A personalized plate self-adaptive robot polishing method, system and electronic equipment

Abstract

The application provides a personalized plate adaptive robot polishing method, system and electronic equipment, and the personalized plate adaptive robot polishing method comprises the following steps: obtaining a plate model; matching the point cloud of the plate to be polished with the point cloud of the plate model to determine the pose of the plate to be polished, so as to generate a polishing processing track; and sending the polishing processing track to a robot. Through the application, the labor operation cost can be reduced. Through the matching of the point cloud of the plate to be polished with the point cloud of the plate model, the adaptive adjustment of the polishing processing track is realized, and then the polishing processing track is sent to the robot, so that the manual operation of the operator is not required.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of plate polishing, in particular to a personalized plate self-adaptive robot polishing method and system and electronic equipment. BACKGROUND

[0002] In modern manufacturing, the processing quality of the surface of the plate has an important influence on the appearance, service life and subsequent process treatment (such as spraying, bonding, etc.) of the product. Polishing, as a key process of plate surface treatment, directly affects the flatness and smoothness of the plate. At present, the common polishing methods in industrial production mainly include traditional mechanical polishing, robot polishing, CNC numerical control polishing and laser scanning assisted polishing, etc. Although these technologies have been widely applied, there are still many limitations in the personalized plate processing scene, and it is difficult to meet the demand of high-precision and self-adaptive polishing.

[0003] Robot polishing technology. In recent years, with the development of industrial automation technology, polishing systems based on robot control have been applied in the plate processing industry. Such systems usually combine force control sensors and pre-set trajectories to enable robots to perform automated polishing tasks. Its main advantage is to reduce manual intervention and improve processing consistency. However, this technology still faces the following problems: dependence on pre-set trajectories, poor adaptability: traditional robot polishing relies on CAD / CAM pre-set trajectories and cannot perceive and adjust the polishing path in real time, making it difficult to handle local minor defects of the plate. Limited intelligence: although force control sensors can adjust the polishing pressure, they lack precise perception of the surface morphology of the plate, making it difficult to achieve truly adaptive polishing. Difficult to efficiently process non-standard plates: for complex-shaped and irregular-sized personalized plates, existing robot polishing systems require additional modeling, long debugging period and poor adaptability.

[0004] CNC numerical control polishing. CNC numerical control equipment can perform high-precision numerical control polishing. Some enterprises obtain the three-dimensional information of the plate surface through laser scanning technology or contact type measuring heads, and then perform accurate polishing according to the modeling data. Although this method has high precision, it still has the following limitations: expensive equipment, limited application: high-precision CNC equipment is expensive and complex to maintain, making it difficult to promote to small and medium-sized enterprises. Low processing efficiency: CNC systems usually prioritize high precision, resulting in relatively slow processing speed, which makes it difficult to meet the demand of high efficiency production. Lack of adaptive adjustment capability: most CNC polishing solutions rely on pre-set parameters and cannot be adjusted in real time according to the personalized characteristics of the plate, resulting in poor adaptability. In summary, the existing plate polishing technologies mainly have the following problems:

[0005] Lack of high-precision adaptive capability: current polishing systems mostly rely on pre-set trajectories or fixed processing parameters, making it difficult to intelligently adjust according to the actual surface characteristics of the plate, resulting in limited processing precision.

[0006] Personalized processing difficulty: for non-standard plates (such as curved plates, special-shaped plates, etc.), the prior art needs additional modeling or manual intervention, and it is difficult to achieve efficient personalized production.

[0007] Insufficient degree of automation and intelligence: existing robot polishing and CNC systems mainly rely on prior programming, lack real-time perception and adaptive optimization capabilities, and cannot flexibly respond to subtle changes on the surface of the plate.

[0008] Serious environmental pollution problem: traditional polishing methods can easily generate a large amount of dust, posing a threat to the health of operators and the production environment, and are difficult to meet the requirements of environmentally friendly production. SUMMARY

[0009] In order to overcome the above technical defects, the present application provides a personalized plate adaptive robot polishing method, system and electronic equipment, which can reduce the cost of manual operation.

[0010] In order to solve the above problems, the present application is implemented according to the following technical solutions:

[0011] A personalized plate adaptive robot polishing method, comprising the steps of:

[0012] Obtaining a plate model;

[0013] Matching the point cloud of the plate to be polished with the point cloud of the plate model to determine the pose of the plate to be polished to generate a polishing processing trajectory;

[0014] Sending the polishing processing trajectory to the robot.

