Rotary table precision alignment control method based on multi-modal sensor fusion

Abstract

The present application relates to the technical field of mechanical engineering, disclose a kind of based on multi-modal sensor fusion's accurate alignment control method of rotary table, comprising: first obtaining target rotary table multi-modal sensing data and fusion as multi-modal vector, real-time pose is identified;The deviation amount of current and target position is calculated, control stage is determined, and hierarchical control procedure is constructed;Then extract control positioning data in procedure, and the adjustment sequence is obtained by division, and rotary table servo motor driving instruction is generated;After that, the high-frequency disturbance value when instruction is executed is collected, the anti-interference control amount is calculated, and the precision imbalance point is detected;Finally, the precision offset value of imbalance point is identified, and the parameter group is obtained by cooperating with the preset precision index alignment, and alignment control strategy is generated.The present application can improve the control precision and adaptive ability of rotary table in complex application environment.

Description

Technical Field

[0001] This invention relates to a method for precise positioning control of a turntable based on multimodal sensor fusion, belonging to the field of mechanical engineering technology. Background Technology

[0002] Turntables, as high-precision rotary positioning devices, are widely used in industrial manufacturing, aerospace and precision measurement and other fields. Their core function is to achieve precise angular positioning of the object being supported.

[0003] Currently, turntable positioning control mostly relies on a single encoder or photoelectric sensor to obtain position information, and uses preset motion curves or PID control methods for positioning adjustment. However, these methods are difficult to cope with complex working conditions under multi-source disturbances. Due to problems such as single sensor data, insufficient dynamic error compensation, and weak anti-interference ability, the positioning process is prone to overshoot, oscillation, or decreased accuracy. Especially in scenarios such as load changes, external vibration, or thermal deformation, stable and high-precision positioning cannot be achieved. Therefore, a turntable precision positioning control method based on multimodal sensor fusion is needed to improve the control accuracy and adaptability of the turntable in complex application environments. Summary of the Invention

[0004] This invention provides a turntable precision alignment control method based on multimodal sensor fusion, the main purpose of which is to improve the control accuracy and adaptability of the turntable in complex application environments.

[0005] To achieve the above objectives, the present invention provides a method for precise turntable alignment control based on multimodal sensor fusion, comprising:

[0006] Acquire turntable sensing data of the target turntable in a multimodal scenario, perform data fusion on the turntable sensing data to obtain a multimodal fusion vector, and identify the real-time pose state of the target turntable based on the multimodal fusion vector.

[0007] Based on the actual pose state, the position deviation between the current position and the target position of the target turntable is calculated. Based on the position deviation, the control stage corresponding to the target turntable is determined. Based on the control stage, the hierarchical control process corresponding to the target turntable is constructed.

[0008] Extract the control positioning data from the hierarchical control process, adjust and divide the control positioning data to obtain the adjustment control sequence, and generate the drive command corresponding to the turntable servo motor in the target turntable based on the adjustment control sequence.

[0009] Collect high-frequency disturbance values ​​during the execution of the drive command, calculate the anti-disturbance control quantity corresponding to the high-frequency disturbance value, and detect the precision imbalance point in the operation of the target turntable based on the anti-disturbance control quantity;

[0010] Identify the accuracy offset value corresponding to the accuracy imbalance point, coordinate the accuracy offset value with the preset accuracy index to obtain the alignment parameter set, and generate the alignment control strategy corresponding to the target turntable based on the alignment parameter set.

[0011] Optionally, identifying the real-time pose state corresponding to the target turntable based on the multi-mode fusion vector includes:

[0012] Extract the multimodal pose features from the multimodal fusion vector;

[0013] Analyze the sensed pose components associated with the multimodal pose features;

[0014] The sensing pose components are synchronized in real time to obtain synchronized component data;

[0015] Based on the synchronous component data, the current tilt angle and center coordinates of the target turntable are analyzed.

[0016] Based on the tilt angle and the center coordinates, the real-time pose state of the target turntable is identified.

[0017] Optionally, the real-time synchronization of the sensing pose components to obtain synchronized component data includes:

[0018] Analyze the component timing corresponding to the sensing pose components;

[0019] Determine the synchronization time axis corresponding to the component timing sequence;

[0020] Query the data timestamps corresponding to each component in the synchronized timeline;

[0021] The data timestamps are synchronized and aligned to obtain a synchronized dataset.

[0022] The consistent data in the synchronization dataset is extracted as the synchronization component data.

[0023] Optionally, the step of constructing the hierarchical control flow corresponding to the target turntable based on the control stage includes:

[0024] Query the phase division logic in the control phase;

[0025] Based on the stage division logic, the interlayer associated modules between the control stages are determined;

[0026] Generate the topology hierarchical framework corresponding to the inter-layer association module;

[0027] Configure the hierarchical control data corresponding to the hierarchical control framework;

[0028] Based on the hierarchical control data, a hierarchical control process corresponding to the target turntable is constructed.

[0029] Optionally, generating the topology hierarchical framework corresponding to the hierarchical association module includes:

[0030] Parse the hierarchical mapping relationship in the hierarchical association module;

[0031] Based on the hierarchical mapping relationship, identify the associated control unit corresponding to the hierarchical association module;

[0032] Extract the key control nodes from the associated control unit;

[0033] Reconstruct the topology control chain corresponding to the key control node;

[0034] Generate the topology hierarchical framework corresponding to the topology control chain.

[0035] Optionally, adjusting and dividing the control positioning data to obtain an adjustment control sequence includes:

[0036] Parse the control positioning tags in the control positioning data;

[0037] Based on the control positioning tag, query the positioning dominant mode in the preset control mode library;

[0038] Extract the dominant performance indicators corresponding to the positioning dominant mode;

[0039] Based on the dominant performance indicators, the temporal scope corresponding to the control positioning data is determined;

[0040] The temporal domain is adjusted and divided to obtain the adjustment control sequence.

[0041] Optionally, determining the temporal scope corresponding to the control positioning data based on the dominant performance index includes:

[0042] Extract the dynamic parameter sequence from the dominant performance indicators;

[0043] The dynamic parameter sequence is collaboratively mapped with the control positioning data to obtain a mapped dataset.

[0044] Analyze the effective time span corresponding to the mapped dataset;

[0045] Identify the key points of action within the effective time span;

[0046] Based on the key action points, the temporal action domain corresponding to the control positioning data is determined.

[0047] Optionally, calculating the positional deviation between the current position and the target position of the target turntable based on the actual pose state includes:

[0048] Analyze the pose state type corresponding to the actual pose state;

[0049] Based on the pose state type, the current position and target position corresponding to the target turntable are analyzed;

[0050] Divide the current position into position axis components;

[0051] Identify the axis deviation value between the position axis component and the target position;

[0052] Based on the axis deviation value, the positional deviation between the current position and the target position of the target turntable is calculated using the following formula:

[0053] ;

[0054] in, This indicates the positional deviation between the current position and the target position of the target turntable. This indicates the total number of control axes in the target turntable. This indicates the control axis index corresponding to the target turntable. Indicates the first Deviation values ​​on each control axis Indicates the first Weighting coefficients for each control axis.

[0055] Optionally, the step of collecting the high-frequency disturbance value during the execution of the drive command includes:

[0056] Identify the execution time interval corresponding to the driving instruction;

[0057] Extract the driver execution data within the execution time interval;

[0058] Statistically analyze the high-frequency oscillation index corresponding to the high-frequency components in the drive execution data;

[0059] Determine the disturbance amplitude level corresponding to the high-frequency oscillation index;

[0060] Based on the disturbance amplitude level, the high-frequency disturbance value of the drive command during execution is collected;

[0061] The anti-interference control quantity corresponding to the high-frequency disturbance value is calculated using the following formula:

[0062] ;

[0063] in, This represents the disturbance rejection control quantity corresponding to the high-frequency disturbance value. This represents the proportionality coefficient. Indicates the level of disturbance. Indicates the execution time interval. This represents the high-frequency disturbance value.

[0064] Optionally, the step of coordinating the precision offset value with a preset precision index to obtain a set of alignment parameters includes:

[0065] Compare the accuracy difference between the accuracy offset value and the preset accuracy index;

[0066] Based on the precision difference, the alignment relationship set in the preset alignment rule library is traversed;

[0067] Filter out collaborative matching pairs within the alignment relationship set that meet the precision threshold;

[0068] Analyze the collaborative compensation amount of the collaborative matching pair;

[0069] Based on the aforementioned collaborative compensation amount, the precision offset value is aligned with a preset precision index to obtain an alignment parameter set.

[0070] Compared to the problems described in the background technology, this invention, by acquiring turntable sensing data of the target turntable in multimodal scenarios, can overcome the information limitations of a single sensor, covering multi-dimensional state information such as position, vibration, and temperature during turntable operation. This enables comprehensive perception of the turntable's operating conditions, avoiding misjudgments caused by missing single data dimensions, and ensuring the accuracy and stability of turntable alignment control from the source. Based on the actual pose state, this invention calculates the positional deviation between the current position and the target position of the target turntable, accurately quantifying the specific difference between the current and target positions. This provides clear numerical basis for subsequent control decisions, avoiding inaccurate control direction or improper adjustment due to the inability to grasp the degree of deviation, and reducing problems such as overshoot and oscillation caused by inaccurate deviation estimation. This ensures the accuracy of the final turntable alignment control from a critical point. Furthermore, by extracting the control positioning data from the hierarchical control process, this invention can provide a basis for subsequent adjustment and division of the control positioning data and generation of turntable servo motor drives. The drive commands provide precise and specific information, avoiding a lack of clear direction in subsequent control stages due to missing or ambiguous data. This helps improve the long-term stability and accuracy of turntable alignment control. Furthermore, by collecting high-frequency disturbance values ​​during the execution of the drive commands and calculating the corresponding anti-disturbance control quantities, this invention can promptly capture the impact of external interference or motor fluctuations on turntable movement, preventing the accumulation of disturbances that could lead to increased posture deviation. This provides a data foundation for precise anti-disturbance and further strengthens the reliability of alignment control. Finally, by identifying the precision offset value corresponding to the precision imbalance point, this invention can accurately quantify the degree of turntable posture deviation from the target, providing clear numerical basis for subsequent correction actions. This allows for more targeted anti-disturbance or adjustment strategies, avoiding resource waste and efficiency loss caused by indiscriminate adjustments. It also allows for tracing the root cause of the imbalance, providing key references for optimizing turntable control logic and improving long-term operating accuracy, thereby ensuring the stability and accuracy of turntable alignment. Therefore, the turntable precision alignment control method based on multimodal sensor fusion provided by this invention can improve the control accuracy and adaptability of the turntable in complex application environments. Attached Figure Description

[0071] Figure 1 This is a flowchart illustrating a method for precise turntable alignment control based on multimodal sensor fusion, provided in an embodiment of the present invention.

[0072] Figure 2 This is a layered schematic diagram of turntable pose control in a turntable precise alignment control method based on multimodal sensor fusion, provided in an embodiment of the present invention.

