Weld tracking and real-time compensation system for automotive welding jigs

Abstract

The application discloses a welding seam tracking and real-time compensation system of an automobile welding fixture, relates to the technical field of data processing, and obtains real-time data of the surface of an automobile workpiece, obtains a welding path, stores the welding path into a welding robot, obtains an actual path of welding, transmits the actual path to the welding robot through a laser vision sensor, compares the welding path and the actual path to determine whether there is deviation, adjusts and moves the welding fixture and a welding gun posture according to the actual path if there is deviation, and performs deviation compensation. The application obtains real-time data of the surface of the automobile workpiece through image acquisition and processing, real-time data before welding starts and during welding, combines segmentation numbering and repeated iteration, improves recognition accuracy and confidence output, calculates a deviation value based on the welding path and the actual path, performs real-time compensation, improves the accuracy of automobile welding of the welding fixture, and reduces human interference and material waste.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a welding seam tracking and real-time compensation system of an automobile welding fixture. BACKGROUND

[0002] Welding is a necessary process method in modern mechanical manufacturing industry, and is widely used in vehicle manufacturing. Unlike the current traditional spot welding process, brazing can achieve molecular bonding between two steel plates. In common terms, the hardness of the welded steel plate is equivalent to a whole steel plate, thereby improving the strength of the vehicle body by 30%, and greatly improving the bonding accuracy of the vehicle body. Generally speaking, the bumps from the ground when the vehicle is running on the road will be converted into thousands of times of torsional motion per minute to test the vehicle body. If the bonding accuracy and strength of the vehicle body are not enough, the vehicle will have frequent abnormal noise and loud noise, and in severe cases, the damage of the parts installed on the vehicle, such as the transmission and front and rear axles, or the rupture of the vehicle body may occur. Therefore, brazing has great significance compared to traditional spot welding.

[0003] At present, the Chinese invention patent with the application number CN107020452 discloses a welding robot, which mainly comprises a base, a rotating motor, a rotating shaft, a large arm rod, a small arm rod, a welding head and the like. One end of the large arm rod is matched with the base, and the other end of the large arm rod is used for being matched with the small arm rod and the welding head of the welding robot. The large arm rod drives the small arm rod and the welding head to make multi-dimensional movement in space, so as to implement welding on the joint of the whole plate. The end of the welding head is provided with a welding wire. The welding wire is melted by using the high-temperature welding arc generated by the welding head, and then the welding on the whole plate is implemented. Although the precision of automobile welding is improved, the temperature is too high during welding, which can cause the offset of the welding seam. The welding seam is not tracked, and real-time compensation is not implemented when the welding seam is offset. This is not conducive to the improvement of the accuracy of welding, and there is no remedy when the offset occurs, which has limitations. SUMMARY

[0004] The technical problem solved by the present application is that in the related art, the welding robot makes multi-dimensional movement in space to implement welding on the joint of the whole plate, so as to improve the precision of automobile welding. However, real-time tracking of the welding seam is not implemented, and compensation measures are not implemented when the welding position is offset. This is not conducive to the improvement of the accuracy of welding, and there is still a situation that the welding path is offset due to high temperature changes. The problem has not been fundamentally solved, and there are limitations.

[0005] To solve the above technical problems, the present application provides the following technical scheme: a welding seam tracking and real-time compensation system of an automobile welding fixture, comprising a collection module, an analysis module and a compensation module:

[0006] The collection module is used for obtaining real-time data of a welded automobile workpiece.

[0007] The analysis module is used to obtain a welding path, store the welding path into a welding robot, obtain an actual path of welding, and transmit the actual path of welding to the welding robot through a laser vision sensor;

[0008] The compensation module is used to compare the welding path and the actual path to determine whether there is deviation, and adjust and move a welding fixture and a welding gun pose according to the actual path to compensate for the deviation.

[0009] As a preferred scheme of the automobile welding fixture seam tracking and real-time compensation system, the acquisition module comprises an acquisition unit.

