Linear module high-precision positioning control method and system

Abstract

The invention relates to the technical field of high-precision positioning, and discloses a linear module high-precision positioning control method and system, and the method comprises the steps: carrying out the feature point extraction and feature point tracking of a motion image sequence of a linear module, and obtaining displacement measurement data and vibration feature data; performing nonlinear compensation on a linear module according to the displacement measurement data and the vibration characteristic data, and constructing a standard linear state equation; performing sliding mode surface design and gain parameter calculation based on the standard linear state equation to obtain an initial sliding mode controller parameter; track interpolation and motion feedback are carried out on the linear module according to the initial sliding mode controller parameters, and real-time position deviation data are obtained; the initial sliding mode controller parameters are iteratively optimized based on the real-time position deviation data, and target sliding mode controller parameters are obtained.According to the method, impact in the acceleration and deceleration process is reduced, system vibration is effectively reduced, and the positioning precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of high-precision positioning technology, and in particular to a high-precision positioning control method and system for a linear module. Background Art

[0002] Traditional linear module control methods face many challenges in high-speed and high-precision positioning. Due to the nonlinear characteristics of the mechanical structure, interference from environmental factors, and load changes, linear modules are prone to vibration, friction, and temperature drift during movement, which seriously restricts further improvement of positioning accuracy.

[0003] The current mainstream linear module control method mainly relies on traditional sensors such as grating scales and magnetic scales for position feedback. These sensors are not only expensive, but also difficult to obtain multi-dimensional information such as displacement and vibration at the same time. At the same time, most of the existing control algorithms use linear control theory, which has limitations in dealing with system nonlinearity and external interference, and it is difficult to meet the requirements of sub-micron positioning accuracy. Summary of the invention

[0004] The present invention provides a high-precision positioning control method and system for a linear module, which reduces the impact during acceleration and deceleration, effectively reduces system vibration, and improves positioning accuracy.

[0005] In a first aspect, the present invention provides a linear module high-precision positioning control method, the linear module high-precision positioning control method comprising: Extract and track the feature points of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data; Perform nonlinear compensation on the linear module according to the displacement measurement data and the vibration characteristic data, and construct a standard linear state equation; Based on the standard linear state equation, sliding surface design and gain parameter calculation are performed to obtain initial sliding mode controller parameters; Perform trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data; The initial sliding mode controller parameters are iteratively optimized based on the real-time position deviation data to obtain target sliding mode controller parameters.

[0006] In a second aspect, the present invention provides a linear module high-precision positioning control system, the linear module high-precision positioning control system comprising: The feature point tracking module is used to extract and track the feature points of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data; A nonlinear compensation module, used for performing nonlinear compensation on the linear module according to the displacement measurement data and the vibration characteristic data, and constructing a standard linear state equation; A calculation module, used for performing sliding surface design and gain parameter calculation based on the standard linear state equation to obtain initial sliding mode controller parameters; A feedback module, used for performing trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data; The optimization module is used to iteratively optimize the initial sliding mode controller parameters based on the real-time position deviation data to obtain target sliding mode controller parameters.

[0007] In the technical solution provided by the present invention, a method combining visual measurement and double matching tracking is adopted to realize high-precision measurement of the motion state of the linear module. Through the design of feature point array and sub-pixel feature extraction algorithm, the limitations of traditional sensors are overcome, and displacement and vibration information are obtained at the same time, thereby improving the measurement accuracy. Based on nonlinear dynamic modeling and feedback linearization method, the nonlinear characteristics of the system are effectively compensated. By establishing a complete dynamic model, the influence of various factors such as inertial force, Coriolis force, friction force, etc. is considered, so that the system can still maintain good linear characteristics when moving in a large range. An adaptive gain sliding mode controller is designed to enhance the anti-interference ability of the system. The continuous reaching law is used instead of the traditional sign function to effectively suppress the chattering caused by control switching. At the same time, the adaptive gain mechanism enables the controller to automatically adapt to load changes. A parameter optimization method based on multi-dimensional performance indicators is proposed to realize automatic optimization of controller parameters. Through innovative designs such as hierarchical weight calculation, dynamic step size adjustment and decoupling compensation, the coupling problem in the parameter optimization process is solved to ensure continuous improvement of system performance. The seventh-order polynomial trajectory planning and third-order continuity processing are adopted to ensure the smoothness of the motion trajectory. Through reasonable trajectory design, the impact during acceleration and deceleration is reduced, the system vibration is effectively reduced, and the positioning accuracy is improved.

[0008] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0009] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of an embodiment of a high-precision positioning control method for a linear module in an embodiment of the present invention; Figure 2 It is a schematic diagram of an embodiment of a high-precision positioning control system for a linear module in an embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0012] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0013] To facilitate understanding of this embodiment, a high-precision positioning control method for a linear module disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps: 101. Extract and track feature points of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data; It is understandable that the execution subject of the present invention may be a linear module high-precision positioning control system, or a terminal or a server, which is not specifically limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0014] Specifically, the 5×5 feature point array preset on the surface of the moving part in the linear module is subjected to Gaussian filtering and noise reduction to effectively remove high-frequency noise in the image, reduce the influence of noise on the subsequent feature point extraction accuracy, and obtain filtered image data. For the filtered image data, Hough transform is used to detect circular contours. Hough transform is an image processing method that effectively identifies and extracts circular features in images. In this process, the circular contour of each feature point in the image is located by Hough transform, and then the specific position of these feature points is determined. Through this step, the contour data of the feature point is obtained. The contour data is refined. The contour data is fitted by two-dimensional Gaussian curve fitting technology to more accurately determine the center coordinates of each feature point. Gaussian curve fitting can eliminate errors caused by image edge blur or noise, and obtain the center coordinate data of the feature point. The center coordinate data of the feature point is converted to the coordinate system, and the local coordinate data is converted to the global coordinate system. By introducing the global coordinate system conversion matrix, the local coordinates of the feature point are converted to the displacement coordinate data in the global coordinate system, ensuring that the obtained motion data can match the global motion state of the linear module. Perform time series differential operation on the displacement coordinate data to calculate the instantaneous velocity and acceleration data of the linear module. Fuse the displacement coordinate data, velocity and acceleration data with the motor torque, load mass and ambient temperature of the linear module. The motor torque is an important factor affecting the motion state of the linear module. The load mass affects the acceleration and stability of the module, while the ambient temperature affects the performance of the motor and the friction characteristics of the module. By fusing these various data, a motion state data set is obtained. Based on the motion state data set, feature point tracking and analysis is performed. Through time series-based kinematic models or Kalman filtering methods, the motion trajectory of each feature point is tracked and corrected to obtain the real-time displacement measurement data and vibration feature data of the linear module.

