Compensation methods, devices and media for robotic machining of large thin-walled parts

Abstract

This invention discloses a method, device, and medium for compensating for the machining of large thin-walled parts using a robot. The method includes: acquiring industrial robot parameters, large thin-walled part parameters, and machining process parameters; establishing a kinematic model based on the industrial robot parameters, and identifying the robot's joint stiffness matrix based on the kinematic model; establishing an end-effector stiffness performance evaluation index based on the joint stiffness matrix; constructing a deformation prediction model based on the machining process parameters and the large thin-walled part parameters; predicting the robot's tool tip deformation and the large thin-walled part machining deformation based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively; and adjusting the robot's pre-determined theoretical machining pose according to the robot's tool tip deformation and the large thin-walled part machining deformation to determine the compensated actual machining pose. This method achieves a significant improvement in the dimensional and positional accuracy of large thin-walled part machining.

Description

Technical Field

[0001] This invention relates to the field of robotic processing technology, and in particular to a method, apparatus and medium for compensating for robotic processing of large thin-walled parts. Background Technology

[0002] Aluminum alloy thin-walled parts possess advantages such as light weight and high structural strength, and are currently widely used in aerospace and other fields. However, their complex shapes and large dimensional variations are significant. Taking aircraft skin as an example, its large dimensions and complex curved surfaces, coupled with manufacturing and assembly errors, necessitate a certain machining allowance during the processing of the skin blank for trimming with adjacent skins during assembly. As a key component in the aircraft's aerodynamic shape, its machining accuracy is crucial for improving the aircraft's aerodynamic performance. Industrial robots offer a large processing range and high flexibility, making them suitable for machining large components. However, their serial structure results in insufficient rigidity, leading to significant deformation at the end of the machining process and consequently, lower machining accuracy. Furthermore, the inherent low rigidity of large thin-walled parts during milling also contributes to significant deformation. These combined effects severely reduce milling accuracy and greatly limit the application of robotic milling technology for large thin-walled parts.

[0003] Currently, many scholars have conducted in-depth theoretical research and practical exploration on improving the machining accuracy of robotic milling. However, most of these studies are limited because the workpieces themselves are small in size or have high inherent rigidity, and the deformation of the workpieces during machining has little to no impact on the machining accuracy. Therefore, they cannot meet the machining accuracy requirements of large thin-walled parts, which limits the application of robotic milling in large thin-walled parts. Summary of the Invention

[0004] This invention provides a method, device, and medium for compensating for the machining of large thin-walled parts by robots, so as to achieve accurate prediction and synchronous compensation for both the tool tip deformation caused by insufficient rigidity of the robot itself and the machining deformation caused by the low inherent rigidity of large thin-walled parts.

[0005] According to one aspect of the present invention, a method for compensating for robotic machining of large thin-walled parts is provided, the method comprising:

[0006] Obtain parameters for industrial robots, large thin-walled parts, and processing techniques;

[0007] A kinematic model is established based on the parameters of the industrial robot, and the joint stiffness matrix of the robot is identified based on the kinematic model; wherein, the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector.

[0008] An end-effector stiffness performance evaluation index is established based on the joint stiffness matrix. A deformation prediction model is constructed based on the machining process parameters and the parameters of the large thin-walled part. The robot tool tip deformation and the machining deformation of the large thin-walled part are predicted based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively.

[0009] The robot's pre-determined theoretical machining pose is adjusted based on the deformation of the robot's cutting edge and the deformation of the large thin-walled part to determine the compensated actual machining pose.

[0010] According to another aspect of the present invention, a compensation device for robotic machining of large thin-walled parts is provided, the device comprising:

[0011] The parameter acquisition module is used to acquire parameters of industrial robots, parameters of large thin-walled parts, and processing parameters.

[0012] The stiffness parameter identification module is used to establish a kinematic model based on the industrial robot parameters and to identify the robot joint stiffness matrix based on the kinematic model; wherein, the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector.

[0013] The deformation prediction module is used to establish an end-effector stiffness performance evaluation index based on the joint stiffness matrix, construct a deformation prediction model based on the processing parameters and the parameters of the large thin-walled part, and predict the deformation amount of the robot's cutting edge and the processing deformation amount of the large thin-walled part based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively.

[0014] The pose compensation module is used to adjust the robot's pre-determined theoretical machining pose based on the deformation of the robot's tool tip and the machining deformation of the large thin-walled part, so as to determine the actual machining pose after compensation.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor;

[0017] and memory that is communicatively connected to at least one processor;

[0018] The memory stores a computer program that can be executed by at least one processor, which enables the at least one processor to perform the large thin-walled part robotic machining compensation method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the large thin-walled part robot machining compensation method of any embodiment of the present invention.

[0020] The technical solution of this invention involves acquiring industrial robot parameters, large thin-walled part parameters, and machining process parameters; establishing a kinematic model based on the industrial robot parameters; identifying the robot joint stiffness matrix based on the kinematic model; wherein the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector; establishing an end effector stiffness performance evaluation index based on the joint stiffness matrix; constructing a deformation prediction model based on the machining process parameters and the large thin-walled part parameters; predicting the robot's tool tip deformation and the large thin-walled part machining deformation based on the end effector stiffness performance evaluation index and the deformation prediction model, respectively; and adjusting the robot's pre-determined machining theoretical pose according to the robot's tool tip deformation and the large thin-walled part machining deformation to determine the compensated actual machining pose. This solves the technical problem of low machining accuracy of large thin-walled parts by robot milling due to deformation coupling, and achieves a significant improvement in the dimensional and positional accuracy of large thin-walled parts machining.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a robot machining compensation method for large thin-walled parts, provided as an embodiment of the present invention;

[0024] Figure 2a A flowchart illustrating another robot machining compensation method for large thin-walled parts provided in an embodiment of the present invention;

[0025] Figure 2b A flowchart illustrating an alternative example of a robot machining compensation method for large thin-walled parts provided in an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of a compensation device for machining large thin-walled parts using a robot, provided in an embodiment of the present invention.