[0073] Figure 3 This is a schematic diagram of a module for implementing a turntable precision alignment control system based on multimodal sensor fusion, as provided in an embodiment of the present invention.

[0074] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0075] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0076] This application provides a method for precise turntable alignment control based on multimodal sensor fusion. The executing entity of this method includes, but is not limited to, at least one electronic device configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0077] Reference Figure 1 The diagram shown is a flowchart illustrating a turntable precise alignment control method based on multimodal sensor fusion according to an embodiment of the present invention. In this embodiment, the turntable precise alignment control method based on multimodal sensor fusion includes:

[0078] S1. Acquire turntable sensing data of the target turntable in a multimodal scenario, perform data fusion on the turntable sensing data to obtain a multimodal fusion vector, and identify the real-time pose state corresponding to the target turntable based on the multimodal fusion vector.

[0079] This invention, by acquiring turntable sensing data of the target turntable in multimodal scenarios, can overcome the information limitations of a single sensor, covering multi-dimensional state information such as position, vibration, and temperature during turntable operation, achieving comprehensive perception of turntable operating conditions, avoiding misjudgment of state due to the lack of a single data dimension, and ensuring the accuracy and stability of turntable alignment control from the source.

[0080] The target turntable refers to the core rotating device for achieving precise angle positioning and alignment control. It typically possesses the function of bearing a specific load and responding to control commands to complete rotational movements. It is widely used in fields requiring high-precision positioning, and its performance directly determines the operational accuracy of the object it supports. For example, a satellite attitude simulation turntable in the aerospace field has a designed load capacity of 80 kg, a rated rotation range of 0° to 360°, and an initial positioning accuracy requirement of ±0.005°. It needs to utilize multimodal sensing and control methods to ensure stable maintenance of the target angle during satellite attitude calibration operations, meeting the high-precision requirements of satellite on-orbit attitude simulation. The multimodal scenario refers to the comprehensive environment faced by the target turntable during operation, encompassing various complex working conditions and interference factors. These factors can individually or cumulatively affect the turntable's positioning accuracy, covering different dimensions of influence conditions such as temperature fluctuations, external vibrations, electromagnetic interference, and load changes. For example, a parts inspection turntable in a precision industrial machining scenario operates in an environment with a temperature fluctuation range of -5℃ to 50℃, while also being affected by 10Hz to 25Hz interference from surrounding CNC machine tools. z. Vibration interference with an amplitude of 0.03mm, as well as electromagnetic radiation from the 380V industrial electricity in the workshop, require multimodal sensing to capture these combined effects. The turntable sensing data refers to multidimensional data collected by various types of sensors to reflect the real-time operating status of the target turntable, covering key information such as the turntable's position, attitude, vibration, and temperature, providing the original basis for subsequent data fusion and status identification. For example, a sensing system configured for a high-precision measurement turntable can collect real-time angle data output by a high-precision encoder (resolution 262144 lines / revolution), vibration acceleration data output by a MEMS accelerometer (sampling rate 2000Hz, measurement range ±10g), and turntable shell temperature data output by an infrared temperature sensor (measurement accuracy ±0.2℃). These data together constitute the turntable sensing data set. Optionally, the acquisition of the target turntable sensing data in multimodal scenarios can be achieved through multimodal sensor integration methods, such as using the ROS framework to integrate the data streams of LiDAR, IMU, and encoder to obtain the turntable sensing data.

[0081] Furthermore, by fusing the turntable sensor data, the present invention obtains a multi-mode fusion vector, which can effectively integrate sensor information from different dimensions, break through the limitations of a single sensor in terms of monitoring range, accuracy, or anti-interference capability, avoid the one-sided perception of the turntable status due to the lack or deviation of certain types of data, achieve a more comprehensive and three-dimensional characterization of the turntable's operating status, and ensure the smoothness of the entire control process.

[0082] The multi-mode fusion vector refers to a structured numerical set formed by processing the multi-dimensional sensor raw data of the turntable (such as position, vibration, temperature, motor operating parameters, etc.) after data calibration, feature extraction, and collaborative fusion. It integrates the effective information of each sensor, eliminates data redundancy and single sensor errors, has a unified data format and clear physical meaning, and can centrally and accurately reflect the real-time operating status of the turntable. For example, the multi-mode fusion vector of a turntable can be represented as [45.327°, 0.011mm, 29.6℃, 4.9A], which corresponds to the fused angle, vibration amplitude, shell temperature, and motor operating current parameters, respectively. This provides a unified and high-quality information input for subsequent turntable pose recognition and control strategy formulation. Optionally, the data fusion of the turntable sensor data can be achieved through a multi-sensor Kalman filter algorithm, such as using an extended Kalman filter to fuse the pose data of the inertial measurement unit and the encoder to obtain the multi-mode fusion vector.

[0083] Furthermore, based on the multi-mode fusion vector, the present invention identifies the real-time pose state corresponding to the target turntable, which can break through the limitations of single sensor data, integrate multi-dimensional information related to pose, and fully present the current angle, attitude deviation and operation correlation characteristics of the turntable, avoid pose perception bias caused by one-sided information, and achieve a more comprehensive control over the turntable state to ensure the overall effect of turntable alignment control.

[0084] The real-time pose state refers to the overall state description that comprehensively reflects the current spatial position and attitude of the turntable, which integrates the tilt angle, center coordinates and other related parameters (such as the current rotation angle and attitude stability). It is the core basis for determining whether adjustment is needed in the turntable alignment control. For example, the real-time pose state of a turntable can be described as: current rotation angle 45.032°, X-axis tilt angle 0.08°, Y-axis tilt angle 0.05°, center coordinates (120.53mm, 80.41mm), and attitude fluctuation amplitude ≤0.005° / s. This state clarifies whether the current pose of the turntable meets the positioning accuracy requirement of ±0.01°.

[0085] As an embodiment of the present invention, the step of identifying the real-time pose state corresponding to the target turntable based on the multi-mode fusion vector includes: extracting multi-mode pose features from the multi-mode fusion vector; analyzing the sensing pose components associated with the multi-mode pose features; synchronizing the sensing pose components in real time to obtain synchronization component data; parsing the current tilt angle and center coordinates of the target turntable based on the synchronization component data; and identifying the real-time pose state corresponding to the target turntable based on the tilt angle and the center coordinates.

[0086] The multi-modal pose features refer to the set of key features extracted from the multi-modal fusion vector that directly reflect the pose state of the target turntable. These features encompass core information related to the turntable's angle, position, and attitude, reflecting its real-time position and attitude tendency in space. They form the basis for subsequent pose analysis. For example, the multi-modal pose features of a turntable may include the turntable's current absolute angle deviation (e.g., ±0.02°), the tilt trend of the platform relative to the reference plane (e.g., ±0.01° in the X-axis direction), and the offset feature value of the rotation center (e.g., ±0.015mm). These features collectively point to the core state of the turntable's pose, providing crucial evidence for accurate identification. The sensing pose components refer to the features in the multi-modal pose features that correspond to data collected by different types of sensors and are related to the turntable's pose state. Specific data components directly related to pose are derived from the pose monitoring results of a single sensor, reflecting the local pose information of the turntable from the sensor's perspective. For example, the sensor pose components of a turntable may include the current turntable angle (e.g., 45.032°) acquired by a high-precision encoder, the table tilt data (e.g., 0.12° in the Y-axis direction) acquired by a MEMS gyroscope, and the rotation center offset (e.g., +0.018mm) acquired by a laser displacement sensor. These components carry pose monitoring results from different sensors and require further processing to integrate them into unified information. The synchronous component data refers to the consistent data extracted from the synchronous dataset, which, after screening, forms a pose data set that is time-consistent, reliable, and without contradictions. Time synchronization and data validity directly provide high-quality data support for subsequent analysis of turntable tilt angles and center coordinates. For example, the synchronization component data of a certain turntable, under the node "1695210800.002ms", contains consistent data such as "angle 45.031°, tilt 0.119°, offset 0.017mm", and all data timestamps are consistent and the deviations meet the requirements, which can be directly used for pose state recognition. The tilt angle refers to the degree of tilt of the target turntable's bearing surface relative to the preset horizontal reference plane (or designated reference plane) in different axes of space (such as the X-axis and Y-axis). It is quantified by angle values ​​and reflects whether the platform is in a horizontal or preset tilt state. It is an important indicator affecting the positioning accuracy of the turntable. For example, a certain precision The precision measurement turntable is designed to maintain a horizontal platform (tilt angle of 0°). In actual operation, analysis reveals a tilt angle of 0.08° along the X-axis and 0.05° along the Y-axis. This angle directly reflects the platform's attitude deviation; if it exceeds 0.1°, attitude correction must be initiated. The center coordinates refer to the specific coordinates of the target turntable's rotation center in a preset three-dimensional coordinate system (or planar coordinate system). This quantifies the position of the rotation center in space, reflecting the spatial positioning of the turntable's rotation axis. It is a key parameter for determining whether the turntable is in a normal rotation reference state. For example, if the preset rotation center coordinates of a turntable in a planar coordinate system are (120.5mm, 80.4mm), analysis reveals the current center coordinates to be (120.53mm, 80.4mm).(41mm), with coordinate deviation controlled within ±0.02mm, ensuring that the object carried by the turntable can rotate around the preset center and avoiding positioning deviation due to center offset.

[0087] Furthermore, the extraction of multi-modal pose features from the multi-modal fusion vector can be achieved through principal component analysis (PCA), such as using PCA to reduce the dimensionality of multi-source sensor fusion data to obtain multi-modal pose features; the analysis of the sensor pose components associated with the multi-modal pose features can be achieved through blind source separation algorithms, such as using Independent Component Analysis (ICA) to decouple the coupled sensor signal components to obtain sensor pose components; the real-time synchronization of the sensor pose components can be achieved through timestamp alignment methods, such as using the Precision Time Protocol (PTP) to synchronize multi-source sensor data streams to obtain synchronized component data; the... The current tilt angle of the target turntable can be analyzed using a Kalman filter algorithm, such as using a complementary filter to fuse accelerometer and gyroscope data to calculate the tilt angle. The current center coordinates of the target turntable can be analyzed using a least squares estimation method, such as using the Levenberg-Marquardt algorithm to optimize the spatial coordinates of visual markers to obtain the center coordinates. The real-time pose state of the target turntable can be identified using a state machine modeling method, such as constructing a finite state machine model and combining it with threshold judgment logic to process pose data to obtain the real-time pose state.

[0088] As another embodiment of the present invention, the real-time synchronization of the sensing pose components to obtain synchronized component data includes: parsing the component time series corresponding to the sensing pose components; determining the synchronization time axis corresponding to the component time series; querying the data timestamps corresponding to each component in the synchronization time axis; synchronizing and aligning the data timestamps to obtain a synchronization dataset; and extracting consistent data from the synchronization dataset as synchronized component data.