[0015] 102. Perform nonlinear compensation on the linear module based on displacement measurement data and vibration characteristic data, and construct a standard linear state equation; Specifically, the motion state data set is input into the spatial matching layer, and the spatial change relationship between each feature point in the image is captured by calculating the affine transformation matrix of the feature point array in adjacent image frames. The calculation of the affine transformation matrix is ​​based on the key point matching in the image, and its spatial transformation is determined by comparing the relative positions of the same feature points in the current frame and the previous frame. In order to improve the tracking accuracy, the optimal matching pairs are screened out, that is, the feature point combinations with the smallest transformation error. These matching pairs reflect the precise displacement of the module in adjacent moments. Through this step, the spatial matching displacement data is obtained, which reflects the position change of the module in actual motion. The spatial matching displacement data is input into the time matching layer for state vector construction. The time matching layer combines the spatial matching displacement data with the time information, and constructs a state vector, which contains two important components of displacement and velocity. The state vector describes the motion state of the linear module at any time. In this way, the motion trajectory of the module is updated and tracked in real time. Based on the state vector, the state model of the filter is constructed. The filter filters out noise and errors by processing these state vectors, optimizes the accuracy of the data, and ensures accurate displacement and velocity estimation in real-time control. Trajectory prediction and smoothing are performed based on spatial matching displacement data and filter state models. Trajectory prediction uses known kinematic models and historical displacement data to predict future motion trajectories, while smoothing helps reduce deviations caused by measurement noise or other uncertainties. By combining spatial matching displacement data and filter state models, trajectory prediction can provide the system with future displacement estimates, and reduce instantaneous fluctuations in a short period of time through smoothing to obtain displacement measurement data. The vibration amplitude is calculated for the displacement measurement data to obtain vibration intensity information. The vibration amplitude reflects the mechanical vibration or instability that occurs in the module during movement. The frequency characteristics of the vibration are extracted to analyze the dynamic response characteristics of the system. The extraction of frequency characteristics uses frequency domain analysis methods such as fast Fourier transform to convert displacement data from the time domain to the frequency domain to obtain vibration frequency data, revealing the vibration mode of the module at different frequencies, which helps to identify potential faults or instability of the system. The vibration amplitude data and vibration frequency data are feature fused to obtain vibration feature data.

[0016] The displacement measurement data is subjected to motion equation parameter identification and dynamic modeling to obtain the mapping relationship between displacement and external force. In the dynamic system, the movement of the linear module is affected by the inertial force, as well as the friction, gravity and other external interferences. Through experimental measurement or data analysis, the displacement-force mapping relationship is obtained, which describes the dynamic response of the module under different force conditions. At the same time, in order to characterize the nonlinear characteristics of the system, the vibration characteristic data is analyzed in the frequency domain to extract the damping characteristics of the system. The damping characteristic is an important parameter to measure the vibration attenuation ability of the system. Through methods such as Fourier transform or wavelet transform, the damping coefficient is identified from the vibration spectrum, and combined with the previously obtained displacement-force mapping relationship, a complete nonlinear dynamic model is constructed. Through mathematical modeling, a nonlinear dynamic equation containing inertia, Coriolis force, gravity, friction and external disturbance is established. The basic form of the equation is: ; in, is the inertia matrix, which describes the mass inertia characteristics of the system. is the displacement vector, which indicates the position state of the system, is the Coriolis force matrix, which is used to describe the Coriolis force effect caused by the non-inertial motion of the system. is the gravity term, reflecting the influence of gravity on the system at different positions. Represents the friction term. Friction usually depends on speed and takes different forms such as Coulomb friction and viscous friction. is the external disturbance term, which indicates the external impact force or environmental disturbance that the system may be subjected to in actual operation, and u is the control input, which represents the control force applied by the system. In order to construct the standard linear state equation, the state variables of the nonlinear dynamic model are defined. By introducing the state vector: ; The state variables of the system are expressed as a combination of displacement and velocity, and the original second-order dynamic equation is converted into a first-order state equation. Since the original dynamic equation contains nonlinear terms, in order to achieve effective control, the nonlinear terms are compensated so that the system is converted into a linear form. Through feedback linearization technology, a nonlinear compensation control law is designed so that the dynamic behavior of the system is expressed in a linear form under the control coordinate. The expression of the control law is: ; in, is the new control input, which is used to compensate the nonlinear part of the original system so that the input-output relationship of the system can be transformed into a linear relationship. When the control law is substituted into the original nonlinear dynamic equation, the dynamic equation of the system is rewritten as: ; in, and are the system matrix and the input matrix respectively. This form of expression is a standard linear state equation. In order to ensure the integrity of the linearized system, the system matrix is ​​constructed and finally the standard linear state equation is obtained: ; in, represents the derivative vector of the state variables, yes dimensional system matrix, describing the internal dynamic characteristics of the system, is the input matrix, which reflects the influence of control input on the system state, and is the disturbance vector, which contains all unmodeled external disturbances or system uncertainties. Through this series of nonlinear compensation and linearization steps, a linear state equation that conforms to the standard form is finally constructed.

[0017] 103. Based on the standard linear state equation, the sliding surface design and gain parameter calculation are performed to obtain the initial sliding mode controller parameters; Specifically, the displacement tracking error equation is defined. The tracking error describes the deviation between the current state of the system and the expected state, so by setting the error vector Quantify this deviation, where is the current state vector of the system, and is the desired state vector. The sliding surface is constructed based on the tracking error equation. The sliding surface is the key part of sliding mode control. Its design determines how the system approaches the target state and remains stable in the presence of disturbances. The sliding surface is set by To define, is the sliding surface vector, is the sliding surface coefficient matrix of the positive definite diagonal matrix, and represents the derivative of the error. The function of this sliding surface expression is to construct a hyperplane for error convergence, so that the system can maintain steady-state tracking when running on the hyperplane without being affected by small disturbances. and The value of is used to adjust the dynamic characteristics of the sliding mode control, such as convergence speed and stability. In order to ensure that the system can converge quickly and maintain sliding mode motion when constructing the sliding surface, a continuous reaching law is designed to control the evolution of the sliding mode variable. The purpose of the reaching law is to guide the system state into the sliding surface and maintain stable motion on the sliding surface. Therefore, the reaching law is defined as ,in is the sliding surface derivative vector, is the approach velocity coefficient matrix of the positive definite diagonal matrix, which controls the rate at which the system approaches the sliding surface. is the integral coefficient matrix of the positive definite diagonal matrix, which determines the influence of the cumulative error in the approach process, and Represents the sliding surface integral vector, reflecting the cumulative change of the sliding state over time. This reaching law ensures that the system can quickly approach the sliding surface and maintain stable tracking behavior on the sliding surface, while ensuring strong robustness to external disturbances. After defining the reaching law, it is then substituted into the standard linear state equation for solution to obtain the specific sliding mode control law. The purpose of the sliding mode control law is to enable the system state to maintain stability on the sliding surface and overcome external disturbances through the design of the control input. Therefore, the control input vector is expressed as ,in is the equivalent control term vector, which is used to compensate for the known system dynamics to ensure that the system moves along the expected trajectory, while is the switching control term vector, which is used to suppress unknown disturbances and enhance the robustness of the system. The equivalent control term is determined by the system dynamics model, while the switching control term is defined by the sliding mode control strategy, which uses a sign function or a saturation function to suppress the influence of external noise on the control system. Based on the sliding mode control law, an adaptive gain update law is designed so that the controller adjusts the control gain according to the real-time error. Define the gain parameter dynamic equation ,in represents the derivative of the gain matrix, is a positive definite diagonal matrix of learning rate coefficients, which determines the speed of gain adjustment, and is the absolute value vector of the sliding surface, which is used to guide the gain update process. The function of the gain update law is to allow the sliding mode control system to dynamically adjust the gain parameter according to the actual error level, so that when the error is large, the control gain increases to improve the convergence speed, and when the error is small, the gain decreases to reduce the control chattering problem of the system. The sliding mode control law is combined with the gain parameter dynamic equation to obtain the initial sliding mode controller parameters.

