Grinding and polishing method based on three-dimensional vision guidance and variable stiffness force impedance control

Abstract

This application provides a grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control, relating to the field of automatic control. The control equipment acquires the prior estimated stiffness of the workpiece at its current trajectory position as the initial control basis, ensuring uninterrupted grinding. It then dynamically corrects this prior estimated stiffness using real-time observed stiffness measured by a stiffness sensor, obtaining a real-time comprehensive stiffness that more closely approximates the actual contact state. Finally, based on this real-time comprehensive stiffness, the grinding force originally determined by the prior estimated stiffness is adjusted in real time. Thus, even under complex working conditions such as abrupt changes in surface curvature or material variations, the grinding force can still quickly and smoothly adapt to changes in environmental stiffness, significantly improving the real-time performance and stability of constant force grinding and polishing.

Description

Technical Field

[0001] This application relates to the field of automatic control, and more specifically, to a grinding and polishing method based on three-dimensional vision guidance and variable stiffness force resistance control. Background Technology

[0002] In high-volume, high-precision manufacturing scenarios such as automotive body panels, aerospace large skins, rail transit vehicle bodies, and high-end home appliance casings, surface grinding is a crucial pre-painting treatment or finishing process. Manual grinding is limited by the operator's physical strength, experience, and attention span, making it difficult to ensure consistency in force application and movement trajectory. This easily leads to problems such as localized over-grinding causing paint damage or missed areas resulting in substandard surface roughness. In contrast, machine grinding offers advantages such as programmability, repeatability, and monitorability, providing significant advantages in operational efficiency, path accuracy, and force control stability. Therefore, it has been widely used in industrial production.

[0003] It should be understood that the actual effect of machine polishing depends not only on the accuracy of trajectory planning, but more importantly on the ability to maintain stable polishing pressure under complex working conditions. Stable polishing pressure means that the normal contact force applied by the robot's end effector to the paint film surface of the workpiece must remain strictly constant throughout the polishing process. This ensures a uniform distribution of material removal rate and avoids paint tearing, exposure of the substrate, or even scratches on the substrate caused by instantaneous overload.

[0004] However, the control methods currently used by mainstream machine grinding systems to achieve the above goals still have a prominent shortcoming: dynamic response lag. Summary of the Invention

[0005] To overcome at least one deficiency in the prior art, this application provides a grinding and polishing method based on three-dimensional vision guidance and variable stiffness force resistance control, which can quickly and smoothly adapt the grinding force to environmental stiffness changes under complex working conditions such as abrupt changes in curved surfaces or material changes, significantly improving the real-time performance and stability of constant force grinding and polishing.

[0006] In a first aspect, this application provides a grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control, the method comprising: When the grinding tool grinds the workpiece along the reference trajectory, the prior estimated stiffness of the workpiece at the current trajectory position is obtained; The prior estimated stiffness is corrected using the observed stiffness measured by a stiffness sensor to obtain the real-time comprehensive stiffness; The grinding force is adjusted based on the real-time integrated stiffness, which is determined by the prior estimated stiffness.

[0007] Compared with the prior art, this application has the following beneficial effects: The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control provided in this application obtains the prior estimated stiffness of the workpiece at the current trajectory position as the initial control basis to ensure that the grinding process is not interrupted. The prior estimated stiffness is then dynamically corrected using the observed stiffness measured in real time by a stiffness sensor to obtain a real-time comprehensive stiffness that more closely approximates the actual contact state. Finally, based on this real-time comprehensive stiffness, the grinding force originally determined by the prior estimated stiffness is adjusted in real time. Thus, even under complex working conditions such as abrupt changes in curvature or material variations, the grinding force can still quickly and smoothly adapt to changes in environmental stiffness, significantly improving the real-time performance and stability of constant force grinding and polishing. Attached Figure Description

[0008] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic flowchart of the grinding and polishing method based on three-dimensional vision guidance and variable stiffness force resistance control provided in the embodiments of this application; Figure 2 A schematic diagram illustrating the coordinate transformation principle provided in the embodiments of this application; Figure 3 This is one of the schematic diagrams illustrating the effect of the polishing method provided in the embodiments of this application; Figure 4 This is the second schematic diagram illustrating the effect of the polishing method provided in the embodiments of this application; Figure 5 This is a schematic diagram of the control model provided in the embodiments of this application; Figure 6 A schematic diagram of the mechanical structure of the grinding tool provided in the embodiments of this application; Figure 7 A schematic diagram of the structure of the grinding and polishing device based on three-dimensional vision guidance and variable stiffness force resistance control provided in the embodiments of this application; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of this application (hereinafter referred to as "the embodiments") clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0011] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0012] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0013] In the description of this application, it should be noted that the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0014] Based on the above statement, as introduced in the background technology, the control method currently used by the machine to achieve the above polishing requirements still has obvious shortcomings, namely the problem of dynamic response lag.

