Automobile rear longitudinal beam assembly welding system based on 3D visual guidance and multi-clamp cooperation

Abstract

The invention relates to the technical field of automatic welding, and particularly discloses an automobile rear longitudinal beam assembly welding system based on 3D visual guidance and multi-clamp collaboration, which comprises an incoming material transportation module for transporting an automobile rear longitudinal beam incoming material from a feeding area to a table garage position of a welding production line based on an automatic guide vehicle; the grabbing track correction module automatically corrects the grabbing track of the robot based on the space position and posture data of the automobile rear longitudinal beam incoming material, and the high-precision corrected grabbing track of the robot is obtained; the supplied material grabbing module is used for controlling a robot to grab an automobile rear longitudinal beam supplied material based on a high-precision correction grabbing track and placing the automobile rear longitudinal beam supplied material on a spot welding clamp; the spot welding operation moving module moves the spot welding clamp and the automobile rear longitudinal beam incoming material to a welding position based on a servo sliding table and completes spot welding operation; after spot welding is completed, the arc welding operation moving module controls the robot to grab the part obtained after spot welding is completed and place the part on an arc welding clamp located on the positioner, and arc welding operation is completed. And automation of the whole process from incoming material transportation to welding and discharging is achieved.

Description

Technical Field

[0001] The present invention relates to the field of automated welding technology, and in particular to an automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration. Background Art

[0002] In the automotive manufacturing industry, welding processes are a critical step in vehicle body production, directly impacting the overall quality and safety of the vehicle. As a crucial component of the vehicle body structure, the welding quality of the rear longitudinal beam assembly plays a crucial role in energy absorption and occupant protection during a collision. The development of the automotive manufacturing industry has placed higher demands on the automation, precision, and efficiency of rear longitudinal beam welding. Traditional welding systems for rear longitudinal beam welding rely primarily on manual experience and basic positioning techniques.

[0003] In terms of incoming material transportation, the degree of automation of material transportation is low, and the transportation route planning is not flexible enough, making it difficult to adapt to the needs of different production rhythms. In the grasping process, the handling and clamping of incoming materials need to be done manually, which has low production efficiency and high labor intensity. Manual placement of parts may cause position deviations, which in turn affects the consistency of welding quality and the accuracy of subsequent welding. During spot welding and arc welding operations, manual operation is required for fixture switching and adjustment, which affects the production rhythm. The coordination between fixtures and welding equipment is poor, and the connection between each process is not smooth enough, which reduces production efficiency. At the same time, due to the lack of real-time monitoring and intelligent adjustment of the welding process, it is difficult to ensure the stability of welding quality. Although the industry has been exploring improvements, the existing welding system still faces many challenges in coping with complex and changing production tasks and improving welding accuracy and efficiency. It is urgent to introduce new technologies and methods for optimization and upgrading.

[0004] Therefore, the present invention proposes an automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration. Summary of the Invention

[0005] The present invention provides an automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration, including: an incoming material transportation module using an automatic guided vehicle to realize the automated transportation of automobile rear longitudinal beam incoming materials from a specific loading area to the welding production line station garage, efficiently supplying materials, reducing manpower and avoiding transportation deviations. The grabbing trajectory correction module automatically corrects the robot grabbing trajectory based on the incoming material spatial position and posture data, improves the grabbing accuracy, and ensures the accurate welding position of the incoming material. The incoming material grabbing module places the incoming material to the spot welding fixture according to the corrected trajectory, provides a precise starting position for the spot welding operation, and ensures the welding quality. The spot welding operation moving module uses a servo slide to accurately control the spot welding position, realize an automated spot welding process, and improve the quality stability and efficiency of spot welding. After spot welding, the arc welding operation moving module accurately transfers the parts to the arc welding fixture to complete arc welding, realizing seamless connection between spot welding and arc welding, ensuring welding continuity, improving overall welding quality and production efficiency, reducing the uncertainty of manual intervention, and promoting the automation and intelligence of the welding process.

[0006] The present invention provides an automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration, comprising: The incoming material transportation module is used to transport the incoming rear longitudinal beams of automobiles from the loading area to the pallet garage of the welding production line using an automatic guided vehicle; The grasping trajectory correction module is used to automatically correct the robot's grasping trajectory based on the spatial position and posture data of the incoming rear longitudinal beam of the automobile, obtaining a high-precision corrected grasping trajectory of the robot; The incoming material grabbing module is used to control the robot to grab the incoming car rear longitudinal beam based on the high-precision correction grabbing trajectory and place it on the spot welding fixture; The spot welding operation moving module is used to move the spot welding fixture and the rear longitudinal beam of the automobile to the welding position based on the servo slide and complete the spot welding operation; The arc welding operation moving module is used to control the robot to grab the spot-welded parts and place them on the arc welding fixture on the positioner after spot welding is completed, and complete the arc welding operation.

[0007] Preferably, the incoming material transport module includes: The initial path generation submodule is used to generate the initial transportation path between the loading area and the pallet garage of the welding production line; The obstacle avoidance mechanism triggering submodule is used to control the AGV to pick up the incoming rear longitudinal beam from the loading area and transport it to the pallet garage of the welding production line according to the initial transportation path. At the same time, the millimeter-wave radar on the AGV detects obstacles within a preset range centered on the AGV in real time. When the millimeter-wave radar detects an obstacle, the obstacle avoidance strategy is triggered, and a new obstacle-avoiding transportation path is generated. The rear longitudinal beam is then transported to the pallet garage of the welding production line according to the new obstacle-avoiding transportation path. The obstacle avoidance strategies include: Generate multiple initial obstacle avoidance transport paths based on the obstacle positions detected by the millimeter-wave radar; Calculate the path cost value of each initial obstacle avoidance transportation path based on the path cost function: Where, is the path cost of a single initial obstacle avoidance transport path, is the total number of position points in a single initial obstacle avoidance transport path, is the first location points, is the first location points, is the first Position point and The Euclidean distance between the locations, is the total number of obstacles detected by the millimeter-wave radar, is the first The horizontal coordinate value of the position point, is the first The vertical coordinate value of the position point, For the The horizontal coordinate value of the obstacle, For the The vertical coordinate value of each obstacle; The initial obstacle-avoiding transport path with the minimum path cost among all the initial obstacle-avoiding transport paths is regarded as the latest obstacle-avoiding transport path.

[0008] Preferably, the grabbing trajectory correction module includes: The incoming material posture scanning submodule is used to scan and identify the spatial position and posture data of the incoming rear longitudinal beam of the automobile through a depth camera; The spatial pose matching submodule is used to match the spatial position and pose data of the automobile rear longitudinal beam material with the preset material 3D model based on the point cloud registration algorithm to generate a spatial pose deviation matrix; The grasping trajectory correction submodule is used to automatically correct the robot's grasping trajectory based on the spatial posture deviation matrix to obtain a high-precision corrected grasping trajectory of the robot.

[0009] Preferably, the grabbing trajectory correction submodule includes: An initial correction unit, used to automatically correct the robot's grasping trajectory based on the spatial posture deviation matrix to obtain the robot's initial corrected grasping trajectory; a dynamic model generation unit, configured to generate a state vector of the robot at each moment based on the dynamic vibration offset and the rate of change of the dynamic vibration offset of the robot at each moment, and to generate a dynamic model of the robot in combination with the state transfer matrix, the control input matrix, and the process noise vector; a measurement model generating unit, configured to generate a measurement model of the robot based on a dynamic model of the robot, a measurement matrix, and a measurement noise vector; a vibration offset estimation unit, configured to estimate the dynamic vibration offset of the robot based on a dynamic model and a measurement model of the robot and a Kalman filter; The reverse compensation unit is used to perform reverse displacement compensation on the initial correction grasping trajectory of the robot based on the dynamic vibration offset of the robot to obtain a high-precision correction grasping trajectory of the robot.

[0010] Preferably, the initial correction unit includes: A position correction subunit is used to perform position correction on each position point in the grasping trajectory of the correction robot based on the translation vector in the spatial posture deviation matrix to obtain the position correction grasping trajectory of the robot; The posture correction subunit is used to adjust the posture of each position point in the robot's position correction grasping trajectory based on the rotation matrix in the spatial posture deviation matrix to obtain the robot's initial correction grasping trajectory.

[0011] Preferably, it also includes: The fixture contact pressure detection module is used to detect the real-time contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture based on the pressure sensor array inside the spot welding fixture; The fixture support posture adjustment module is used to adjust the fixture support posture based on the real-time contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture.