[0010] The acquisition unit is used to scan the surface of the automobile workpiece in real time according to the laser vision sensor bound on the automobile welding fixture, leave a stripe on the surface of the workpiece through a linear laser, form a deformation, and acquire the deformation by the laser vision sensor.

[0011] As a preferred scheme of the automobile welding fixture seam tracking and real-time compensation system, the image acquisition and preprocessing module further comprises a calculation unit.

[0012] The real-time data comprises shape data, welding fixture state data, welding gun pose data, position data and feature data of the automobile workpiece weld.

[0013] The calculation unit is used to capture the stripe in the deformation by a high-frame-rate camera, obtain pixel coordinates, convert the pixel coordinates into three-dimensional point cloud data by laser triangulation, and acquire the shape data, the welding fixture state data, the position data and the feature data of the automobile workpiece weld according to the three-dimensional point cloud data.

[0014] The position data comprises weld position deviation, weld center point coordinates, weld normal vector and weld tangent vector.

[0015] The feature data comprises groove depth data, angle data, gap data and thermal deformation rate data.

[0016] The shape data comprises curved surface curvature change data and thermal deformation displacement data.

[0017] The welding fixture state data comprises welding fixture shape data and positioning reference data.

[0018] The welding gun pose data comprises approach angle data and working angle data.

[0019] As a preferred scheme of the automobile welding fixture seam tracking and real-time compensation system, the analysis module comprises a welding path unit.

[0020] The welding path unit is used to obtain a workpiece number according to automobile workpiece data and feature data, and obtain a welding path in a corresponding welding gun console system according to the workpiece number, wherein the automobile workpiece comprises an automobile workpiece shape and an automobile workpiece material.

[0021] As a preferred scheme of the welding seam tracking and real-time compensation system of the automobile welding fixture, the welding path unit comprises key point coordinates.

[0022] The welding path unit is used to set first time information according to the key point coordinates and position data, divide the welding path into a plurality of welding path segments according to the first time information, number the welding path segments, and obtain first number information, wherein the first time information comprises a start time point, a welding seam center time point and an end time point.

[0023] According to the first number information, a welding gun posture is adjusted by using a welding robot to control the welding gun, and posture data is obtained.

[0024] According to the face curvature change data, the thermal deformation displacement data and the posture data, a welding path segment direction is obtained by using a laser vision sensor of the welding fixture.

[0025] The welding path unit is repeated until the welding seam is completed, different welding path segment directions in the first number information are obtained, the different welding path segment directions are iteratively merged according to the first time information to obtain a welding path, and the welding path is stored in the welding robot.

[0026] As a preferred scheme of the welding seam tracking and real-time compensation system of the automobile welding fixture, the analysis module further comprises an actual path unit.

[0027] The actual path unit is used to input the posture data into a welding gun control system in the welding robot before formal welding of the automobile welding fixture.

[0028] The laser vision sensor is used to scan the surface of the automobile workpiece in real time to obtain accurate position data, feature data and shape data of the welding seam of the automobile workpiece.

[0029] According to the position data, second time information is set, and an actual welding path is obtained in real time according to the second time information, the feature data and the shape data, and is transmitted to the welding robot by the laser vision sensor, wherein the second time information comprises a start time point, a welding seam center time point and an end time point.

[0030] As a preferred scheme of the welding seam tracking and real-time compensation system of the automobile welding fixture, the compensation module comprises a deviation compensation unit.

[0031] The deviation compensation unit is used for the welding robot to extract the welding path stored in advance when obtaining the actual path, and the welding path is segmented and numbered according to the second time information to obtain a second welding path.

[0032] According to the second time information, the actual path is compared with the second welding path one by one to obtain position deviation three-dimensional data.

[0033] If the position deviation three-dimensional data is zero, there is no deviation, and the actual path is correct.

[0034] If the position deviation three-dimensional data is not zero, there is deviation, and adjustment is made according to the position deviation three-dimensional data.

[0035] If the position deviation three-dimensional data is greater than zero, the automobile workpiece is merged inward by using the welding fixture to drag the automobile workpiece, the welding gun posture is adjusted according to the position deviation three-dimensional data, and deviation compensation is performed.