[0089] The component timing sequence refers to the dynamic data sequence of different sensor pose components changing over time. It records the temporal order and corresponding values ​​of pose information collected by each sensor, reflecting the changing patterns of the sensor pose components in the time dimension. This serves as the basis for subsequent synchronization processing. For example, in the component timing sequence of a turntable, the encoder-collected angle data timing sequence is "1695210800.001ms: 45.030°, 1695210800.003ms: 45.031°", and the gyroscope-collected tilt data timing sequence is "1695210800.002ms: 0.118°, 1695210800.004ms: 0.120°", clearly presenting... The two types of data are correlated in terms of time and value. The synchronization timeline refers to a standard time frame constructed to unify the time reference of each sensor pose component. It sets fixed time intervals and key time nodes to align the time dimensions of different components and eliminate time deviations in the data acquisition of each sensor. For example, for the synchronization requirements of a turntable, the constructed synchronization timeline uses 1ms as the time interval and marks key time points such as "1695210800.001ms, 1695210800.002ms, 1695210800.003ms...". The time series data of all sensor pose components must correspond to the nodes of this timeline. The data timestamp refers to the timestamp for each sensor pose component. The unique time stamp that identifies the data acquisition time is usually presented in the form of "year-month-day hour:minute:second, millisecond" or a timestamp value. It is used to accurately locate the data acquisition time and is a core identifier for determining whether the data is synchronized. For example, if a turntable encoder acquires angle data of "45.031°" at 1695210800.002ms, then "1695210800.002ms" is the timestamp of that data. If a gyroscope acquires tilt data of "0.119°" at the same timestamp, the consistency of the timestamps indicates that the data is from the same period. The synchronized dataset refers to the original data of each sensor pose component, matched with the synchronization time axis based on the data timestamp. The dataset formed after matching and alignment ensures that the pose data of different sensors correspond at the same time node, eliminating the data misalignment problem caused by time delay. For example, at the "1695210800.002ms" node of the synchronization time axis, the synchronization dataset integrates the "45.031°" angle data of the encoder, the "0.119°" tilt data of the gyroscope, and the "0.017mm" offset data of the laser displacement sensor, forming a multi-dimensional pose data set at the same moment. The consistent data refers to the fact that in the synchronization dataset, the data collected by different sensors for the same pose dimension (such as angle and tilt) of the turntable, the deviation between them is within a preset reasonable range (such as angle deviation ≤ 0).Data with no logical contradictions (e.g., 0.001°) accurately reflects the actual position and posture of the turntable, eliminating interference from abnormal or erroneous data. For example, in a synchronized dataset, if the encoder and laser angle sensor show angles of "45.031°" and "45.0315°" at time "1695210800.002ms" respectively, with a deviation of 0.0005° ≤ 0.001°, then both sets of data are considered consistent.

[0090] Furthermore, the analysis of the component time series corresponding to the sensing pose components can be achieved through time series analysis methods, such as using Fourier transform to extract the frequency domain features of the sensor data to obtain the component time series; the determination of the synchronization time axis corresponding to the component time series can be achieved through dynamic time warping algorithms, such as using the DTW algorithm to align sensor time series data with different sampling rates to obtain the synchronization time axis; the query of the data timestamps corresponding to each component in the synchronization time axis can be achieved through timestamp extraction methods, such as using the Linux system clock function gettimeofday() to obtain time stamps accurate to microseconds to obtain the data timestamps; the synchronization alignment of the data timestamps can be achieved through interpolation synchronization methods, such as using a cubic spline interpolation algorithm to resample the asynchronously acquired data points on the time axis to obtain the synchronization dataset; the extraction of consistent data from the synchronization dataset as synchronization component data can be achieved through data consistency testing methods, such as using a sliding window variance detection algorithm to remove abnormal data points and retain stable data segments to obtain the synchronization component data.

[0091] S2. Based on the actual pose state, calculate the position deviation between the current position and the target position of the target turntable. Based on the position deviation, determine the control stage corresponding to the target turntable. Based on the control stage, construct the hierarchical control process corresponding to the target turntable.

[0092] Based on the actual pose state, this invention calculates the positional deviation between the current position and the target position of the target turntable. It can accurately quantify the specific difference between the current position and the target position, providing a clear numerical basis for subsequent control decisions. This avoids inaccurate control direction or improper adjustment due to the inability to grasp the degree of deviation, and reduces problems such as overshoot and oscillation caused by inaccurate deviation estimation. It ensures the accuracy of the final alignment control of the turntable from a key point.

[0093] The current position refers to the set of specific parameters reflecting the turntable's current spatial positioning in the real-time pose state derived from multi-mode fusion vector analysis. This includes the turntable's current rotation angle, rotation center coordinates, and positional relationship relative to the reference plane. It represents the actual spatial position of the turntable at a given moment, directly reflecting the real-time result of pose recognition. For example, at 1695210800.5ms, the current position of a precision inspection turntable, obtained through synchronous component data analysis, is: rotation angle 45.032°, rotation center coordinates (120.53mm, 80.41mm), and vertical distance relative to the reference platform 200.05mm. These parameters together constitute the turntable's actual spatial positioning at this moment. The target position refers to the set of spatial positioning parameters preset according to the turntable's operational requirements (such as part machining positioning, satellite attitude simulation calibration), which is expected to be achieved by the turntable. This includes the target rotation angle, target rotation center coordinates, and target attitude parameters. It serves as the benchmark and final target for turntable alignment control, directly determining the direction and accuracy of the control process. For example, the aforementioned precision inspection turntable needs to complete the inspection of the center hole of a certain part. The preset target position is: rotation angle 45.000°, rotation center coordinates (120.50mm, 80.40mm), and vertical distance relative to the reference stage 200.00mm. This position is the core basis for judging whether the turntable has completed precise alignment. The position deviation refers to the axis deviation value of all position axis components, calculated by a preset algorithm (such as Euclidean distance formula, weighted summation method), reflecting the current position of the turntable. The quantitative indicators of the overall deviation from the target position are the key basis for judging whether the turntable needs adjustment and what control strategy to choose. For example, the turntable mentioned above uses the Euclidean distance method to calculate the position deviation. If the units of the X and Y axes are uniformly mm, and the Z axis deviation is converted into a linear deviation (assuming a rotation radius of 100 mm, 0.032° corresponds to a linear deviation of about 0.056 mm), then the deviation is √[(0.03)²+(0.01)²+(0.056)²]≈0.065 mm, which intuitively reflects the overall deviation.

[0094] As an embodiment of the present invention, the step of calculating the position deviation between the current position and the target position of the target turntable based on the actual pose state includes: analyzing the pose state type corresponding to the actual pose state; parsing the current position and the target position of the target turntable based on the pose state type; dividing the position axis component corresponding to the current position; identifying the axis deviation value between the position axis component and the target position; and calculating the position deviation between the current position and the target position based on the axis deviation value.

[0095] The pose state type refers to the specific category of the real-time pose state of the target turntable based on its operational requirements, motion characteristics, and pose performance. Different types correspond to different operating modes and control priorities of the turntable, directly determining the subsequent methods for analyzing the current position and calculating the deviation. For example, a satellite attitude simulation turntable in the aerospace field can be divided into "static positioning type" (such as simulating the geostationary orbit attitude of a satellite, with the turntable angle fixed at 0°) and "dynamic tracking type" (such as simulating satellite orbit change, with the turntable rotating at an angular velocity of 30° / s). The pose analysis logic and deviation calculation dimensions of the two types are significantly different. The position axis component refers to the decomposition of the current position of the target turntable into independent position parameters along different axes according to a preset coordinate system (such as a three-dimensional rectangular coordinate system or a polar coordinate system). Each component corresponds to the spatial positioning information of the turntable along a single axis, which is the basis for subsequent calculation of single-axis deviation and integration of overall deviation. For example, a precision machining turntable in the industry uses a two-dimensional planar coordinate system (X-axis, Y-axis). The positioning system uses a rotary axis (Z-axis) and its current position (X: 120.53mm, Y: 80.41mm, Z: 45.032°) to be decomposed into three independent position axis components: X-axis component 120.53mm, Y-axis component 80.41mm, and Z-axis component 45.032°. The axis deviation value refers to the difference between each position axis component and the corresponding axial component of the target position, quantifying the degree to which the turntable deviates from the target position along a single axis. This deviation can be positive or negative (positive indicates deviation beyond the target position). The target value (a negative value indicates that the target value has not been reached) is the core unit constituting the overall positional deviation. For example, if the target position of the turntable mentioned above is (X: 120.50mm, Y: 80.40mm, Z: 45.000°), then the X-axis deviation is 120.53mm-120.50mm=+0.03mm, the Y-axis deviation is 80.41mm-80.40mm=+0.01mm, and the Z-axis deviation is 45.032°-45.000°=+0.032°.

[0096] Furthermore, the analysis of the pose state type corresponding to the actual pose state can be achieved through pattern recognition algorithms, such as using a Support Vector Machine (SVM) classifier to classify the pose data to obtain the pose state type; the parsing of the current position corresponding to the target turntable can be achieved through visual SLAM methods, such as using an ORB-SLAM3 system to calculate the spatial coordinates of the turntable in the environment in real time to obtain the current position; the parsing of the target position corresponding to the target turntable can be achieved through path planning algorithms, such as using the A* algorithm to calculate the endpoint coordinates of the turntable's motion trajectory to obtain the target position; the division of the position axis components corresponding to the current position can be achieved through coordinate system projection methods, such as using a three-dimensional coordinate transformation matrix to decompose the global coordinates into the motion axis vectors of the turntable to obtain the position axis components; the identification of the axis deviation value between the position axis component and the target position can be achieved through an error calculation model, such as using the Euclidean distance formula to calculate the difference between the current position and the target position axis by axis to obtain the axis deviation value; the calculation of the position deviation between the current position and the target position of the target turntable can be achieved through the following formula.

[0097] In another embodiment of the present invention, the positional deviation between the current position and the target position of the target turntable is calculated based on the axis deviation value using the following formula:

[0098] ;

[0099] in, This indicates the positional deviation (in mm) between the current position and the target position of the target turntable. This indicates the total number of control axes in the target turntable. This indicates the control axis index corresponding to the target turntable. Indicates the first Deviation values ​​on each control axis (unit: mm). Indicates the first Weighting coefficients for each control axis.