[0012] Preferably, the fixture support posture adjustment module includes: The sliding trend prediction submodule is used to analyze the sliding trend of the automobile rear longitudinal beam material in the fixture based on the real-time contact pressure distribution data between the automobile rear longitudinal beam material and the fixture; The support posture adjustment submodule is used to adjust the support posture of the fixture based on the sliding trend of the automobile rear longitudinal beam material in the fixture.

[0013] Preferably, the sliding trend prediction submodule includes: A normal pressure calculation unit, configured to calculate a total normal pressure in a contact area between the rear longitudinal beam material and the fixture based on real-time contact pressure distribution data between the rear longitudinal beam material and the fixture; A resultant force analysis unit is used to calculate the friction force on the fixture of the rear longitudinal beam material based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture, and to determine the resultant force on the fixture of the rear longitudinal beam material in combination with other external forces on the fixture; A maximum static friction force analysis unit is used to calculate the maximum static friction force between the rear longitudinal beam material and the fixture based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture; The sliding trend analysis unit is used to analyze the sliding trend of the automobile rear longitudinal beam material in the fixture based on the tangential component of the resultant force and the maximum static friction force exerted on the automobile rear longitudinal beam material on the fixture.

[0014] Preferably, the support posture adjustment submodule includes: An optimization objective function construction unit is used to construct an objective function with the goal of minimizing the sliding tendency when the sliding tendency of the automobile rear longitudinal beam material in the fixture exceeds a sliding tendency threshold; The posture optimization unit is used to solve the optimal support posture vector of the fixture based on the objective function and the upper and lower limits of the support posture vector of the fixture, and control the support posture of the fixture based on the optimal support posture vector of the fixture.

[0015] Preferably, the arc welding fixture comprises: The clamping force is controlled by a proportional valve through a pneumatic clamping mechanism and a thermocouple array that are adjustable within a preset range; The thermocouple array monitors the real-time temperature of the welding area on the part after spot welding is completed in real time, and dynamically adjusts the welding current based on the real-time temperature of the welding area on the part after spot welding is completed.

[0016] The beneficial effects of the present invention compared to the prior art are as follows: the incoming material transportation module uses an automatic guided vehicle to realize the automated transportation of the incoming material of the rear longitudinal beam of the automobile from a specific loading area to the welding production line station garage, efficiently supplying materials, reducing manpower and avoiding transportation deviations. The grabbing trajectory correction module automatically corrects the robot grabbing trajectory based on the spatial position and posture data of the incoming material, improves the grabbing accuracy, and ensures the accurate welding position of the incoming material. The incoming material grabbing module places the incoming material to the spot welding fixture according to the corrected trajectory, provides a precise starting position for the spot welding operation, and ensures the welding quality. The spot welding operation mobile module uses a servo slide to accurately control the spot welding position, realize an automated spot welding process, and improve the quality stability and efficiency of spot welding. After spot welding, the arc welding operation mobile module accurately transfers the parts to the arc welding fixture to complete arc welding, achieving seamless connection between spot welding and arc welding, ensuring welding continuity, improving overall welding quality and production efficiency, reducing the uncertainty of manual intervention, and promoting the automation and intelligence of the welding process.

[0017] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0018] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 Schematic diagram of the internal functional modules of an automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration in an embodiment of the present invention; Figure 2 Schematic diagram of the overall layout of an automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration in an embodiment of the present invention; Figure 3 This is a schematic diagram of the overall layout of another automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration in an embodiment of the present invention; Figure 4 3D vision system structure diagram in an embodiment of the present invention; Figure 5 Schematic diagram of the spot welding fixture and servo slide in an embodiment of the present invention; Figure 6 1 is a structural diagram of an arc welding fixture and a positioner in an embodiment of the present invention; Figure 7 This is an assembly welding flow chart in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0021] Example 1: The present invention provides a welding system for automobile rear longitudinal beam assembly based on 3D vision guidance and multi-fixture collaboration, referring to Figures 1 to 7 ,include: The incoming material transportation module is used to transport the incoming rear longitudinal beams of automobiles from the loading area to the pallet garage of the welding production line using an automatic guided vehicle; The grasping trajectory correction module is used to automatically correct the robot's grasping trajectory based on the spatial position and posture data of the incoming rear longitudinal beam of the automobile, obtaining a high-precision corrected grasping trajectory of the robot; The incoming material grabbing module is used to control the robot to grab the incoming car rear longitudinal beam based on the high-precision correction grabbing trajectory and place it on the spot welding fixture; The spot welding operation moving module is used to move the spot welding fixture and the rear longitudinal beam of the automobile to the welding position based on the servo slide and complete the spot welding operation; The arc welding operation moving module is used to control the robot to grab the spot-welded parts and place them on the arc welding fixture on the positioner after spot welding is completed, and complete the arc welding operation.

[0022] In this embodiment, the spatial position and posture data of the incoming rear longitudinal beam of an automobile is obtained by scanning and identifying it with a depth camera. This data reflects the specific position of the incoming rear longitudinal beam in three-dimensional space (e.g., its coordinates in the workshop coordinate system) and its posture (e.g., tilt and rotation angle), which are key foundational data for precise grasping and welding. For example, if the depth camera scans the coordinates of one end of the rear longitudinal beam as (x1, y1, z1) and the other end as (x2, y2, z2), and it is rotated by a certain angle around each axis, these coordinates and angles constitute the spatial position and posture data.

[0023] In this embodiment, the robot's grasping trajectory is the path the robot's arm follows when grasping an incoming automotive rear longitudinal beam. The initial trajectory is planned according to a pre-set standard model, but the actual position and posture of the incoming material may deviate, often requiring subsequent correction. For example, the robot moves from its initial position, point A, along a planned curve to point B to grasp the incoming material. This curve from A to B is the grasping trajectory. However, the actual incoming material may not be at point B, and the trajectory may require adjustment.

[0024] In this embodiment, the robot's high-precision corrected grasping trajectory is based on the spatial position and posture data of the incoming material, the rear longitudinal beam of the automobile. This precise path is obtained by correcting the initial grasping trajectory. This correction comprehensively considers factors such as deviations between the actual incoming material and the preset model, as well as the robot's own dynamic vibration offset, to ensure accurate grasping and placement, and precise welding position. For example, after obtaining the actual incoming material data, a series of corrections are performed to determine the precise motion path from point A to the actual incoming material position, point C, which is the high-precision corrected grasping trajectory.

[0025] In this embodiment, the servo slide moves the spot welding fixture and the incoming rear longitudinal beam material to the welding position and completes the spot welding operation. This means that the servo slide can precisely control the position movement. After the spot welding fixture clamps the incoming rear longitudinal beam material, the servo slide accurately moves them to the specific welding position according to a preset program. Once in place, the spot welding equipment is activated to complete the spot welding and achieve the initial connection and fixation of the parts. Similar to a production line, the servo slide acts as a precise transport track, carrying the spot welding fixture and incoming material, moving them smoothly and accurately from their initial position to the designated welding position below the welding robot, where the welding robot then completes the spot welding.

[0026] In this embodiment, after spot welding is completed, the robot is controlled to grab the spot-welded parts and place them on the arc welding fixture on the positioner to complete the arc welding operation. This means that after the initial connection of the rear longitudinal beam of the automobile is completed by spot welding, the robot is controlled to clamp the parts according to the high-precision correction grasping trajectory and place them on the arc welding fixture on the positioner for fixation. The positioner adjusts the position angle of the fixture and parts to provide suitable welding conditions. The robot then completes the arc welding, strengthens the welding part, and improves the quality. For example, in actual production, after spot welding is completed, the robot accurately grabs the spot-welded parts and places them on the arc welding fixture on the positioner. The positioner adjusts the angle according to the program, and the arc welding robot performs arc welding to make the weld more solid.

[0027] In this embodiment, the system consists of the following subsystems or structures: AGV transport system: Automatic guided vehicles (AGVs) are responsible for transporting incoming materials from the loading area to the pallet garage on the welding production line, realizing unmanned logistics distribution.

[0028] 3D vision guidance system: Installed above the robot's grasping position, it uses a depth camera to scan and identify the spatial position and posture data of the incoming material, guiding the robot to automatically correct the grasping trajectory.