[0018] 104. Perform trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data; Specifically, trajectory planning is performed for the positioning task of the linear module to ensure that the module remains stable throughout the entire motion process and meets the constraints of position, velocity and acceleration. In the trajectory generation stage, the seventh-order polynomial planning method is used to ensure the smoothness and executability of the trajectory, while ensuring that the module will not experience sudden or drastic changes in velocity and acceleration during the motion process. By setting the starting point, end point and corresponding motion boundary conditions, an ideal trajectory is generated, so that the system can perform high-precision motion control according to the established path, and meet the dynamic constraints throughout the motion process to ensure the smoothness of the operation. According to the initial sliding mode controller parameters, the expected trajectory data is processed for third-order continuity, that is, to ensure that the trajectory remains continuous at the three levels of position, velocity and acceleration. The purpose of continuity processing is to reduce the uncertainty of the trajectory and prevent shock and oscillation caused by the uneven trajectory, thereby improving the dynamic performance of the system. In actual calculations, methods such as high-order spline interpolation or Bezier curve interpolation are used to make the trajectory have a smooth transition at each time point, conform to the physical characteristics of the linear module, and obtain smooth trajectory interpolation data. At the same time, in order to ensure the control effect of the system, the actual motion state of the linear module is monitored in real time. Through high-precision sensors such as grating rulers, laser rangefinders or inertial measurement units, the position information of the module during operation is collected, and the real-time speed data is obtained in combination with speed sensors or other calculation methods. These real-time motion feedback data provide the control system with current state information, so that the controller can make necessary adjustments according to the actual situation, ensuring that the module can strictly run according to the planned trajectory and can quickly correct the motion error when it is disturbed by external interference. The smooth trajectory interpolation data is compared and calculated with the real-time motion feedback data to calculate the real-time position deviation data. The position deviation reflects the gap between the current system state and the ideal state, and is obtained by comparing the measured real-time position with the target position in the trajectory interpolation. At the same time, the speed deviation is determined by comparing the actual speed with the expected speed. The calculation results of the position deviation and speed deviation can directly reflect the current motion error of the module and provide it to the control system for further adjustment and optimization. The real-time position deviation data will be input into the sliding mode controller as feedback to optimize the control strategy and dynamically adjust the control parameters so that the linear module can still maintain high-precision positioning effect under external interference and system parameter changes.

[0019] 105. Iteratively optimize the initial sliding mode controller parameters based on the real-time position deviation data to obtain the target sliding mode controller parameters.

[0020] Specifically, the real-time position deviation data is subjected to positioning accuracy calculation and response time statistics, the dynamic performance of the system is evaluated, and the performance evaluation index is obtained through data analysis. The positioning accuracy calculation focuses on the error between the actual position and the expected position of the module when performing motion control tasks, including instantaneous error and cumulative error, while the response time statistics measure the time it takes for the system to reach the target position and stabilize after receiving the control command. These data together constitute the performance evaluation index, reflecting the quality of the control system. Based on these performance evaluation indicators, they are compared with the preset thresholds to determine the optimization direction of the current system parameters and clarify which parameters need to be adjusted to improve the overall control effect. After determining the parameter optimization direction, the adjustment range of each control parameter is quantified to construct a weight allocation matrix. This matrix is ​​used to weight the key control parameters in the optimization direction data, including the sliding surface coefficient, the approach speed coefficient, and the integral coefficient. Since different parameters have different influences on the system performance, a hierarchical weight calculation method is adopted, that is, different optimization weights are set according to the contribution of each parameter to the positioning accuracy and response speed. Through the weight calculation process, the optimization weight matrix is ​​obtained to ensure that the relative influence relationship between different variables in the parameter adjustment process is reasonably distributed, so as to avoid some parameters being adjusted too large or too small during the optimization process, affecting the stability of the system. The optimization weight matrix is ​​matrix multiplied with the parameter optimization direction data to calculate the initial gradient vector. The function of the gradient vector is to indicate the adjustment direction and adjustment amplitude of the current control parameter. By weighted processing of the optimization weight matrix, the gradient calculation can more accurately reflect the optimization requirements of the system. The initial gradient vector is dynamically calculated. The initial gradient vector is grouped according to different error types, including displacement error, velocity error and acceleration error, and the maximum allowable step length of each error group is calculated respectively to obtain adaptive step length data. During the optimization process, the adjustment amplitude of different parameters is dynamically adjusted according to the specific situation of the error, so as to ensure the efficiency and stability of the optimization. The initial gradient vector is segmented linearly mapped based on the adaptive step length data to achieve adaptive optimization. In different error intervals, the sensitivity of the system is different, and different mapping functions are used to adjust the gradient calculation results, so that the adjustment amplitude is larger when the error is large, and the adjustment amplitude is smaller when the error is small. Through piecewise linear mapping, the optimization process can be ensured to converge quickly without causing over-adjustment and resulting in system instability, and the optimized mapping gradient vector is obtained. The mapping gradient vector is orthogonally decomposed to extract the coupling components between the control parameters and construct a decoupling compensation matrix. The function of the decoupling compensation matrix is ​​to eliminate the mutual influence between different parameters, so that the optimization of each parameter can be carried out independently, improving the efficiency and accuracy of the optimization. Through decoupling processing, a decoupled gradient vector is obtained, which can achieve more accurate optimization adjustment while reducing parameter interference.In order to ensure the stability and convergence of the optimization process, the decoupled gradient vector is dynamically attenuated to prevent the system from being unstable due to excessive parameter adjustment during the optimization process. The attenuation coefficient is dynamically adjusted based on the historical number of iterations, so that in the early stage of optimization, the parameter adjustment amplitude is large to accelerate convergence, and in the later stage of optimization, the adjustment amplitude is gradually reduced to improve the stability of the system. Through the attenuation process, the final parameter adjustment amount data is obtained, and it is iterated with the initial sliding mode controller parameters to obtain the final target sliding mode controller parameters.

[0021] In the embodiment of the present invention, a method combining visual measurement and double matching tracking is adopted to achieve high-precision measurement of the motion state of the linear module. Through the design of feature point array and sub-pixel feature extraction algorithm, the limitations of traditional sensors are overcome, and displacement and vibration information are obtained at the same time, thereby improving the measurement accuracy. Based on nonlinear dynamic modeling and feedback linearization method, the nonlinear characteristics of the system are effectively compensated. By establishing a complete dynamic model, the influence of various factors such as inertial force, Coriolis force, friction force, etc. is considered, so that the system can still maintain good linear characteristics when moving in a large range. An adaptive gain sliding mode controller is designed to enhance the anti-interference ability of the system. The continuous reaching law is used instead of the traditional sign function to effectively suppress the chattering caused by control switching. At the same time, the adaptive gain mechanism enables the controller to automatically adapt to load changes. A parameter optimization method based on multi-dimensional performance indicators is proposed to achieve automatic optimization of controller parameters. Through innovative designs such as hierarchical weight calculation, dynamic step size adjustment and decoupling compensation, the coupling problem in the parameter optimization process is solved to ensure continuous improvement of system performance. The seventh-order polynomial trajectory planning and third-order continuity processing are adopted to ensure the smoothness of the motion trajectory. Through reasonable trajectory design, the impact during acceleration and deceleration is reduced, the system vibration is effectively reduced, and the positioning accuracy is improved.

