Positioning method for welding starting point of clamping-free workpiece

Abstract

The invention discloses a positioning method for a welding starting point of a clamping-free workpiece, which comprises the following steps: importing a three-dimensional model of a to-be-welded workpiece into a visual simulation system, performing off-line trajectory planning on a welding process, and generating an operation trajectory for welding operation, the operation trajectory comprising a reverse scanning trajectory and a welding trajectory; in the actual operation process, the visual simulation module conducts global positioning on a workpiece to be welded, obtains the pose transformation relation between the pose of the workpiece and the initial pose, and corrects the operation trajectory planned in the trajectory planning module. And accurate tracking and recording of the welding seam points are ensured through the consistency of simulation welding seam numbers, and the tail end point in the candidate welding seam points is determined as the welding starting point. Visual simulation information is introduced, workpiece clamping is not depended on, adaptability is high, and the application working range of the structured light visual sensor is expanded.

Description

Technical Field

[0001] This invention relates to the field of industrial welding, and in particular to a method for locating the starting point of welding on a workpiece without clamping. Background Technology

[0002] In clamp-free welding, the uncertain placement of the workpiece on the worktable makes it difficult to directly apply the welding trajectory obtained through offline programming to actual operations, especially when there is a significant deviation in the welding start point, which can easily lead to welding failure. Existing technologies mostly rely on manual teaching or simple visual inspection to obtain the welding start point, which struggles to ensure stability and consistency under complex weld morphologies and interference conditions. Therefore, it is necessary to propose a positioning method suitable for clamp-free workpiece conditions that can stably determine the welding start point. Summary of the Invention

[0003] In order to overcome the above-mentioned shortcomings and deficiencies of the prior art, the purpose of this invention is to provide a method for positioning the welding start point of a workpiece without clamping.

[0004] This invention combines a visual simulation system with line structured light weld seam tracking information under clamping conditions. By using a weld seam number consistency criterion during reverse scanning, the welding starting point is stably and reliably determined. Based on this, the welding trajectory is corrected, thereby improving the automation level, positioning accuracy, and engineering applicability of welding operations.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A method for locating the starting point of welding on a workpiece without clamping includes a line structured light sensor and a vision simulation system, wherein the vision simulation system includes a trajectory planning module, a vision simulation module, and a weld seam tracking module.

[0007] The specific positioning method is as follows:

[0008] Import the 3D model of the workpiece to be welded into the visual simulation system. During the initialization phase, ensure that the initial pose settings of the trajectory planning module and the workpiece to be welded in the visual simulation module are consistent.

[0009] For the workpiece to be welded in its initial pose, offline trajectory planning is performed on the welding process to generate a working trajectory for welding operations. The working trajectory includes a reverse scanning trajectory and a welding trajectory.

[0010] In actual operation, the visual simulation module performs global positioning of the workpiece to be welded and obtains the pose transformation relationship between the workpiece pose and the initial pose.

[0011] Based on the pose transformation relationship, the planned operation trajectory in the trajectory planning module is corrected;

[0012] The reverse scanning is performed according to the corrected reverse scanning trajectory. During the reverse scanning process, the visual simulation module generates simulated weld points in real time in the simulation environment and assigns numbers to the acquired simulated weld points as the initial weld numbers. Meanwhile, in actual operation, the weld tracking module detects the actual weld points.

[0013] When the actual weld point obtained by the current scan is consistent with the initial weld number recorded at the start of the reverse scan, the corresponding weld point is recorded as a candidate weld point.

[0014] After the reverse scan is completed, the end point of the candidate weld point is determined as the welding start point, and the welding trajectory is offset and corrected based on the welding start point. The corrected welding trajectory is then sent to the robot to perform the welding operation.

[0015] Furthermore, the reverse scanning trajectory moves in the opposite direction to the welding trajectory and is used to cover the welding start area.

[0016] Furthermore, the pose transformation relationship is represented in the form of a homogeneous transformation matrix.

[0017] Furthermore, based on the aforementioned pose transformation relationship, the planned work trajectory in the trajectory planning module is corrected, specifically as follows:

[0018] The roll angle, pitch angle, and yaw angle of the workpiece to be welded are calculated by the pose transformation relationship, and then the attitude in the working trajectory is corrected.

[0019] Furthermore, during the reverse scanning process, the visual simulation module generates simulated laser lines and simulated weld points in real time based on the robot's current pose, and obtains the weld number corresponding to the simulated weld points.

[0020] Furthermore, the offset correction of the welding trajectory based on the welding starting point specifically adopts rigid offset correction.

