Three-axis moving part control method and system based on machine vision

Abstract

The invention discloses a three-axis moving part control method and system based on machine vision. The method comprises the following steps: obtaining a trajectory feature data set from a motion trajectory original data stream; analyzing the difference between the trajectory feature data set and a theoretical trajectory to obtain a deviation distribution report, and calculating the variation amplitude of deviation in combination with historical data and environmental fluctuation conditions to obtain a deviation value matrix; extracting spatial displacement and angle offset data from the deviation value matrix to obtain a basic adjustment amount, and calculating an initial parameter adjustment scheme of each axis in combination with a multi-axis cooperative control constraint condition to obtain a parameter optimization result; and fusing a compensation factor into the parameter optimization result, recalculating the compensation factor for the parameter of which the parameter adjustment result exceeds a preset precision range to obtain a final parameter adjustment scheme, and adjusting the motion parameter of each axis according to the scheme. The method can effectively improve the motion precision and stability of the multi-axis motion part.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation control, and in particular to a three-axis motion component control method and system based on machine vision. BACKGROUND

[0002] Currently, in the field of modern industrial manufacturing and automation, precise control of three-axis motion components is crucial for improving production efficiency and product quality. With the development of industrial vision, visual intelligence, AI vision and other technologies, more and more motion control systems are trying to use visual detection and visual measurement methods to monitor and adjust the motion trajectory in real time.

[0003] In one prior art, traditional detection methods such as encoders and sensors are usually used to realize the positioning and control of three-axis motion components. These methods can meet the basic needs in static and ideal environments, but in complex working conditions, due to the lack of precise perception of dynamic changes and slight deviations, it is difficult to cope with the wear and tear of the equipment after long-term operation or external environmental disturbances.

[0004] In existing multi-axis motion control systems, the motion trajectory data acquisition and deviation identification link is easily affected by factors such as equipment wear and tear and environmental disturbances, making it difficult to capture the slight deviations of the motion trajectory in a timely and accurate manner, resulting in a lag in parameter adjustment schemes and affecting the control efficiency of multi-axis collaborative motion and the long-term stability of system operation. SUMMARY

[0005] The present application provides a three-axis motion component control method and system based on machine vision to solve the problem of low control efficiency and stability of multi-axis collaborative motion.

[0006] In a first aspect, to solve the above technical problems, the present application provides a three-axis motion component control method based on machine vision, comprising: Obtaining a motion trajectory original data stream and pre-processing to obtain a trajectory feature data set; According to the trajectory feature data set, analyzing the difference between the actual trajectory and the theoretical trajectory to obtain a deviation distribution report; According to the deviation analysis report, combining historical data and current environmental fluctuation conditions, calculating the real-time change amplitude of the deviation to obtain a deviation value matrix; According to the deviation value matrix, extracting spatial displacement and angular displacement data to obtain a basic adjustment amount; According to the basic adjustment amount, combining the constraint conditions of multi-axis collaborative control, calculating the initial parameter adjustment scheme of each axis to obtain a parameter optimization result; Integrating a compensation factor into the parameter optimization result, recalculating the compensation factor for parameters whose parameter adjustment results exceed the preset accuracy range to obtain a final parameter adjustment scheme; According to the final parameter adjustment scheme, the motion parameters of each axis are synchronously adjusted.

[0007] Preferably, the motion trajectory original data stream is acquired and preprocessed to obtain the trajectory feature dataset, including: Acquiring a motion trajectory original data stream containing position, velocity and acceleration information; Performing denoising processing on the original data stream to obtain a first data stream; Dividing the first data stream into time periods, and marking time periods in which data fluctuations exceed a preset fluctuation threshold to obtain a second data stream; Performing motion trajectory simulation on the second data stream to obtain a trajectory feature set; Correcting the trajectory feature set using a pre-established trajectory feature template to obtain a trajectory feature dataset.

[0008] Preferably, according to the trajectory feature dataset, the difference between the actual trajectory and the theoretical trajectory is analyzed to obtain a deviation distribution report, including: Obtaining trajectory feature values in the trajectory feature dataset; Comparing the trajectory feature values with a pre-established trajectory template to obtain a deviation distribution record table; Dividing the deviation distribution record table according to time sequence, and statistically analyzing the deviation fluctuation range in each time sequence to obtain a deviation trend dataset; Correcting the deviation trend dataset to obtain a deviation correction dataset; Simulating future trajectory deviations from the deviation correction dataset to obtain a deviation distribution report.