[0027] Figure 4 A schematic diagram of the structure of an electronic device for implementing a robot processing compensation method for large thin-walled parts according to an embodiment of the present invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Figure 1 This is a flowchart illustrating a compensation method for machining large thin-walled parts using a robot, as provided in an embodiment of the present invention. This embodiment is applicable to the machining of large thin-walled parts using a robot. The method can be executed by a compensation device for machining large thin-walled parts using a robot. This device can be implemented in hardware and / or software and can be configured in an electronic device. For example... Figure 1 As shown, the method specifically includes the following steps:

[0031] S110: Obtain industrial robot parameters, large thin-walled part parameters, and processing parameters.

[0032] Industrial robot parameters can be understood as the core characteristic data of the industrial robot itself, including the number of joints, link length, and joint range of motion. Large thin-walled part parameters can be understood as the key attribute data of the large thin-walled part, including wall thickness, overhang length, and distance from the fixture fixing point. Machining process parameters can be understood as the operating parameters set during milling, including spindle speed, depth of cut, feed rate, and number of milling cutter teeth.

[0033] Specifically, we collect various key data related to the robot's own characteristics, the properties of the thin-walled parts to be processed, and the processing operations, so as to provide basic data support for subsequent modeling and prediction.

[0034] Optionally, before obtaining the industrial robot parameters, large thin-walled part parameters, and processing parameters, the method further includes: determining the processing area of ​​the large thin-walled part, and defining the robot processing range to be covered by the robot end effector based on the actual location, actual size, and preset processing path of the processing area.

[0035] The area to be processed can be understood as a specific area on a large, thin-walled part that requires milling. The robot's processing range can be understood as the spatial range that the robot's end effector needs to cover to complete the processing task, which is determined by the actual situation of the area to be processed and the preset processing path.

[0036] Specifically, determine the specific areas to be processed on the large thin-walled parts, and then, based on the actual location and size of the areas to be processed and the pre-planned processing path, define the spatial range that the robot's end effector must cover to ensure that the robot can accurately reach the required processing location.

[0037] S120. Establish a kinematic model based on the industrial robot parameters, and identify the robot joint stiffness matrix based on the kinematic model.

[0038] The kinematic model is a mathematical model system that describes the relationship between the motion of each joint of the robot and the pose of the end effector.

[0039] The kinematic model can be understood as a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector. The joint stiffness matrix can be understood as matrix data characterizing the stiffness characteristics of each joint of the robot, used to reflect the joint's ability to resist deformation.

[0040] Specifically, a mathematical model is constructed using the collected industrial robot parameters. Then, by combining this model with relevant testing methods, matrix data reflecting the joint stiffness of the robot is identified, and the joint's resistance to deformation is determined.

[0041] For example, the kinematic model of a robot can be represented as:

[0042] ;

[0043] in, Represents the number of joints. Represents the joint variable vector. This represents the secondary transformation matrix of the end effector coordinate system relative to the base coordinate system. Represents the end-effector attitude rotation matrix. Represents the end position vector. It represents a homogeneous transformation.

[0044] For example, the stiffness matrix of the robot is:

[0045] ;

[0046] in, Represents the joint stiffness matrix. The Jacobian matrix representing the robot.

[0047] Optionally, the step of identifying the robot joint stiffness matrix based on the kinematic model includes: selecting multiple poses within the robot's processing range, applying a force to the robot's end effector in each pose, recording the target coordinates before and after loading, and calculating the end effector deformation; based on the robot's static stiffness model and the end effector deformation, calling the Jacobian matrix in the kinematic model, and identifying the joint stiffness matrix using the linear least squares method.

[0048] In this context, multiple poses can be understood as the postures and positions corresponding to multiple different combinations of robot joint angles selected within the robot's machining range. A target can be understood as a high-precision marker fixed to the robot's end effector. End effector deformation can be understood as the change in position of the robot's end effector before and after the application of force. The robot's static stiffness model can be understood as a mathematical model describing the relationship between force and deformation under static stress on the robot. The Jacobian matrix can be understood as a key matrix in the kinematic model, used to establish the mapping relationship between robot joint motion and end effector pose changes. The linear least squares method can be understood as solving for the parameter values ​​that best reflect the actual situation from a set of measurement data; in this embodiment, it is used to identify the joint stiffness matrix.

[0049] Specifically, multiple poses are selected within the defined robot processing range. A force is applied to the robot end effector in each pose, and the coordinates of the target before and after loading are recorded and the end effector deformation is calculated. Then, based on the robot static stiffness model, the Jacobian matrix in the kinematic model is called, and the linear least squares method is used to process the measurement data to identify the robot joint stiffness matrix, ensuring that the stiffness data can be adapted to the processing range.

[0050] S130. Establish an end-effector stiffness performance evaluation index based on the joint stiffness matrix, construct a deformation prediction model based on the machining process parameters and the parameters of the large thin-walled part, and predict the deformation amount of the robot's cutting edge and the machining deformation amount of the large thin-walled part based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively.