[0022] In a specific embodiment, the process of executing step 101 may specifically include the following steps: Perform Gaussian filtering and noise reduction on the 5×5 feature point array preset on the surface of the moving part in the linear module to obtain filtered image data; Perform Hough transform circular contour detection on the filtered image data to obtain feature point contour data, and perform two-dimensional Gaussian curve fitting on the feature point contour data to obtain feature point center coordinate data; Input the feature point center coordinate data into the global coordinate system conversion matrix for coordinate transformation to obtain displacement coordinate data; Perform time series differentiation operation on the displacement coordinate data to obtain velocity and acceleration data; The displacement coordinate data, velocity and acceleration data are fused with the motor torque, load mass and ambient temperature of the linear module to obtain a motion state data set; The feature points of the motion state data set are tracked to obtain the displacement measurement data and vibration feature data of the linear module.

[0023] Specifically, the image data is preprocessed to improve the accuracy of subsequent feature point extraction. Gaussian filtering is used to reduce the noise of the image. The mathematical expression of Gaussian filtering is: ; in, represents the pixel value of the image after filtering, and are the pixel coordinates of the image, is the standard deviation of the Gaussian filter, which determines the smoothness of the filter. Through this filtering operation, high-frequency noise in the image is removed, making the edges of feature points smoother. Feature point detection is performed on the image data. Since the feature points are circular marks, Hough transform is used for circular contour detection. Hough transform transforms the circular edge points in the image into voting points in the parameter space through parameterized space transformation, and identifies the most likely circular contour by cumulative voting. The circle detection formula of Hough transform is as follows: ; in, are the coordinates of the center of the circle, is the radius of the circle, and is the detected edge point coordinate. By traversing different The value is calculated and the point with the highest number of votes is found in the cumulative space to identify the contour of the feature point in the image. After completing the circular contour detection, in order to improve the accuracy of the center coordinates of the feature points, a two-dimensional Gaussian curve fitting is performed. Since the grayscale distribution of the circular feature points in the image in the local area is approximately a two-dimensional Gaussian distribution, a two-dimensional Gaussian function is used for fitting, and its mathematical expression is: ; in, is the gray value of the pixel, represents the peak intensity, is the precise center coordinate of the feature point, are the standard deviations in the horizontal and vertical directions, respectively, and Represents the background noise intensity. Solve the parameters by the least squares method , obtain the precise center coordinates of the feature points. Convert the center coordinate data of the feature points to the global coordinate system of the linear module so as to match the motion data of the module. Assume that there is a rotation matrix between the imaging coordinate system of the camera and the global coordinate system of the module. and translation vector , then the global coordinates Calculated by the following transformation formula: ; in, is the coordinate of the feature point in the camera coordinate system, for Rotation matrix, for Translation vector. By calibrating the camera parameters, we can get and , thus realizing coordinate transformation. After obtaining the global coordinates of the feature points, calculate the motion parameters of the module. Through time series differential operation, the velocity and acceleration of the module are obtained. The velocity calculation uses the first-order backward difference: ; The acceleration is calculated using second-order backward difference: ; in, For modules in time The displacement of time, is the sampling time interval, and The displacement, velocity and acceleration data are combined with external factors such as motor torque, load mass and ambient temperature. Measured by the current sensor, the load mass and ambient temperature Obtained through sensors. The final motion state dataset is expressed as: ; in, It is a complete motion state data set, including all key variables that affect the motion of the module. In order to obtain the displacement measurement data and vibration characteristic data of the linear module, the motion state data set is tracked for feature points. Feature point tracking uses methods such as Kalman filtering or particle filtering to improve the stability of displacement measurement. The vibration characteristics are extracted through frequency domain analysis, such as using fast Fourier transform to calculate the spectrum of the vibration signal and obtain the main vibration frequency. and amplitude The vibration characteristic data is expressed as: ; in, Represents the module at time Vibration characteristic data at the moment.

[0024] In a specific embodiment, the process of performing feature point tracking on the motion state data set to obtain displacement measurement data and vibration feature data of the linear module may specifically include the following steps: The motion state data set is input into the spatial matching layer, the affine transformation matrix of the feature point array in adjacent image frames is calculated, and the optimal matching pair is selected to obtain the spatial matching displacement data; The spatial matching displacement data is input into the time matching layer to construct the state vector, and the filter state model is obtained. The state vector includes position and velocity components. Based on the spatial matching displacement data and the filter state model, trajectory prediction and smoothing are performed to obtain displacement measurement data; Calculating the vibration amplitude of the displacement measurement data to obtain vibration amplitude data, and extracting the frequency characteristics of the displacement measurement data to obtain vibration frequency data; The vibration amplitude data and the vibration frequency data are feature fused to obtain vibration feature data.

[0025] Specifically, the motion state dataset is input into the spatial matching layer to calculate the affine transformation matrix of the feature point array in adjacent image frames. Since the coordinates of the feature points of the linear module in the continuous image frames will change during the motion process, the affine transformation matrix is ​​calculated to describe these changes. The basic form of the affine transformation matrix is: ; in, and Respectively represent the coordinates of a feature point in the previous frame and the current frame, yes The transformation matrix describes the effects of rotation and scaling, and yes The translation vector represents the global displacement of the feature point. In order to solve the affine transformation matrix, a linear equation system is established through multiple matching point pairs based on the least squares method, and then solved and . After obtaining the affine transformation matrix of all feature points, in order to improve the matching accuracy, the RANSAC (random sampling consensus) algorithm is used to screen the optimal matching pairs, that is, by iteratively randomly extracting feature point pairs and calculating their affine transformation errors, the feature point pairs with the smallest error are screened out to obtain the final spatial matching displacement data. The spatial matching displacement data is input into the temporal matching layer to construct the filter state model. The purpose of temporal matching is to use time series data to establish a stable state estimation model to make motion estimation more accurate. Constructing the state vector ,in, is the state vector, containing the position and speed Two components, and the velocity is obtained by time differentiation of the position data, that is: ; This state vector is input to the Kalman filter for filtering to eliminate measurement noise and improve estimation accuracy. The prediction equation of the Kalman filter is: ; in, is the current state, is the state transfer matrix, is the input matrix, is the control input, is the process noise. The measurement update equation is: ; in, is the measured value, is the measurement matrix, To measure noise. Through Kalman filtering, smoother and more stable displacement data is obtained in a noisy environment. Trajectory prediction and smoothing are performed based on spatial matching displacement data and filter state model. Trajectory prediction estimates future movement trends based on the current state, while smoothing reduces jitter in a short period of time and improves the continuity of measurement data. Trajectory prediction uses a kinematic model: ; in, is the acceleration, calculated by the difference of the velocities: ; In order to reduce the jitter in trajectory prediction, the Savitzky-Golay filter is used for smoothing. This method eliminates high-frequency noise by fitting a polynomial to the data in the local window and obtains smooth displacement measurement data. The vibration amplitude is calculated for the displacement measurement data. The vibration amplitude is obtained by calculating the root mean square value of the displacement: ; in, represents the vibration amplitude, For the The displacement value at a moment, is the average value of displacement, is the total number of sampling points. This value reflects the vibration intensity of the module during movement. The frequency characteristics of the displacement measurement data are extracted to analyze the main frequency components of the module vibration. The displacement signal is converted from the time domain to the frequency domain using fast Fourier transform to calculate the main frequency of the vibration. : ; in, is the spectrum of the displacement signal, and the frequency corresponding to the maximum value is taken as the main vibration frequency. This value reflects the main vibration mode of the module, which helps to determine whether there is resonance or external interference. The vibration amplitude data and vibration frequency data are feature fused to obtain vibration feature data. Comprehensively consider the different dimensions of vibration information to improve the system's ability to judge the vibration state. Fusion is performed using feature vectors: ; in, The final vibration characteristic data includes the amplitude and main frequency information of the vibration. Through this data, the vibration state of the linear module is evaluated and the optimization basis is provided for the control system.

