning fixture, and the two positioning points set on the part determine the position of the part. The scanning path takes a certain position on the positioning fixture as the starting point and another positioning point as the end point, and passes through the two positioning points of the part in sequence. After scanning, the scanning path is transmitted to the welding robot.

2. A spot welding flow processing method according to claim 1, characterized in that: The detection signals of the photoelectric sensor are visible light-plasma spectrum, near infrared-laser reflection spectrum and far infrared-molten pool spectrum.

3. A spot welding flow processing method according to claim 1, characterized in that: The laser position finder is composed of a CCD camera, a semiconductor laser and a laser protection lens.

4. A spot welding flow processing method according to claim 1, characterized in that: It also includes a comparison module, which is communicatively connected with the laser positioning instrument and the visual detection mechanism, and compares the position of the positioning point in the laser positioning instrument with the position of the positioning point detected by the visual detection mechanism.

5. A spot welding flow processing method according to claim 1, characterized in that: The visual detection mechanism includes an ARC system, a visual sensor and an ultrasonic sensor, and the welding robot performs automatic welding through the ARC system.

6. A spot welding flow processing method according to claim 5, characterized in that: The visual detection mechanism also includes a fill light, and the irradiation area of ​​the fill light coincides with the shooting area of ​​the visual sensor.

7. A spot welding flow processing method according to claim 1, characterized in that: The welding robot body is a robot with six degrees of freedom joints.

Images