[0051] Among them, the end-effector stiffness performance evaluation index can be understood as a standard built based on the joint stiffness matrix, used to measure the robot end effector's resistance to deformation under different poses. The deformation prediction model can be understood as a model used to predict the amount of deformation during machining, including a robot tool tip deformation prediction model and a large thin-walled part machining deformation prediction model. Robot tool tip deformation can be understood as the positional offset of the tool tip caused by insufficient stiffness of the robot itself during machining. Large thin-walled part machining deformation can be understood as the shape or positional deformation of the large thin-walled part due to its inherent low stiffness during machining.

[0052] Specifically, an end-effector stiffness evaluation standard is formulated based on the joint stiffness matrix, and two types of deformation prediction models are built in combination with the machining process parameters. Then, the deformation of the tool tip and the thin-walled part during the machining process is calculated by the two types of deformation prediction models respectively.

[0053] Optionally, the step of identifying the robot joint stiffness matrix based on the kinematic model includes: selecting multiple poses within the robot's processing range, applying a force to the robot's end effector in each pose, recording the target coordinates before and after loading, and calculating the end effector deformation; based on the robot's static stiffness model and the end effector deformation, calling the Jacobian matrix in the kinematic model, and identifying the joint stiffness matrix using the linear least squares method.

[0054] The Cartesian compliance matrix can be understood as a matrix derived from the joint stiffness matrix, used to characterize the compliance properties of the robot's end effector in the Cartesian coordinate system. The translational compliance matrix, a component of the Cartesian compliance matrix, reflects the compliance of the robot's end effector in the translational directions (x-axis, y-axis, z-axis). The rotational compliance matrix, also a component of the Cartesian compliance matrix, reflects the compliance of the robot's end effector in the rotational direction. The coupling compliance matrix, also a component of the Cartesian compliance matrix, reflects the compliance properties of the interaction between the translational and rotational motions of the robot's end effector.

[0055] Specifically, the pre-identified joint stiffness matrix is ​​transformed into a Cartesian compliance matrix containing translational, rotational, and coupling compliance information. Then, an end-effector stiffness performance evaluation index is constructed based on these three types of compliance matrices. This end-effector stiffness performance evaluation index clearly reflects the robot's ability to resist deformation under different poses.

[0056] Optionally, the step of constructing a deformation prediction model based on the machining process parameters and the parameters of the large thin-walled part includes: establishing a micro-element milling force model based on the cutting depth, feed rate, and number of milling cutter teeth in the machining process parameters, combined with the milling force coefficient, and using the micro-element milling force model as a prediction model for the deformation of the robot's cutting edge; and constructing a thin-walled part deformation prediction model based on a deep belief network based on the spindle speed, cutting depth, and feed rate in the machining process parameters, combined with the thin-wall thickness, overhang length, and distance from the fixture fixing point in the parameters of the large thin-walled part, and using the thin-walled part deformation prediction model as a prediction model for the machining deformation of the large thin-walled part.

[0057] The micro-element milling force model can be understood as breaking down the milling cutter edge into micro-elements, calculating the force on each micro-element and superimposing them to predict the total cutting force, used for robot tool tip deformation prediction. The milling force coefficient can be understood as a fixed coefficient predetermined based on the tool and workpiece materials, used to calculate the micro-element forces during the milling process; this embodiment does not impose specific limitations on it. The deep belief network can be understood as a deep learning model; in this embodiment, it is used to construct a thin-walled part deformation prediction model. The thin-walled part deformation prediction model can be understood as a model built based on a deep belief network, used to predict the deformation amount during the machining of large thin-walled parts.

[0058] Specifically, based on the cutting depth, feed rate, and number of milling cutter teeth in the machining process parameters, and combined with the preset milling force coefficient, a micro-element milling force model is established to predict the tool tip deformation; at the same time, combined with the machining process parameters and the parameters of large thin-walled parts, a thin-walled part deformation prediction model based on deep belief network is built to achieve accurate prediction of the machining deformation of the thin-walled part itself.

[0059] Optionally, the step of constructing a deformation prediction model based on the machining process parameters and the parameters of the large thin-walled part includes: establishing a micro-element milling force model based on the cutting depth, feed rate, and number of milling cutter teeth in the machining process parameters, combined with the milling force coefficient, and using the micro-element milling force model as a prediction model for the deformation of the robot's cutting edge; and constructing a thin-walled part deformation prediction model based on a deep belief network based on the spindle speed, cutting depth, and feed rate in the machining process parameters, combined with the thin-wall thickness, overhang length, and distance from the fixture fixing point in the parameters of the large thin-walled part, and using the thin-walled part deformation prediction model as a prediction model for the machining deformation of the large thin-walled part.

[0060] Specifically, the cutting force data calculated by the micro-element milling force model is combined with the pre-established end stiffness performance evaluation index. Through the correspondence between force and stiffness, the deformation of the robot tool tip at each point on the machining path is predicted. At the same time, the machining process parameters are input into the thin-walled part deformation prediction model based on deep belief network. Based on the model output results, the machining deformation of large thin-walled parts at each point on the machining path is determined.

[0061] S140. Adjust the robot's pre-determined theoretical machining pose based on the deformation of the robot's cutting edge and the deformation of the large thin-walled part to determine the compensated actual machining pose.

[0062] The theoretical machining pose can be understood as the predetermined theoretical posture and position of the robot during machining, i.e., the theoretical angle combination of each joint. The actual machining pose can be understood as the posture and position of the robot when actually performing the machining task after deformation compensation adjustment.