[0026] In a specific embodiment, the process of executing step 102 may specifically include the following steps: The displacement measurement data is subjected to motion equation parameter identification and dynamic modeling to obtain the displacement-force mapping relationship. At the same time, the vibration characteristic data is subjected to frequency domain analysis and damping characteristic extraction to obtain the damping coefficient. The displacement-force mapping relationship and the damping coefficient are combined to construct the nonlinear dynamic model M(p)d 2 p / dt 2 +C(p,dp / dt)(dp / dt)+G(p)+F(dp / dt)+W(t)=u, where M(p) is the inertia matrix, p is the displacement vector, C(p,dp / dt) is the Coriolis force matrix, G(p) is the gravity term, F(dp / dt) is the friction term, W(t) is the external disturbance term, u is the control input, and t is time; Define the state variables of the nonlinear dynamic model and obtain the state vector expression X(t)=[p(t),dp / dt] T , where X(t) is the state vector, p(t) is the displacement component state variable, and dp / dt is the velocity component state variable; The nonlinear term compensation processing is performed on the state vector expression to obtain the feedback linearization control law, which is expressed as: u=M(p)[vM -1 (p)(C(p,dp / dt)(dp / dt)+G(p)+F(dp / dt)+W(t))], where v is the new control input; Substitute the feedback linearization control law into the nonlinear dynamics model to obtain the linearized system expression dX / dt=AX(t)+Bv(t); The system matrix of the linearized system expression is constructed to obtain the standard linear state equation X'(t)=AX(t)+Bv(t)+E(t), where X'(t) is the derivative vector of the state variable, A is the n×n dimensional system matrix, B is the input matrix, E(t) is the disturbance vector, and n is the system order.

[0027] Specifically, using the measured displacement data and externally applied forces Perform dynamic modeling. During the motion of the linear module, its dynamic characteristics are represented as a controlled rigid body system, which includes inertia, Coriolis force, gravity, friction and external disturbance. In order to obtain the dynamic model of the system, the parameter identification method is used to fit the mapping relationship between displacement and force based on the input-output relationship of the system. In the dynamic system, the acceleration of the object The control force exerted And the physical characteristics of the system itself, so it is expressed as: ; in, is the inertia matrix, which represents the mass inertia characteristics of the system and is the displacement Function of is the Coriolis force matrix, describing the inertial coupling effect caused by motion; is the gravity term, which indicates the influence of gravity on the module at different positions; represents the friction term, which is a function of velocity and is expressed as static friction, dynamic friction, or viscous friction; It is an external interference item, including external vibration, airflow resistance, etc. is the control input of the system, that is, the force applied by the drive motor. In order to improve the dynamic model, the vibration characteristic data is analyzed in the frequency domain, and the damping characteristics of the system are extracted to obtain the damping coefficient. Convert to the frequency domain via Fourier transform: ; in is the frequency spectrum of the displacement, is the frequency. By analyzing the main vibration frequency and amplitude of the system, the damping ratio of the system is calculated. : ; in is the damping coefficient, is the equivalent mass, is the stiffness of the system. The damping coefficient is used to compensate for the vibration effect of the system and improve the motion accuracy. After the nonlinear dynamic model is established, the state variables of the system are defined for control design. The state vector is defined as: ; in, is the state vector, containing the displacement component and velocity components The original second-order dynamic equation is converted to a first-order state equation. Since the dynamic model contains nonlinear terms, nonlinear compensation is performed. The feedback linearization method is used to design the control law so that the behavior of the system becomes linear in the transformed coordinate system. The feedback linearization control law is expressed as: ; in, is a new control input used to replace the nonlinear force term in the nonlinear system; Represents the inverse of the inertia matrix, so that the control law can directly act on the acceleration term of the system. Through this feedback linearization control law, the original nonlinear system is converted into a standard linear system, which is expressed as: ; in, It is the system matrix, which determines the inherent dynamic characteristics of the system; is the input matrix, describing the influence of the control input on the system state. In order to finally obtain the standard linear state equation, the system matrix is ​​constructed. The final standard linear state equation is expressed as: ; in, is the derivative vector of the state variables; for dimensional system matrix, describing the intrinsic characteristics of the system; is the input matrix, describing the influence of control input on the system; is the disturbance vector, which includes the influence of the external environment.

[0028] In a specific embodiment, the process of executing step 103 may specifically include the following steps: The displacement tracking error is defined for the standard linear state equation, and the tracking error equation is obtained as e(t)=X(t)-Xd(t), where e(t) is the tracking error vector, X(t) is the state vector, and Xd(t) is the desired state vector; The sliding surface is constructed based on the tracking error equation, and the sliding surface expression s(t)=ce(t)+de(t) / dt is obtained, where s(t) is the sliding surface vector, c is the sliding surface coefficient matrix of the positive definite diagonal matrix, and de(t) / dt is the tracking error derivative vector; The sliding surface expression is designed with a continuous reaching law, and the reaching law equation ds / dt=-η|s|-k∫|s|dt is obtained, where ds / dt is the sliding surface derivative vector, η is the reaching velocity coefficient matrix of the positive definite diagonal matrix, k is the integral coefficient matrix of the positive definite diagonal matrix, and ∫|s|dt is the sliding surface integral vector; Substituting the reaching law equation into the standard linear state equation for solution, we can obtain the sliding mode control law v(t)=vs(t)+vn(t), where v(t) is the control input vector, vs(t) is the equivalent control term vector used to compensate for known dynamics, and vn(t) is the switching control term vector used to suppress unknown disturbances. Based on the sliding mode control law, an adaptive gain update law is designed to obtain the gain parameter dynamic equation dk / dt=γ|s|, where dk / dt is the derivative of the gain matrix, γ is the learning rate coefficient matrix of the positive definite diagonal matrix, and |s| is the absolute value vector of the sliding surface; The sliding mode control law and the gain parameter dynamic equation are combined to obtain the initial sliding mode controller parameters.