[0015] In related technologies, fuzzy controllers rely on expert experience to design rule bases, while variable impedance control requires dynamic adjustment of stiffness / damping parameters to adapt to the environment. When faced with unknown curved surfaces (e.g., weld burrs) or abrupt material changes (e.g., at the interface between metal and composite materials), fuzzy logic struggles to accurately map real-time force-position relationships, resulting in lag or inaccuracy in impedance parameter adjustment. Simultaneously, stiffness estimation and impedance parameter updates themselves require computation time. In high-speed grinding or areas with abrupt curvature changes, this delay further exacerbates force control deviations, ultimately causing grinding pressure fluctuations to exceed process tolerances, affecting paint film integrity and processing stability.

[0016] It should be noted that the defects in the solutions in the prior art are the result of practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the embodiments of this application in the following text should be regarded as contributions to this application in the process of invention and creation, and should not be understood as technical content known to those skilled in the art.

[0017] Based on the discovery of the above-mentioned technical problems, this embodiment provides a grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control. For example... Figure 1 As shown, the method includes: S1, when the grinding tool grinds the workpiece along the reference trajectory, obtain the prior estimated stiffness of the workpiece at the current trajectory position.

[0018] S2 uses the observed stiffness measured by the stiffness sensor to correct the prior estimated stiffness, thus obtaining the real-time comprehensive stiffness.

[0019] S3 adjusts the grinding force determined by the prior estimated stiffness based on the real-time comprehensive stiffness.

[0020] This embodiment can be understood as follows: First, the prior estimated stiffness of the workpiece at the current trajectory position is obtained as the initial control basis to ensure that the grinding process is not interrupted. Then, the prior estimated stiffness is dynamically corrected using the observed stiffness measured in real time by a stiffness sensor to obtain a real-time comprehensive stiffness that is closer to the actual contact state. Finally, based on this real-time comprehensive stiffness, the grinding force originally determined by the prior estimated stiffness is adjusted in real time. In this way, under complex working conditions such as abrupt changes in curvature or material changes, the grinding force can still quickly and smoothly adapt to changes in environmental stiffness, significantly improving the real-time performance and stability of constant force grinding and polishing.

[0021] In this embodiment, the control device for implementing the method can be a control device integrated into the grinding robot, or it can be a tablet computer, laptop computer, desktop computer, etc., that communicates with the robot as a host computer.

[0022] To make the solution provided in this embodiment clearer, the following is combined with... Figure 1 Each step of the method is described in detail. However, it should be understood that the operations in the flowchart may not be implemented in sequence, and steps without logical contextual relationships may be reversed in order or implemented simultaneously. Furthermore, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowchart, or remove one or more operations from the flowchart. Figure 1 As shown, the method includes: S1, when the grinding tool grinds the workpiece along the reference trajectory, obtain the prior estimated stiffness of the workpiece at the current trajectory position.

[0023] As an optional implementation, the grinding tool is located at the end of the robot, and the robot, under the control of the control device, causes the grinding tool located at the end of the robot to grind the workpiece along a reference trajectory.

[0024] It should be noted that during related procedures, control equipment typically uses fixed stiffness values ​​or relies on force sensor feedback before adjustments are made, lacking the ability to proactively perceive the geometric characteristics of the workpiece surface. Curvature characteristic maps reflect the degree of surface unevenness; for example, curvature in planar areas approaches zero, while curvature increases significantly at edges or corners. Since the actual contact stiffness varies greatly across different curvature regions, if the control equipment cannot know this local curvature in advance, it is difficult to pre-set a reasonable stiffness reference. This can easily lead to excessive pressure being applied in the initial grinding phase of high-curvature areas, damaging the paint film, while insufficient pressure may affect grinding uniformity in the initial grinding phase of low-curvature areas.