[0063] Specifically, taking into account both the deformation of the tool tip and the thin-walled part, the pre-set theoretical pose of the robot machining is adjusted to obtain the actual machining pose that can offset the deformation error and ensure machining accuracy.

[0064] Optionally, before adjusting the robot's predetermined machining theoretical pose based on the deformation of the robot's cutting edge and the machining deformation of the large thin-walled part, the method further includes: acquiring three-dimensional contour data of the area to be machined; generating theoretical machining path points within the area to be machined based on preset machining rules; establishing an optimization model with task redundancy parameters as independent variables and optimal robot end-effector stiffness performance as the objective, combined with robot joint angle range constraints, joint velocity threshold constraints, and dexterity constraints; wherein, the task redundancy parameters include the rotation angle around the cutting axis; and solving the optimization model based on a genetic algorithm to obtain the machining theoretical pose corresponding to each theoretical machining path point.

[0065] Among these, 3D contour data can be understood as the 3D spatial coordinate data of the area to be processed. Theoretical machining path points can be understood as machining reference points generated based on preset machining rules, continuously distributed within the area to be processed, covering the entire area. Task redundancy can be understood as the robot having more degrees of freedom than required to complete the predetermined machining task. Task redundancy parameters can be understood as parameters that the robot can flexibly adjust beyond meeting the basic requirements of the end effector's position and posture when completing a predetermined machining task (such as milling along a preset path). The rotation angle around the tool axis can be understood as the angle at which the robot's end effector rotates around the tool's own axis. Joint angle range constraints can be understood as the limit range of angles that each joint of the robot cannot exceed during rotation. Joint velocity threshold constraints can be understood as the safe velocity threshold that each joint of the robot cannot exceed during rotation. Dexterity constraints can be understood as constraints that prevent the robot joints from being in awkward postures, preventing motion jamming or singularities, represented by the F-norm after homogenization of the Jacobian matrix. The optimization model can be understood as a mathematical model established with the goal of optimal end effector stiffness, combined with various constraints, used to solve for the optimal theoretical pose. Genetic algorithms can be understood as a type of intelligent optimization algorithm used to solve optimization models and find the optimal combination of joint angles.

[0066] Specifically, the three-dimensional contour data of the area to be processed is acquired. Based on preset rules such as row spacing and step distance, continuous theoretical processing path points that completely cover the area are discretized within the processing area. Task redundancy parameters are used as independent variables for posture adjustment. Simultaneously, the optimization process must satisfy three constraints: the rotation angle of each robot joint does not exceed the limit range, the joint rotation speed does not exceed the safety threshold, and dexterity constraints prevent joints from being in strange postures or experiencing motion jams. Based on the above independent variables, optimization objectives, and constraints, a mathematical optimization model is constructed. Finally, a genetic algorithm is used to solve the model to obtain the optimal theoretical robot pose for each theoretical processing path point.

[0067] The technical solution of this invention involves acquiring industrial robot parameters, large thin-walled part parameters, and machining process parameters; establishing a kinematic model based on the industrial robot parameters; identifying the robot joint stiffness matrix based on the kinematic model; wherein the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector; establishing an end effector stiffness performance evaluation index based on the joint stiffness matrix; constructing a deformation prediction model based on the machining process parameters and the large thin-walled part parameters; predicting the robot's tool tip deformation and the large thin-walled part machining deformation based on the end effector stiffness performance evaluation index and the deformation prediction model, respectively; and adjusting the robot's pre-determined machining theoretical pose according to the robot's tool tip deformation and the large thin-walled part machining deformation to determine the compensated actual machining pose. This solves the technical problem of low machining accuracy of large thin-walled parts by robot milling due to deformation coupling, and achieves a significant improvement in the dimensional and positional accuracy of large thin-walled parts machining.

[0068] Figure 2a This is a flowchart illustrating another robot-assisted machining compensation method for large thin-walled parts, provided as an embodiment of the present invention. Based on the above embodiments, this embodiment is a further refinement of the above embodiments, and its specific implementation can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here. Figure 2a As shown, the method specifically includes the following steps:

[0069] S210. Obtain industrial robot parameters, large thin-walled part parameters, and processing parameters.

[0070] S220. Establish a kinematic model based on the industrial robot parameters, and identify the robot joint stiffness matrix based on the kinematic model; wherein, the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector.

[0071] S230. Establish an end-effector stiffness performance evaluation index based on the joint stiffness matrix, construct a deformation prediction model based on the machining process parameters and the parameters of the large thin-walled part, and predict the deformation amount of the robot's cutting edge and the machining deformation amount of the large thin-walled part based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively.

[0072] S240. Calculate the total compensation amount for each point on the machining path based on the deformation amount of the robot's cutting tip and the machining deformation amount of the large thin-walled part.

[0073] The total compensation amount can be understood as the sum of the deformation at the robot's cutting edge and the deformation during the machining of large thin-walled parts; it is the total deformation error that needs to be offset.

[0074] Specifically, the tool tip deformation and thin-walled part deformation corresponding to each machining path point are superimposed to obtain the total deformation error that needs to be offset at that point.

[0075] S250. Perform homogeneous transformation on the total compensation amount to obtain the compensation amount of the robot end effector, and convert it into joint angle compensation amount through the robot Jacobian matrix.