[0036] If the position deviation three-dimensional data is less than zero, the automobile workpiece is separated outward by using the welding fixture to drag the automobile workpiece, the welding gun posture is adjusted according to the position deviation three-dimensional data, and deviation compensation is performed.

[0037] The actual path of the next step is welded according to the welding path.

[0038] As a preferred scheme of the welding seam tracking and real-time compensation system of the automobile welding fixture, wherein the three-dimensional point cloud data:

[0039] According to the linear laser generator in the laser vision sensor, laser is emitted to leave a stripe on the surface of the automobile workpiece, so that the surface of the automobile workpiece is deformed, the high-frame-rate camera in the laser vision sensor captures the stripe in the deformation, the reflected light of the stripe enters the camera, photoelectric conversion is performed, a pixel voltage signal is obtained, the voltage signal is analog-digital converted to obtain a pixel coordinate, and the pixel coordinate is extracted by laser triangulation to obtain second three-dimensional point cloud.

[0040] As a preferred scheme of the welding seam tracking and real-time compensation system of the automobile welding fixture, wherein the actual path of welding is obtained:

[0041] Before formal welding of the automobile welding fixture, the posture data is input into the welding gun control system in the welding robot, when formal welding of the automobile welding fixture starts, the welding fixture scans the surface of the automobile workpiece in real time through the laser vision sensor, the image acquisition preprocessing module is repeated, real-time welding seam tracking is performed, the position data, feature data and shape data of the automobile workpiece weld at each moment before welding are obtained, the actual path unit is repeated according to the position data, feature data and shape data, and the actual path of welding is obtained in real time.

[0042] As a preferred scheme of the weld tracking and real-time compensation system of the automobile welding fixture according to the application, wherein: the actual path of the next step is welded according to the welding path.

[0043] After the welding robot receives the actual path from the laser vision sensor, deviation calculation is performed by comparing the actual path with the welding path, and position deviation three-dimensional data is obtained. According to the position deviation three-dimensional data, the welding robot will immediately adjust the welding fixture and the welding gun posture, so that the actual path of the next step is welded according to the welding path. The whole welding process is repeated according to the image calculation unit, the path analysis unit and the deviation compensation unit from the formal start until the end of the welding.

[0044] The beneficial effects of the application are: through the laser vision sensor of the acquisition module, the shape data before welding and during welding, the welding fixture state data, the welding gun posture data, the position data and the feature data of the automobile workpiece weld are accurately obtained, which helps to accurately calculate the position deviation three-dimensional data and lays a foundation for subsequent compensation. The analysis module obtains the subsequent trend of the weld through the laser vision sensor on the welding fixture, compares it with the welding path trend, obtains the position deviation three-dimensional data, can accurately control the weld trend, greatly reduces the manual intervention, improves the welding accuracy, reduces the waste of resources, and at the same time, the compensation module adjusts the welding gun and the welding fixture according to the obtained position deviation three-dimensional data, compensates the error from the welding source, improves the welding accuracy, greatly speeds up the production rhythm, significantly optimizes the production efficiency, and greatly reduces the comprehensive cost. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 The basic flowchart of the weld tracking and real-time compensation system of the automobile welding fixture is provided for an embodiment of the application. DETAILED DESCRIPTION

[0046] In order to make the above-mentioned objects, features and advantages of the application more obvious and easy to understand, the specific embodiments of the application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments.

[0047] Embodiments, with reference to Figure 1 , the weld tracking and real-time compensation system of the automobile welding fixture is provided, which comprises: an acquisition module, an analysis module and a compensation module:

[0048] The acquisition module is used to obtain real-time data of the welded automobile workpiece.

[0049] The analysis module is used to obtain the welding path, store the welding path into the welding robot, obtain the actual path of welding, and transmit the actual path of welding to the welding robot through the laser vision sensor;

[0050] The compensation module is used to compare whether there is deviation between the welding path and the actual path, and if there is deviation, adjust and move the welding fixture and the welding gun attitude according to the actual path to compensate for the deviation.