[0015] As a further improvement of the present application, the step of matching the point cloud of the plate to be polished with the point cloud of the plate model to determine the pose of the plate to be polished to generate a polishing processing trajectory comprises:

[0016] Generating a plate model point cloud according to the plate model;

[0017] Obtaining the point cloud of the plate to be polished by a camera;

[0018] Matching the plate model point cloud with the point cloud of the plate to be polished to obtain a translation vector and a rotation vector;

[0019] Obtaining the pose of the plate to be polished based on the translation vector and the rotation vector to generate a polishing processing trajectory.

[0020] As a further improvement of the present application, the step of matching the plate model point cloud with the point cloud of the plate to be polished comprises:

[0021] Taking the plate model point cloud as the source point cloud and the point cloud of the plate to be polished as the target point cloud;

[0022] continuously aligning the source point cloud to the position of the target point cloud;

[0023] performing registration using the point cloud alignment function, and storing the registration result;

[0024] outputting the registration result;

[0025] extracting a translation vector and a rotation vector from the registration result.

[0026] As a further improvement of the present application, in the step of continuously aligning the source point cloud to the position of the target point cloud, if the transformation amount between two consecutive iterations is less than a convergence threshold, it is determined to be converged; if the error is less than a mean square error threshold, the iteration is terminated; if it is converged within a maximum number of iterations, the iteration is forcibly stopped.

[0027] As a further improvement of the present application, the registration result includes: whether the registration is converged, a final mean square error score, and a 4x4 pose transformation matrix.

[0028] The step of extracting a translation vector and a rotation vector from the registration result includes:

[0029] extracting a 3x3 sub-matrix at the upper left corner of the 4x4 pose transformation matrix, representing the rotation part;

[0030] assigning the 3x3 sub-matrix;

[0031] generating a quaternion object using the 4x4 pose transformation matrix;

[0032] extracting a translation vector and a rotation vector from the quaternion object and the 4x4 pose transformation matrix.

[0033] As a further improvement of the present application, the present application further includes the steps of:

[0034] installing the camera in an eye-in-hand manner, and calibrating the positional relationship between the camera and the robot.

[0035] As a further improvement of the present application, the present application further includes the steps of:

[0036] If further processing is needed, returning to the step of matching the point cloud of the plate to be polished with the plate model point cloud.

[0037] The present application also provides a personalized plate adaptive robot polishing system for implementing the personalized plate adaptive robot polishing method described above, comprising:

[0038] a bottom three-dimensional engine module for obtaining a plate model;

[0039] The point cloud analysis processing module is configured to match the point cloud of the plate to be polished with the point cloud of the plate model, determine the pose of the plate to be polished, and generate a polishing processing track.

[0040] The user interaction module is configured to send the polishing processing track to the robot.

[0041] The application further provides an electronic device, which comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to implement the personalized plate adaptive robot polishing method.

[0042] Compared with the prior art, the application has the following beneficial effects: by matching the point cloud of the plate to be polished with the point cloud of the plate model, the adaptive adjustment of the polishing processing track is realized, and then the polishing processing track is sent to the robot, without the need for manual operation by an operator. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 A flowchart of the personalized plate adaptive robot polishing method described in Embodiment 1;

[0044] Figure 2 Another flowchart of the personalized plate adaptive robot polishing method described in Embodiment 1;

[0045] Figure 3 A schematic diagram of the point cloud of the plate model described in Embodiment 1;

[0046] Figure 4 A schematic diagram of the point cloud of the plate to be polished described in Embodiment 1;

[0047] Figure 5 A schematic diagram of the point cloud of the plate model described in Embodiment 1, and a schematic diagram before matching of the point cloud of the plate to be polished;

[0048] Figure 6 A schematic diagram of the point cloud of the plate model described in Embodiment 1, and a schematic diagram after matching of the point cloud of the plate to be polished. DETAILED DESCRIPTION

[0049] The preferred embodiments of the application are described below in conjunction with the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the application, and are not used to limit the application.

[0050] Embodiment 1

[0051] The embodiment provides a personalized plate adaptive robot polishing method, which comprises the following steps as shown in Figure 1 and Figure 2 ​

[0052] S1, build the polishing environment, first, use the camera with the eye outside the hand to calibrate. Communicate with the robot and PLC. Import the working model of the plate part to ensure the accuracy of the machining data. Place the plate to be polished and prepare for the subsequent process.

[0053] S2, match the point cloud of the plate to be polished with the plate model point cloud to determine the pose of the plate to be polished to generate the polishing machining trajectory:

[0054] S21, generate the plate model point cloud according to the plate model, as shown in Figure 3 .

[0055] S22, obtain the point cloud of the plate to be polished by the camera, as shown in Figure 4 .