[0076] Homogeneous transformation can be understood as a standardized coordinate transformation method used to convert the total compensation amount into compensation parameters that the robot's end effector can recognize. The end effector compensation amount can be understood as the position and orientation parameters that the robot's end effector needs to adjust. Joint angle compensation amount can be understood as the angle value that each robot joint needs to adjust, obtained after converting the end effector compensation amount through a Jacobian matrix.

[0077] Specifically, a homogeneous transformation method is used to convert the total compensation amount into compensation parameters that the robot end effector can recognize. Then, using the mapping relationship of the Jacobian matrix, the end effector compensation amount is converted into the angle adjustment amount of each joint, so that the robot can achieve compensation by adjusting the joint angle.

[0078] S260. Adjust the joint angle corresponding to the theoretical machining pose according to the joint angle compensation amount to determine the actual machining pose after compensation.

[0079] Specifically, the calculated joint angle compensation is superimposed on the joint angle corresponding to the theoretical machining pose to correct the robot's theoretical posture, ultimately obtaining the actual machining pose that can offset the double deformation and ensure machining accuracy.

[0080] The technical solution of this invention obtains a total compensation amount by superimposing the deformation amount of the robot tool tip with the deformation amount of the large thin-walled part, converts it into a joint angle compensation amount through homogeneous transformation and Jacobian matrix, and then adjusts the joint angle corresponding to the theoretical machining pose. This achieves precise cancellation of the double deformation, solves the technical problem of low machining accuracy caused by double deformation in the robot milling of large thin-walled parts, and achieves the technical effect of significantly improving the machining pose accuracy and ensuring the machining quality of large thin-walled parts.

[0081] Figure 2b A flowchart illustrating an alternative example of a robot machining compensation method for large thin-walled parts provided by an embodiment of the present invention; as shown below. Figure 2b As shown, the method includes:

[0082] Step 1: Establish a kinematic model of the robot, fix a high-precision target at the end of the robot, and use a laser tracker to identify joint stiffness and establish evaluation indicators for the stiffness performance of the robot's end effector.

[0083] The robot's processing range specifically refers to the spatial range of the workpiece's processing area. This spatial range should be covered when identifying the robot's joint stiffness, using the robot joint stiffness matrix. This indicates that `diag()` represents a diagonal matrix consisting of the elements within the parentheses, and its identification method is as follows:

[0084] First, N sets of poses with good dexterity are selected within the working area where the robot performs milling. At each robot pose, a certain force F is applied to the robot's end effector. The magnitude of force F is the weight of the applied load. The direction of force F can be measured using a laser tracker. First, the coordinates of the pivot point are measured, then the coordinates of the force application point at the robot's end effector are measured, and the direction of force F is obtained by subtracting the two. The deformation of the robot's end effector under load is obtained by recording the coordinate values ​​of the target before and after loading, and subtracting the two values. These are the forces and moments along the x, y, and z axes. These represent the positional and orientation deformations along the x, y, and z axes.

[0085] Secondly, according to the robot's static stiffness model:

[0086] ;

[0087] Where J is the robot Jacobian matrix. This represents the robot's joint flexibility matrix.

[0088] Will Extracting from this, and linearizing the above equation, we get:

[0089] ;

[0090] The expression for the coefficient matrix A is as follows:

[0091] ;

[0092] in, Let i be the element in the i-th row and j-th column of the robot's Jacobian matrix. Let F be the i-th element;

[0093] Substituting the N sets of measured data into the above formula, we get:

[0094]

[0095] in, , , where Ai is the coefficient matrix obtained from the i-th set of measurement data, and Xi is the end deformation amount of the i-th measurement;

[0096] Obtained by linear least squares method Complete the identification of robot joint stiffness parameters.

[0097] Finally, the joint flexibility matrix Transformed into a robot Cartesian compliance matrix:

[0098] ;

[0099] in, The translational compliance matrix, For rotational compliance matrix, The coupling compliance matrix is ​​further used to define the robot end effector stiffness performance evaluation index as follows:

[0100] ;

[0101] Step 2: Using the redundant degrees of freedom of the six-axis robot as independent variables, optimize to obtain the theoretical pose for robot processing.

[0102] Specifically, considering robot joint range constraints, robot joint velocity constraints, and robot dexterity constraints, and aiming at optimizing the robot's end effector stiffness performance, the theoretical pose for robot machining is obtained. The following mathematical model is established:

[0103] ;

[0104] in, Let be the minimum value of the j-th joint angle. The maximum value of the j-th joint angle. The maximum velocity at the j-th joint angle is... denoted by F-norm, it represents the homogenized Jacobian matrix of the robot. A larger value indicates that the Jacobian matrix is ​​closer to singularity. The optimal theoretical pose can be calculated using intelligent optimization algorithms such as genetic algorithms.

[0105] Step 3: Establish a micro-element milling force model and predict the deformation at the tool tip based on the robot end effector stiffness.

[0106] Specifically, the instantaneous total cutting force is:

[0107] ;

[0108] Where N is the number of milling cutter teeth. The tool tooth position angle, Let be the total cutting force generated by cutting edge j, and:

[0109] ;

[0110] Where 'a' is the cutting height. Let be the infinitesimal milling force in the x, y, and z directions, and:

[0111] ;

[0112] in, Let be the axial, radial, and tangential infinitesimal forces acting on the blade element, and:

[0113] ;

[0114] Where Ktc, Kte, Krc, Kre, Kac, and Kae are milling force coefficients, and dz is the tool edge micro-element. For instantaneous cutting thickness, It is a unit step function.

[0115] Step 4: Conduct cutting experiments to establish a deformation prediction model for thin-walled part milling based on deep belief networks.