[0029] Spot welding fixture and servo slide mechanism: The spot welding fixture is moved to the welding station by a servo slide mechanism to ensure high precision and consistency of fixture positioning.

[0030] Positioner: The arc welding fixture is installed on the positioner, which can quickly switch positions to ensure high consistency.

[0031] Robot grasping and welding module: Industrial robots complete the entire process of material grabbing, fixture switching, spot welding and arc welding.

[0032] Unloading platform: The final product is placed by the robot on the unloading table for easy subsequent transportation.

[0033] How it works: Material transportation and loading: After the AGV transports the materials to the pallet garage, the 3D vision system identifies and guides the robot to grab the materials and place them on the spot welding fixture.

[0034] Spot welding operation: The spot welding fixture is moved to the welding position by the servo slide to complete the spot welding operation.

[0035] Arc welding operation: After spot welding is completed, the robot grabs the part and places it on the arc welding fixture, which is located on the positioner to complete the welding.

[0036] Unloading and transfer: The welded assembly is grabbed by the robot and placed on the unloading table, waiting for transfer.

[0037] The beneficial effects of the above technical solutions are as follows: the incoming material transportation module uses an automatic guided vehicle to realize the automated transportation of the incoming materials of the rear longitudinal beam of the automobile from a specific loading area to the welding production line station garage, efficiently supplying materials, reducing manpower and avoiding transportation deviations. The grasping trajectory correction module automatically corrects the robot grasping trajectory based on the spatial position and posture data of the incoming materials, improves the grasping accuracy, and ensures the accurate welding position of the incoming materials. The incoming material grasping module places the incoming materials on the spot welding fixture according to the corrected trajectory, provides a precise starting position for the spot welding operation, and ensures the welding quality. The spot welding operation mobile module uses a servo slide to accurately control the spot welding position, realize the automated spot welding process, and improve the quality stability and efficiency of spot welding. After spot welding, the arc welding operation mobile module accurately transfers the parts to the arc welding fixture to complete the arc welding, realizing seamless connection between spot welding and arc welding, ensuring welding continuity, improving the overall welding quality and production efficiency, reducing the uncertainty of manual intervention, and promoting the automation and intelligence of the welding process.

[0038] Example 2: Based on Example 1, the incoming material transportation module includes: The initial path generation submodule is used to generate the initial transportation path between the loading area and the pallet garage of the welding production line; The obstacle avoidance mechanism triggering submodule is used to control the AGV to pick up the incoming rear longitudinal beam from the loading area and transport it to the pallet garage of the welding production line according to the initial transportation path. At the same time, the millimeter-wave radar on the AGV detects obstacles within a preset range centered on the AGV in real time. When the millimeter-wave radar detects an obstacle, the obstacle avoidance strategy is triggered, and a new obstacle-avoiding transportation path is generated. The rear longitudinal beam is then transported to the pallet garage of the welding production line according to the new obstacle-avoiding transportation path. The obstacle avoidance strategies include: Generate multiple initial obstacle avoidance transport paths based on the obstacle positions detected by the millimeter-wave radar; Calculate the path cost value of each initial obstacle avoidance transportation path based on the path cost function: Where, is the path cost of a single initial obstacle avoidance transport path, is the total number of position points in a single initial obstacle avoidance transport path, is the first location points, is the first location points, is the first Position point and The Euclidean distance between the locations, is the total number of obstacles detected by the millimeter-wave radar, is the first The horizontal coordinate value of the position point, is the first The vertical coordinate value of the position point, For the The horizontal coordinate value of the obstacle, For the The vertical coordinate value of each obstacle; The initial obstacle-avoiding transport path with the minimum path cost among all the initial obstacle-avoiding transport paths is regarded as the latest obstacle-avoiding transport path.

[0039] In this embodiment, generating an initial transport route from the loading area to the pallet bays of the welding production line means that before material transportation begins, the system pre-plans a route from the loading area, where incoming rear longitudinal beams are stored, to the pallet bays of the welding production line. This provides directional guidance for the AGV's initial travel. For example, based on information such as the workshop layout and equipment location, a relatively direct route that aligns with the production process can be planned as the initial transport route.

[0040] In this embodiment, the preset range, centered around the AGV, refers to a specific spatial area surrounding the AGV, used to determine the millimeter-wave radar's detection range. This range must be determined based on factors such as the AGV's speed, braking capability, and the potential distribution of obstacles within the vehicle, ensuring timely detection of obstacles that could affect its movement. For example, a circular area with a radius of 5 meters centered around the AGV can be used as the preset range.

[0041] In this embodiment, the millimeter-wave radar on the AGV detects obstacles within a preset range centered on the AGV in real time. Specifically, the millimeter-wave radar installed on the AGV continuously monitors the presence of obstacles within the set area. The millimeter-wave radar transmits and receives millimeter-wave signals, analyzing the characteristics of the reflected waves to determine the location, speed, and other information of obstacles, providing real-time data support for the AGV's obstacle avoidance decisions. For example, while the AGV is in motion, the millimeter-wave radar continuously scans the surrounding area within a 5-meter radius. If an object enters this range, it immediately feeds relevant information back to the control system.

[0042] In this embodiment, the latest obstacle-avoiding transport path refers to a new path replanned by the system based on specific algorithms and strategies after the millimeter-wave radar detects an obstacle, avoiding the current obstacle and allowing the AGV to continue smoothly transporting the rear longitudinal beam to the welding line's parking garage. This path must bypass obstacles while ensuring transportation efficiency and avoiding excessive increases in transportation time and distance. For example, if the AGV is originally traveling in a straight line and detects an obstacle ahead, the system will plan a curved path to the left as the latest obstacle-avoiding transport path.

[0043] In this embodiment, multiple initial obstacle avoidance transport paths are generated based on the obstacle positions detected by the millimeter-wave radar. This means that after knowing the obstacle position information, the system uses a specific algorithm to generate multiple different candidate paths that can avoid obstacles, starting from the current position of the AGV and ending at the parking garage. These paths will take into account different detour directions, distances, and other factors, providing multiple possibilities for subsequent selection of the optimal path. For example, if the obstacle is located to the right in front of the AGV, the system may generate multiple different initial obstacle avoidance transport paths, such as detouring to the left, detouring backward, and then forward.

[0044] In this embodiment, the path cost of the initial obstacle avoidance transport path is calculated using a specific path cost function, which is used to measure the quality of each initial obstacle avoidance transport path. The path cost takes into account factors such as the length of the path and the distance to obstacles. A lower path cost indicates that the path has greater advantages in terms of safety and transportation efficiency. The system will tend to select the path with the smallest path cost as the final latest obstacle avoidance transport path. For example, according to a given path cost function, the cost of one initial obstacle avoidance transport path is calculated to be 0.8, and another is calculated to be 0.6. Relatively speaking, the path with a cost of 0.6 is better.

[0045] In this embodiment, the preset adjustment factor is a parameter in the path cost function that is used to adjust the calculation of the path cost value, balancing the relative importance of factors such as path length and obstacle distance in path evaluation. For example, when calculating the path cost value of the initial obstacle-avoiding transport path, if the preset adjustment factor is set to a large value, the influence of the distance factor on the path cost value in the path evaluation will increase, making the system more inclined to select paths farther away from obstacles. Conversely, if the preset adjustment factor is set to a small value, the influence of the path length factor will be relatively greater, and the system may prefer to select a shorter path, even if it is slightly closer to the obstacle. The specific value needs to be pre-set based on the different requirements of safety and transportation efficiency in the actual scenario.

[0046] The beneficial effects of the above technical solutions are as follows: the initial path generation submodule can autonomously generate the initial transportation path between the loading area and the welding production line garage, clarifying the starting direction for material transportation and allowing transportation to start in an orderly manner. The obstacle avoidance mechanism triggering submodule enables the automatic guided vehicle to use the millimeter-wave radar to monitor obstacles within the preset range in real time when transporting according to the initial path. Once an obstacle is found, the obstacle avoidance strategy is immediately triggered to quickly generate the latest obstacle avoidance transportation path to avoid collisions, ensure the safety and continuity of material transportation, and reduce production delays. The obstacle avoidance strategy generates multiple initial obstacle avoidance transportation paths and uses the path cost function to calculate the path cost value. It comprehensively considers the relationship between the path length and the obstacle distance, and finally selects the path with the minimum path cost value as the latest obstacle avoidance transportation path, which not only achieves safe obstacle avoidance, but also takes into account the path economy and efficiency, reduces the path growth and time consumption caused by obstacle avoidance, ensures transportation efficiency, and supplies materials to the welding production line in a timely manner.