[0029] Specifically, we first define the tracking error of the system in order to quantify the deviation between the current state of the system and the desired state. In control theory, the tracking error is expressed as: ; in, is the tracking error vector, which represents the current state of the system and expected state The difference between is the current state vector of the system, and is the desired state vector corresponding to the target trajectory. The error vector is used to guide the adjustment of the controller to ensure that the system can move along the desired trajectory and compensate for the error when there is a disturbance. After defining the error equation, the sliding surface is constructed so that the system remains stable when running on the sliding surface and effectively suppresses the error. The sliding surface is defined by the following expression: ; in, is the sliding surface vector, which represents the state variable of the system in sliding mode control. is the sliding mode surface coefficient matrix of the positive definite diagonal matrix, which is used to adjust the sliding mode dynamic characteristics of the system. is the derivative vector of the tracking error, which describes the rate of change of the error. and Matrix, adjust the sliding mode control performance of the system to ensure that the error can converge quickly and slide stably on the sliding surface. In order to make the system state converge quickly to the sliding surface, the reaching law is designed so that the error can be reduced quickly according to the expected trajectory, thereby realizing robust trajectory tracking control. The design of the reaching law adopts the exponential reaching strategy, and its mathematical expression is: ; in, is the sliding surface derivative vector, which represents the sliding state change rate of the system. is a positive definite diagonal matrix of approaching speed coefficients, which determines the speed at which the system approaches the sliding surface. is the integral coefficient matrix of the positive definite diagonal matrix, the error accumulation influence of the control system, and is the sliding surface integral vector, reflecting the cumulative impact of system errors. The reaching law ensures that the system state can quickly enter the sliding surface and maintain stable motion on the sliding surface, thereby improving the robustness and dynamic response capability of the system. After constructing the reaching law, it is substituted into the standard linear state equation and solved to obtain the sliding mode control law. The form of the standard linear state equation is: ; In order to design a sliding mode controller, a sliding mode control law is constructed so that the system can be controlled according to the sliding mode surface. Therefore, the sliding mode control law is expressed as: ; in, is the control input vector, is the equivalent control term vector, which is used to compensate for the known dynamic characteristics of the system so that the system can move according to the expected trajectory, and is the switching control term vector, which is used to suppress the unknown disturbance of the system and improve the robustness of the system. Determined by the system's dynamics model, the switching control term Use sign function or saturation function to adjust to reduce control jitter and ensure system stability. Design an adaptive gain update law so that the controller can automatically adjust the control gain according to the error. The dynamic adjustment of the gain parameter is expressed as: ; in, represents the derivative of the gain matrix, indicating the rate of change of the system gain parameters, is a positive definite diagonal matrix of learning rate coefficients, which determines the speed of gain adjustment, and is the absolute value vector of the sliding surface, which is used to guide the gain update process. The purpose of adaptive gain update is to increase the control gain when the error is large to improve the convergence speed of the system, and to reduce the gain when the error is small to reduce the control chattering problem. The sliding mode control law is combined with the gain parameter dynamic equation to obtain the initial sliding mode controller parameters.

[0030] In a specific embodiment, the process of executing step 104 may specifically include the following steps: Perform seventh-order polynomial trajectory planning on the positioning task of the linear module to obtain the expected trajectory data of position, velocity and acceleration; Performing third-order continuity processing on the desired trajectory data according to the initial sliding mode controller parameters to obtain smooth trajectory interpolation data; Collect the position and speed of the real-time motion state of the linear module to obtain real-time motion feedback data; The smooth trajectory interpolation data is compared and calculated with the real-time motion feedback data to obtain the real-time position deviation data.

[0031] Specifically, a mathematical model for trajectory planning is established. The goal of seventh-order polynomial trajectory planning is to ensure the smoothness, continuity and feasibility of the trajectory, so that the linear module can move with high precision along the set path during the movement process. The basic form of the seventh-order polynomial trajectory is expressed as: ; in, For in time The expected position at the time, are the trajectory parameters that need to be determined. In order to ensure that the trajectory meets the boundary conditions, the initial position, final position, initial velocity, final velocity, initial acceleration, final acceleration, and higher-order derivatives such as jerk are set so that the trajectory has a high degree of smoothness at each time point. The boundary conditions of trajectory planning are set as: ; in, represent the initial position and the final position respectively, represent the initial velocity and final velocity respectively, denote the initial acceleration and the final acceleration respectively, and is the total duration of trajectory planning. By solving the above equations, the coefficients of the seventh-order polynomial are determined arrive , we get the complete trajectory planning function. We can calculate the expected value of velocity and acceleration by taking derivatives: ; Through this step, the complete expected trajectory data, including location information, is obtained. , speed information And acceleration information , providing an ideal motion path for the motion control of the linear module. The expected trajectory data is processed for third-order continuity to ensure the smoothness of the track edge. Third-order continuity means that the trajectory remains continuous at the three levels of position, velocity and acceleration, that is, the following conditions are met: ; in, is the acceleration, which is used to describe the rate of change of acceleration of the system. In order to ensure the smoothness of the trajectory, the third-order spline interpolation method is used to further optimize the seventh-order polynomial trajectory. The core idea of ​​the spline interpolation method is to construct a third-order polynomial between known data points so that the interpolation curve meets the continuity requirements in each interval, that is: ; in, Indicates The interpolation function of the segment trajectory, are the coefficients that need to be solved, and these coefficients are constrained by boundary conditions to ensure a smooth transition of the trajectory. The final smooth trajectory interpolation data can improve the control accuracy of the system and reduce the impact and vibration caused by sudden changes in the trajectory. At the same time, in order to ensure that the system can accurately execute the planned trajectory, the motion state of the linear module is acquired in real time to collect the position and speed. The position data of the module is measured in real time through high-precision sensors such as grating rulers, laser rangefinders or inertial measurement units. The velocity data is obtained by direct measurement or numerical differentiation, namely: ; in, is the sampling time interval. Through this process, the real-time motion feedback data of the linear module is obtained, thereby providing the control system with information on the actual operating status. After obtaining the smooth trajectory interpolation data and real-time motion feedback data, the data is compared to calculate the real-time position deviation data. The position deviation is calculated as follows: ; in, is the position error, which indicates the difference between the actual measured position and the target trajectory. Similarly, the velocity error is calculated: ; in, Reflects the deviation between the actual speed and the target speed. The calculation results of position deviation and speed deviation are used to adjust the control strategy in real time to ensure that the linear module can strictly run according to the planned trajectory and make appropriate compensation when it is subject to external interference.

[0032] In a specific embodiment, the process of executing step 105 may specifically include the following steps: Perform positioning accuracy calculation and response time statistics on real-time position deviation data to obtain performance evaluation index data, and perform index threshold judgment based on the performance evaluation index data to obtain parameter optimization direction data; The weight distribution matrix is ​​constructed for the parameter optimization direction data, and the sliding surface coefficient, approach speed coefficient and integral coefficient in the parameter optimization direction data are calculated by layered weights to obtain the optimization weight matrix; Perform matrix multiplication operation on the optimization weight matrix and the parameter optimization direction data to obtain the initial gradient vector; Perform dynamic step length calculation on the initial gradient vector, group the initial gradient vector according to displacement error, velocity error, and acceleration error, calculate the maximum allowable step length of each group, and obtain adaptive step length data; Based on the adaptive step size data, the initial gradient vector is piecewise linearly mapped, and different mapping functions are used for different error intervals to obtain the mapped gradient vector; The mapped gradient vector is orthogonally decomposed, the coupling components between the control parameters are extracted, a decoupling compensation matrix is ​​constructed, and a decoupling gradient vector is obtained; The decoupled gradient vector is dynamically attenuated, the attenuation coefficient is dynamically adjusted based on the historical number of iterations, the parameter adjustment data is obtained, and the parameter adjustment data is iteratively calculated with the initial sliding mode controller parameters to obtain the target sliding mode controller parameters.