[0100] In detail, the position deviation can represent the overall degree of deviation of the current position of the target turntable from the target position. It is a quantitative value obtained by weighting, averaging, and then taking the square root of the deviations of each control axis. It can comprehensively reflect the position accuracy under multi-axis coordination. For example, for a three-control-axis turntable (n=3, including the X-axis, Y-axis, and rotation axis), the X-axis deviation... Weight (X-axis positioning accuracy requirements are higher), Y-axis deviation Weight After the rotation axis deviation is converted into a linear deviation (Corresponding angular deviation 0.028°, deviation 2mm per degree), weighting Substituting into the formula, we get This visually reflects the degree to which the overall position of the turntable deviates from the target. The control axis represents an independent motion axis that enables the target turntable to adjust its posture. It is the core dimension constituting the turntable's spatial posture. Each control axis corresponds to a motion or positioning direction (such as the X-axis, Y-axis, rotation axis, etc.), and n is the total number of control axes. For example, in the three-axis turntable mentioned above, the X-axis controls horizontal lateral displacement, the Y-axis controls horizontal longitudinal displacement, and the rotation axis controls the surface angle. The three axes together determine the turntable's spatial position, therefore n=3. The deviation value represents the difference between the current position and the target position of the turntable on the i-th control axis, quantifying the degree of deviation in a single axis direction. If it is a rotation axis, it needs to be converted to a unit consistent with the linear axis (such as mm). For example, the current position of the turntable on the X-axis is 120.53 mm, and the target position is 120.50 mm. The target angle of the rotation axis is 45.000°, and the current angle is 45.028°. Since the rotation radius is 100mm, the arc length (linear deviation) corresponding to an angle deviation of 0.028° is 2π × 100 × ≈0.0489mm, approximately taken This is used as the unit of linear deviation; the weighting coefficient represents the importance of the i-th control axis in the overall pose accuracy of the turntable. It is set according to the axis's positioning accuracy requirements and the degree of dependence of the operation on that axis. The larger the weight, the stronger the influence of the axis's deviation on the overall positional deviation. For example, the aforementioned turntable X-axis is responsible for the lateral positioning of key holes in the part, which requires high accuracy. The Y-axis positioning requirement is secondary, therefore The rotation axis mainly assists in angular alignment, and its direct impact on the final positional deviation is relatively weak. The importance of each axis is reflected by weighting.

[0101] Based on the position deviation, this invention determines the control stage corresponding to the target turntable, which can accurately match the adaptability of different deviation levels and control strategies, avoiding inefficiency or insufficient accuracy caused by a uniform control method; it can clearly divide stages such as coarse adjustment and fine adjustment, making the control target of each stage more focused, and ensuring that the turntable can quickly and accurately complete the pose adjustment.

[0102] The control phase refers to the operational phases divided for the target turntable based on the magnitude of the position deviation, corresponding to different control strategies and accuracy requirements. Different phases are matched with targeted control parameters (such as driving force, adjustment step size, etc.) to efficiently and accurately complete the pose adjustment, avoiding the inefficiency of a single control strategy across the entire deviation range. For example, when the position deviation of a turntable is greater than 0.5mm, it is in the "coarse adjustment phase," using a fast adjustment strategy with large driving force and large step size; when the deviation is between 0.05mm and 0.5mm, it is the "transitional adjustment phase," reducing the driving force and decreasing the step size; when the deviation is less than 0.05mm, it is the "fine adjustment phase," achieving micron-level precise alignment with minimal driving force and step size. Optionally, the control phase corresponding to the target turntable can be determined using a finite state machine (FSM) modeling method, such as using an FSM framework to define the state transition conditions for turntable start-up, operation, and stop, thereby obtaining the control phase.

[0103] Furthermore, based on the control stage, the present invention constructs a hierarchical control process corresponding to the target turntable, which enables precise matching of different deviation levels with appropriate control strategies, avoiding the inefficiency of a single strategy across the entire deviation range; it can also optimize the dynamic allocation of control resources, ensuring both convergence efficiency under large deviations and alignment accuracy under small deviations, thereby improving the overall accuracy and efficiency of turntable pose adjustment.

[0104] The layered control process refers to a turntable control process based on a topological layered framework and layered control data, which is executed in stages and operates collaboratively in each stage. Through the sequential or collaborative action of each stage, precise adjustment from the current pose to the target pose is achieved. For example, the turntable layered control process is as follows: when the position deviation is >0.5mm, coarse adjustment is entered, with a 10N driving force and a step size of 0.2mm for rapid adjustment; when the deviation drops to 0.5mm, the process switches to transition adjustment through the interlayer correlation module, with a 7N driving force and a step size of 0.05mm for continued adjustment; when the deviation is <0.05mm, the process switches to fine adjustment, with a 3N driving force and a step size of 0.005mm to complete micron-level alignment. The stages are connected in an orderly manner to complete the control.

[0105] As an embodiment of the present invention, the step of constructing the hierarchical control flow corresponding to the target turntable based on the control stage includes: querying the stage division logic in the control stage; determining the inter-layer association modules between the control stages based on the stage division logic; generating the topology hierarchical framework corresponding to the inter-layer association modules; configuring the hierarchical control data corresponding to the hierarchical control framework; and constructing the hierarchical control flow corresponding to the target turntable based on the hierarchical control data.

[0106] The aforementioned stage division logic refers to a rule system that classifies and defines control stages based on core indicators such as position deviation. It clarifies the triggering conditions, deviation range, and core control objectives for each control stage, serving as the foundation for hierarchical control. For example, a turntable might use 0.5mm and 0.05mm as position deviation thresholds to divide the stage into three phases: "coarse adjustment (deviation > 0.5mm, target rapidly converges to the deviation), transition adjustment (0.05mm - 0.5mm, target smoothly transitions to adjustment), and fine adjustment (deviation < 0.05mm, target precisely aligned at the micrometer level)." This series of classification rules constitutes the stage division logic. The inter-layer correlation module refers to the functional module responsible for stage switching, data transmission, and control strategy connection between different control stages. It ensures smooth transition and coordinated operation of control processes at each stage, avoiding control gaps between stages. For example, in the inter-layer correlation module for a turntable transitioning from "coarse adjustment" to "transition adjustment," when the position deviation drops to 0.5mm, it triggers parameter switching logic of "driving force attenuation by 30%, adjustment step size reduced to half of the original," while simultaneously transmitting real-time data of the current pose to ensure control continuity after the transition. The topology layering framework refers to the integration of all control stages, associated control units, and corresponding topologies. The control chain, presented as a visually hierarchical structure, provides an overall control framework that intuitively displays the location of each control layer (coarse adjustment, transition adjustment, and fine adjustment), the connection relationships of the topology control chains, and the data interaction paths. For example, the turntable's topology hierarchical framework is presented horizontally as "coarse adjustment layer → transition adjustment layer → fine adjustment layer," and vertically connected to each layer by two topology control chains (coarse adjustment-transition adjustment chain, transition adjustment-fine adjustment chain). Key control nodes of each chain are also labeled (such as the 0.5mm threshold node and the 7N parameter node), forming a three-in-one structured framework of "layer-chain-node," providing an intuitive execution basis for the hierarchical control process. Layered control data refers to the set of parameters configured for each control stage and inter-layer interconnected modules in a layered control framework, supporting the operation of control logic. It covers control parameters (such as driving force and adjustment step size), stage switching thresholds, data interaction formats, etc. for each stage. For example, for layered control of a turntable, the coarse adjustment layer is configured with "driving force 10N, adjustment step size 0.2mm, switching threshold 0.5mm", the transition adjustment layer is configured with "driving force 7N, adjustment step size 0.05mm, switching threshold 0.05mm", and the fine adjustment layer is configured with "driving force 3N, adjustment step size 0.005mm". These parameters together constitute the layered control data.

[0107] Furthermore, the querying of the stage division logic in the control stage can be implemented through state transition analysis methods, such as using Petri net modeling tools to analyze the conditional triggering relationships between stages to obtain the stage division logic; the determination of the inter-layer association modules between the control stages can be implemented through interface mapping methods, such as using the SysML system modeling language to define the data exchange interface specifications between each level to obtain the inter-layer association modules; the generation of the topology hierarchical framework corresponding to the inter-layer association modules can be implemented through hierarchical modeling algorithms, such as using a directed acyclic graph (DAG) to construct the dependency topology structure of the control layer to obtain the topology hierarchical framework; the configuration of the hierarchical control data corresponding to the hierarchical control framework can be implemented through parameterized configuration methods, such as using XML configuration files to define the control parameters and constraints of each level to obtain the hierarchical control data; the construction of the hierarchical control flow corresponding to the target turntable can be implemented through a workflow engine, such as using the Activiti workflow engine to arrange the execution order of multi-level control instructions to obtain the hierarchical control flow.

[0108] As another embodiment of the present invention, generating the topology hierarchical framework corresponding to the hierarchical association module includes: parsing the hierarchical mapping relationship in the hierarchical association module; identifying the associated control unit corresponding to the hierarchical association module based on the hierarchical mapping relationship; extracting the key control nodes in the associated control unit; reconstructing the topology control chain corresponding to the key control node; and generating the topology hierarchical framework corresponding to the topology control chain.

[0109] The hierarchical mapping relationship refers to the corresponding connection rules between inter-layer associated modules and different control stages. It clarifies the upper and lower level control stages, data interaction directions, and triggering conditions that a certain associated module needs to interface with, forming the logical foundation for constructing the topology. For example, in turntable control, the "coarse adjustment-transition adjustment" associated module (M1) maps "coarse adjustment stage (deviation > 0.5mm)" to the upper level and "transition adjustment stage (0.05mm - 0.5mm)" to the lower level, with the triggering condition being that the position deviation decreases to 0.5mm; the "transition adjustment-fine adjustment" associated module (M2) maps "transition adjustment stage" to the upper level and "fine adjustment stage (deviation < 0.05mm)" to the lower level, with the triggering condition being that the position deviation decreases to 0.5mm; the "transition adjustment-fine adjustment" associated module (M2) maps "transition adjustment stage" to the upper level and "fine adjustment stage (deviation < 0.05mm)" to the lower level. At the next level, the trigger condition is that the deviation drops to 0.05mm. These two sets of corresponding rules constitute the hierarchical mapping relationship. The associated control unit refers to the specific functional unit that realizes the connection between specific upper and lower level control stages according to the hierarchical mapping relationship. It integrates core logic such as deviation monitoring, parameter conversion, and command transmission, and is the execution carrier for stage switching. For example, the associated control unit corresponding to the "coarse adjustment-transition adjustment" mapping relationship includes a deviation real-time monitoring chip with a sampling rate of 1kHz (used to capture whether the deviation drops to 0.5mm), a parameter conversion module (converting the 10N driving force of coarse adjustment into the 7N driving force of transition adjustment), and a command synchronization module (sending switching commands to the turntable servo system). When the triggering conditions are met, the unit can automatically complete the two-stage control transition. The key control nodes refer to the core components in the associated control unit that determine stage switching, data processing, or command output. They are the core support points of the topology control chain and directly affect the accuracy and timeliness of the control transition. For example, in the aforementioned "coarse adjustment-transition adjustment" associated control unit, key control nodes include: a deviation threshold comparison node (preset 0.5mm threshold to determine if the deviation meets the standard), a control parameter storage node (pre-stores the 7N driving force and 0.05mm adjustment step size parameters for the transition adjustment stage), and a command sending node (sends a parameter switching command to the servo motor after the deviation meets the standard). Missing or faulty points can cause phase switching to fail. The topology control chain refers to a chain structure formed by connecting key control nodes corresponding to the same associated module in the logical order of "data input → processing → output". It clearly shows the transmission path of control signals and data between nodes. For example, the topology control chain of the "coarse adjustment-transition adjustment" associated module is: coarse adjustment layer data output node → deviation monitoring node (sampling deviation value) → threshold comparison node (comparing with 0.5mm) → parameter call node (retrieving transition adjustment parameters) → instruction generation node (generating switching instructions) → transition adjustment layer data receiving node. Each node transmits information in sequence to ensure that the phase switching process is executed in an orderly manner.