[0047] Example 3: Based on Example 1, the grasping trajectory correction module includes: The incoming material posture scanning submodule is used to scan and identify the spatial position and posture data of the incoming rear longitudinal beam of the automobile through a depth camera; The spatial pose matching submodule is used to match the spatial position and pose data of the automobile rear longitudinal beam material with the preset material 3D model based on the point cloud registration algorithm to generate a spatial pose deviation matrix; The grasping trajectory correction submodule is used to automatically correct the robot's grasping trajectory based on the spatial posture deviation matrix to obtain a high-precision corrected grasping trajectory of the robot.

[0048] In this embodiment, the depth camera is used to scan and identify the spatial position and posture data of the incoming rear longitudinal beam material. This involves using the depth camera to scan the incoming rear longitudinal beam material. The depth camera can obtain three-dimensional spatial information of an object. By emitting and receiving light signals and calculating the round-trip time of the light, it determines the distance between each point on the object's surface and the camera. It then identifies the specific spatial position of the incoming rear longitudinal beam material, such as its coordinate value in a certain coordinate system, as well as its own posture information, such as its tilt angle and rotation direction, providing basic data for subsequent precise manipulation of the incoming material. For example, the depth camera can scan the coordinates (x, y, z) of a certain end point of the rear longitudinal beam in a specific coordinate system and derive the angle of rotation of the incoming material around a certain axis. These data are the spatial position and posture data of the incoming rear longitudinal beam material.

[0049] In this embodiment, the spatial position and posture data of the incoming rear longitudinal beam material are matched with a preset 3D model of the material based on a point cloud registration algorithm to generate a spatial pose deviation matrix. This involves presenting the spatial position and posture data of the incoming rear longitudinal beam material, captured by a depth camera, in the form of a point cloud. The point cloud contains coordinate information representing numerous surface feature points. The point cloud registration algorithm is then used to match this measured point cloud data with the point cloud data of the preset 3D model of the rear longitudinal beam. The algorithm calculates the differences in spatial position and posture between the actual incoming material and the preset model, quantifying these differences into a matrix, the spatial pose deviation matrix. This matrix clearly reflects the deviations of the actual incoming material from the standard model in terms of translation and rotation, providing a key basis for subsequent correction of the robot's grasping trajectory. For example, after calculation using the point cloud registration algorithm, the resulting spatial pose deviation matrix can indicate information such as whether the actual incoming material has been translated a certain distance in the x-direction or rotated a certain angle around the y-axis.

[0050] The beneficial effects of the above technical solutions are as follows: the incoming material posture scanning submodule uses a depth camera to accurately scan and identify the spatial position and posture data of the incoming material of the rear longitudinal beam of the automobile, providing accurate original information for subsequent correction, making the acquisition of the incoming material status more accurate. The spatial posture matching submodule matches the scanned data with the preset incoming material three-dimensional model based on the point cloud registration algorithm and generates a spatial posture deviation matrix. Through this comparative analysis, the difference between the actual incoming material and the standard model can be clearly obtained, providing a quantitative basis for trajectory correction. The grasping trajectory correction submodule automatically corrects the robot grasping trajectory based on the spatial posture deviation matrix to obtain a high-precision corrected grasping trajectory, greatly improving the accuracy of the robot's grasping, enabling the robot to accurately grasp the incoming material, reducing problems such as inaccurate welding position caused by grasping deviation, effectively ensuring welding quality, improving production reliability and stability, and helping the automobile rear longitudinal beam assembly welding system to operate more efficiently and accurately.

[0051] Example 4: Based on Example 3, the grasping trajectory correction submodule includes: An initial correction unit, used to automatically correct the robot's grasping trajectory based on the spatial posture deviation matrix to obtain the robot's initial corrected grasping trajectory; a dynamic model generation unit, configured to generate a state vector of the robot at each moment based on the dynamic vibration offset and the rate of change of the dynamic vibration offset of the robot at each moment, and to generate a dynamic model of the robot in combination with the state transfer matrix, the control input matrix, and the process noise vector; a measurement model generating unit, configured to generate a measurement model of the robot based on a dynamic model of the robot, a measurement matrix, and a measurement noise vector; a vibration offset estimation unit, configured to estimate the dynamic vibration offset of the robot based on a dynamic model and a measurement model of the robot and a Kalman filter; The reverse compensation unit is used to perform reverse displacement compensation on the initial correction grasping trajectory of the robot based on the dynamic vibration offset of the robot to obtain a high-precision correction grasping trajectory of the robot.

[0052] In this embodiment, the robot's initial corrected grasping trajectory is derived by making preliminary adjustments to the robot's original grasping trajectory based on the spatial pose deviation matrix. This makes preliminary corrections to the robot's grasping path based on the deviation between the incoming material's actual spatial position and pose and the preset model. However, it does not yet account for the effects of dynamic vibration during robot operation, laying the foundation for further precise corrections.

[0053] In this embodiment, based on the dynamic vibration offset of the robot at each moment and the rate of change of dynamic vibration offset Generate the state vector of the robot at each moment, and combine the state transition matrix, control input matrix, and process noise vector to generate the dynamic model of the robot. This means that during the operation of the robot, its dynamic vibration offset (the deviation between the actual position of the robot and the ideal position) and the rate of change of the dynamic vibration offset change continuously over time. These time-varying quantities are combined to form a state vector, which is used to describe the state of the robot at each moment. The state transition matrix describes the natural evolution of the robot's state from one moment to the next, the control input matrix reflects the impact of external control instructions on the robot's state, and the process noise vector covers the various random interference factors that are inevitable in the operation of the robot. The dynamic model constructed by combining these elements can accurately characterize the dynamic behavior of the robot under the influence of various factors, providing a basis for more accurate analysis and correction of the robot's grasping trajectory; in, The state vector at time ,in, is the n-dimensional dynamic vibration offset vector, is its rate of change vector; The dynamic model of the robot can be expressed as ; Where, It is a 2n×2n state transition matrix, which describes the system state from time At the time For example, for a simple linear system, It can be expressed as ,in is the n-dimensional identity matrix, is the time interval, is an n-dimensional zero matrix; yes The state vector at the moment; is a 2n×m control input matrix; is an m-dimensional control input vector. In this scenario, if there is no external control input, Item can be ignored; is a 2n-dimensional process noise vector with a mean of zero and a covariance matrix of Gaussian distribution, is the variance of the Gaussian distribution.

[0054] In this embodiment, a measurement model of the robot is generated based on the robot's dynamic model, measurement matrix, and measurement noise vector. This is based on the established robot dynamic model and takes into account the actual measurement process. The measurement matrix links the robot's internal state with the actual measurable physical quantity, and the measurement noise vector represents the inevitable errors and interference in the measurement process. By combining the dynamic model, measurement matrix, and measurement noise vector to generate a measurement model, the relationship between the actual measurement value and the robot's true state can be more accurately reflected, thereby providing an effective way to estimate the robot's true state and facilitating more precise correction of the robot's grasping trajectory. The measurement model can be expressed as: Where, is a p-dimensional measurement vector, which can be measured, for example, by an accelerometer or displacement sensor mounted on the robot; is a p×2n measurement matrix that describes the relationship between the system state and the measurement value; is a p-dimensional measurement noise vector with zero mean and covariance matrix Gaussian distribution, is the covariance matrix of the measurement noise.

[0055] In this embodiment, the robot's dynamic vibration offset is estimated based on the robot's dynamic model, measurement model, and Kalman filter, utilizing the algorithmic tool of the Kalman filter. The Kalman filter can predict the robot's state at the next moment based on the robot's dynamic model, and, combined with the actual measurement values obtained from the measurement model, optimally estimates the robot's dynamic vibration offset through continuous iterative calculations. It can effectively handle the uncertainty in the dynamic model and measurement model, filter out noise interference, and thus obtain a relatively accurate estimate of the robot's dynamic vibration offset, providing key data for the subsequent precise correction of the grasping trajectory, including: Prediction steps: Where, It is at the moment Based on the moment The estimated value of the prediction of the system state; is the covariance matrix of the predicted state; It's time The best estimate of ; It's time The covariance matrix of the optimal estimate; for The transposed matrix of .