[0056] S23, match the plate model point cloud with the plate to be polished point cloud to get the translation vector and rotation vector. The pose of the plate generated by the software and the pose of the real camera shooting are not the same, and it cannot be guaranteed to be consistent every time, as shown in Figure 5 . Therefore, the matching between the generated plate model point cloud and the real plate to be polished point cloud shot by the camera is needed, which includes the following steps:

[0057] This embodiment performs point cloud registration (alignment) based on the ICP algorithm (Iterative Closest Point) of the PCL library, and converts the transformation matrix obtained by registration into the representation form (gp_Trsf and gp_Quaternion) in OpenCASCADE, so as to be used in the subsequent geometric modeling system.

[0058] Initialize the ICP object: take the plate model point cloud as the source point cloud and the plate to be polished point cloud as the target point cloud, and let the plate model point cloud go to the position of the real point cloud.

[0059] Specify the source point cloud (Source Cloud) and the target point cloud (Target Cloud) of the registration: the source point cloud is the object to be transformed and aligned; the target point cloud is the reference object and remains unchanged; the ICP will continuously iterate the alignment of the source point cloud to the position of the target point cloud.

[0060] Configure ICP parameters: To ensure the registration effect and efficiency, several key parameters of ICP are set: Convergence threshold (Transformation Epsilon) is set to 1e-10, which means that when the transformation amount between two consecutive iterations is less than this value, the ICP algorithm is considered to have converged. The smaller the value, the more stringent the convergence, but the longer the calculation time. The maximum correspondence distance (Max Correspondence Distance) is set to 100 (in millimeters), that is, when establishing point pair matching relationship, if the distance between two points is greater than 10 centimeters, it will not be considered as a matching point pair. This parameter can effectively exclude false matches and improve accuracy. The mean square error threshold (Euclidean Fitness Epsilon) is set to 0.001, which means that when the error is lower than this value, the iteration can be terminated in advance. The maximum number of iterations (Maximum Iterations) is set to 500, which means that ICP will perform a maximum of 500 iterations, and if it does not converge within this range, it will be forced to stop.

[0061] Perform point cloud matching: Use the point cloud alignment function icp.align() to perform registration, and store the registration result to a new point cloud Cloud_Final, which is the transformed source point cloud that has been fitted to the target point cloud as much as possible.

[0062] Output registration results: Whether the registration converges (Boolean value), the final mean square error score (the smaller the better), the 4x4 pose transformation matrix, which represents the spatial transformation from the source point cloud to the target point cloud, including rotation and translation.

[0063] Extract translation vector and rotation vector from registration results.

[0064] 4x4 pose transformation matrix in OpenCASCADE format: In order to use the point cloud transformation result in the OpenCASCADE geometry system, the program converts the obtained registration transformation matrix into the following two components: gp_Mat (rotation matrix) extracts the 3x3 submatrix in the upper left corner of the 4x4 registration transformation matrix, which represents the rotation part; assign it to the matrix format gp_Mat of OpenCASCADE. gp_Quaternion (quaternion) generates a quaternion object using the rotation matrix just constructed; quaternion represents a more concise and stable description of rotation information. gp_Trsf (complete transformation) finally, the quaternion and the translation vector, rotation vector (last column first three components) extracted from the transformation matrix are passed into gp_Trsf; gp_Trsf is a standard class in OpenCASCADE for describing rigid body transformation (rotation + translation).

[0065] By the above steps, the pose difference between the source point cloud and the target point cloud is obtained.

[0066] S25, based on the translation vector and the rotation vector, obtaining the pose of the plate to be polished, and generating a polishing processing track.

[0067] S3, sending the polishing processing track to the robot.

[0068] S4, if further processing is needed, returning to the step of matching the plate to be polished with the plate model point cloud.

[0069] In summary, the embodiment has the following beneficial effects: the polishing environment is built by calibrating the camera outside the hand. By communicating with the robot and PLC, the working model of the plate can be imported to ensure the accuracy of the processing data. The camera obtains the point cloud of the plate to be polished and performs preprocessing (removing noise and smoothing). According to the plate model, the plate model point cloud is generated, and the point cloud matching is performed to determine the working pose of the plate, providing accurate information for path planning. Then, the polishing processing track is adaptively generated and sent to the robot, which is converted into a program executable by the robot. The robot automatically performs the polishing operation according to the processing program. According to the processing result, it is judged whether further processing is needed: if needed, the processing program is repeatedly run. If not, the running is ended.

[0070] The embodiment ensures that the robot can adaptively generate a polishing path according to different shapes and sizes of the plate, improving the accuracy and efficiency of polishing.