[0110] Furthermore, the parsing of the hierarchical mapping relationships in the hierarchical association modules can be achieved through graph theory analysis methods, such as using adjacency matrix modeling tools to parse the bidirectional dependencies between levels, thereby obtaining the hierarchical mapping relationships; the identification of the associated control units corresponding to the hierarchical association modules can be achieved through modular decomposition methods, such as using UML component graph tools to identify control unit clusters with independent functions, thereby obtaining the associated control units; the extraction of key control nodes in the associated control units can be achieved through centrality measurement algorithms, such as using the PageRank algorithm to calculate the influence weight of nodes in the control network, thereby obtaining the key control nodes; the reconstruction of the topology control chain corresponding to the key control nodes can be achieved through path optimization algorithms, such as using the Dijkstra algorithm to reconstruct the shortest control path between key nodes, thereby obtaining the topology control chain; the generation of the topology hierarchical framework corresponding to the topology control chain can be achieved through hierarchical topology sorting methods, such as using the Kahn algorithm to generate a hierarchical topology structure based on control chain dependencies, thereby obtaining the topology hierarchical framework.

[0111] Specifically, for a more intuitive understanding of the execution logic and data flow relationships corresponding to the target turntable pose control in this solution, please refer to [reference needed]. Figure 2 The layered diagram of turntable pose control, Figure 2 As the core process framework of the turntable pose control system, it clearly presents the complete link from actual pose input to hierarchical control process output: The input layer focuses on the actual pose state of the turntable, which is the basis for subsequent deviation calculation and control strategy generation; the processing layer completes position deviation quantification, control stage matching, topology hierarchical framework and process generation in sequence through the hierarchical logic of "data preparation and deviation calculation → control stage analysis → process construction and output", transforming pose data into an executable hierarchical control strategy; the output layer takes the "hierarchical control process" as the final result. It should be noted that the connection between each link in the flowchart is essentially an abstract refinement of the turntable pose control logic. In actual scenarios, the complexity of pose analysis (such as the real-time requirements of multi-mode fusion vector analysis) and the diversity of stage adaptation (different deviation amounts correspond to differentiated control parameter rules) are far greater than what is shown in the diagram. This architecture is only a concise display of the core logic to provide an intuitive reference for understanding the systematic thinking of turntable pose control.

[0112] S3. Extract the control positioning data from the hierarchical control process, adjust and divide the control positioning data to obtain an adjustment control sequence, and generate the drive command corresponding to the turntable servo motor in the target turntable based on the adjustment control sequence.

[0113] This invention extracts control positioning data from the hierarchical control process, providing accurate and specific information for subsequent adjustment and division of control positioning data and generation of turntable servo motor drive commands. This avoids the lack of clear direction in subsequent control links due to missing or ambiguous data, and helps improve the long-term stability and accuracy of turntable positioning control.

[0114] The control positioning data refers to the set of key data directly related to positioning control extracted from each stage of the turntable's layered control process. It covers control parameters (such as driving force and adjustment step size), real-time positioning results (such as current position coordinates), and deviation change data for each control stage. This data is the core basis for subsequently dividing the adjustment control sequence and generating servo motor drive commands. For example, in a turntable's layered control, the coarse adjustment stage extracts "driving force 10N, adjustment step size 0.2mm, current X-axis position 120.4mm, deviation 0.4mm," the transition adjustment stage extracts "driving force 7N, step size 0.05mm, X-axis position 120.52mm, deviation 0.02mm," and the fine adjustment stage extracts "driving force 3N, step size 0.005mm, X-axis position 120.50mm, deviation 0.001mm." These data collectively constitute the control positioning data. Optionally, the extraction of control positioning data from the layered control process can be achieved through data mining methods, such as using association rule mining algorithms to extract the position parameters of each control node in the process, thereby obtaining the control positioning data.

[0115] Furthermore, by adjusting and dividing the control positioning data, the present invention obtains an adjustment control sequence, which can transform the scattered positioning data into an ordered and structured set of control steps. This provides a clear and executable basis for the subsequent generation of servo motor drive commands, avoids disordered or deviated drive command generation due to data clutter, and improves the precision of turntable alignment.

[0116] The aforementioned adjustment and control sequence refers to an ordered set of control steps formed by segmenting the time domain according to the dominant performance indicators and decomposing the actions. Each step includes specific adjustment parameters (such as driving force and step size), execution time point, and target deviation, which can directly guide the servo motor action. For example, the adjustment and control sequence of the above transition adjustment stage is: "1695210800.0s: driving force 7N, step size 0.05mm, target deviation 0.04mm; 1695210802.0s: driving force 6.5N, step size 0.03mm, target deviation 0.03mm; 1695210805.0s: driving force 6N, step size 0.02mm, target deviation 0.02mm". The steps are ordered and meet the performance indicator requirements.

[0117] As an embodiment of the present invention, the step of adjusting and dividing the control positioning data to obtain an adjustment control sequence includes: parsing the control positioning tags in the control positioning data; querying the positioning dominant mode in the preset control mode library according to the control positioning tags; extracting the dominant performance index corresponding to the positioning dominant mode; determining the temporal domain corresponding to the control positioning data based on the dominant performance index; and adjusting and dividing the temporal domain to obtain the adjustment control sequence.

[0118] The control positioning label refers to a classification label added to the control positioning data to identify its core attributes. It covers key information such as the control stage to which the data belongs, the corresponding control axis, and the positioning accuracy requirements. It is the core basis for matching preset control modes. For example, the control positioning data label for a turntable can be marked as "Stage: Coarse Adjustment, Axis: X-axis, Accuracy Level: Fast Convergence Level". This label allows for quick identification of the control scenario corresponding to the data, avoiding directional deviations during subsequent mode matching and ensuring that the appropriate control logic is called from the mode library. The preset control mode library refers to a pre-built database storing various turntable positioning control modes. Each mode corresponds to a specific control scenario (such as coarse adjustment and fine adjustment), including the control strategy, parameter range, and... The execution logic provides adaptation schemes for positioning data of different tags. For example, the mode library stores "coarse adjustment - fast convergence mode" (driving force 8-12N, step size 0.1-0.3mm, sampling rate 500Hz) and "fine adjustment - high precision stable mode" (driving force 2-5N, step size 0.001-0.01mm, sampling rate 2000Hz), and the corresponding mode can be directly retrieved according to the positioning tag. The positioning dominant mode refers to the core control mode that is most suitable for the current positioning scenario and is matched from the preset control mode library according to the control positioning tag. It determines the core strategy and parameter direction of the positioning adjustment at this stage and is the basis for subsequent division of the adjustment sequence. For example, when the control positioning tag is "stage: transition adjustment, axis type: Y-axis, accuracy level: In the "smooth transition level," a "transitional adjustment-smooth transition mode" (driving force 6-8N, step size 0.03-0.08mm, switching threshold 0.05mm) is matched from the mode library. This mode is the current dominant positioning mode, guiding subsequent adjustment divisions. The dominant performance indicators refer to the key performance parameters that determine the adjustment effect in the dominant positioning mode, covering adjustment speed, accuracy threshold, stability requirements, etc., directly constraining the division of the temporal action domain and the design of adjustment steps, ensuring that the adjustment process meets the core objectives of the mode. For example, the dominant performance indicators of the "transitional adjustment-smooth transition mode" are "convergence time ≤ 5s, deviation fluctuation ≤ 0.02mm, step size adjustment gradient ≤ 0.02mm / time." These indicators clearly define the speed and accuracy that the adjustment needs to achieve. Precision standards are used to avoid deviations from the pattern requirements in the divided sequences. The time-series action domain refers to a specific time sub-interval within the effective time span, based on the key action point. Each sub-interval corresponds to a continuous adjustment action and includes the dynamic parameters and positioning data within that interval, clearly defining the effective time range of a set of adjustment strategies. For example, in the aforementioned transition adjustment phase, based on the key action point, it is divided into three time-series action domains: "1695210800.0s-1695210802.0s (effective with a step size of 0.05mm), 1695210802.0s-1695210805.0s (effective with a step size of 0.03mm to 0.02mm)". The parameters within each action domain are stable, providing clear time boundaries for the adjustment division.

[0119] Furthermore, the parsing of the control positioning labels in the control positioning data can be achieved through semantic analysis methods, such as using Natural Language Processing (NLP) to extract key identifiers from the data to obtain control positioning labels; the querying of the dominant positioning mode in the preset control mode library can be achieved through pattern matching algorithms, such as using the K-Nearest Neighbor (KNN) classifier to match the most similar positioning control mode to obtain the dominant positioning mode; the extraction of the dominant performance index corresponding to the dominant positioning mode can be achieved through feature extraction methods, such as using Principal Component Analysis (PCA) dimensionality reduction technology to obtain the core performance parameters of the mode to obtain the dominant performance index; the determination of the temporal domain corresponding to the control positioning data can be achieved through time window analysis methods, such as using a sliding window algorithm to divide the effective time interval of the data to obtain the temporal domain; the adjustment and division of the temporal domain can be achieved through dynamic segmentation methods, such as using an adaptive threshold algorithm to divide the time segment according to control requirements to obtain the adjustment control sequence.

[0120] As another embodiment of the present invention, determining the temporal domain corresponding to the control and positioning data based on the dominant performance index includes: extracting the dynamic parameter sequence from the dominant performance index; performing a collaborative mapping between the dynamic parameter sequence and the control and positioning data to obtain a mapping dataset; analyzing the effective time span corresponding to the mapping dataset; identifying key action points in the effective time span; and determining the temporal domain corresponding to the control and positioning data based on the key action points.