[0021] Furthermore, the weld seam tracking module detects actual weld seam points by processing images acquired by the line structured light sensor.

[0022] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0023] (1) This invention proposes a welding start point detection process based on reverse scanning. The method does not depend on workpiece clamping, is suitable for flexible manufacturing scenarios, and has strong engineering applicability.

[0024] (2) This invention introduces a weld numbering mechanism through a visual simulation system, which realizes the consistency constraint of weld identity, stably obtains the welding start point, and can be seamlessly integrated with existing weld tracking algorithms and trajectory planning systems;

[0025] (3) The method proposed in this invention introduces visual simulation information, does not rely on workpiece clamping, and has strong adaptability. The method of this invention expands the application range of structured light vision sensors. Attached Figure Description

[0026] Figure 1 This is a flowchart of the welding initial point positioning method of the present invention;

[0027] Figure 2 yes Figure 1 Schematic diagram of the starting point identification sub-process;

[0028] Figure 3 This is the visual inspection diagram of the present invention;

[0029] Figure 4 This is a schematic diagram of the reverse scanning and weld number consistency queue mechanism of the present invention. Detailed Implementation

[0030] The present invention will be further described in detail below with reference to the embodiments, but the implementation of the present invention is not limited thereto.

[0031] Example

[0032] like Figures 1-4 As shown, a method for locating the starting point of welding on a workpiece without clamping includes a line structured light sensor and a vision simulation system. The vision simulation system includes a trajectory planning module, a vision simulation module, and a weld seam tracking module. The specific method is as follows:

[0033] The visual simulation system utilizes open-source libraries such as OpenCASCADE, Robop (RoboticsObject-Oriented Package), and Orocos-KDL (Kinematics and Dynamics Library) to assist in kinematic modeling and scene creation. The scene is built by inputting the geometric models (CAD data) of objects in the work environment, robot model parameters (DH parameters, joint limits), and sensor calibration parameters. The robot's motion trajectory is generated offline using the RRT algorithm (Rapidly-exploring RandomTree). The simulation space scene is corrected in real-time by reading the robot's pose and kinematic model, and the acquisition process of the line structured light sensor is simulated to obtain the acquisition results under the simulation environment.

[0034] Import the 3D model of the workpiece to be welded into the visual simulation system, and establish corresponding workpiece models in the trajectory planning module and the visual simulation module respectively. During the system initialization phase, set the initial pose of the workpiece model to be welded in the trajectory planning module and the visual simulation module to be consistent, that is, both use the same spatial position and attitude parameters, thereby ensuring that there is a unified coordinate reference between the offline trajectory planning and the simulation environment.

[0035] In the trajectory planning module, based on the initial pose of the workpiece to be welded, offline trajectory planning is performed on the welding process to generate a working trajectory for the welding operation. The working trajectory includes at least two parts: a reverse scanning trajectory and a welding trajectory. The reverse scanning trajectory moves in the opposite direction to the welding trajectory, and is configured to cover the welding start area to ensure that the welding start point can be scanned during subsequent reverse scanning.

[0036] In actual operation, the workpiece is placed on the worktable without clamping. Since the actual placement posture of the workpiece deviates from the initial pose set in the simulation system, the visual simulation module performs global localization of the actual workpiece based on the collected point cloud information to obtain its spatial pose. Let the actual workpiece pose be:

[0037] ;

[0038] The initial pose set in the simulation system is:

[0039] ;

[0040] The pose transformation relationship between the two is described by the homogeneous transformation matrix 𝑇.

[0041] Based on the homogeneous transformation matrix , the planned work trajectory in the trajectory planning module is corrected. For any point in the work trajectory... Its corrected spatial location is:

[0042] ;

[0043] in, The corrected position point is shown below. Simultaneously, by performing an inverse solution on the homogeneous transformation matrix, the corresponding attitude correction amount is obtained, and the attitude parameters in the original trajectory are synchronously corrected, thereby obtaining a corrected work trajectory consistent with the actual workpiece pose.

[0044] The robot is controlled to perform a reverse scan following the corrected reverse scan trajectory. During the reverse scan, the vision simulation module reads the robot's current pose in real time and generates corresponding simulated laser lines and simulated weld points in the simulation environment based on this pose. Simultaneously, it obtains the unique weld number for each scanned weld point. At the start of the reverse scan, the corresponding weld number is recorded as the initial weld number.