[0056] Update steps: Where, is the Kalman gain matrix; for The transposed matrix of is the optimal estimate at time k; It's time The covariance matrix of the optimal estimate; By continuously performing prediction and update steps, the Kalman filter can estimate the dynamic vibration offset of the robot in real time. , it is The first n elements of .

[0057] In this embodiment, the robot's initial correction grasping trajectory is subjected to reverse displacement compensation based on the robot's dynamic vibration offset to obtain a high-precision correction grasping trajectory of the robot. This means that after obtaining the robot's dynamic vibration offset, the robot's grasping trajectory will deviate due to the offset. To eliminate this deviation, the initial correction grasping trajectory is subjected to corresponding displacement compensation in the direction opposite to the vibration offset direction. For example, if dynamic vibration causes the robot to deviate in a certain direction, it is moved the same distance in the opposite direction, thereby obtaining a high-precision correction grasping trajectory after taking into account the influence of dynamic vibration, ensuring that the robot can accurately grasp the incoming rear longitudinal beam of the automobile, thereby improving the accuracy and quality of welding, wherein: Assume the original trajectory planning position is , the dynamic vibration offset estimated by the Kalman filter is , then the corrected grasping trajectory position It can be expressed as: ; Where, and are all n-dimensional position vectors, in meters (m); is the n-dimensional dynamic vibration offset vector, also in meters (m).

[0058] The beneficial effects of the above technical solutions are as follows: the initial correction unit performs preliminary correction on the robot's grasping trajectory based on the spatial posture deviation matrix, quickly narrowing the gap between the grasping trajectory and the ideal state, laying the foundation for subsequent more accurate corrections. The dynamic model generation unit comprehensively considers the dynamic vibration offset and the rate of change of the dynamic vibration offset of the robot at each moment, combines the state transfer matrix, the control input matrix and the process noise vector to generate a dynamic model, comprehensively and accurately describes the dynamic characteristics of the robot, and provides accurate model support for further optimization of the grasping trajectory. The measurement model generation unit generates a measurement model based on the robot dynamic model, the measurement matrix and the measurement noise vector, making the measurement of the robot state more scientific and reasonable, and improving the accuracy and reliability of the measurement. The vibration offset estimation unit uses a Kalman filter to accurately estimate the dynamic vibration offset of the robot based on the dynamic model and the measurement model, which can effectively remove noise interference and extract true vibration offset information. The reverse compensation unit performs reverse displacement compensation on the initial corrected grasping trajectory based on the estimated dynamic vibration offset. This targeted compensation method effectively eliminates trajectory deviations caused by robot vibration, ultimately obtaining a high-precision corrected grasping trajectory. This greatly improves the accuracy of the robot's grasping action, thereby significantly improving the positioning accuracy of the vehicle's rear longitudinal beam grasping and placement, further ensuring welding quality, improving the stability and reliability of the entire welding system, and reducing product defects and production delays caused by inaccurate grasping.

[0059] Example 5: On the basis of embodiment 4, the initial correction unit includes: A position correction subunit is used to perform position correction on each position point in the grasping trajectory of the correction robot based on the translation vector in the spatial posture deviation matrix to obtain the position correction grasping trajectory of the robot; The posture correction subunit is used to adjust the posture of each position point in the robot's position correction grasping trajectory based on the rotation matrix in the spatial posture deviation matrix to obtain the robot's initial correction grasping trajectory.

[0060] In this embodiment, the translation vector in the spatial pose deviation matrix describes the difference in spatial translation between the actual position of the incoming material for the rear longitudinal beam of the vehicle and the preset model position. It consists of three components, corresponding to the x, y, and z axes in three-dimensional space, and intuitively reflects the amount of translation of the incoming material along each coordinate axis relative to the standard position in space. For example, a translation vector of (10, -5, 3) indicates that the incoming material has moved 10 units more along the x-axis than the preset position, 5 units less along the y-axis, and 3 units more along the z-axis.

[0061] In this embodiment, the position of each position point in the robot's grasping trajectory is corrected based on the translation vector in the spatial pose deviation matrix to obtain the robot's position-corrected grasping trajectory. This is done by adjusting each position point on the robot's grasping trajectory accordingly based on the incoming material position deviation reflected by the translation vector. Each point on the grasping trajectory is displaced and adjusted in the x, y, and z axes according to the corresponding components of the translation vector so that the robot's grasping trajectory can match the actual position of the incoming material, thereby obtaining a position-corrected grasping trajectory. For example, if the initial coordinates of a point on the grasping trajectory are (x0, y0, z0) and the translation vector is (Δx, Δy, Δz), then the coordinates of the point after correction become (x0+Δx, y0+Δy, z0+Δz). Similarly, all position points are corrected to obtain a position-corrected grasping trajectory.

[0062] In this embodiment, the robot's position-corrected grasping trajectory is derived from a position correction of the original grasping trajectory based on the translation vector. This trajectory accounts for the deviation of the incoming rear longitudinal beam material's spatial position relative to the pre-set model, allowing the robot's grasping path to more closely align with the material's actual position, providing a positional foundation for subsequent precise grasping. However, the grasping posture is not yet adjusted.

[0063] In this embodiment, the rotation matrix within the spatial pose deviation matrix describes the rotational difference between the actual pose of the incoming rear longitudinal beam material and the pre-set model pose. This is a 3×3 matrix that, through matrix operations, implements coordinate system rotation transformations. This matrix reflects the material's rotation about the x, y, and z axes in three-dimensional space, determining the material's rotation angle and direction relative to the standard pose.

[0064] In this embodiment, the robot's initial corrected grasping trajectory is obtained by adjusting the posture of each position point in the position-corrected grasping trajectory based on the rotation matrix in the spatial pose deviation matrix. This involves using the rotation matrix to correct the posture of each point on the grasping trajectory that has already completed position correction. By performing matrix operations on the rotation matrix and the coordinates of each point on the position-corrected grasping trajectory, each point is rotated according to the actual posture of the incoming material relative to the preset model posture, thereby adjusting the robot's grasping posture. Ultimately, an initial corrected grasping trajectory is obtained that takes into account both position deviation and posture deviation, further improving the accuracy and stability of the robot's grasping.

[0065] The above technical solution has the following beneficial effects: the position correction subunit uses the translation vectors in the spatial pose deviation matrix to perform position correction on each position point in the correction robot's grasping trajectory, precisely adjusting the robot's grasping position and effectively compensating for the deviation of the incoming material's spatial position from the preset model. This ensures that the robot's grasping position more closely matches the actual incoming material position, providing a foundation for subsequent accurate grasping. The posture correction subunit uses the rotation matrix in the spatial pose deviation matrix to perform posture adjustment on each position point in the position-corrected grasping trajectory, meticulously correcting the robot's grasping posture and ensuring that the robot grasps the incoming material at the appropriate angle, avoiding unstable grasping or inaccurate placement caused by improper posture. Through the orderly execution of position correction and posture correction, not only is the spatial pose deviation meticulously addressed step by step, but the robot's grasping trajectory is also gradually optimized. The resulting initial corrected grasping trajectory is more precise, further improving the accuracy and stability of the robot's grasping, effectively ensuring the quality of the grasping and placement process before welding the rear longitudinal beam of the automobile, reducing welding defects caused by grasping problems, and improving the overall production efficiency and product quality of the welding system.

[0066] Example 6: On the basis of Example 1, the following further aspects are included: The fixture contact pressure detection module is used to detect the real-time contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture based on the pressure sensor array inside the spot welding fixture; The fixture support posture adjustment module is used to adjust the fixture support posture based on the real-time contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture.

[0067] In this embodiment, the pressure sensor array inside the spot welding fixture consists of multiple pressure sensors installed on the inside of the spot welding fixture where it contacts the incoming material from the rear longitudinal beam of the vehicle. These sensors work together to detect the pressure at the contact point between the fixture and the incoming material in real time, providing data support for determining the placement of the incoming material in the fixture. For example, after the incoming material from the rear longitudinal beam of the vehicle is placed on the spot welding fixture, the pressure sensor array begins operating, collecting pressure information at each contact point.

[0068] In this embodiment, real-time contact pressure distribution data between the incoming rear longitudinal beam material and the fixture is acquired in real time by a pressure sensor array inside the spot welding fixture. This data reflects the magnitude and distribution of pressure at different locations on the contact surface between the incoming rear longitudinal beam material and the fixture. By analyzing this data, it is possible to determine whether the incoming material is positioned stably in the fixture and whether the force is evenly distributed. For example, excessive or insufficient pressure in a particular area may indicate tilted placement of the incoming material or a problem with that area of the fixture. This data provides a basis for subsequent adjustments to the fixture's support posture.