[0121] The dynamic parameter sequence refers to the set of key parameters in the dominant performance indicators that change dynamically over time. Recording the values ​​and trends of these parameters in chronological order reflects the dynamic characteristics of performance requirements during positioning adjustment and serves as the core link for coordinating control and positioning data. For example, the dynamic parameter sequence for the "transitional adjustment-smooth transition mode" is "0s: step size 0.05mm, driving force 7N; 2s: step size 0.03mm, driving force 6.5N; 5s: step size 0.02mm, driving force 6N," clearly presenting the parameter changes at different times and providing a time-dimensional reference for matching control and positioning data. The mapping number... A dataset refers to a dataset formed by associating and matching dynamic parameter sequences with control and positioning data along the time dimension. This ensures a one-to-one correspondence between dynamic parameters (such as step length and driving force) at each time point and the corresponding positioning data (such as current position and deviation value), eliminating time misalignment between data. For example, the mapping dataset for the transition adjustment phase of a turntable might be: "0s: step length 0.05mm + position 120.4mm + deviation 0.4mm; 2s: step length 0.03mm + position 120.48mm + deviation 0.02mm; 5s: step length 0.02mm + position 120.50mm + deviation 0.001mm". This achieves time alignment between parameters and positioning data. The effective time span refers to the time interval within the mapped dataset where both the dynamic parameter sequence and the control positioning data are valid and mutually matched. Specifically, it's the time range from the start of the adjustment action to the end point when the dominant performance index requirement is met. Data exceeding this range is considered invalid. For example, in the mapped dataset during the transition adjustment phase, the dynamic parameters are valid from 0s, and the deviation decreases to 0.001mm by 5s (meeting the ≤0.02mm requirement). Therefore, the effective time span is "1695210800.0s-1695210805.0s," ensuring that subsequent analysis only... Based on data within the effective time period; the key action points refer to the time nodes in the effective time span where dynamic parameters change significantly or positioning data reaches important thresholds. These nodes mark the turning point of the adjustment strategy (such as parameter switching) or the achievement of the stage goal, and are the core basis for dividing the time-series action domain. For example, within the effective time span of the above transition adjustment stage, the key action points include "0s (adjustment begins, initial parameters take effect), 2s (step size decreases from 0.05mm to 0.03mm), 5s (deviation reaches the target, adjustment ends)". Each node corresponds to an important turning point in the adjustment process and determines the segmentation logic of the time-series action domain.

[0122] Furthermore, the extraction of the dynamic parameter sequence from the dominant performance index can be achieved through time series decomposition methods, such as using the STL seasonal decomposition algorithm to extract the trend and periodic components from the index, thereby obtaining the dynamic parameter sequence; the co-mapping of the dynamic parameter sequence with the control and positioning data can be achieved through data association analysis methods, such as using the Dynamic Time Warping (DTW) algorithm to align the time series features of the two types of data, thereby obtaining the mapped dataset; the analysis of the effective time span corresponding to the mapped dataset can be achieved through sliding window statistical methods, such as using the Autocorrelation Function (ACF) to analyze the significance interval of the data in the time dimension, thereby obtaining the effective time span; the identification of key action points in the effective time span can be achieved through extreme value detection algorithms, such as using the peak detection algorithm to identify local maxima and abrupt change points in the time series, thereby obtaining key action points; the determination of the temporal domain of the control and positioning data can be achieved through time domain partitioning methods, such as using the density-based clustering algorithm DBSCAN to perform cluster analysis on timestamps, thereby obtaining the temporal domain.

[0123] Based on the aforementioned adjustment and control sequence, this invention generates drive commands corresponding to the turntable servo motor in the target turntable. This transforms the ordered adjustment steps into precise signals that the motor can directly execute, avoiding motor action deviations caused by ambiguous or disordered commands. It provides a clear execution basis for turntable posture adjustment and further solidifies the precision foundation of alignment control.

[0124] The turntable servo motor refers to a drive actuator specifically designed for target turntable pose adjustment, possessing high-precision closed-loop control capabilities. It can precisely adjust speed, torque, and rotation angle according to the input control signal, adapting to the power requirements of different stages such as coarse and fine adjustment of the turntable. It is the core power source for realizing the movement of the turntable from its current pose to the target pose. For example, a turntable equipped with a servo motor model MS-200 has a rated power of 2.0kW, a rated speed of 3600r / min, and a positioning accuracy of ±0.0005°. In the coarse adjustment stage, it can output a large torque of 12N·m to achieve rapid rotation, while in the fine adjustment stage, it maintains micron-level angle control with a small torque of 0.5N·m, ensuring the accuracy and efficiency of turntable pose adjustment. The drive command refers to a digital control signal generated based on the adjustment control sequence, which the turntable servo motor can recognize and execute, including motor movements. The type (such as position adjustment, speed adjustment), specific parameters (speed, torque, target angle), and execution time can directly guide the motor to complete the posture adjustment action according to the preset logic. For example, for the MS-200 motor mentioned above, a certain drive command is "Action type: position adjustment; target speed: 1800 r / min; output torque: 5 N·m; execution time: 1.5 s; target angle: 45.000°". After receiving the command, the motor will rotate at a speed of 1800 r / min and a torque of 5 N·m within 1.5 seconds, and finally stop at the target angle of 45.000°, matching the positioning requirements of the turntable fine-tuning stage. Optionally, the drive command corresponding to the turntable servo motor in generating the target turntable can be implemented by a PID control algorithm, such as using a position-type PID controller to calculate the motor speed and steering control quantity, thereby obtaining the drive command.

[0125] S4. Collect the high-frequency disturbance value during the execution of the drive command, calculate the anti-disturbance control quantity corresponding to the high-frequency disturbance value, and detect the precision imbalance point in the operation of the target turntable based on the anti-disturbance control quantity.

[0126] This invention collects high-frequency disturbance values ​​during the execution of the drive commands and calculates the anti-disturbance control quantities corresponding to the high-frequency disturbance values. It can promptly capture the impact of external interference or the motor's own fluctuations on the turntable's movement, avoid the accumulation of disturbances leading to an increase in position and posture deviation, provide a data foundation for accurate anti-disturbance, and further solidify the reliability of alignment control.

[0127] The high-frequency disturbance value refers to a specific numerical value quantifying the magnitude of high-frequency interference, collected from the high-frequency components of the drive execution data based on the disturbance amplitude level. Low-level disturbances are collected with conventional accuracy, while medium- and high-level disturbances are collected with higher accuracy to ensure that the data can truly reflect the actual impact of disturbances of different intensities. For example, the aforementioned "medium-level" speed disturbance is collected with an accuracy of 0.01 r / min, resulting in a high-frequency disturbance value of "±0.82 r / min"; the "low-level" torque disturbance is collected with an accuracy of 0.05 N·m, resulting in a high-frequency disturbance value of "±0.38 N·m". These values ​​are directly used for subsequent anti-disturbance control quantity calculations.

[0128] As an embodiment of the present invention, the step of collecting the high-frequency disturbance value of the driving instruction during execution includes: identifying the execution time interval corresponding to the driving instruction; extracting the driving execution data within the execution time interval; statistically analyzing the high-frequency oscillation index corresponding to the high-frequency components in the driving execution data; determining the disturbance amplitude level corresponding to the high-frequency oscillation index; and collecting the high-frequency disturbance value of the driving instruction during execution based on the disturbance amplitude level.

[0129] The execution time interval refers to the complete time range from the issuance of the drive command to the completion of the command action by the servo motor. This clearly defines the time boundary for disturbance value collection, ensuring that analysis is performed only on data generated during command execution and avoiding interference from irrelevant time data. For example, a turntable servo motor's "position adjustment" drive command has a preset execution time of 1.5 seconds. If the command is received starting at 10:00:00.000 and the target angle adjustment is completed at 10:00:01.500, then the execution time interval for this command is "10:00:00.000-10:00:01.500". All disturbance data collection operations... All actions must be completed within this time interval. The drive execution data refers to the key parameter data collected in real time within the execution time interval, reflecting the status of the turntable servo motor executing drive commands. This data includes motor speed, output torque, stator current, actual rotation angle, etc., and serves as the basic data source for extracting high-frequency disturbance characteristics. For example, within the aforementioned 1.5s execution interval, the collected drive execution data includes: speed (fluctuation between 1798r / min and 1802r / min), torque (change between 4.8N・m and 5.2N・m), and actual angle (adjustment between 44.995° and 45.005°). This data comprehensively records the motor's execution... The dynamic process of executing instructions; the high-frequency oscillation index refers to an index that extracts high-frequency components (usually fluctuations with a frequency > 100Hz) from the drive execution data through signal processing (such as Fourier transform) and quantifies their oscillation intensity. It can be calculated through the standard deviation or peak difference of the amplitude of the high-frequency components, intuitively reflecting the severity of high-frequency disturbances. For example, after processing the above speed data, a 120Hz high-frequency fluctuation component is separated, whose amplitude varies between 1798r / min and 1802r / min. The calculated standard deviation is 0.8r / min, and this "0.8r / min" is the high-frequency oscillation index corresponding to the speed. The strength of the high-frequency disturbance at the rotational speed is considered. The disturbance amplitude level refers to the intensity level of the high-frequency disturbance, which is divided according to the numerical range of the high-frequency oscillation index. It is usually divided into three levels: low, medium, and high. Different levels correspond to different acquisition accuracy and subsequent anti-interference strategies, providing a classification basis for accurately acquiring high-frequency disturbance values. For example, the preset high-frequency oscillation index <0.5 is "low level", 0.5-1.0 is "medium level", and >1.0 is "high level". The above-mentioned high-frequency oscillation index of rotational speed of 0.8 r / min belongs to "medium level", and the high-frequency oscillation index of torque of 0.4 N·m belongs to "low level". The two correspond to different disturbance acquisition accuracy requirements.

[0130] Furthermore, identifying the execution time interval corresponding to the drive command can be achieved through timestamp parsing methods, such as using a command cycle analysis tool to extract the effective time period of the PWM waveform to obtain the execution time interval; extracting the drive execution data within the execution time interval can be achieved through a data acquisition card, such as using the NIDAQmx tool to record the current and voltage waveforms of the motor driver in real time to obtain the drive execution data; statistically analyzing the high-frequency oscillation index corresponding to the high-frequency components in the drive execution data can be achieved through spectrum analysis methods, such as using Fast Fourier Transform (FFT) to calculate the energy integral value of the frequency band above 5kHz to obtain the high-frequency oscillation index; determining the disturbance amplitude level corresponding to the high-frequency oscillation index can be achieved through threshold grading methods, such as using the ISO 10816 vibration standard to divide the danger level range of the oscillation index to obtain the disturbance amplitude level; collecting the high-frequency disturbance value of the drive command during execution can be achieved through bandpass filtering methods, such as using a digital filter to extract the effective value of the vibration signal in the 2kHz-10kHz frequency band to obtain the high-frequency disturbance value.

[0131] In another embodiment of the present invention, the anti-interference control quantity corresponding to the high-frequency disturbance value is calculated using the following formula:

[0132] ;

[0133] in, This represents the disturbance rejection control quantity (unit: Nm) corresponding to the high-frequency disturbance value. This represents the proportionality coefficient. Indicates the level of disturbance. Indicates the execution time interval (unit: seconds). The value of the high-frequency disturbance is expressed in Nm.