[0033] Specifically, the real-time position deviation data is used to calculate the positioning accuracy and response time. Positioning accuracy is an important indicator to measure whether the linear module can accurately reach the target position, while the response time determines the speed at which the system reaches a steady state. Positioning accuracy is measured by calculating the root mean square value of the position error, and its expression is as follows: ; in, represents the root mean square error, For the The actual location at a point in time, is the expected position, is the number of sampling points. The smaller the RMS error, the higher the positioning accuracy of the system. Response time It is the time required for the system to enter the steady state from the initial state. By analyzing the error signal of the system, it converges to a certain threshold The time within is calculated as: ; in, is the time when the system starts to move, is the error threshold. By calculating and , obtain the system performance evaluation index data. Compare these data with the preset thresholds to determine whether the current system parameters need to be optimized and determine the direction of parameter optimization. Weight allocation is performed on the parameter optimization direction data to reasonably allocate optimization resources. Weight allocation is to allocate appropriate optimization weights according to the degree of influence of different control parameters on system performance. Set the optimization weight matrix As a diagonal matrix, where each element represents the optimization weight of the corresponding parameter: ; in, represents the optimization weight of the sliding surface coefficient, represents the optimization weight of the approach speed coefficient, Represents the optimization weights of the integral coefficients. These weights are set based on empirical values ​​or online adaptive adjustments to ensure that the optimization process can effectively improve system performance. Optimize direction data with parameters Perform a matrix multiplication to obtain the initial gradient vector: ; in, Represents the initial gradient vector, describing the optimization trend of each control parameter. Dynamic step length calculation is performed on the initial gradient vector to ensure that the optimization process can be reasonably adjusted under different error conditions. The basic principle of step length calculation is to set different optimization step lengths according to different types of errors (such as position error, velocity error, acceleration error). Setting the step length matrix for: ; in, represents the step size corresponding to the position error, represents the step size corresponding to the speed error, Represents the step length corresponding to the acceleration error. The step length is calculated using the maximum allowable error method: ; in, are the maximum allowable errors of position, velocity and acceleration respectively. After the step length is calculated, the initial gradient vector With the step size matrix Perform operations to obtain adaptive step size data: ; in, Represents the gradient vector after adaptive step size adjustment. In order to optimize the parameter update process, Perform piecewise linear mapping to ensure that the parameter adjustment of different error intervals is reasonable to set the error interval And define the mapping function: ; in, is the error mapping function, using piecewise linear mapping, such as: ; in, is the proportionality coefficient, Is the error threshold. Mapping gradient vector More effectively guide parameter updates. After completing the gradient mapping, in order to reduce the mutual interference between parameters, orthogonal decomposition is performed to extract the coupling components between the control parameters and construct a decoupling compensation matrix. : ; in, is the decoupling compensation matrix, which is obtained by principal component analysis or singular value decomposition, so that the updates of different parameters do not affect each other. Perform dynamic attenuation processing to ensure that the optimization process gradually converges in the later stages of iteration. Set the dynamic attenuation factor : ; in, is the decay rate, is the number of iterations. The final parameter adjustment data is: ; and the initial sliding mode controller parameters To update: ; Finally, the target sliding mode controller parameters are obtained.

[0034] The above describes the high-precision positioning control method of the linear module in the embodiment of the present invention. The following describes the high-precision positioning control system of the linear module in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a linear module high-precision positioning control system includes: The feature point tracking module 201 is used to extract and track feature points of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data; The nonlinear compensation module 202 is used to perform nonlinear compensation on the linear module according to the displacement measurement data and the vibration characteristic data, and construct a standard linear state equation; A calculation module 203 is used to perform sliding surface design and gain parameter calculation based on a standard linear state equation to obtain initial sliding mode controller parameters; Feedback module 204, used to perform trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data; The optimization module 205 is used to iteratively optimize the initial sliding mode controller parameters based on the real-time position deviation data to obtain target sliding mode controller parameters.

[0035] Through the cooperation of the above components, the method of combining visual measurement with double matching tracking is used to achieve high-precision measurement of the motion state of the linear module. Through the design of feature point array and sub-pixel feature extraction algorithm, the limitations of traditional sensors are overcome, and displacement and vibration information are obtained at the same time, which improves the measurement accuracy. Based on nonlinear dynamic modeling and feedback linearization method, the nonlinear characteristics of the system are effectively compensated. By establishing a complete dynamic model, the influence of various factors such as inertial force, Coriolis force, friction force, etc. is considered, so that the system can still maintain good linear characteristics when moving in a large range. An adaptive gain sliding mode controller is designed to enhance the anti-interference ability of the system. The continuous reaching law is used instead of the traditional sign function to effectively suppress the chattering caused by control switching. At the same time, the adaptive gain mechanism enables the controller to automatically adapt to load changes. A parameter optimization method based on multi-dimensional performance indicators is proposed to achieve automatic optimization of controller parameters. Through innovative designs such as hierarchical weight calculation, dynamic step size adjustment and decoupling compensation, the coupling problem in the parameter optimization process is solved to ensure continuous improvement of system performance. The seventh-order polynomial trajectory planning and third-order continuity processing are used to ensure the smoothness of the motion trajectory. Through reasonable trajectory design, the impact during acceleration and deceleration is reduced, the system vibration is effectively reduced, and the positioning accuracy is improved.

[0036] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0037] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0038] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A high-precision positioning control method for a linear module, characterized in that: The method comprises: Extract and track the feature points of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data; Perform nonlinear compensation on the linear module according to the displacement measurement data and the vibration characteristic data, and construct a standard linear state equation; Based on the standard linear state equation, sliding surface design and gain parameter calculation are performed to obtain initial sliding mode controller parameters; Perform trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data; The initial sliding mode controller parameters are iteratively optimized based on the real-time position deviation data to obtain target sliding mode controller parameters.

2. The linear module high-precision positioning control method according to claim 1, characterized in that: The feature point extraction and feature point tracking of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data includes: Perform Gaussian filtering and noise reduction on the 5×5 feature point array preset on the surface of the moving part in the linear module to obtain filtered image data; Performing Hough transform circular contour detection on the filtered image data to obtain feature point contour data, and performing two-dimensional Gaussian curve fitting on the feature point contour data to obtain feature point center coordinate data; Input the feature point center coordinate data into the global coordinate system conversion matrix for coordinate transformation to obtain displacement coordinate data; Performing time series differentiation operation on the displacement coordinate data to obtain velocity and acceleration data; The displacement coordinate data, the velocity and acceleration data are fused with the motor torque, load mass and ambient temperature of the linear module to obtain a motion state data set; Feature point tracking is performed on the motion state data set to obtain displacement measurement data and vibration feature data of the linear module.