[0051] In this embodiment, the laser vision sensor of the acquisition module is used to accurately obtain the shape data before and during welding, the welding fixture state data, the welding gun attitude data, the position data and the feature data of the automobile workpiece weld, which helps to accurately calculate the position deviation three-dimensional data and lays a foundation for subsequent compensation. The analysis module obtains the subsequent trend of the weld through the laser vision sensor on the welding fixture in real time, compares the trend with the welding path, obtains the position deviation three-dimensional data, and can accurately control the trend of the weld, greatly reduces the manual intervention, improves the accuracy of welding, reduces the waste of resources, and at the same time, the compensation module adjusts the welding gun and the welding fixture according to the obtained position deviation three-dimensional data, compensates for the error from the welding source, improves the accuracy of welding, greatly speeds up the production rhythm, and significantly optimizes the production efficiency, greatly reduces the comprehensive cost.

[0052] The acquisition unit is used to scan the surface of the automobile workpiece in real time according to the laser vision sensor bound on the automobile welding fixture, leave a stripe on the surface of the workpiece through a linear laser, form a deformation, and obtain the deformation through the laser vision sensor;

[0053] The real-time data includes shape data, welding fixture state data, welding gun attitude data, position data and feature data of the automobile workpiece weld;

[0054] The calculation unit is used to capture the stripe in the deformation through the high-frame-rate camera, obtain the pixel coordinates, convert the pixel coordinates into three-dimensional point cloud data through the laser triangulation method, and obtain the shape data, welding fixture state data, position data and feature data of the automobile workpiece weld according to the three-dimensional point cloud data;

[0055] The position data includes weld position deviation, weld center point coordinates, weld normal vector and weld tangent vector;

[0056] The feature data includes groove depth data, angle data, gap data and thermal deformation rate data;

[0057] The shape data includes curved surface curvature change data and thermal deformation displacement data;

[0058] The welding fixture state data includes welding fixture shape data and positioning reference data;

[0059] The welding gun attitude data includes approach angle data and working angle data.

[0060] In specific implementation, the surface of the automobile workpiece is scanned in real time by the laser vision sensor bound on the automobile welding fixture, real-time data before and during welding is acquired in real time, the welding seam is tracked in real time, the welding direction is monitored, the foundation for subsequent real-time compensation is laid, the human resources are greatly reduced, the calculation accuracy is improved, and resource waste is avoided.

[0061] The welding path unit is used to obtain the workpiece number according to the automobile workpiece data and the feature data, and obtain the welding path in the welding gun console system according to the workpiece number, the automobile workpiece including the automobile workpiece shape and the automobile workpiece material.

[0062] The welding path unit is used to set first time information according to the key point coordinates and the position data, divide the welding path into a plurality of welding path segments according to the first time information, number the welding path segments, and obtain first number information, wherein the first time information includes a start time point, a welding seam center time point and an end time point.

[0063] According to the welding gun controlled by the welding robot, the welding gun posture is adjusted according to the first number information, and posture data is obtained.

[0064] According to the surface curvature change data, the thermal deformation displacement data and the posture data, the welding path segment direction is obtained by using the welding fixture laser vision sensor.

[0065] The welding path unit is repeated until the welding seam is completed, the directions of different welding path segments in the first number information are obtained, the directions of different welding path segments are iteratively merged according to the first time information to obtain a welding path, and the welding path is stored in the welding robot.

[0066] The actual path unit is used to input the posture data into the welding gun control system in the welding robot before the automobile welding fixture is formally welded.

[0067] The surface of the automobile workpiece is scanned in real time by the laser vision sensor, and the accurate position data, the feature data and the shape data of the automobile workpiece welding seam are acquired.

[0068] According to the position data, second time information is set, and the actual welding path is obtained in real time according to the second time information, the feature data and the shape data, and the laser vision sensor is transmitted to the welding robot, wherein the second time information includes a start time point, a welding seam center time point and an end time point.