[0025] Therefore, this embodiment provides the following optional implementation methods for step S1: S1-1, Obtain the curvature feature map of the workpiece surface.

[0026] It should be understood here that in related technologies, control equipment often directly uses unprocessed raw point clouds or performs only simple filtering. However, raw 3D point clouds are often affected by ambient lighting, material reflections, and ranging errors, containing a large number of noise points and outliers. If the control equipment does not effectively preprocess the raw point cloud, the curvature calculated subsequently based on the distorted point set will be severely distorted. For example, a surface that should have a smooth transition may be misjudged as a sharp edge, or a real pit may be masked by noise and become unrecognizable. Thus, the resulting curvature feature map cannot truly reflect the geometric characteristics of the workpiece surface.

[0027] Therefore, the control device can acquire the original three-dimensional point cloud of the workpiece; preprocess the original three-dimensional point cloud to obtain the preprocessed optimized point cloud; for each target point in the preprocessed optimized point cloud, acquire the set of points within a preset range from the target point; and obtain the curvature corresponding to the target point based on the geometric information in the point set.

[0028] This embodiment can be understood as follows: it can use a three-dimensional vision sensor to collect geometric information of the workpiece surface, and after systematic noise reduction and structural alignment processing, construct a curvature feature map that can characterize the local surface bending characteristics.

[0029] In practical applications, the control equipment first controls the 3D structured light camera to take pictures of the workpiece and obtain the original three-dimensional point cloud of the workpiece. The original three-dimensional point cloud is subject to noise interference, mainly due to changes in ambient lighting, the reflective properties of the workpiece material, and the technical limitations of the ranging principle itself. Therefore, the original point cloud needs to be preprocessed to improve data reliability.

[0030] Preprocessing includes three types of operations: size adjustment, coordinate transformation, and data simplification, collectively referred to as point cloud preprocessing in this embodiment. The control device can employ any of the following methods to achieve data simplification: voxel grid downsampling, uniform sampling, farthest point sampling, or random sampling. This embodiment selects voxel grid downsampling due to its advantages of both computational efficiency and spatial uniformity. The concept of a voxel is analogous to a pixel in an image. The control device uses the AABB (Axis-Aligned Bounding Box) algorithm to divide the three-dimensional space into a uniformly sized cubic grid, with each cube defined as a voxel unit. For all point cloud data falling within the same voxel, the control device calculates its spatial centroid as the representative point of that voxel. The centroid coordinates are determined by the following formula:

[0031] In the formula, This indicates the number of point clouds contained within the voxel. The centroid of the voxel is located to replace the original point group, thereby achieving downsampling and effectively filtering out outliers and noise, while maintaining the overall geometric structure basically unchanged.

[0032] Furthermore, the control equipment can further improve the point cloud quality by combining radius filtering optimization and coordinate system transformation. It should be understood that the necessity of coordinate system transformation lies in resolving the inconsistency between the visual coordinate system and the robot's base coordinate system. Since the trajectory planned in the camera coordinate system cannot be directly used for robot motion control, the control equipment needs to perform hand-eye calibration to establish a mathematical mapping relationship between the two.

[0033] like Figure 2 As shown, depending on the installation method, eye-hand calibration is divided into Eye-in-Hand (camera fixed to the robot's end effector coordinate system). There are two types of calibration methods: Eye-in-Hand and Eye-to-Hand (camera fixed to the robot's base coordinate system). In Eye-in-Hand mode, the control device controls the robot to change multiple sets of end-effector poses, sequentially capturing images of the checkerboard calibration board. In Eye-to-Hand mode, the control device only adjusts the robot's pose, allowing the camera fixed to the base to acquire images of the calibration board from multiple angles. Regardless of the mode, the calibration process solves the coordinate system from the camera coordinate system. To the robot's base coordinate system The rigid body transformation matrix is ​​used to accurately map the point cloud data obtained from visual recognition to the work space that the robot can perform.

[0034] In this process, the control device performs a neighborhood search on each target point based on the preprocessed optimized point cloud to obtain a set of neighboring points within a preset range from the target point; then, based on the spatial distribution geometric information of each point in the set, it calculates the curvature value corresponding to the target point; and so on, it traverses point by point to finally generate a curvature feature map covering the entire surface of the workpiece.