3. The high-precision positioning control method of a linear module according to claim 2, characterized in that: The tracking of feature points of the motion state data set to obtain displacement measurement data and vibration feature data of the linear module includes: Input the motion state data set into the spatial matching layer, calculate the affine transformation matrix of the feature point array in adjacent image frames, and select the best matching pair to obtain spatial matching displacement data; Inputting the spatial matching displacement data into a time matching layer to construct a state vector to obtain a filter state model, wherein the state vector includes position and velocity components; Perform trajectory prediction and smoothing processing based on the spatial matching displacement data and the filter state model to obtain displacement measurement data; Calculating the vibration amplitude of the displacement measurement data to obtain vibration amplitude data, and extracting frequency features of the displacement measurement data to obtain vibration frequency data; The vibration amplitude data and the vibration frequency data are feature fused to obtain vibration feature data.

4. The linear module high-precision positioning control method according to claim 1, characterized in that: The nonlinear compensation of the linear module is performed according to the displacement measurement data and the vibration characteristic data to construct a standard linear state equation, including: The displacement measurement data is subjected to motion equation parameter identification and dynamic modeling to obtain a displacement-force mapping relationship. At the same time, the vibration characteristic data is subjected to frequency domain analysis and damping characteristic extraction to obtain a damping coefficient. The displacement-force mapping relationship and the damping coefficient are combined to construct a nonlinear dynamic model M(p)d 2 p / dt 2 +C(p,dp / dt)(dp / dt)+G(p)+F(dp / dt)+W(t)=u, where M(p) is the inertia matrix, p is the displacement vector, C(p,dp / dt) is the Coriolis force matrix, G(p) is the gravity term, F(dp / dt) is the friction term, W(t) is the external disturbance term, u is the control input, and t is time; The nonlinear dynamic model is defined by state variables to obtain the state vector expression X(t)=[p(t),dp / dt] T , where X(t) is the state vector, p(t) is the displacement component state variable, and dp / dt is the velocity component state variable; The nonlinear term compensation processing is performed on the state vector expression to obtain the feedback linearization control law, which is expressed as: u=M(p)[vM -1 (p)(C(p,dp / dt)(dp / dt)+G(p)+F(dp / dt)+W(t))], where v is the new control input; Substituting the feedback linearization control law into the nonlinear dynamics model, a linearized system expression dX / dt=AX(t)+Bv(t) is obtained; The linearized system expression is subjected to system matrix construction to obtain the standard linear state equation X'(t)=AX(t)+Bv(t)+E(t), wherein X'(t) is the derivative vector of the state variable, A is the n×n dimensional system matrix, B is the input matrix, E(t) is the disturbance vector, and n is the system order.

5. The high-precision positioning control method of a linear module according to claim 1, characterized in that: The sliding surface design and gain parameter calculation based on the standard linear state equation to obtain the initial sliding mode controller parameters include: The displacement tracking error is defined for the standard linear state equation to obtain the tracking error equation e(t)=X(t)-Xd(t), where e(t) is the tracking error vector, X(t) is the state vector, and Xd(t) is the desired state vector; A sliding surface is constructed based on the tracking error equation to obtain a sliding surface expression s(t)=ce(t)+de(t) / dt, wherein s(t) is a sliding surface vector, c is a sliding surface coefficient matrix of a positive definite diagonal matrix, and de(t) / dt is a tracking error derivative vector; A continuous reaching law design is performed on the sliding surface expression to obtain a reaching law equation ds / dt=-η|s|-k∫|s|dt, wherein ds / dt is a sliding surface derivative vector, η is a reaching velocity coefficient matrix of a positive definite diagonal matrix, k is an integral coefficient matrix of a positive definite diagonal matrix, and ∫|s|dt is a sliding surface integral vector; Substituting the reaching law equation into the standard linear state equation for solution, the sliding mode control law v(t)=vs(t)+vn(t) is obtained, wherein v(t) is the control input vector, vs(t) is the equivalent control term vector used to compensate for known dynamics, and vn(t) is the switching control term vector used to suppress unknown disturbances; Based on the sliding mode control law, an adaptive gain update law is designed to obtain a gain parameter dynamic equation dk / dt=γ|s|, where dk / dt is the gain matrix derivative, γ is the learning rate coefficient matrix of the positive definite diagonal matrix, and |s| is the absolute value vector of the sliding mode surface; The sliding mode control law and the gain parameter dynamic equation are combined to obtain initial sliding mode controller parameters.

6. The linear module high-precision positioning control method according to claim 1, characterized in that: The performing trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data includes: Performing seventh-order polynomial trajectory planning on the positioning task of the linear module to obtain expected trajectory data of position, velocity and acceleration; Performing third-order continuity processing on the desired trajectory data according to the initial sliding mode controller parameters to obtain smooth trajectory interpolation data; Collecting the position and speed of the real-time motion state of the linear module to obtain real-time motion feedback data; The smooth trajectory interpolation data is compared and calculated with the real-time motion feedback data to obtain real-time position deviation data.

7. The linear module high-precision positioning control method according to claim 1, characterized in that: The iterative optimization of the initial sliding mode controller parameters based on the real-time position deviation data to obtain target sliding mode controller parameters includes: Performing positioning accuracy calculation and response time statistics on the real-time position deviation data to obtain performance evaluation index data, and performing index threshold judgment based on the performance evaluation index data to obtain parameter optimization direction data; A weight distribution matrix is ​​constructed for the parameter optimization direction data, and a hierarchical weight calculation is performed on the sliding surface coefficient, the approach speed coefficient, and the integral coefficient in the parameter optimization direction data to obtain an optimization weight matrix; Performing a matrix multiplication operation on the optimization weight matrix and the parameter optimization direction data to obtain an initial gradient vector; Performing dynamic step length calculation on the initial gradient vector, grouping the initial gradient vector according to displacement error, velocity error, and acceleration error, respectively calculating the maximum allowable step length of each group, and obtaining adaptive step length data; Performing piecewise linear mapping on the initial gradient vector based on the adaptive step size data, using different mapping functions for different error intervals, to obtain a mapping gradient vector; Orthogonally decomposing the mapping gradient vector, extracting coupling components between various control parameters, constructing a decoupling compensation matrix, and obtaining a decoupling gradient vector; The decoupling gradient vector is dynamically attenuated, the attenuation coefficient is dynamically adjusted based on the historical number of iterations to obtain parameter adjustment data, and the parameter adjustment data and the initial sliding mode controller parameters are iteratively calculated to obtain target sliding mode controller parameters.

8. A linear module high-precision positioning control system, characterized in that: The system is used to execute the high-precision positioning control method of a linear module according to any one of claims 1 to 7, and comprises: The feature point tracking module is used to extract and track the feature points of the motion image sequence of the linear module to obtain displacement measurement data and vibration feature data; A nonlinear compensation module, used for performing nonlinear compensation on the linear module according to the displacement measurement data and the vibration characteristic data, and constructing a standard linear state equation; A calculation module, used for performing sliding surface design and gain parameter calculation based on the standard linear state equation to obtain initial sliding mode controller parameters; A feedback module, used for performing trajectory interpolation and motion feedback on the linear module according to the initial sliding mode controller parameters to obtain real-time position deviation data; The optimization module is used to iteratively optimize the initial sliding mode controller parameters based on the real-time position deviation data to obtain target sliding mode controller parameters.

Images