[0009] Preferably, according to the deviation analysis report, the real-time change amplitude of the deviation is calculated in combination with historical data and current environmental fluctuation conditions to obtain a deviation value matrix, including: Obtaining deviation feature values in the deviation analysis report; Statistically analyzing the fluctuation of the deviation distribution from the deviation feature values to obtain deviation analysis data; Dividing the deviation analysis data into time periods, and segmenting and marking deviation data that has periodic fluctuations to obtain deviation trend features; Matching the deviation trend features with real-time data, and performing fluctuation range analysis to obtain a deviation amplitude record; Comparing the deviation amplitude record and the deviation analysis report with a preset deviation amplitude threshold, and performing normalization processing on data that exceeds the preset deviation amplitude threshold to obtain a deviation value matrix.

[0010] Preferably, according to the deviation value matrix, spatial displacement and angle offset data are extracted to obtain a basic adjustment amount, including: The offset parameters and rotation angles in the deviation value matrix are extracted, and the offset parameters and the rotation angles are classified to obtain a classified deviation record; The classified deviation record is subjected to coordinate transformation processing, and data beyond a preset threshold are subjected to boundary correction processing to obtain a displacement angle set; Spatial displacement data and angle offset data in the displacement angle set are extracted, and the spatial displacement data and the angle offset data are subjected to data mapping processing to obtain a trajectory deviation distribution record; The trajectory deviation distribution record is subjected to motion trajectory correction and error-prone point parameters are adjusted to obtain a basic adjustment amount.

[0011] Preferably, according to the basic adjustment amount, a constraint condition of multi-axis collaborative control is combined to calculate an initial parameter adjustment scheme of each axis to obtain a parameter optimization result, including: Real-time motion state data of each axis are obtained; The real-time motion state data of each axis are compared with the constraint condition of multi-axis collaborative control to obtain a state data group; The state data group is analyzed for inter-axis coupling relationship to generate an associated data table, and key influence indicators are extracted from the associated data table.

[0012] Preferably, the constraint condition of multi-axis collaborative control includes: The constraint condition of multi-axis collaborative control includes a preset upper limit of motion speed of each axis, a position tolerance range, and a dynamic response time threshold, which are used as a basis for constraint judgment and parameter adjustment of real-time motion state of each axis.

[0013] Preferably, a compensation factor is integrated into the parameter optimization result, and a compensation factor is recalculated for a parameter that exceeds a preset precision range in a parameter adjustment result to obtain a final parameter adjustment scheme, including: Mechanical wear history information of a motion component is extracted according to the parameter optimization result and is subjected to comparison processing to obtain mechanical wear influence data; The mechanical wear influence data are subjected to compensation factor calculation and comparison with a preset compensation factor threshold, and the compensation factor is recalculated when the threshold is exceeded to obtain a compensation factor data set; The parameter optimization result is dynamically compensated according to the compensation factor data set to obtain a final parameter adjustment scheme.

[0014] Preferably, according to the final parameter adjustment scheme, motion parameters of each axis are synchronously adjusted, including: Obtain real-time motion trajectory data of each axis, perform deviation analysis on the real-time motion trajectory data of each axis, and obtain error detection results; Classify the error detection results, adjust the parameter scheme of error data exceeding the preset error threshold, and obtain and update the motion instructions of each axis in real time; Synchronously adjust the motion parameters of each axis according to the motion instructions of each axis, monitor the motion state of each axis, and compare the motion state with the running standard to obtain a control output result.