[0069] In specific implementation, the welding path is segmented according to the first time information, the welding path segments are numbered, and first numbering information is obtained, wherein the first time information includes a starting time point, a weld center time point and an ending time point, the entire welding process is covered according to the first time information, the data obtained is more real and accurate, and the entire welding process is complete, the welding path is segmented and numbered according to the first time information, which makes the entire welding process orderly, facilitates the extraction of the welding path segment trend, provides convenience for subsequent comparison, greatly reduces the waste of resources and human intervention, makes the cumbersome steps simple, and provides data support for managers. According to the position data, the second time information is set, and the actual welding path is obtained in real time according to the second time information, the feature data and the shape data. When the welding officially starts, the laser vision sensor obtains real-time data, sets the second time information according to the position data, and the second time information divides the weld trend into multiple segments. When welding the weld, the laser vision sensor obtains the second time information, the feature data and the shape data of the current segment, so as to judge and obtain the actual welding path of the next step. In this way, the segmented second time information, the feature data and the shape data are obtained, and the actual welding path is judged. This method uses simple iteration to obtain the actual path, saves a lot of time and calculation cost, reasonably allocates resources, improves the efficiency of weld tracking, and has strong emergency response capability.

[0070] The deviation compensation unit is used for the welding robot to extract the welding path stored in advance when obtaining the actual path, and the welding path is segmented and numbered according to the second time information to obtain a second welding path.

[0071] According to the second time information, the actual path is compared with the second welding path one by one to obtain position deviation three-dimensional data.

[0072] If the position deviation three-dimensional data is zero, there is no deviation, and the actual path is correct.

[0073] If the position deviation three-dimensional data is not zero, there is deviation, and adjustment is made according to the position deviation three-dimensional data.

[0074] If the position deviation three-dimensional data is greater than zero, the automobile workpiece is merged inward by using the welding fixture, the welding gun posture is adjusted according to the position deviation three-dimensional data, and deviation compensation is performed.

[0075] If the position deviation three-dimensional data is less than zero, the automobile workpiece is separated outward by using the welding fixture, the welding gun posture is adjusted according to the position deviation three-dimensional data, and deviation compensation is performed.

[0076] The next actual path is welded according to the welding path.

[0077] In the embodiment, the welding path is segmented and numbered according to the second time information to obtain a second welding path, the first time information in the welding path unit is replaced by the second time information, the welding path unit is repeated to obtain a new welding path segment direction, the actual path is compared with the second welding path one by one according to the second time information to obtain position deviation three-dimensional data, and the segments of the actual path and the second welding path are compared according to the second time information, so that whether the position deviation three-dimensional data exists is more intuitively obtained, and only each segment needs to be calculated, so that the calculation amount is greatly reduced, the calculation efficiency is improved, accurate data are provided for subsequent implementation compensation, and the cost of manpower and materials is reduced.

[0078] According to the linear laser generator in the laser vision sensor, laser is emitted to leave a stripe on the surface of the automobile workpiece, the surface of the automobile workpiece is deformed, the high-frame-rate camera in the laser vision sensor captures the stripe in the deformation, the reflected light of the stripe enters the camera, photoelectric conversion is performed, a pixel voltage signal is acquired, the voltage signal is analog-digital converted, a pixel coordinate is obtained, and the pixel coordinate is extracted by laser triangulation to obtain three-dimensional point cloud data.

[0079] Before formal welding of the automobile welding fixture, the posture data is input into a welding gun control system in the welding robot, when the automobile welding fixture starts formal welding, the welding fixture scans the surface of the automobile workpiece in real time through the laser vision sensor, the image acquisition and preprocessing module is repeated, real-time weld tracking is performed, position data, feature data and shape data of the automobile workpiece weld before each moment in the welding process are acquired, the actual path unit is repeated according to the position data, the feature data and the shape data, and the actual path of welding is obtained in real time.

[0080] After the welding robot receives the actual path transmitted by the laser vision sensor, the actual path is compared with the welding path, deviation calculation is performed, position deviation three-dimensional data is obtained, according to the position deviation three-dimensional data, the welding robot immediately adjusts the posture of the welding fixture and the welding gun, so that the actual path in the following is welded according to the welding path, and the whole welding process is repeated according to the image calculation unit, the path analysis unit and the deviation compensation unit from the formal start until the welding is finished.