[0035] For example, OpenCV's curvature estimation algorithm can be used:

[0036] In the formula, For the normal vector field, These are the surface parameterized coordinates.

[0037] In addition, the control equipment analyzes the surface features of the workpiece based on the pre-processed optimized point cloud, including geometric attributes such as contour boundaries, planar regions, and transition surfaces; based on the analysis results, a spatial curve covering the area to be polished and satisfying the continuity and accessibility constraints is generated, namely the reference trajectory.

[0038] Based on the above description of the curvature feature map and reference trajectory in the embodiments, step S1 further includes: S1-2, Obtain the local curvature corresponding to the current trajectory position from the curvature feature map.

[0039] In this embodiment, the constructed curvature feature map can be regarded as a digital map recording the degree of curvature at every point on the workpiece surface. The control device, based on the current reference trajectory position of the grinding tool, directly finds the curvature value matching the spatial coordinates of that position in the map. In practical applications, the control device can use conventional spatial mapping methods such as bilinear interpolation or nearest neighbor search to locate the corresponding pixel or voxel unit in the curvature feature map at the current trajectory position, and read the stored curvature value to obtain the local curvature.

[0040] S1-3, based on the local curvature, obtain the prior estimate of stiffness at the current trajectory position.

[0041] As an optional implementation, the control device can input the local curvature into a preset stiffness mapping function to obtain the prior estimated stiffness corresponding to the current trajectory position, wherein the stiffness mapping function includes at least one workpiece grinding process parameter.

[0042] This embodiment can be understood as follows: by converting the curvature value obtained from visual detection into a priori estimated stiffness that can be used to adjust the polishing force through a mathematical function.

[0043] In practical applications, the control equipment follows the visual surface curvature detection process, acquires point clouds of the workpiece surface through a binocular camera, calculates curvature using OpenCV's curvature estimation algorithm, and determines, based on the curvature-stiffness mapping relationship, that high curvature areas (such as edges and pits) correspond to high stiffness, and low curvature areas (planes) correspond to low stiffness.

[0044] Specifically, the control device will control the local curvature Substitute into the stiffness mapping function:

[0045] In the formula, , The material coefficients are used, and the output is the prior estimate of stiffness. .

[0046] Based on the above embodiments' explanation of prior estimation stiffness, please refer to... Figure 1 Next, we will continue with... Figure 1 Step S2 will be explained below: S2 uses the observed stiffness measured by the stiffness sensor to correct the prior estimated stiffness, thus obtaining the real-time comprehensive stiffness.

[0047] In practical applications, the stiffness sensor used for observing stiffness here can be a six-dimensional force sensor. This six-dimensional force sensor is installed between the robot's end effector and the grinding tool to measure the three-dimensional contact force in real time. and three-dimensional torque Based on the measured change in contact force The tiny displacement of the tool at the corresponding time point The observed stiffness is calculated using the following expression:

[0048] In the formula, This refers to the minute displacement of the tool in the contact direction, which can be obtained by forward kinematics calculation from the position data fed back by the robot joint encoder, or directly measured by a high-resolution displacement sensor.

[0049] In addition, as an optional implementation, the controller can determine the corresponding fusion weights based on the uncertainty measures of the prior estimated stiffness and the observed stiffness, and obtain the real-time integrated stiffness by weighting the prior estimated stiffness and the observed stiffness.

[0050] This embodiment can be understood as employing a weighted Bayesian fusion method to combine stiffness information from different sources by assigning weights according to their reliability.

[0051] In practical applications, the control equipment first determines the corresponding fusion weights based on the uncertainty measures of the prior estimated stiffness and the observed stiffness. , The weights are dynamically adjusted based on sensor confidence levels; they need to be reduced when force jitter is significant. Secondly, the stiffness is estimated based on prior knowledge. With observed stiffness and the corresponding weights , The weighted average yields the real-time comprehensive stiffness, and this relationship can be expressed as:

[0052] In the formula, Indicates real-time composite stiffness. Indicates the observed stiffness. This indicates the prior estimate of stiffness. , This indicates the corresponding weight.