[0116] Specifically, a robotic milling experiment was conducted on thin-walled parts, selecting machining parameters such as robot joint angles at machining path points, spindle speed, feed rate, depth of cut, wall thickness, overhang length, and distance from the fixture fixing point. Deformation data of the thin-walled parts during milling was measured. First, the data was preprocessed to remove a small amount of outlier data. Second, a random forest algorithm was used to select the feature variables with the greatest impact on the deformation of the thin-walled parts, which were then used as input to a deep belief network to establish a deformation prediction model for thin-walled parts milled by the robot. A genetic algorithm was introduced to optimize the initial parameters of the prediction model, such as the pre-training learning rate, the number of RBN iterations, and the number of DBN fine-tuning iterations, to obtain the optimal parameter combination. Finally, a deformation prediction model for thin-walled parts milled by the robot based on a deep belief network was established.

[0117] For example, milling experiments were conducted using key parameters such as robot joint angles, spindle speed, feed rate, depth of cut, thin-wall thickness, overhang length, and distance from the fixture fixing point at the machining path point, while simultaneously measuring the deformation data of the thin-walled part. After preprocessing the raw data to remove outliers, a random forest algorithm was used to select core feature variables that significantly affect deformation, which were then used as model inputs. The model adopted an architecture of "input layer - multi-layer stacked restricted Boltzmann machine (RBM) hidden layer - single-node linear output layer". Initial parameters such as the pre-training learning rate, the number of RBM iterations, and the number of DBN fine-tunings were first optimized using a genetic algorithm. Then, unsupervised greedy layer-by-layer pre-training combined with supervised backpropagation fine-tuning was performed. During the pre-training stage, the contrastive divergence algorithm was used to train the RBM layer by layer to extract data features. During the fine-tuning stage, backpropagation was used to optimize the weights of the entire network. The model performance was ensured by combining training sets, validation sets, and test sets. Finally, a deep belief network model that can accurately predict the deformation of thin-walled parts was constructed, providing reliable data support for subsequent deformation compensation.

[0118] Step 5: Based on the milling deformation prediction model, predict the deformation of thin-walled parts during milling according to process parameters such as robot joint angles and spindle speed.

[0119] Specifically, the robot joint angles corresponding to the optimal theoretical pose of the robot at each machining path point during this machining process, as well as machining process parameters such as spindle speed, feed rate, depth of cut, thin-wall thickness, overhang length, and distance from the fixture fixing point, are input into the established deformation prediction model for thin-walled part robot milling to obtain the deformation amount of the thin-walled part at any machining path point pi. .

[0120] Step 6: Consider the deformation of the tool tip and the machining deformation of thin-walled parts to comprehensively adjust the robot's machining theoretical pose.

[0121] Specifically, for any processing path point p i At this point, based on the deformation value of the robot's cutting edge. Deformation of thin-walled parts Obtain any processing path point p during the robot processing. i The amount of compensation required :

[0122]

[0123] Secondly, a homogeneous transformation is performed to obtain the compensation amount that the robot end effector needs to perform. :

[0124]

[0125] The amount of compensation required for the robot's joint angles is calculated using the robot's Jacobian matrix.

[0126]

[0127] When planning the machining of thin-walled parts by robot milling, the robot joint compensation amount is calculated for all machining path points, the joint angles corresponding to the theoretical pose of the robot are compensated, and the compensated robot joint angles are input into the controller to execute the corresponding thin-walled part milling task.

[0128] The technical solution of this invention, after identifying the robot's joint stiffness, calculates the Cartesian compliance matrix of the robot's end effector, and establishes an evaluation index for the robot's end effector stiffness performance. It optimizes the robot pose solution under constraints of robot joint range, robot joint velocity, and robot dexterity, reducing deformation during robot milling. Using a deep belief network, it can accurately and quickly predict the deformation during the milling of thin-walled parts based on machining parameters such as spindle speed, feed rate, and depth of cut set during machining planning. Simultaneously considering the deformation of the robot's end effector and the deformation generated by the thin-walled part itself during robot milling, it compensates for the robot's theoretical machining pose using the predicted deformation, effectively improving the machining accuracy of robot milling of thin-walled parts.

[0129] Figure 3 This is a schematic diagram of a compensation device for machining large thin-walled parts using a robot, provided as an embodiment of the present invention. Figure 3 As shown, the device includes: a parameter acquisition module 310, a stiffness parameter identification module 320, a deformation prediction module 330, and a pose compensation module 340.

[0130] The system includes the following modules: a parameter acquisition module 310, used to acquire industrial robot parameters, large thin-walled part parameters, and processing parameters; a stiffness parameter identification module 320, used to establish a kinematic model based on the industrial robot parameters and identify the robot joint stiffness matrix based on the kinematic model; wherein the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector; a deformation prediction module 330, used to establish an end effector stiffness performance evaluation index based on the joint stiffness matrix, construct a deformation prediction model based on the processing parameters and the large thin-walled part parameters, and predict the robot tool tip deformation and the large thin-walled part processing deformation based on the end effector stiffness performance evaluation index and the deformation prediction model, respectively; and a pose compensation module 340, used to adjust the robot's pre-determined theoretical processing pose according to the robot tool tip deformation and the large thin-walled part processing deformation to determine the compensated actual processing pose.