[0134] In detail, the disturbance rejection control quantity can be expressed as the additional control quantity (unit: Nm) needed to counteract the influence of high-frequency disturbances on the turntable servo motor. This quantity guides the motor to output a reverse force for stable operation. For example, when a turntable servo motor is subjected to a high-frequency disturbance, the calculated... =1.2Nm, the motor will output a reverse torque of 1.2Nm to counteract the disturbance; the proportional coefficient can represent an adjustment coefficient set according to the mechanical characteristics of the turntable (such as inertia, stiffness), used to adapt to the disturbance rejection intensity and avoid excessive or insufficient disturbance rejection. For example, if the turntable has a small inertia and a fast response, K=0.5 is set to match the disturbance rejection control quantity with the response capability of the mechanical system; the disturbance amplitude level can represent the classification of the high-frequency disturbance intensity (such as low level 1, medium level 2, high level 3). The higher the level, the more urgent the disturbance rejection requirement. The formula will amplify or reduce the disturbance rejection control quantity through this value. For example, if the disturbance is at a high level at a certain moment, =3, which makes the calculated disturbance rejection control quantity more significant; the execution time interval can represent the execution duration of the drive instruction (unit: s), which determines the time range of the integral in the formula, used to statistically analyze the cumulative effect of the disturbance within that time period. For example, if the drive instruction execution time is T=1s, the integral extends from t=0 to t=1, covering the disturbance during the entire instruction execution process; the high-frequency disturbance value can represent the high-frequency interference value (unit: Nm) that changes with time, reflecting the magnitude of the disturbance at different times. By integrating its square, the energy accumulation of the disturbance can be reflected. For example, within T=2s, =0.3sin(100πt) (simulating 100Hz high-frequency oscillation), substituting into the formula, the energy contribution of the disturbance can be calculated, and then the disturbance rejection control quantity can be obtained.

[0135] Based on the aforementioned anti-disturbance control quantity, this invention detects the accuracy imbalance point during the operation of the target turntable, which can accurately locate the root cause of the pose deviation caused by the disturbance and avoid aimless accuracy investigation; it can capture the potential trend of accuracy decline in advance and intervene before the imbalance intensifies to prevent serious deviations; it also provides a basis for subsequent optimization of anti-disturbance strategies and continuous improvement of the long-term operating accuracy of the turntable, effectively ensuring the stability of alignment control.

[0136] The precision imbalance point refers to a specific position or moment during the operation of the target turntable where the actual pose accuracy exceeds a preset allowable threshold due to factors such as high-frequency disturbances, mechanical wear, or control delays. This is a key indicator of decreased turntable positioning stability, and the concentrated outbreak point of the positioning deviation needs to be deduced through the anti-disturbance control quantity. For example, a precision turntable has a preset allowable X-axis positioning deviation of ±0.002mm. When executing drive commands, monitoring through the anti-disturbance control quantity reveals that when the turntable moves to the X-axis position of 120.50mm, the actual deviation remains stable at 0.005mm (exceeding the threshold), and the anti-disturbance control quantity needs to be increased to 1.8Nm to barely maintain this position. This 120.50mm position is a typical precision imbalance point, reflecting a significant disturbance or mechanical coordination problem. Optionally, the detection of the precision imbalance point during the operation of the target turntable can be achieved through laser interferometry, such as using a Renishaw XL-80 laser interferometer to monitor the turntable positioning deviation and repeatability accuracy, thereby obtaining the precision imbalance point.

[0137] S5. Identify the precision offset value corresponding to the precision imbalance point, coordinate the precision offset value with the preset precision index to obtain the alignment parameter group, and generate the alignment control strategy corresponding to the target turntable based on the alignment parameter group.

[0138] This invention can accurately quantify the degree of turntable pose deviation from the target by identifying the accuracy offset value corresponding to the accuracy imbalance point, providing a clear numerical basis for subsequent correction actions; it can make anti-disturbance or adjustment strategies more targeted, avoiding resource waste and efficiency loss caused by indiscriminate adjustment; it can also trace the root cause of the imbalance, providing a key reference for optimizing the turntable control logic and improving long-term operating accuracy, thereby ensuring the stability and accuracy of turntable alignment.

[0139] The precision offset value refers to the deviation between the actual pose (such as position, angle, etc.) of the target turntable and the preset target pose at the precision imbalance point. It quantifies the degree of pose deviation at the imbalance point and provides a specific basis for subsequent deviation correction. For example, if the target position of a turntable on the X-axis is 120.500mm and the actual position at the precision imbalance point is 120.505mm, its X-axis precision offset value is 0.005mm; if the target position on the Y-axis is 80.200mm and the actual position is 80.197mm, the Y-axis precision offset value is -0.003mm. This intuitively reflects the deviation of each axis at the imbalance point. Optionally, the identification of the precision offset value corresponding to the precision imbalance point can be achieved through deviation quantification analysis methods, such as using the least squares method to fit the coordinate deviation between the theoretical trajectory and the actual trajectory to obtain the precision offset value.

[0140] Furthermore, the present invention coordinates the accuracy offset value with the preset accuracy index to obtain an alignment parameter set, which can establish a quantitative correlation between the actual position and posture deviation and the standard requirements, clearly define whether the offset exceeds the index range and the specific degree of exceedance, provide a clear reference benchmark for subsequent processing, and help stabilize the high-precision operation of the turntable in the long term.

[0141] The alignment parameter set refers to a set of parameters formed by integrating accuracy offset value, accuracy difference, collaborative compensation amount, preset accuracy index, and target correction value. It contains all the key data required for turntable alignment and can directly guide subsequent accuracy correction actions. For example, the alignment parameter set for the X-axis is: "Accuracy offset value: 0.005mm; Preset accuracy index: ±0.002mm; Accuracy difference: 0.003mm; Collaborative compensation amount: 0.0039mm; Target corrected position: 120.500mm-0.0039mm=120.4961mm", which fully presents the alignment logic and correction target, providing a clear basis for turntable adjustment.

[0142] As an embodiment of the present invention, the step of coordinating the precision offset value with a preset precision index to obtain a alignment parameter set includes: comparing the precision difference between the precision offset value and the preset precision index; based on the precision difference, traversing the alignment relationship set in the preset alignment rule base; filtering out coordinating matching pairs that meet the precision threshold within the alignment relationship set; parsing the coordinating compensation amount matched by the coordinating matching pair; and based on the coordinating compensation amount, coordinating the precision offset value with the preset precision index to obtain the alignment parameter set.

[0143] The precision difference refers to the absolute or relative difference between the precision offset value and the preset precision index. It quantifies the degree to which the actual pose of the turntable deviates from the standard requirements and is the core quantitative basis for determining whether compensation is needed and the extent of compensation. For example, if the preset precision index for the X-axis of a turntable is ±0.002mm (i.e., the allowable deviation range), and the precision offset measured at the precision imbalance point is 0.005mm, then the precision difference is 0.005mm - 0.002mm = 0.003mm, clearly reflecting the specific value by which the actual offset exceeds the standard, providing a basis for subsequent matching compensation rules. The preset alignment rule library refers to a pre-built database that stores the corresponding rules for precision differences and compensation strategies, covering the coordination corresponding to different precision difference ranges. The matching logic, compensation method, and parameter calculation method provide a standardized basis for the coordinated alignment of precision offset values ​​and preset indicators. For example, the rule base stores: "Precision difference 0.001-0.003mm: linear compensation with a compensation coefficient of 1.2; precision difference 0.003-0.005mm: nonlinear compensation with a compensation coefficient of 1.5; difference > 0.005mm: trigger emergency calibration process." The corresponding rule can be directly retrieved based on the calculated precision difference. The alignment relationship set refers to the specific rule set in the preset alignment rule base, classified according to the precision difference range, containing "precision difference range - compensation type - compensation coefficient." It is the direct object for traversing and matching. Each relationship corresponds to a clear set of coordinated alignment logic. For example, the alignment set for a "slight offset" scenario is: "Difference 0.001-0.002mm: linear compensation, coefficient 1.0; difference 0.002-0.003mm: linear compensation, coefficient 1.2"; the alignment set for a "moderate offset" scenario is: "Difference 0.003-0.004mm: non-linear compensation, coefficient 1.3; difference 0.004-0.005mm: non-linear compensation, coefficient 1.5", facilitating precise matching based on the difference. The collaborative matching pair refers to the "precision difference range - compensation rule" pair selected from the alignment set that perfectly matches the current precision difference, satisfying the preset precision threshold condition (i.e., the current difference falls within a certain interval of the alignment set), which determines the amount of collaborative compensation. Based on direct evidence, for example, if the currently calculated X-axis accuracy difference is 0.003mm, after traversing the alignment relationship set, it is found that it falls within the relationship range of "difference 0.003-0.004mm: nonlinear compensation, coefficient 1.3". This combination of "0.003-0.004mm (difference range) - nonlinear compensation + 1.3 (compensation rule)" is the cooperative matching pair. The cooperative compensation amount refers to the specific value used to correct the accuracy offset value, calculated based on the compensation rules (such as compensation type and coefficient) in the cooperative matching pair. It enables the actual pose of the turntable to approach the preset accuracy index and is the core execution parameter of cooperative alignment. For example, the above cooperative matching pair is "nonlinear compensation, coefficient 1.3", and the current accuracy difference is 0.The calculated compensation amount is 0.003mm × 1.3 = 0.0039mm. This value is the compensation amount that needs to be adjusted towards the target position on the X-axis to compensate for the 0.005mm accuracy offset.

[0144] Furthermore, the comparison of the precision difference between the precision offset value and the preset precision index can be achieved through a difference calculation method, such as using an absolute value difference algorithm to calculate the numerical difference between the measured offset value and the standard index, thereby obtaining the precision difference; the traversal of the alignment relationship set in the preset alignment rule base can be achieved through a graph database query method, such as using the Cypher language of the Neo4j graph database to traverse all node relationship paths, thereby obtaining the alignment relationship set; the filtering of collaborative matching pairs that meet the precision threshold within the alignment relationship set can be achieved through a filtering algorithm, such as using a Bloom filter to quickly filter matching combinations that meet the precision tolerance range, thereby obtaining collaborative matching pairs; the parsing of the collaborative compensation amount of the collaborative matching pairs can be achieved through a regression analysis method, such as using a multiple linear regression model to calculate the compensation coefficient matrix of each matching pair, thereby obtaining the collaborative compensation amount; the collaborative alignment of the precision offset value with the preset precision index can be achieved through a data registration method, such as using the ICP iterative nearest point algorithm to achieve spatial alignment between the offset data and the standard data, thereby obtaining the alignment parameter set.

[0145] Based on the alignment parameter set, this invention generates an alignment control strategy corresponding to the target turntable. This strategy can closely match the actual accuracy deviation and compensation requirements of the turntable, avoid blind control due to lack of quantitative data, provide a clear direction for accurate pose correction, and provide data support for subsequent iterative optimization of control logic and adaptation to alignment requirements under different working conditions, thus helping to maintain the high-precision operation of the turntable in the long term.