[0053] It should be noted that this real-time composite stiffness can also be used to construct a real-time environmental stiffness map; this real-time environmental stiffness map is implemented using a hash voxel storage method, that is, dividing the space into a sparse voxel grid, with each voxel storing the average stiffness value. and update count New data points Upon entry, the stiffness value of the voxel corresponding to that point changes accordingly. The specific update method is as follows: and order

[0054] Based on the description of real-time integrated stiffness in the above embodiments, the following will continue to discuss... Figure 1 Step S3 will be explained below: S3 adjusts the grinding force determined by the prior estimated stiffness based on the real-time comprehensive stiffness.

[0055] To make the following explanation easier to understand, before explaining how to adjust the polishing force, it is necessary to introduce the theoretical analysis of the polishing process for flat and curved surfaces.

[0056] It should be understood that the polishing process does not change the original shape of the workpiece. By ensuring the uniformity of material removal during each polishing process, the polishing quality of the workpiece surface can be controlled. Equation, material removal rate during grinding and polishing h t ( x , y The relationship between the parameters and the relevant process parameters can be expressed as follows:

[0057] In the formula, The polishing point on the workpiece surface Material removal rate It refers to the amount of material removed from the machined surface. yes proportionality coefficient It is the relative speed between the grinding head and the processing point. It refers to the contact pressure during the polishing process.

[0058] The material removal rate is related to the polishing pressure and the relative polishing speed. Assume the relative polishing speed is constant. The material removal rate is determined solely by the polishing pressure. If the polishing pressure can be controlled to remain constant... This ensures a constant material removal rate on the workpiece surface, thereby guaranteeing uniform removal and significantly improving the quality of grinding and polishing.

[0059] like Figure 3 as well as Figure 4 As shown, in actual grinding processes, the workpiece typically consists of large-area flat surfaces and small-curvature curved surfaces, which constitute the main part of robotic grinding. For the flat surface, since the normal to each machining point on the workpiece surface is the same, the contact area between the grinding head and the workpiece remains constant during grinding. When the grinding head rotation speed and grinding feed speed are constant, as long as the output force of the grinding tool remains constant, the pressure between the grinding head and the workpiece can be kept constant during grinding, and the material removal rate can be maintained at a certain level.

[0060] Based on the above theoretical analysis of the grinding and polishing process of planar and curved surfaces, as an optional implementation method, this embodiment provides the following optional implementation method for step S3: S3-1, obtain the force error of the current polishing force.

[0061] S3-2, If the real-time comprehensive stiffness is higher than the preset stiffness threshold, then reduce the current stiffness parameter of the variable impedance controller and increase the current damping parameter to obtain the adjusted variable impedance controller.

[0062] S3-3, If the real-time comprehensive stiffness is lower than the preset stiffness threshold, then increase the current stiffness parameter of the variable impedance controller and decrease the current damping parameter to obtain the adjusted variable impedance controller.

[0063] S3-4: The force error is processed by the adjusted variable impedance controller to obtain the compensation amount for the current trajectory position.

[0064] S3-5, adjust the position of the polishing tool according to the compensation amount to adjust the polishing force.

[0065] This embodiment can be understood as providing a compliant grinding and polishing control mechanism. This mechanism achieves adaptive adjustment of the contact state between the grinding tool and the workpiece by simulating the dynamic response of the mass-spring-resistance system in the physical world, thereby maintaining a constant and safe grinding force in complex curved surfaces and areas with abrupt changes in stiffness.

[0066] like Figure 5 As shown, in practical applications, it is necessary to first establish an impedance control model for robot compliant polishing. This model includes two cooperating levels: an inner motion control loop and an outer force control loop. The outer loop is responsible for determining the reference trajectory based on the environmental position and stiffness, and receiving actual contact force information from the force sensor. The inner loop receives the robot command position generated by the outer loop, converts it into joint trajectories via the inverse kinematics module, and then inputs it to the motion controller based on fast end-effector sliding mode to output the control torque required for each joint.

[0067] In this process, the control equipment models the dynamic characteristics of the interaction between the robot and the workpiece as a second-order mass-spring-damped system. For example... Figure 6 As shown, the mathematical essence of the mass-spring-damped second-order system is described by a one-dimensional impedance model, expressed as:

[0068] In the formula, This represents the external force acting on the robot's end effector. and These represent the desired position and the current actual position, respectively. This constitutes a positional error, and , , These correspond to the equivalent mass, damping coefficient, and stiffness coefficient, respectively. Together, they describe the response characteristics of the mass-spring-damped second-order system to external disturbances.