[0131] The technical solution of this invention involves acquiring industrial robot parameters, large thin-walled part parameters, and machining process parameters; establishing a kinematic model based on the industrial robot parameters; identifying the robot joint stiffness matrix based on the kinematic model; wherein the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector; establishing an end effector stiffness performance evaluation index based on the joint stiffness matrix; constructing a deformation prediction model based on the machining process parameters and the large thin-walled part parameters; predicting the robot's tool tip deformation and the large thin-walled part machining deformation based on the end effector stiffness performance evaluation index and the deformation prediction model, respectively; and adjusting the robot's pre-determined machining theoretical pose according to the robot's tool tip deformation and the large thin-walled part machining deformation to determine the compensated actual machining pose. This solves the technical problem of low machining accuracy of large thin-walled parts by robot milling due to deformation coupling, and achieves a significant improvement in the dimensional and positional accuracy of large thin-walled parts machining.

[0132] Optionally, the device further includes:

[0133] The processing range delineation module is used to determine the processing area of ​​the large thin-walled part before acquiring the industrial robot parameters, large thin-walled part parameters and processing technology parameters. Based on the actual location, actual size and preset processing path of the processing area, the module delineates the robot processing range that the robot end effector needs to cover.

[0134] Optionally, the stiffness parameter identification module includes:

[0135] The end-effector deformation calculation unit is used to select multiple poses within the robot's processing range, apply a force to the robot's end in each pose, record the target's coordinate values ​​before and after loading, and calculate the end-effector deformation.

[0136] The joint stiffness matrix identification unit is used to identify the joint stiffness matrix based on the robot's static stiffness model and end-effector deformation by calling the Jacobian matrix in the kinematic model and using the linear least squares method.

[0137] Optionally, the deformation prediction module includes:

[0138] A matrix transformation unit is used to transform the joint stiffness matrix into a Cartesian compliance matrix; wherein the Cartesian compliance matrix includes a translational compliance matrix, a rotational compliance matrix, and a coupling compliance matrix;

[0139] The evaluation index construction unit is used to construct an end-effector stiffness performance evaluation index based on the translational compliance matrix, the rotational compliance matrix, and the coupling compliance matrix; wherein, the stiffness performance evaluation index is used to characterize the robot end-effector's resistance to deformation under different poses.

[0140] Optionally, the deformation prediction module includes:

[0141] The first model building unit is used to establish a micro-element milling force model based on the cutting depth, feed rate and number of milling cutter teeth in the machining process parameters, combined with the milling force coefficient, and to use the micro-element milling force model as a prediction model for the deformation of the robot tool tip.

[0142] The second model construction unit is used to construct a deformation prediction model for thin-walled parts based on the spindle speed, depth of cut, and feed rate in the machining process parameters, combined with the thin-wall thickness, overhang length, and distance from the fixture fixing point in the parameters of the large thin-walled parts. The deformation prediction model of the thin-walled parts is then used as the prediction model for the machining deformation of large thin-walled parts.

[0143] Optionally, the deformation prediction module includes:

[0144] The first deformation prediction unit is used to combine the cutting force data calculated by the micro-element milling force model with the end stiffness performance evaluation index to predict the deformation of the robot tool tip at each point of the machining path.

[0145] The second deformation prediction unit is used to input the machining process parameters into the thin-walled part deformation prediction model based on deep belief network, and output the machining deformation of large thin-walled parts at each point of the machining path.

[0146] Optionally, the pose compensation module includes:

[0147] The total compensation calculation unit is used to calculate the total compensation amount at each point of the machining path based on the deformation amount of the robot's cutting tip and the machining deformation amount of the large thin-walled part.

[0148] The compensation amount determination unit is used to perform homogeneous transformation on the total compensation amount to obtain the compensation amount of the robot end effector, and convert it into joint angle compensation amount through the robot Jacobian matrix.

[0149] The pose adjustment unit is used to adjust the joint angles corresponding to the theoretical machining pose according to the joint angle compensation amount, so as to determine the actual machining pose after compensation.

[0150] Optionally, the device further includes:

[0151] The theoretical machining point determination module is used to obtain the three-dimensional contour data of the area to be machined before adjusting the robot's pre-determined theoretical machining pose according to the deformation of the robot's cutting edge and the deformation of the large thin-walled part, and to generate theoretical machining path points in the area to be machined based on preset machining rules.

[0152] The optimization model construction module is used to establish an optimization model with task redundancy parameters as independent variables and optimal robot end-effector stiffness performance as the objective, combined with robot joint angle range constraints, joint velocity threshold constraints, and dexterity constraints; wherein, the task redundancy parameters include the rotation angle around the tool axis;

[0153] The processing theoretical pose determination module is used to solve the optimization model based on a genetic algorithm to obtain the processing theoretical pose corresponding to each of the theoretical processing path points.

[0154] The large thin-walled part robot processing compensation device provided in the embodiments of the present invention can execute the large thin-walled part robot processing compensation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0155] Figure 4 This is a schematic diagram of an electronic device for implementing the robotic machining compensation method for large thin-walled parts according to embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0156] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0157] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0158] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method of compensation in robotic machining of large thin-walled parts.

[0159] In some embodiments, the method for large thin-walled part robotic machining compensation can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for large thin-walled part robotic machining compensation described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for large thin-walled part robotic machining compensation by any other suitable means (e.g., by means of firmware).