[0015] In a second aspect, the present application provides a three-axis motion component control system based on machine vision, comprising: A data acquisition module is configured to acquire motion trajectory raw data streams of a multi-axis motion component in real time through a sensor network, wherein the raw data streams contain position, speed and acceleration information. A data preprocessing module is configured to preprocess the acquired raw data streams to obtain a trajectory feature data set. A deviation identification and analysis module is configured to compare the trajectory feature data set with a preset theoretical trajectory template, identify and analyze the difference between the actual trajectory and the theoretical trajectory, determine the distribution and variation trend of the deviation. A deviation quantification module is configured to calculate the real-time variation amplitude of the deviation according to the deviation analysis report, combine historical data and current environmental fluctuation conditions, and obtain a deviation value matrix. A parameter extraction and correction module is configured to extract spatial displacement and angular displacement data according to the deviation value matrix, and obtain a basic adjustment amount. A multi-axis coordination and optimization module is configured to calculate an initial parameter adjustment scheme for each axis based on the basic adjustment amount and the constraint conditions of multi-axis coordinated control, and obtain a parameter optimization result. A motion control instruction generation and execution module is configured to incorporate a compensation factor into the parameter optimization result, recalculate the compensation factor for parameters exceeding the preset accuracy range of the parameter adjustment result, obtain a final parameter adjustment scheme, and synchronously adjust the motion parameters of each axis according to the final parameter adjustment scheme.

[0016] In a third aspect, the present application further provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the three-axis motion component control method based on machine vision according to any one of the above.

[0017] In a fourth aspect, the present application also provides a computer readable storage medium comprising a stored computer program, wherein the computer program, when executed, controls a device in which the computer readable storage medium is located to perform any one of the above machine vision-based three-axis motion component control methods.

[0018] Compared with the prior art, the present application has the following beneficial effects:

[0019] (1) The present application obtains the motion trajectory original data stream of the three-axis motion component, pre-processes it, extracts the trajectory feature data set, and then generates the deviation distribution report and the deviation value matrix according to the difference between the actual trajectory and the theoretical trajectory, combined with the historical data and the environmental fluctuation conditions. This process not only improves the accuracy and consistency of the trajectory data, but also enhances the system's ability to identify and adapt to dynamic interference factors, providing high-quality data support for subsequent control strategies and improving the system's response sensitivity and robustness.

[0020] (2) According to the deviation value matrix, the present application extracts the spatial displacement and angular offset data, calculates the basic adjustment amount, and generates an initial parameter adjustment scheme combined with the multi-axis collaborative control constraints, and then introduces a compensation factor to correct the parameters that exceed the accuracy range, and finally synchronously adjusts the motion parameters of each axis. This mechanism realizes high-precision dynamic parameter optimization, effectively solves the poor coordination and error accumulation problems in traditional multi-axis control, and improves the overall control precision and stability of three-axis linkage.

[0021] (3) The present application discloses a multi-axis motion component trajectory deviation dynamic correction method, which collects motion trajectory data in real time through a sensor network, identifies trajectory deviation using a deep learning model, constructs a deviation dynamic quantization mechanism to calculate the real-time change amplitude, decomposes the deviation components to extract relevant parameters, analyzes the inter-axis coupling relationship combined with multi-axis collaborative control constraints, and incorporates a compensation factor for mechanical wear effects to finally obtain a parameter adjustment scheme.

[0022] (4) The present application updates the execution instructions of the multi-axis motion control system in real time, synchronously adjusts the motion parameters of each axis, and completes the dynamic correction of trajectory deviation, effectively improving the motion precision and stability of the multi-axis motion component, and is suitable for high-precision motion control fields, which can significantly improve the motion trajectory control effect in complex environments. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a flowchart of a machine vision-based three-axis motion component control method provided by an embodiment of the present application; Figure 2 is a structural schematic diagram of a machine vision-based three-axis motion component control system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the protection scope of the present application.

[0025] With reference to Figure 1 The first embodiment of the present application provides a three-axis motion component control method based on machine vision, comprising the following steps: S101, acquiring a motion trajectory original data stream and pre-processing to obtain a trajectory feature data set; S102, analyzing the difference between an actual trajectory and a theoretical trajectory according to the trajectory feature data set to obtain a deviation distribution report; S103, calculating the real-time change amplitude of the deviation according to the deviation analysis report, combining historical data and current environmental fluctuation conditions to obtain a deviation value matrix; S104, extracting spatial displacement and angle offset data according to the deviation value matrix to obtain a basic adjustment amount; S105, calculating an initial parameter adjustment scheme for each axis according to the basic adjustment amount, combining the constraint conditions of multi-axis collaborative control to obtain a parameter optimization result; S106, incorporating a compensation factor into the parameter optimization result, recalculating the compensation factor for parameters whose parameter adjustment results exceed the preset precision range to obtain a final parameter adjustment scheme; S107, synchronously adjusting the motion parameters of each axis according to the final parameter adjustment scheme.