[0069] See also Figure 5 Based on the impedance modulus, a variable impedance controller is further constructed. This variable impedance controller is based on a mass-spring-damped system, and its transfer function is constructed in Cartesian space, i.e., the expression is:

[0070] In the formula, , , These are the equivalent mass parameters, damping parameters, and stiffness parameters set for the control equipment, respectively.

[0071] See also Figure 5 This transfer function describes the dynamic mapping relationship from the force error input to the compensation trajectory output, that is, the control device receives the force error calculated in real time. The output compensation value is calculated by this second-order system. Used for reference trajectory Make corrections to generate the final robot command position. .

[0072] During implementation, the control equipment adjusts the parameters of the variable impedance controller online based on the real-time comprehensive stiffness, and the adjustment method is as follows: When the real-time overall stiffness exceeds the preset stiffness threshold, the control device can reduce the current stiffness parameter of the variable impedance controller. To prevent the grinding tool from overloading the workpiece surface due to excessively harsh environments, the damping parameters are increased. To suppress the resulting positional oscillations; When the real-time overall stiffness is lower than the preset stiffness threshold, the control device can increase the stiffness parameter. This enhances the system's ability to correct positional deviations, thereby improving the accuracy of grinding trajectory tracking, while simultaneously reducing damping parameters. This is to accelerate the system response speed and avoid trajectory lag caused by excessive damping.

[0073] It should be noted that the aforementioned adjustment of the damping parameter is not completed independently, but rather in real time by the fuzzy controller based on the force error and its rate of change. Together with the variable impedance controller, it is referred to as the fuzzy variable impedance module. The control device uses the force error and force error rate as input quantities, and fuzzifies them into seven levels of linguistic variables using a triangular membership function, including negative large (NB), negative medium (NM), negative small (NS), zero (ZE), positive small (PS), positive medium (PM), and positive large (PB). Parallel reasoning is then performed using a preset fuzzy rule base, and finally, the centroid method is used to defuzzify the parameters, outputting the adjustment amount of the damping parameter. Ensure that it is always greater than zero.

[0074] Subsequently, the control equipment uses the adjusted variable impedance controller to process the force error of the current grinding force, obtaining the compensation amount required for the current trajectory position; this compensation amount is essentially based on the updated... and Recalculated trajectory correction value The control equipment adjusts the actual position of the grinding tool in space accordingly, so that the contact force continuously approaches and stabilizes at the target force value.

[0075] In this way, without the need for precise pre-modeling of workpiece geometry and material properties, the control equipment can drive the variable impedance controller to dynamically match the real environment simply by fusing the prior stiffness estimation provided by vision with the observed stiffness fed back by the force sensor. This reduces the risk of rigid collisions with the workpiece, ensures that fragile surfaces such as paint films are not damaged during the polishing process, and significantly improves the uniformity of material removal and the consistency of processing.

[0076] Based on the same inventive concept as the grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control provided in this embodiment, this embodiment also provides a grinding and polishing device based on three-dimensional vision guidance and variable stiffness force-resistance control. This device includes at least one functional module that can be stored in a memory or embedded in a control device in software form. The processor in the control device is used to execute the executable module stored in the memory. For example, the software functional modules and computer programs included in the device. Please refer to... Figure 7 Functionally, the device may include: The prior stiffness module 11 is used to obtain the prior estimated stiffness of the workpiece at the current trajectory position when the grinding tool grinds the workpiece along the reference trajectory. The stiffness correction module 12 is used to correct the prior estimated stiffness using the observed stiffness measured by the stiffness sensor to obtain the real-time comprehensive stiffness. The force adjustment module is used to adjust the grinding force determined by the prior estimated stiffness based on the real-time comprehensive stiffness.

[0077] In this embodiment, the prior stiffness module 11 is used to implement Figure 1 In step S1, the force adjustment module is used to achieve... Figure 1 In step S2, the force adjustment module is used to achieve... Figure 1 Step S3 in the above process. Therefore, for a detailed description of each of the above modules, please refer to the specific methods described in the corresponding steps above.