[0081] In the embodiment, the image acquisition preprocessing module is repeated, and real-time data of the current weld is obtained through the laser vision sensor without interruption, thereby laying a foundation for obtaining the actual path, making the data based on reality, not virtual, and improving the authenticity and operability of the welding path. According to the one-word laser generator in the laser vision sensor, laser is emitted, and a stripe is left on the surface of the automobile workpiece. The one-word laser adopts a high-frame-rate camera arranged symmetrically left and right, and through the laser diode, collimating lens and cylindrical mirror in the one-word laser generator, the point laser is diffused into a uniform straight line to obtain the stripe. According to the position deviation three-dimensional data, the welding robot will immediately adjust the welding fixture and welding gun posture, so that the actual path for the next welding is in accordance with the welding path. The position deviation three-dimensional data is divided into positive value, negative value and zero value. According to different position deviation three-dimensional data, the welding robot is controlled to operate correspondingly, so that the welding quality is greatly improved, the welding defect rate is greatly reduced, and the production efficiency is significantly improved.

[0082] The laser vision sensor of the acquisition module accurately obtains the shape data before and during welding, the welding fixture state data, the welding gun posture data, the position data and the feature data of the automobile workpiece weld, which helps to accurately calculate the position deviation three-dimensional data and lays a foundation for subsequent compensation. The analysis module obtains the subsequent trend of the weld through the laser vision sensor on the welding fixture, compares it with the welding path trend, obtains the position deviation three-dimensional data, accurately controls the weld trend, greatly reduces manual intervention, improves the accuracy of welding, reduces resource waste, and the compensation module adjusts the welding gun and the welding fixture according to the obtained position deviation three-dimensional data to compensate for errors from the welding source, improve the accuracy of welding, greatly speed up the production rhythm, significantly optimize the production efficiency, and greatly reduce the comprehensive cost.

[0083] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A weld seam tracking and real-time compensation system for automotive welding fixtures, characterized in that, include: Acquisition module, analysis module, and compensation module: The acquisition module is used to obtain real-time data of welded automotive workpieces; The analysis module is used to obtain the welding path, store the welding path in the welding robot, obtain the actual welding path, and transmit it to the welding robot through the laser vision sensor. The compensation module is used to compare whether there is a deviation between the welding path and the actual path. If there is a deviation, the welding fixture and welding torch posture are adjusted and moved according to the actual path to compensate for the deviation. The acquisition module includes an acquisition unit: The acquisition unit is used to scan the surface of the automotive workpiece in real time according to the laser vision sensor attached to the automotive welding fixture, leave stripes on the surface of the workpiece with a linear laser to form deformation, and acquire the deformation through the laser vision sensor. The acquisition module also includes a computing unit: The real-time data includes shape data, welding fixture status data, welding torch posture data, and the position and feature data of the weld seam on the automotive workpiece. The computing unit is used by the high frame rate camera to capture the stripes in the deformation, obtain pixel coordinates, convert the pixel coordinates into three-dimensional point cloud data through laser triangulation, and obtain shape data, welding fixture status data, position data and feature data of automotive workpiece weld seam based on the three-dimensional point cloud data. The location data includes weld position deviation, weld center point coordinates, weld normal vector, and weld tangent vector; The feature data includes bevel depth data, angle data, gap data, and thermal deformation rate data; The shape data includes surface curvature variation data and thermal deformation displacement data; The welding fixture status data includes welding fixture shape data and positioning reference data; The welding torch attitude data includes approach angle data and working angle data; The analysis module includes a welding path unit: The welding path unit is used to obtain the workpiece number based on the automotive workpiece data and feature data, and to obtain the welding path in the corresponding welding gun control console system based on the workpiece number. The automotive workpiece includes the automotive workpiece shape and automotive workpiece material. The welding path unit includes key point coordinates: The welding path unit is used to set first time information according to the key point coordinates and position data, divide the welding path into multiple welding path segments according to the first time information, number the welding path segments, and obtain first number information, wherein the first time information includes the start time point, the weld center time point, and the end time point. The welding robot controls the welding torch and adjusts the torch posture according to the first number information to obtain posture data. Based on the surface curvature change data, thermal deformation displacement data, and attitude data, the welding path segmentation direction is obtained using the welding fixture and the laser vision sensor. Repeat the welding path unit until the weld is finished, obtain the different welding path segments in the first number information, iteratively merge the different welding path segments according to the first time information to obtain the welding path, and store the welding path in the welding robot. The compensation module includes a deviation compensation unit: The deviation compensation unit is used by the welding robot to extract the pre-stored welding path when acquiring the actual path, and to divide and number the welding path according to the second time information to obtain the second welding path. Based on the second time information, the actual path is compared one-to-one with the second welding path to obtain three-dimensional data of positional deviation. If the three-dimensional data of the positional deviation is zero, then there is no deviation and the actual path is correct; If the three-dimensional position deviation data is not zero, then there is a deviation, and adjustments should be made according to the three-dimensional position deviation data. If the three-dimensional positional deviation data is greater than zero, the automotive workpiece is dragged inward using the welding fixture to merge, and the welding torch posture is adjusted according to the three-dimensional positional deviation data to compensate for the deviation. If the three-dimensional positional deviation data is less than zero, the automotive workpiece is pulled outward using a welding fixture, and the welding torch posture is adjusted according to the three-dimensional positional deviation data to compensate for the deviation. The actual path in the next step is then welded according to the welding path.