[0045] During the reverse scanning process, the weld seam tracking module processes the images acquired by the line structured light sensor using a weld seam tracking method to detect actual weld seams in real time. These weld seams are obtained through a 3D CAD model and related algorithms. When the weld seam number obtained in the current scan matches the initial weld seam number, the corresponding weld seam is recorded and sequentially stored in the weld seam queue; when the weld seam number obtained in the current scan does not match the initial weld seam number, the corresponding weld seam is not recorded. This method achieves weld seam selection based on the consistency of weld seam numbers.

[0046] Specifically, the weld seam tracking method uses the visual information of simulated weld seam images as training samples to train and generate a kernel correlation filter. After preprocessing the actual weld seam image by combining simulated laser stripe information and the Gabor algorithm, the kernel correlation filter is used to detect weld seam point targets in real time.

[0047] To further explain, when the numbers are the same, it means that they belong to the same weld segment, which is the weld segment that needs to be welded.

[0048] After the reverse scan is completed, the last weld point in the weld point queue is taken as the welding start point. Based on the welding start point, the welding trajectory is corrected a second time, that is, the position points in the welding trajectory are corrected. After performing rigid offset correction, the corrected welding trajectory points are obtained:

[0049] ;

[0050] The corrected welding trajectory is sent to the robot, which then performs subsequent welding and welding correction processes according to the trajectory.

[0051] This method imports a 3D model into a simulation system, plans the welding trajectory in the trajectory planning module, and corrects the trajectory using the positioning results from the visual simulation module. This ensures that even without a clamped workpiece, the accurate welding starting point can be obtained through reverse scanning during actual welding. The consistency of the simulated weld seam numbering ensures accurate tracking and recording of weld points. Finally, by correcting the offset of the welding trajectory, high precision and reliability of the welding process are ensured. This invention solves the problem of starting point positioning in welding without a clamped workpiece, and has strong anti-interference capabilities and wide applicability.

[0052] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the embodiments described above. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for locating the welding start point of a workpiece without clamping, characterized in that, It includes a line structured light sensor and a vision simulation system, wherein the vision simulation system includes a trajectory planning module, a vision simulation module, and a weld seam tracking module; The specific positioning method is as follows: Import the 3D model of the workpiece to be welded into the visual simulation system, and set the initial pose of the workpiece to be welded in the trajectory planning module to be consistent with that in the visual simulation module. For the workpiece to be welded in its initial pose, offline trajectory planning is performed on the welding process to generate a working trajectory for welding operations. The working trajectory includes a reverse scanning trajectory and a welding trajectory. In actual operation, the visual simulation module performs global positioning of the workpiece to be welded and obtains the pose transformation relationship between the workpiece pose and the initial pose. Based on the pose transformation relationship, the planned operation trajectory in the trajectory planning module is corrected; The reverse scanning is performed according to the corrected reverse scanning trajectory. During the reverse scanning process, the visual simulation module generates simulated weld points in real time in the simulation environment and assigns numbers to the acquired simulated weld points as the initial weld numbers. Meanwhile, in actual operation, the weld tracking module detects the actual weld points. When the actual weld point obtained by the current scan is consistent with the initial weld number recorded at the start of the reverse scan, the corresponding weld point is recorded as a candidate weld point. After the reverse scan is completed, the end point of the candidate weld point is determined as the welding start point, and the welding trajectory is offset and corrected based on the welding start point. The corrected welding trajectory is then sent to the robot to perform the welding operation.

2. The positioning method according to claim 1, characterized in that, The reverse scanning trajectory moves in the opposite direction to the welding trajectory and is used to cover the welding start area.

3. The positioning method according to claim 1, characterized in that, The pose transformation relationship is represented in the form of a homogeneous transformation matrix.

4. The positioning method according to claim 1, characterized in that, Based on the aforementioned pose transformation relationship, the planned work trajectory in the trajectory planning module is corrected, specifically as follows: The roll angle, pitch angle, and yaw angle of the workpiece to be welded are calculated by the pose transformation relationship, and then the attitude of the workpiece to be welded in the working trajectory is corrected.

5. The positioning method according to claim 1, characterized in that, During the reverse scanning process, the visual simulation module generates simulated laser lines and simulated weld points in real time based on the robot's current pose, and obtains the weld number corresponding to the simulated weld points.

6. The positioning method according to claim 1, characterized in that, The offset correction of the welding trajectory based on the welding starting point specifically employs rigid offset correction.

7. The positioning method according to claim 1, characterized in that, The weld seam tracking module detects actual weld seam points by processing images acquired by the line structured light sensor.

Images