[0078] Optionally, the grinding force of the grinding tool is controlled by a variable impedance controller. The force adjustment module adjusts the grinding force determined by the prior estimated stiffness based on the real-time comprehensive stiffness, including: Obtain the force error of the current polishing force; If the real-time overall stiffness is higher than the preset stiffness threshold, then the current stiffness parameter of the variable impedance controller is reduced and the current damping parameter is increased to obtain the adjusted variable impedance controller. If the real-time overall stiffness is lower than the preset stiffness threshold, then increase the current stiffness parameter of the variable impedance controller and decrease the current damping parameter to obtain the adjusted variable impedance controller. The force error is processed by the adjusted variable impedance controller to obtain the compensation amount for the current trajectory position; Adjust the position of the polishing tool according to the compensation amount to adjust the polishing intensity.

[0079] Optionally, the stiffness correction module 12 uses the observed stiffness measured by the stiffness sensor to correct the prior estimated stiffness to obtain the real-time comprehensive stiffness, including: The corresponding fusion weights are determined based on the uncertainty measures of the prior estimated stiffness and the observed stiffness. The real-time composite stiffness is obtained by weighting the prior estimated stiffness and the observed stiffness.

[0080] Optionally, the relationship between the real-time integrated stiffness and the prior estimated stiffness, observed stiffness, and their corresponding weights is as follows:

[0081] In the formula, Indicates real-time composite stiffness. Indicates the observed stiffness. This indicates the prior estimate of stiffness. , This indicates the corresponding weight.

[0082] Optionally, the prior stiffness module 11 obtains the prior estimated stiffness of the workpiece at the current trajectory position in the following ways: Obtain the curvature feature map of the workpiece surface; Obtain the local curvature corresponding to the current trajectory position from the curvature feature map; Based on the local curvature, the prior estimate of stiffness at the current trajectory position is obtained.

[0083] Optionally, the prior stiffness module 11 obtains the prior estimated stiffness of the current trajectory position based on the local curvature in the following ways: The local curvature is input into a preset stiffness mapping function to obtain the prior estimated stiffness corresponding to the current trajectory position. The stiffness mapping function includes at least one grinding process parameter of the workpiece, and its expression is:

[0084] In the formula, This indicates the prior estimate of stiffness. This indicates the grinding process parameters of the workpiece. Indicates local curvature.

[0085] Optionally, the prior stiffness module 11 acquires the curvature feature map of the workpiece surface in the following ways: Obtain the original 3D point cloud of the workpiece; The original 3D point cloud is preprocessed to obtain the preprocessed optimized point cloud; For each target point in the preprocessed optimized point cloud, obtain a set of points within a preset range from the target point; Based on the geometric information in the point set, the curvature corresponding to the target point is obtained.

[0086] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0087] It should also be understood that if the above embodiments are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0088] Therefore, this embodiment also provides a storage medium, which is a computer-readable storage medium. This storage medium stores a computer program, which, when executed by a processor, implements the grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control provided in this embodiment. The storage medium can be any medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0089] Please refer to Figure 8 This embodiment provides a control device, which may include a processor 22 and a memory 21. The memory 21 stores a computer program, and the processor reads and executes the computer program in the memory 21 corresponding to the above-described embodiments to implement the grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control provided in this embodiment.

[0090] See also Figure 8 The control device also includes a communication unit 23. The memory 21, processor 22 and communication unit 23 are electrically connected to each other directly or indirectly through system bus 24 to realize data transmission or interaction.

[0091] The memory 21 can be an information recording device based on any electronic, magnetic, optical, or other physical principles, used to record execution instructions, data, etc. In some embodiments, the memory 21 can be, but is not limited to, volatile memory, non-volatile memory, memory drive, etc.

[0092] In some embodiments, the volatile memory may be random access memory (RAM); in some embodiments, the non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, etc.; in some embodiments, the storage drive may be a disk drive, solid-state drive, any type of storage disk (such as optical disc, DVD, etc.), or similar storage media, or a combination thereof.