[0069] In this embodiment, the fixture's support posture refers to the position and posture of the spot welding fixture when supporting the incoming material for the automotive rear longitudinal beam. This includes the fixture's spatial location (e.g., coordinates within the welding station) as well as the fixture's own posture information, such as its angle and inclination. A suitable support posture ensures that the incoming material for the automotive rear longitudinal beam remains stable during the spot welding process, preventing displacement or deformation caused by external forces, thereby ensuring high-quality spot welding. For example, the fixture must be adjusted to a specific position and angle based on the shape and size of the incoming material to ensure that it is securely supported within the fixture.

[0070] The beneficial effects of the above technical solution are as follows: the fixture contact pressure detection module can accurately detect the contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture in real time through the pressure sensor array on the inside of the spot welding fixture. These data intuitively reflect the degree of fit between the incoming material and the fixture and the stress conditions, providing an accurate basis for subsequent adjustments. The fixture support posture adjustment module adjusts the support posture of the fixture in a targeted manner based on the real-time contact pressure distribution data. This helps to ensure that the incoming material of the automobile rear longitudinal beam is evenly and stably supported in the fixture, avoiding deformation, displacement and other problems caused by excessive or insufficient local pressure. By timely adjusting the fixture support posture, the incoming material is in the optimal position and posture before spot welding, which greatly improves the accuracy and quality of spot welding, reduces welding defects caused by poor contact between the fixture and the incoming material, improves the overall quality and stability of the welding of the automobile rear longitudinal beam assembly, ensures the smooth progress of the welding production process, improves production efficiency, and reduces the defective rate.

[0071] Example 7: Based on Example 6, the fixture support posture adjustment module includes: The sliding trend prediction submodule is used to analyze the sliding trend of the automobile rear longitudinal beam material in the fixture based on the real-time contact pressure distribution data between the automobile rear longitudinal beam material and the fixture; The support posture adjustment submodule is used to adjust the support posture of the fixture based on the sliding trend of the automobile rear longitudinal beam material in the fixture.

[0072] In this embodiment, the sliding tendency of the incoming material of the rear longitudinal beam of the automobile in the fixture is obtained based on the analysis of the real-time contact pressure distribution data between the incoming material of the rear longitudinal beam of the automobile and the fixture, reflecting the possibility and direction of the incoming material sliding tendency in the fixture. By calculating the total normal pressure, friction, resultant force and maximum static friction in the contact area, and comparing the tangential component of the resultant force with the maximum static friction, it is determined whether the incoming material has a sliding risk and the direction in which it may slide. For example, when the tangential component of the resultant force is greater than the maximum static friction, the incoming material has a sliding tendency. Analyzing these data can also know the size and general direction of the sliding tendency, providing a basis for adjusting the fixture support posture in advance and preventing the incoming material from sliding during the spot welding process, thereby ensuring the welding quality.

[0073] The beneficial effects of the above technical solution are as follows: the sliding trend prediction submodule analyzes the sliding trend of the incoming material of the automobile rear longitudinal beam in the fixture based on the real-time contact pressure distribution data, which enables the system to predict the possible sliding of the incoming material in advance and change the passive response to active prevention. By deeply analyzing the pressure distribution data, the potential sliding direction and possibility of the incoming material are accurately captured, providing highly forward-looking information for subsequent adjustments. The support posture adjustment submodule adjusts the support posture of the fixture in time according to the predicted sliding trend, effectively avoiding the position offset problem caused by the sliding of the incoming material. This targeted adjustment can better constrain the incoming material, keep it stable during the spot welding process, and further improve the quality of spot welding. Since the welding defects that may be caused by sliding are prevented in advance, the production of defective products is reduced, and production efficiency is improved. At the same time, the stability of the welding process and the consistency of product quality are guaranteed, making the automobile rear longitudinal beam assembly welding system more reliable and efficient in actual production.

[0074] Example 8: Based on Example 7, the sliding trend prediction submodule includes: A normal pressure calculation unit, configured to calculate a total normal pressure in a contact area between the rear longitudinal beam material and the fixture based on real-time contact pressure distribution data between the rear longitudinal beam material and the fixture; A resultant force analysis unit is used to calculate the friction force on the fixture of the rear longitudinal beam material based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture, and to determine the resultant force on the fixture of the rear longitudinal beam material in combination with other external forces on the fixture; A maximum static friction force analysis unit is used to calculate the maximum static friction force between the rear longitudinal beam material and the fixture based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture; The sliding trend analysis unit is used to analyze the sliding trend of the automobile rear longitudinal beam material in the fixture based on the tangential component of the resultant force and the maximum static friction force exerted on the automobile rear longitudinal beam material on the fixture.

[0075] In this embodiment, the total normal pressure of the contact area between the incoming material of the rear longitudinal beam of the automobile and the fixture is calculated based on the real-time contact pressure distribution data between the incoming material and the fixture, that is, the pressure data of each contact point obtained by the pressure sensor array is integrated and calculated according to a certain mathematical method. Since these pressure data reflect the normal pressure conditions at different positions, the total normal pressure value of the entire contact area can be obtained by accumulation or other related operations. For example, the pressure values measured by each pressure sensor are added together, and the sum obtained is the total normal pressure. This value reflects the magnitude of the force in the vertical direction between the incoming material and the fixture. For another example, in the contact area, the normal pressure can be obtained by integrating the contact pressure distribution data. For a tiny area element , normal pressure It can be expressed as: For the entire contact area Integrate to get the total normal pressure : ; Where, represents the coordinate point in the contact area, It is a two-dimensional matrix, each element represents the pressure value of the corresponding coordinate point, in Pascal (Pa).

[0076] In this embodiment, the friction force of the automobile rear longitudinal beam incoming material on the fixture is calculated based on the total normal pressure of the contact area between the automobile rear longitudinal beam incoming material and the fixture, and the resultant force of the automobile rear longitudinal beam incoming material on the fixture is determined in combination with other external forces on the automobile rear longitudinal beam incoming material on the fixture. According to the friction calculation formula, the friction force is usually proportional to the normal pressure, and the friction force can be calculated using the total normal pressure and the friction coefficient (related to factors such as the incoming material and the fixture material). In addition to friction, the incoming material may also be subjected to other external forces on the fixture, such as gravity and the force generated during welding. These external forces are synthesized with the calculated friction force according to the law of force composition (such as the parallelogram law or the triangle law) to determine the resultant force of the automobile rear longitudinal beam incoming material on the fixture. For example, it is known that the total normal pressure is , the friction coefficient is , calculate the friction , combined with other external forces, the magnitude and direction of the resultant force are obtained by vector addition.

[0077] In this embodiment, the maximum static friction between the rear longitudinal beam material and the fixture is calculated based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture. According to the calculation formula of the maximum static friction, the maximum static friction is proportional to the normal pressure. The calculation is also based on the total normal pressure and the static friction coefficient (which depends on the material properties of the material and the fixture, etc.). For example, if the static friction coefficient is , the total normal pressure is , then the maximum static friction F. This maximum static friction value represents the maximum tangential force that the incoming material can withstand when it remains stationary on the fixture. Once the external force exceeds this value, the incoming material may begin to slide.

[0078] In this embodiment, the sliding tendency of the automobile rear longitudinal beam incoming material in the fixture is analyzed based on the tangential component of the resultant force and the maximum static friction force that the automobile rear longitudinal beam incoming material is subjected to on the fixture. The resultant force can be decomposed into a normal component perpendicular to the contact surface and a tangential component along the contact surface. The tangential component of the resultant force is compared with the maximum static friction force. If the tangential component of the resultant force is less than the maximum static friction force, it means that the force that causes the incoming material to slide is not enough to overcome the maximum static friction force, the incoming material remains stationary relative to the fixture, and the sliding tendency is small; on the contrary, if the tangential component of the resultant force is greater than the maximum static friction force, the incoming material has a tendency to slide, and the more the tangential component of the resultant force exceeds the maximum static friction force, the greater the sliding tendency. Through this comparative analysis, the sliding tendency of the incoming material in the fixture can be accurately judged, providing a key basis for timely adjusting the fixture support posture, and avoiding the impact of sliding of the incoming material on the welding quality during the spot welding process.