[0146] The alignment control strategy refers to a systematic control scheme formulated based on core data such as the accuracy offset value and collaborative compensation amount in the alignment parameter set to correct the accuracy deviation of the turntable and bring it back to the preset accuracy index. It includes control objectives, execution steps, parameter configuration and monitoring logic, and can directly guide the turntable to complete the pose correction. For example, if the X-axis alignment parameter set of a turntable is "accuracy offset value 0.005mm, collaborative compensation amount 0.0039mm, target correction position 120.4961mm", the corresponding alignment control strategy is as follows: First, the servo motor performs a compensation movement of 0.0039mm with a torque of 5N·m; second, the real-time deviation is monitored at a frequency of 1kHz; third, when the deviation is ≤0.002mm, the torque is switched to 3N·m to maintain stability, ensuring that the turntable accurately returns to the target position. Optionally, the alignment control strategy corresponding to the target turntable can be generated by a fuzzy logic control method, such as using a Mamdani-type fuzzy inference system to generate a multi-level adjustment strategy based on the alignment parameter set, thereby obtaining the alignment control strategy.

[0147] Compared to the problems described in the background technology, this invention, by acquiring turntable sensing data of the target turntable in multimodal scenarios, can overcome the information limitations of a single sensor, covering multi-dimensional state information such as position, vibration, and temperature during turntable operation. This enables comprehensive perception of the turntable's operating conditions, avoiding misjudgments caused by missing single data dimensions, and ensuring the accuracy and stability of turntable alignment control from the source. Based on the actual pose state, this invention calculates the positional deviation between the current position and the target position of the target turntable, accurately quantifying the specific difference between the current and target positions. This provides clear numerical basis for subsequent control decisions, avoiding inaccurate control direction or improper adjustment due to the inability to grasp the degree of deviation, and reducing problems such as overshoot and oscillation caused by inaccurate deviation estimation. This ensures the accuracy of the final turntable alignment control from a critical point. Furthermore, by extracting the control positioning data from the hierarchical control process, this invention can provide a basis for subsequent adjustment and division of the control positioning data and generation of turntable servo motor drives. The drive commands provide precise and specific information, avoiding a lack of clear direction in subsequent control stages due to missing or ambiguous data. This helps improve the long-term stability and accuracy of turntable alignment control. Furthermore, by collecting high-frequency disturbance values ​​during the execution of the drive commands and calculating the corresponding anti-disturbance control quantities, this invention can promptly capture the impact of external interference or motor fluctuations on turntable movement, preventing the accumulation of disturbances that could lead to increased posture deviation. This provides a data foundation for precise anti-disturbance and further strengthens the reliability of alignment control. Finally, by identifying the precision offset value corresponding to the precision imbalance point, this invention can accurately quantify the degree of turntable posture deviation from the target, providing clear numerical basis for subsequent correction actions. This allows for more targeted anti-disturbance or adjustment strategies, avoiding resource waste and efficiency loss caused by indiscriminate adjustments. It also allows for tracing the root cause of the imbalance, providing key references for optimizing turntable control logic and improving long-term operating accuracy, thereby ensuring the stability and accuracy of turntable alignment. Therefore, the turntable precision alignment control method based on multimodal sensor fusion provided by this invention can improve the control accuracy and adaptability of the turntable in complex application environments.

[0148] like Figure 3 The diagram shown is a functional block diagram of a turntable precision alignment control system based on multimodal sensor fusion according to the present invention.

[0149] The turntable precision alignment control system 200 based on multimodal sensor fusion described in this invention can be installed in an electronic device. Depending on the functions implemented, the turntable precision alignment control system may include a state recognition module 201, a process construction module 202, an instruction generation module 203, an imbalance point detection module 204, and a strategy generation module 205. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0150] In this embodiment of the invention, the functions of each module / unit are as follows:

[0151] The state recognition module 201 is used to acquire turntable sensing data of the target turntable in a multimodal scenario, perform data fusion on the turntable sensing data to obtain a multimodal fusion vector, and identify the real-time pose state of the target turntable based on the multimodal fusion vector.

[0152] The process construction module 202 is used to calculate the position deviation between the current position and the target position of the target turntable based on the actual pose state, determine the control stage corresponding to the target turntable based on the position deviation, and construct the hierarchical control process corresponding to the target turntable based on the control stage.

[0153] The instruction generation module 203 is used to extract control positioning data from the hierarchical control process, adjust and divide the control positioning data to obtain an adjustment control sequence, and generate drive instructions corresponding to the turntable servo motor in the target turntable based on the adjustment control sequence.

[0154] The imbalance point detection module 204 is used to collect the high-frequency disturbance value during the execution of the drive command, calculate the anti-disturbance control quantity corresponding to the high-frequency disturbance value, and detect the precision imbalance point in the operation of the target turntable based on the anti-disturbance control quantity.

[0155] The strategy generation module 205 is used to identify the precision offset value corresponding to the precision imbalance point, coordinate the precision offset value with the preset precision index to obtain the alignment parameter group, and generate the alignment control strategy corresponding to the target turntable based on the alignment parameter group.

[0156] In detail, the modules in the turntable precision alignment control system 200 based on multimodal sensor fusion described in this embodiment of the invention employ the same methods as described above. Figure 1 The method is the same as the turntable precision alignment control method based on multimodal sensor fusion described in the article, and can produce the same technical effect, so it will not be repeated here.

[0157] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. In the above multiple embodiments, each embodiment can be combined with each other or independent. Deleting any one of them will not affect the technical implementation of other embodiments. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for precise alignment control of a rotary table based on multi-modal sensor fusion, characterized in that, The method comprises: Obtaining the target turntable in a multi-modal scene under the turntable sensor data, performing data fusion on the turntable sensor data to obtain a multi-modal fusion vector, and identifying the real-time pose state corresponding to the target turntable based on the multi-modal fusion vector; Based on the real-time pose state, calculate the position deviation between the current position and the target position of the target turntable, according to the position deviation, wherein the position deviation between the current position and the target position of the target turntable is calculated based on the real-time pose state, comprising: Analyze the pose state type corresponding to the real-time pose state; Based on the pose state type, analyze the current position and target position corresponding to the target turntable; Divide the position axis component corresponding to the current position; Identify the axis deviation value of the position axis component relative to the target position; Based on the axis deviation value, the position deviation between the current position and the target position of the target turntable is calculated by the following formula: wherein, represents a position deviation amount between a current position and a target position of the target rotary table, represents a total number of control axes in the target rotary table, represents a control axis index corresponding to the target rotary table, represents a deviation value on a first control axis, represents a weight coefficient of a first control axis, determines a control stage corresponding to the target rotary table, and constructs a hierarchical control process corresponding to the target rotary table based on the control stage; Extract the control positioning data in the hierarchical control process, adjust and divide the control positioning data to obtain an adjustment control sequence, and generate the drive instruction corresponding to the turntable servo motor in the target turntable according to the adjustment control sequence; Collect the high-frequency disturbance value of the drive instruction in the execution process, and calculate the anti-disturbance control amount corresponding to the high-frequency disturbance value, wherein the high-frequency disturbance value of the drive instruction in the execution process comprises: Identify the execution time interval corresponding to the drive instruction; Extract the drive execution data in the execution time interval; Statistical high-frequency component corresponding to the high-frequency oscillation index in the drive execution data; Determine the disturbance amplitude level corresponding to the high-frequency oscillation index; Based on the disturbance amplitude level, collect the high-frequency disturbance value of the drive instruction in the execution process; And calculate the anti-disturbance control amount corresponding to the high-frequency disturbance value by the following formula: wherein, represents the anti-disturbance control amount corresponding to the high-frequency disturbance value, represents a proportional coefficient, represents a disturbance amplitude level, represents an execution time interval, detects the precision imbalance point in the target turntable operation based on the anti-disturbance control amount. Identify the precision offset value corresponding to the precision imbalance point, and align the precision offset value with the preset precision index to obtain an alignment parameter group, and generate the alignment control strategy corresponding to the target turntable based on the alignment parameter group.

2. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 1, wherein, The real-time pose state corresponding to the target turntable is identified based on the multi-modal fusion vector, comprising: Extract the multi-modal pose feature in the multi-modal fusion vector; Analyze the sensing pose component associated with the multi-modal pose feature; Real-time synchronization of the sensing pose component is performed to obtain synchronization component data; Based on the synchronization component data, analyze the current tilt angle and center coordinates of the target turntable; Based on the tilt angle and the center coordinates, identify the real-time pose state corresponding to the target turntable.

3. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 2, characterized in that, The real-time synchronization of the sensing pose component is performed to obtain synchronization component data, comprising: Analyze the component time sequence corresponding to the sensing pose component; Determine the synchronization time axis corresponding to the component time sequence; Query the data timestamp corresponding to each component in the synchronization time axis; Synchronize and align the data timestamp to obtain a synchronization data set; Extract the consistency data in the synchronization data set as the synchronization component data.

4. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 1, wherein, The control stage is based on the construction of the target turntable corresponding hierarchical control process, comprising: Inquiry stage division logic in the control stage; Based on the stage division logic, determine the interlayer association module between the control stage; Generate the topological hierarchical framework corresponding to the interlayer association module; Configure the hierarchical control data corresponding to the hierarchical control framework; Based on the hierarchical control data, construct the hierarchical control process corresponding to the target turntable.

5. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 4, characterized in that, The topological hierarchical framework corresponding to the interlayer association module is generated, comprising: Analysis of the hierarchical mapping relationship in the interlayer association module; Based on the hierarchical mapping relationship, identify the associated control unit corresponding to the interlayer association module; Extract the key control nodes in the associated control unit; Reconstruct the topological control chain corresponding to the key control node; Generate the topological hierarchical framework corresponding to the topological control chain.

6. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 1, wherein, The control positioning data is adjusted and divided to obtain the adjustment control sequence, comprising: Analysis of the control positioning data in the control positioning label; According to the control positioning label, query the positioning dominant mode in the preset control mode library; Extract the dominant performance index corresponding to the dominant performance index; Based on the dominant performance index, determine the time sequence scope corresponding to the control positioning data; Adjust and divide the time sequence scope to obtain the adjustment control sequence.

7. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 6, characterized in that, Based on the dominant performance index, determine the time sequence scope corresponding to the control positioning data, comprising: Extract the dynamic parameter sequence in the dominant performance index; The dynamic parameter sequence is mapped with the control positioning data to obtain the mapping data set; Analysis of the effective time span corresponding to the mapping data set; Identify the key action point in the effective time span; Based on the key action point, determine the time sequence scope corresponding to the control positioning data.

8. The multi-modal sensor fusion-based precision alignment control method of a turntable according to claim 1, wherein, The precision offset value is cooperatively aligned with the preset precision index to obtain the alignment parameter group, comprising: Compare the precision difference between the precision offset value and the preset precision index; Based on the precision difference, traverse the alignment relationship set in the preset alignment rule library; Screen the cooperative matching pair in the alignment relationship set that meets the accuracy threshold; Analysis of the cooperative compensation amount matched by the cooperative matching pair; Based on the cooperative compensation amount, the precision offset value is cooperatively aligned with the preset precision index to obtain the alignment parameter group.

Images