2. The weld seam tracking and real-time compensation system for automotive welding fixtures as described in claim 1, characterized in that, The analysis module also includes an actual path unit: The actual path unit is used to input the posture data into the welding gun control system in the welding robot before the formal welding of the automotive welding fixture. The laser vision sensor scans the surface of automotive workpieces in real time to obtain precise location, feature, and shape data of the weld seams. Based on the location data, second time information is set, and the actual welding path is obtained in real time based on the second time information, feature data, and shape data. This path is then transmitted to the welding robot via a laser vision sensor. The second time information includes the start time point, the weld center time point, and the end time point.

3. The weld seam tracking and real-time compensation system for automotive welding fixtures as described in claim 1, characterized in that, The three-dimensional point cloud data: The laser is emitted by the linear laser generator in the laser vision sensor, leaving stripes on the surface of the car workpiece, causing the surface of the car workpiece to deform. The high frame rate camera in the laser vision sensor captures the stripes in the deformation. The light reflected from the stripes enters the camera, undergoes photoelectric conversion, and obtains the pixel voltage signal. The voltage signal is converted from analog to digital to obtain the pixel coordinates. The pixel coordinates are extracted by laser triangulation to obtain three-dimensional point cloud data.

4. The weld seam tracking and real-time compensation system for automotive welding fixtures as described in claim 3, characterized in that, The actual path to achieve welding: Before the automotive welding fixture officially begins welding, the posture data is input into the welding torch control system in the welding robot. When the automotive welding fixture officially begins welding, the welding fixture scans the surface of the automotive workpiece in real time through a laser vision sensor. The image acquisition and preprocessing module is repeated to perform real-time weld seam tracking and obtain the position data, feature data, and shape data of the automotive workpiece weld seam at each moment in the welding process. Based on the position data, feature data, and shape data, the actual path unit is repeated to obtain the actual path of the welding torch.

5. The weld seam tracking and real-time compensation system for automotive welding fixtures as described in claim 1, characterized in that, The next step involves welding the actual path according to the welding path: After receiving the actual path from the laser vision sensor, the welding robot compares it with the welding path, calculates the deviation, and obtains three-dimensional positional deviation data. Based on the three-dimensional positional deviation data, the welding robot immediately adjusts the welding fixture and welding torch posture so that the actual path is welded according to the welding path. The entire welding process repeats from the start of the process according to the image calculation unit, path analysis unit, and deviation compensation unit until the welding is completed.

Images