[0071] Embodiment 2

[0072] The embodiment provides a personalized plate adaptive robot polishing system for realizing the personalized plate adaptive robot polishing method in embodiment 1, which includes a bottom three-dimensional engine module, a point cloud analysis and processing module, and a user interaction module. The bottom three-dimensional engine module is used to obtain a plate model. The point cloud analysis and processing module is used to match the point cloud of the plate to be polished with the plate model point cloud to determine the pose of the plate to be polished, so as to generate a polishing processing track. The user interaction module is used to send the polishing processing track to the robot.

[0073] Further, the bottom three-dimensional engine module can provide a basic interactive interface, import and export of the plate model, and display and operation of the plate model. The interactive functions include a toolbar, which provides quick access to common tools such as selection, movement, rotation, scaling, etc. The menu system includes file operation, view control, and setting menu options. The file format supports common three-dimensional model file formats such as OBJ, STL, STEP, IGS, etc.

[0074] The user interaction module adds special interaction buttons for the user on the basis of the basic interaction interface, which are used to operate various functions. Intuitive function areas are designed, and commonly used operation buttons (such as import, export, analysis, view switching, etc.) are placed in appropriate positions to facilitate quick operation by the user.

[0075] For other implementation manners of the embodiment, refer to Embodiment 1, which will not be repeated here.

[0076] Embodiment 3

[0077] The embodiment provides an electronic device, which includes a processor and a memory. The memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the personalized plate adaptive robot polishing method in Embodiment 1.

[0078] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0079] The memory can be used to store the computer programs or modules, and the processor realizes various functions of the trajectory planning optimization method by running or executing the computer programs or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system, at least one application required by a function, and the like. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0080] The above describes in detail the personalized plate self-adaptive robot polishing method, system and electronic device provided by the present application. The principles and implementation manners of the present application are described by using specific examples. The above example is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, the specific implementation manners and application ranges can be changed according to the idea of the present application. In summary, the content of the specification should not be understood as a limitation of the present application. The above examples are only preferred examples for fully illustrating the present application, and the protection scope of the present application is not limited thereto. Any equivalent replacement or transformation made by those skilled in the art on the basis of the present application is within the protection scope of the present application.

Claims

1. A personalized adaptive robot sanding method for sheet metal, characterized in that, Including the following steps: Obtain the board model; The point cloud of the material to be polished is matched with the point cloud of the board model to determine the pose of the material to be polished in order to generate the polishing trajectory. This includes: generating the point cloud of the board model based on the material model; acquiring the point cloud of the material to be polished through a camera; matching the point cloud of the board model with the point cloud of the material to be polished to obtain translation vectors and rotation vectors; and obtaining the pose of the material to be polished based on the translation vectors and rotation vectors to generate the polishing trajectory. Send the grinding and processing trajectory to the robot; The step of matching the point cloud of the plate model with the point cloud of the plate to be polished includes: The point cloud of the board model is used as the source point cloud, and the point cloud of the board to be polished is used as the target point cloud. Iteratively align the source point cloud to the position of the target point cloud; Perform registration using the point cloud alignment function and store the registration results; Output the registration results; Extract translation and rotation vectors from the registration results; In the step of iteratively aligning the source point cloud to the target point cloud, if the transformation amount between two consecutive iterations is less than the convergence threshold, then convergence is determined; if the error is less than the mean square error threshold, the iteration is terminated; if convergence is not achieved within the maximum number of iterations, then the iteration is forcibly stopped. The registration results include: whether the registration is converged, the final mean square error score, and the 4×4 pose transformation matrix; The steps for extracting translation and rotation vectors from the registration results include: Extract the top-left 3×3 submatrix from the 4×4 pose transformation matrix to represent the rotation part; Assign values ​​to the 3×3 submatrix; Generate quaternion objects using a 4×4 pose transformation matrix; Extract translation and rotation vectors from quaternion objects and 4×4 pose transformation matrices.

2. The personalized adaptive robot sanding method for sheet metal according to claim 1, characterized in that, It also includes the following steps: The camera was mounted using an eye-out method to calibrate the positional relationship between the camera and the robot.

3. The personalized adaptive robot sanding method for sheet metal according to claim 1, characterized in that, It also includes the following steps: If further processing is required, return to the step of matching the point cloud of the material to be polished with the point cloud of the board model.

4. A personalized adaptive robotic sanding system for sheet materials, characterized in that, The method for implementing the personalized adaptive robotic sanding method for sheet metal as described in any one of claims 1 to 3 includes: The underlying 3D engine module is used to obtain the board model; The point cloud analysis and processing module is used to match the point cloud of the material to be polished with the point cloud of the board model to determine the pose of the material to be polished in order to generate the polishing process trajectory. The user interaction module is used to send the grinding and processing trajectory to the robot.

5. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing at least one instruction, which is loaded and executed by the processor to implement the personalized adaptive robotic sanding method for sheet metal as described in any one of claims 1 to 3.

Images