[0079] The beneficial effects of the above technical solution are as follows: The normal pressure calculation unit processes real-time contact pressure distribution data to accurately calculate the total normal pressure in the contact area between the material and the fixture, providing basic data support for subsequent analysis. This data accurately reflects the vertical force between the material and the fixture and is a key starting point for studying the interaction between the two. The resultant force analysis unit calculates friction based on the total normal pressure and integrates other external forces to determine the resultant force acting on the material in the fixture, comprehensively considering various forces affecting the material's stability. This comprehensive force analysis approach provides a more comprehensive and in-depth understanding of the material's forces and more accurately grasps the material's actual state in the fixture. The maximum static friction force analysis unit calculates the maximum static friction force based on the total normal pressure, clarifying the critical condition for the material to remain stationary in the fixture. This value is crucial for determining whether the material will slip and provides a key reference for sliding trend analysis. The sliding trend analysis unit combines the tangential component of the resultant force with the maximum static friction force to accurately analyze the material's sliding tendency in the fixture. By comparing these two key parameters, it is possible to accurately predict the risk of slippage in the incoming material, as well as the direction and likelihood of slippage. This precise prediction of slip trends enables the fixture support posture adjustment module to react early and accurately, adjusting the fixture support posture in a timely manner, effectively preventing the incoming material from slipping during the spot welding process. This further improves spot welding quality, ensures product consistency and stability, reduces defective products caused by slippage, and enhances the production efficiency and reliability of the entire welding system.

[0080] Example 9: Based on Example 7, the support posture adjustment submodule includes: An optimization objective function construction unit is used to construct an objective function with the goal of minimizing the sliding tendency when the sliding tendency of the automobile rear longitudinal beam material in the fixture exceeds a sliding tendency threshold; The posture optimization unit is used to solve the optimal support posture vector of the fixture based on the objective function and the upper and lower limits of the support posture vector of the fixture, and control the support posture of the fixture based on the optimal support posture vector of the fixture.

[0081] In this embodiment, the sliding tendency threshold is a pre-set standard value used to determine whether the sliding tendency of the incoming material of the automobile rear longitudinal beam in the fixture is within an acceptable range. If the calculated incoming material sliding tendency exceeds this threshold, it indicates a high sliding risk and requires adjustment to the fixture support posture to prevent the incoming material from sliding during the spot welding process and affecting the weld quality. If it does not exceed this threshold, the incoming material sliding risk is considered to be within a controllable range under the current fixture support posture.

[0082] In this embodiment, the objective function is constructed with the goal of minimizing sliding tendency. The goal is to find a method that can effectively reduce the sliding tendency of incoming automotive rear longitudinal beams within the fixture. Through mathematical modeling, various factors influencing the sliding tendency, such as the fixture's support angle and position, are incorporated as variables into the function. The form and specific content of the objective function will be determined based on the actual situation and the mathematical method used, but its core goal is to minimize the sliding tendency, providing a mathematical basis for subsequently solving the optimal support position of the fixture.

[0083] In this embodiment, the upper and lower limits of the support posture vector of the fixture refer to the value range limits of each component in the vector that describes the support posture of the fixture. The support posture of the fixture can be represented by a vector, and each component of the vector corresponds to a certain position or posture parameter of the fixture in space (such as x, y, z coordinates, rotation angle, etc.). The setting of the upper and lower limits is based on factors such as the physical structure limitations of the fixture, the workspace requirements, and the needs of the welding process. For example, the movement distance of the fixture in a certain direction cannot exceed a certain range, or the rotation angle must be within a specific range. These limitations constitute the upper and lower limits of each component of the support posture vector.

[0084] In this embodiment, the optimal fixture support posture vector is determined based on the objective function and the upper and lower limits of the fixture's support posture vector. This involves using a mathematical optimization algorithm to find the fixture support posture vector that minimizes the objective function (i.e., minimizes the sliding tendency) while satisfying the upper and lower limits of the support posture vector. These optimization algorithms can be linear programming, nonlinear programming, or other methods. Through iterative calculation and optimization, a set of fixture support posture parameters—the optimal support posture vector—is ultimately obtained that minimizes the sliding tendency of the incoming material under given conditions.

[0085] In this embodiment, the optimal support posture vector for the fixture is a set of parameter values derived through the aforementioned solution process. It accurately describes the fixture's support posture that minimizes the tendency of the rear longitudinal beam material to slip within the fixture under the current circumstances. Each component of this vector corresponds to the fixture's specific position or posture in space. Adjusting the fixture according to the posture determined by this vector effectively improves the material's stability during spot welding, ensuring weld quality.

[0086] In this embodiment, the fixture's support posture is controlled based on its optimal support posture vector. This involves converting the calculated optimal support posture vector into an actual control signal, which is then transmitted to the fixture's drive system or adjustment mechanism. Based on these control signals, the fixture automatically adjusts its position and posture to the state determined by the optimal support posture vector. This ensures that the incoming rear longitudinal beam material is stably supported during the spot welding process, minimizing the risk of slippage and improving weld accuracy and quality.

[0087] The above technical solution has the following beneficial effects: When the sliding tendency of the incoming material of the rear longitudinal beam exceeds a threshold, the optimization objective function is constructed with the goal of minimizing this tendency. This provides a clear and targeted optimization direction for the fixture posture adjustment, aiming to minimize the sliding tendency of the incoming material, ensure the stability of the incoming material in the fixture, and provide a reliable foundation for subsequent welding processes. The posture optimization unit calculates the optimal support posture vector for the fixture based on the constructed objective function and the upper and lower limits of the fixture support posture vector, and controls the fixture support posture accordingly. This optimal solution fully considers the physical limitations of the fixture and finds the posture that best minimizes sliding within the feasible range, making the adjustment process more scientific and precise. Through precise posture adjustment, incoming material sliding is effectively suppressed, significantly improving spot welding quality, reducing product defects caused by sliding, and thus increasing product qualification rate. Furthermore, the optimized fixture support posture reduces unnecessary adjustments, improves production efficiency, and ensures stable and efficient operation of the welding system, bringing better quality control and economic benefits to the welding production of automotive rear longitudinal beam assemblies.

[0088] Example 10: On the basis of Example 1, the arc welding fixture includes: The clamping force is controlled by a proportional valve through a pneumatic clamping mechanism and a thermocouple array that are adjustable within a preset range; The thermocouple array monitors the real-time temperature of the welding area on the part after spot welding is completed in real time, and dynamically adjusts the welding current based on the real-time temperature of the welding area on the part after spot welding is completed.

[0089] In this embodiment, the clamping force is adjustable within a preset range through the control of a proportional valve, which means that the clamping force of the pneumatic clamping mechanism on the arc welding fixture on the parts can be flexibly changed according to actual needs. As a control element, the proportional valve can proportionally adjust the flow of the fluid passing through according to the input signal, thereby controlling the pneumatic clamping mechanism to generate different amounts of clamping force. The preset range is a clamping force interval pre-set according to the characteristics of the part and the welding process requirements, ensuring that the parts can be firmly clamped to prevent them from moving during arc welding, and that the parts will not be deformed due to excessive clamping force. For example, for parts with different thicknesses or materials after spot welding, the clamping force can be changed within a suitable Newton value range by adjusting the proportional valve.

[0090] In this embodiment, a thermocouple array, comprised of multiple thermocouples, is installed on an arc welding fixture to monitor the temperature of the weld area of a component in real time after spot welding. A thermocouple is a temperature sensor based on the thermoelectric effect, converting temperature signals into electrical signals. This array of multiple thermocouples can collect temperature data from multiple locations, providing a more comprehensive and accurate picture of the temperature distribution in the weld area and providing precise data support for subsequent temperature-based adjustments to the welding current.

[0091] In this embodiment, the real-time weld zone temperature refers to the actual temperature of the welding area on the part at each moment during the arc welding process, after spot welding is complete. This temperature data is acquired in real time via a thermocouple array. Its values are constantly changing, reflecting the real-time heat transfer and distribution during the welding process and playing a key role in controlling weld quality. For example, after welding begins, the real-time weld zone temperature will rise rapidly as the welding arc is applied, and will maintain a certain fluctuation throughout the welding process.