[0160] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0161] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0162] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0163] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0164] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0165] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0166] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0167] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for compensating for robotic machining of large thin-walled parts, characterized in that, include: Obtain parameters for industrial robots, large thin-walled parts, and processing techniques; A kinematic model is established based on the parameters of the industrial robot, and the joint stiffness matrix of the robot is identified based on the kinematic model; wherein, the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector. An end-effector stiffness performance evaluation index is established based on the joint stiffness matrix. A deformation prediction model is constructed based on the machining process parameters and the parameters of the large thin-walled part. The robot tool tip deformation and the machining deformation of the large thin-walled part are predicted based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively. The robot's pre-determined theoretical machining pose is adjusted based on the deformation of the robot's cutting edge and the deformation of the large thin-walled part to determine the compensated actual machining pose.

2. The method according to claim 1, characterized in that, Before obtaining parameters for industrial robots, large thin-walled parts, and processing techniques, the following steps are also included: The processing area of ​​the large thin-walled part is determined, and the robot processing range to be covered by the robot end effector is defined based on the actual location, actual size and preset processing path of the processing area.

3. The method according to claim 2, characterized in that, The process of identifying the robot joint stiffness matrix based on the kinematic model includes: Multiple poses are selected within the robot's processing range. A force is applied to the robot's end effector under each pose. The coordinate values ​​of the target before and after loading are recorded, and the end effector deformation is calculated. Based on the robot's static stiffness model and end-effector deformation, the joint stiffness matrix is ​​obtained by calling the Jacobian matrix in the kinematic model and using the linear least squares method.

4. The method according to claim 1, characterized in that, The establishment of end-effector stiffness performance evaluation index based on the joint stiffness matrix includes: The joint stiffness matrix is ​​transformed into a Cartesian compliance matrix; wherein the Cartesian compliance matrix includes a translational compliance matrix, a rotational compliance matrix, and a coupling compliance matrix; An end-effector stiffness performance evaluation index is constructed based on the translational compliance matrix, the rotational compliance matrix, and the coupling compliance matrix; wherein, the stiffness performance evaluation index is used to characterize the robot end-effector's resistance to deformation under different poses.

5. The method according to claim 1, characterized in that, The process of constructing a deformation prediction model based on the processing parameters and the parameters of the large thin-walled part includes: Based on the cutting depth, feed rate and number of milling cutter teeth in the machining process parameters, and combined with the milling force coefficient, a micro-element milling force model is established, and the micro-element milling force model is used as a prediction model for the deformation of the robot's tool tip. Based on the spindle speed, depth of cut, and feed rate in the machining process parameters, and combined with the thin wall thickness, overhang length, and distance from the fixture fixing point in the parameters of the large thin-walled part, a deformation prediction model for the thin-walled part based on a deep belief network is constructed, and the deformation prediction model for the thin-walled part is used as the prediction model for the machining deformation of the large thin-walled part.

6. The method according to claim 1, characterized in that, The prediction of robot tool tip deformation and large thin-walled part machining deformation based on the end-effector stiffness performance evaluation index and the deformation prediction model includes: By combining the cutting force data calculated by the micro-element milling force model with the end stiffness performance evaluation index, the deformation of the robot tool tip at each point on the machining path can be predicted. The machining process parameters are input into the thin-walled part deformation prediction model based on deep belief network, and the machining deformation of large thin-walled parts at each point of the machining path is output.

7. The method according to claim 1, characterized in that, The step of adjusting the robot's pre-determined theoretical machining pose based on the deformation of the robot's cutting edge and the deformation of the large thin-walled part to determine the compensated actual machining pose includes: The total compensation for each point on the machining path is calculated based on the deformation of the robot's cutting tip and the deformation of the large thin-walled part. The total compensation amount is homogeneously transformed to obtain the compensation amount of the robot end effector, which is then converted into joint angle compensation amount using the robot Jacobian matrix. Adjust the joint angles corresponding to the theoretical machining pose based on the joint angle compensation amount to determine the actual machining pose after compensation.

8. The method according to claim 1, characterized in that, Before adjusting the robot's predetermined machining theoretical pose based on the deformation of the robot's cutting edge and the machining deformation of the large thin-walled part, the method further includes: Acquire the three-dimensional contour data of the area to be processed, and generate theoretical processing path points within the area to be processed based on preset processing rules; An optimization model is established with task redundancy parameters as independent variables and optimal robot end-effector stiffness performance as the objective, combining robot joint angle range constraints, joint velocity threshold constraints, and dexterity constraints; wherein, the task redundancy parameters include the rotation angle around the tool axis; The optimization model is solved using a genetic algorithm to obtain the theoretical processing pose corresponding to each theoretical processing path point.

9. A compensation device for robotic processing of large thin-walled parts, characterized in that, include: The parameter acquisition module is used to acquire parameters of industrial robots, parameters of large thin-walled parts, and processing parameters. The stiffness parameter identification module is used to establish a kinematic model based on the industrial robot parameters and to identify the robot joint stiffness matrix based on the kinematic model; wherein, the kinematic model is a mathematical model system describing the relationship between the motion of each joint of the robot and the pose of the end effector. The deformation prediction module is used to establish an end-effector stiffness performance evaluation index based on the joint stiffness matrix, construct a deformation prediction model based on the processing parameters and the parameters of the large thin-walled part, and predict the deformation amount of the robot's cutting edge and the processing deformation amount of the large thin-walled part based on the end-effector stiffness performance evaluation index and the deformation prediction model, respectively. The pose compensation module is used to adjust the robot's pre-determined theoretical machining pose based on the deformation of the robot's tool tip and the machining deformation of the large thin-walled part, so as to determine the actual machining pose after compensation.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the large thin-walled part robotic machining compensation method according to any one of claims 1-8.

Images