[0093] The communication unit 23 is used to send and receive data over a network. In some embodiments, the network may include a wired network, a wireless network, a fiber optic network, a telecommunications network, an intranet, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), a public switched telephone network (PSTN), a Bluetooth network, a ZigBee network, or a near field communication (NFC) network, or any combination thereof. In some embodiments, the network may include one or more network access points. For example, the network may include wired or wireless network access points, such as base stations and / or network switching nodes, through which one or more components of the service request processing system can connect to the network to exchange data and / or information.

[0094] The processor 22 may be an integrated circuit chip with signal processing capabilities, and may include one or more processing cores (e.g., a single-core processor or a multi-core processor). By way of example only, the processor described above may include a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), an Application Specific Instruction-set Processor (ASIP), a Graphics Processing Unit (GPU), a Physics Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller unit, a Reduced Instruction Set Computing (RISC) computer, or a microprocessor, or any combination thereof.

[0095] Understandable. Figure 8 The structure shown is for illustrative purposes only. The control device may also have more advanced features. Figure 8 Showing more or fewer components, or having with Figure 8 The different configurations shown. Figure 8 The components shown can be implemented using hardware, software, or a combination thereof.

[0096] It should be understood that the apparatus and methods disclosed in the above embodiments can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0097] The above descriptions are merely various embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control, characterized in that, The method includes: When the grinding tool grinds the workpiece along the reference trajectory, the prior estimated stiffness of the workpiece at the current trajectory position is obtained; The prior estimated stiffness is corrected using the observed stiffness measured by a stiffness sensor to obtain the real-time comprehensive stiffness; The grinding force is adjusted based on the real-time integrated stiffness, which is determined by the prior estimated stiffness.

2. The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control according to claim 1, characterized in that, The grinding force of the grinding tool is controlled by a variable impedance controller. Based on the real-time comprehensive stiffness, the grinding force determined by the prior estimated stiffness is adjusted, including: Obtain the force error of the current polishing force; If the real-time integrated stiffness is higher than the preset stiffness threshold, then the current stiffness parameter of the variable impedance controller is reduced and the current damping parameter is increased to obtain the adjusted variable impedance controller. If the real-time integrated stiffness is lower than the preset stiffness threshold, then the current stiffness parameter of the variable impedance controller is increased and the current damping parameter is decreased to obtain the adjusted variable impedance controller. The force error is processed using the adjusted variable impedance controller to obtain the compensation amount for the current trajectory position; Adjust the position of the polishing tool according to the compensation amount to adjust the polishing force.

3. The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control according to claim 1, characterized in that, The prior estimated stiffness is corrected using the observed stiffness measured by a stiffness sensor to obtain the real-time composite stiffness, including: Based on the uncertainty measures of the prior estimated stiffness and the observed stiffness, the corresponding fusion weights are determined; The real-time composite stiffness is obtained by weighting the prior estimated stiffness and the observed stiffness.

4. The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control according to claim 3, characterized in that, The relationship between the real-time integrated stiffness, the prior estimated stiffness, the observed stiffness, and the corresponding weights is as follows: In the formula, This represents the real-time composite stiffness. This indicates the observed stiffness. This represents the prior estimated stiffness. , This indicates the corresponding weight.

5. The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control according to claim 1, characterized in that, Obtaining the prior estimated stiffness of the workpiece at the current trajectory position includes: Obtain the curvature feature map of the workpiece surface; Obtain the local curvature corresponding to the current trajectory position from the curvature feature map; Based on the local curvature, the prior estimate of stiffness at the current trajectory position is obtained.

6. The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control according to claim 5, characterized in that, Based on the local curvature, the prior estimated stiffness of the current trajectory position is obtained, including: The local curvature is input to a preset stiffness mapping function to obtain the prior estimated stiffness corresponding to the current trajectory position. The stiffness mapping function includes at least one grinding process parameter of the workpiece, and its expression is: In the formula, This represents the prior estimated stiffness. This indicates the grinding process parameters of the workpiece. This represents the local curvature.

7. The grinding and polishing method based on three-dimensional vision guidance and variable stiffness force-resistance control according to claim 5, characterized in that, Obtaining the curvature feature map of the workpiece surface includes: Obtain the original three-dimensional point cloud of the workpiece; The original 3D point cloud is preprocessed to obtain the preprocessed optimized point cloud. For each target point in the preprocessed optimized point cloud, obtain a set of points within a preset range from the target point; Based on the geometric information in the point set, the curvature corresponding to the target point is obtained.

Images