[0092] In this embodiment, the welding current is dynamically adjusted based on the real-time weld zone temperature of the part after spot welding. This utilizes real-time weld zone temperature information monitored by a thermocouple array to automatically adjust the arc welding current. Because different welding temperatures require corresponding welding currents to ensure a good welding result, when the real-time weld zone temperature is too high, the welding current is appropriately reduced to avoid problems such as wide welds and metal spatter caused by overheating. When the temperature is too low, the welding current is increased to ensure good weld fusion and prevent defects such as incomplete penetration. This dynamic adjustment enables adaptive control of the welding process, improving the stability and reliability of welding quality.

[0093] The above technical solution has the following beneficial effects: The pneumatic clamping mechanism in the arc welding fixture can flexibly adjust the clamping force within a preset range via a proportional valve. This design allows precise control of the clamping force for different spot-welded parts of varying specifications and materials. This ensures that the parts are securely clamped during the arc welding process, preventing weld position deviations caused by unstable clamping, while also preventing part deformation caused by excessive clamping force. This effectively ensures part stability and precision during arc welding, thereby improving weld quality. Furthermore, a thermocouple array monitors the temperature of the weld area on the spot-welded part in real time and dynamically adjusts the welding current based on this temperature. This feature enables adaptive control during the welding process, as different welding temperatures require a corresponding welding current to ensure optimal welding results. If the temperature is too high or too low, the system can promptly adjust the welding current to avoid weld defects such as excessively wide or narrow welds, or incomplete weld penetration. This further improves the stability and reliability of welding quality, reduces the defect rate caused by temperature and current mismatch, and increases production efficiency, providing more precise and stable processing conditions for the arc welding process of automotive rear longitudinal beam assemblies.

[0094] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is intended to include these modifications and variations.

Claims

1. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration is characterized by: include: The incoming material transportation module is used to transport the incoming rear longitudinal beams of automobiles from the loading area to the pallet garage of the welding production line using an automatic guided vehicle; The grasping trajectory correction module is used to automatically correct the robot's grasping trajectory based on the spatial position and posture data of the incoming rear longitudinal beam of the automobile, obtaining a high-precision corrected grasping trajectory of the robot; The incoming material grabbing module is used to control the robot to grab the incoming car rear longitudinal beam based on the high-precision correction grabbing trajectory and place it on the spot welding fixture; The spot welding operation moving module is used to move the spot welding fixture and the rear longitudinal beam of the automobile to the welding position based on the servo slide and complete the spot welding operation; The arc welding operation moving module is used to control the robot to grab the spot-welded parts and place them on the arc welding fixture on the positioner after spot welding is completed, and complete the arc welding operation.

2. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 1 is characterized in that: Incoming material transport module, including: The initial path generation submodule is used to generate the initial transportation path between the loading area and the pallet garage of the welding production line; The obstacle avoidance mechanism triggering submodule is used to control the AGV to pick up the incoming rear longitudinal beam from the loading area and transport it to the pallet garage of the welding production line according to the initial transportation path. At the same time, the millimeter-wave radar on the AGV detects obstacles within a preset range centered on the AGV in real time. When the millimeter-wave radar detects an obstacle, the obstacle avoidance strategy is triggered, and a new obstacle-avoiding transportation path is generated. The rear longitudinal beam is then transported to the pallet garage of the welding production line according to the new obstacle-avoiding transportation path. The obstacle avoidance strategies include: Generate multiple initial obstacle avoidance transport paths based on the obstacle positions detected by the millimeter-wave radar; Calculate the path cost value of each initial obstacle avoidance transportation path based on the path cost function: Where, is the path cost of a single initial obstacle avoidance transport path, is the total number of position points in a single initial obstacle avoidance transport path, is the first location points, is the first location points, is the first Position point and The Euclidean distance between the locations, is the total number of obstacles detected by the millimeter-wave radar, is the first The horizontal coordinate value of the position point, is the first The vertical coordinate value of the position point, For the The horizontal coordinate value of the obstacle, For the The vertical coordinate value of each obstacle; The initial obstacle-avoiding transport path with the minimum path cost among all the initial obstacle-avoiding transport paths is regarded as the latest obstacle-avoiding transport path.

3. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 1 is characterized in that: Grasping trajectory correction module, including: The incoming material posture scanning submodule is used to scan and identify the spatial position and posture data of the incoming rear longitudinal beam of the automobile through a depth camera; The spatial pose matching submodule is used to match the spatial position and pose data of the automobile rear longitudinal beam material with the preset material 3D model based on the point cloud registration algorithm to generate a spatial pose deviation matrix; The grasping trajectory correction submodule is used to automatically correct the robot's grasping trajectory based on the spatial posture deviation matrix to obtain a high-precision corrected grasping trajectory of the robot.

4. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 3 is characterized in that: Grasping trajectory correction submodule, including: An initial correction unit, used to automatically correct the robot's grasping trajectory based on the spatial posture deviation matrix to obtain the robot's initial corrected grasping trajectory; a dynamic model generation unit, configured to generate a state vector of the robot at each moment based on the dynamic vibration offset and the rate of change of the dynamic vibration offset of the robot at each moment, and to generate a dynamic model of the robot in combination with the state transfer matrix, the control input matrix, and the process noise vector; a measurement model generating unit, configured to generate a measurement model of the robot based on a dynamic model of the robot, a measurement matrix, and a measurement noise vector; a vibration offset estimation unit, configured to estimate the dynamic vibration offset of the robot based on a dynamic model and a measurement model of the robot and a Kalman filter; The reverse compensation unit is used to perform reverse displacement compensation on the initial correction grasping trajectory of the robot based on the dynamic vibration offset of the robot to obtain a high-precision correction grasping trajectory of the robot.

5. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 4 is characterized in that: Initial calibration unit, including: A position correction subunit is used to perform position correction on each position point in the grasping trajectory of the correction robot based on the translation vector in the spatial posture deviation matrix to obtain the position correction grasping trajectory of the robot; The posture correction subunit is used to adjust the posture of each position point in the robot's position correction grasping trajectory based on the rotation matrix in the spatial posture deviation matrix to obtain the robot's initial correction grasping trajectory.

6. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 1 is characterized in that: Also includes: The fixture contact pressure detection module is used to detect the real-time contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture based on the pressure sensor array inside the spot welding fixture; The fixture support posture adjustment module is used to adjust the fixture support posture based on the real-time contact pressure distribution data between the incoming material of the automobile rear longitudinal beam and the fixture.

7. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 6 is characterized in that: Fixture support posture adjustment module, including: The sliding trend prediction submodule is used to analyze the sliding trend of the automobile rear longitudinal beam material in the fixture based on the real-time contact pressure distribution data between the automobile rear longitudinal beam material and the fixture; The support posture adjustment submodule is used to adjust the support posture of the fixture based on the sliding trend of the automobile rear longitudinal beam material in the fixture.

8. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 7 is characterized in that: Sliding trend prediction submodule, including: A normal pressure calculation unit, configured to calculate a total normal pressure in a contact area between the rear longitudinal beam material and the fixture based on real-time contact pressure distribution data between the rear longitudinal beam material and the fixture; A resultant force analysis unit is used to calculate the friction force on the fixture of the rear longitudinal beam material based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture, and to determine the resultant force on the fixture of the rear longitudinal beam material in combination with other external forces on the fixture; A maximum static friction force analysis unit is used to calculate the maximum static friction force between the rear longitudinal beam material and the fixture based on the total normal pressure in the contact area between the rear longitudinal beam material and the fixture; The sliding trend analysis unit is used to analyze the sliding trend of the automobile rear longitudinal beam material in the fixture based on the tangential component of the resultant force and the maximum static friction force exerted on the automobile rear longitudinal beam material on the fixture.

9. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 7 is characterized in that: Support posture adjustment submodule, including: An optimization objective function construction unit is used to construct an objective function with the goal of minimizing the sliding tendency when the sliding tendency of the automobile rear longitudinal beam material in the fixture exceeds a sliding tendency threshold; The posture optimization unit is used to solve the optimal support posture vector of the fixture based on the objective function and the upper and lower limits of the support posture vector of the fixture, and control the support posture of the fixture based on the optimal support posture vector of the fixture.

10. The automobile rear longitudinal beam assembly welding system based on 3D vision guidance and multi-fixture collaboration according to claim 1, characterized in that: Arc welding fixture includes: The clamping force is controlled by a proportional valve through a pneumatic clamping mechanism and a thermocouple array that are adjustable within a preset range; The thermocouple array monitors the real-time temperature of the welding area on the part after spot welding is completed in real time, and dynamically adjusts the welding current based on the real-time temperature of the welding area on the part after spot welding is completed.

Images