[0026] In step S101, the motion trajectory original data stream needs to be acquired and pre-processed to obtain a trajectory feature data set, including: Acquiring a motion trajectory original data stream containing position, velocity and acceleration information; Performing denoising processing on the original data stream to obtain a first data stream; Dividing the first data stream into time periods and marking time periods whose data fluctuations exceed a preset fluctuation threshold to obtain a second data stream; Performing motion trajectory simulation on the second data stream to obtain a trajectory feature set; Correcting the trajectory feature set by using a pre-established trajectory feature template to obtain a trajectory feature data set.

[0027] It should be noted that the motion trajectory original data stream contains position information, speed information and acceleration information of each axis of the multi-axis component, and the multi-axis component is collected in real time through a sensor network; Exemplarily, taking an industrial robot arm as a specific scene, the industrial robot arm performs a precise assembly task on a production line, sensors are deployed at multiple joints of the industrial robot arm, and position, speed and acceleration data of each axis are collected in real time to form a motion trajectory original data stream.

[0028] In a possible implementation, for the position information of the motion trajectory original data, a time window-based smoothing filtering method is used for denoising processing, for example, the noise amplitude of the position information is ±0.5 mm, and after filtering, the noise is reduced to ±0.1 mm, to obtain a more stable first data stream.

[0029] It should be noted that, for the time period division of the first data stream, the data is segmented into a time period every 5 seconds; feature extraction is performed on the data in each time period; data with a feature fluctuation exceeding a preset fluctuation threshold in a time period is marked to form a second data stream. This marking method helps to quickly locate the abnormal motion interval and facilitates subsequent analysis. Curve fitting is performed on the second data stream to simulate the motion trajectory of the multi-axis component under external interference; trajectory features such as curvature change and deviation angle are extracted from the motion trajectory to form a trajectory feature set.

[0030] In a possible implementation, the trajectory feature set is matched with a pre-established template, and a similarity calculation method based on the Euclidean distance can be used to determine whether the corrected trajectory feature meets the expected motion trajectory. Assuming that the ideal trajectory curvature defined in the template is 0.2, and the actual simulated trajectory curvature is 0.25, the system will determine that it is close to the expectation, but still needs to be fine-tuned, and finally generates a corrected trajectory feature data set.

[0031] In step S102, the difference between the actual trajectory and the theoretical trajectory is analyzed according to the trajectory feature data set to obtain a deviation distribution report, including: obtaining a trajectory feature value in the trajectory feature data set; comparing the trajectory feature value with a pre-established trajectory template to obtain a deviation distribution record table; dividing the deviation distribution record table according to a time sequence, statistically analyzing a deviation fluctuation range in each time sequence, and obtaining a deviation trend data set; correcting the deviation trend data set to obtain a deviation correction data set; performing future trajectory deviation simulation on the deviation correction data set to obtain a deviation distribution report.

[0032] It should be noted that when analyzing based on the preliminary trajectory feature data set, the characteristic values such as trajectory curvature and deviation angle can be extracted through a specific software platform. Assuming that the actual trajectory curvature value is 0.3 and the theoretical template value is 0.28, the system will record this difference; the deviation data can be divided into time periods of every 10 seconds. Assuming that the deviation value rises from 0.02 to 0.05 within a certain period of time, the fluctuation range can be determined as 0.03 after analyzing the trend; when correcting the deviation trend data set, if the deviation fluctuation range of a certain time period exceeds the preset threshold, the data needs to be dynamically adjusted through a specific correction algorithm to adjust the deviation value from 0.05 to within 0.03; In one possible implementation, for future trajectory deviation simulation, environmental interference factors such as mechanical vibration on the production line can be combined to simulate future trajectory deviation based on the corrected data. Assuming that the prediction result shows that the deviation may increase to 0.04 within the next 5 seconds, but is still within the acceptable range, the system will determine that it meets the expected trajectory and finally generate a deviation distribution report. This prediction mechanism can identify potential problems in advance and provide a reference for real-time monitoring.

[0033] In step S103, the real-time change amplitude of the deviation needs to be calculated based on the deviation analysis report, combined with historical data and current environmental fluctuation conditions, to obtain a deviation value matrix, including: Obtaining the deviation characteristic value in the deviation analysis report; Statistically analyzing the fluctuation of the deviation distribution based on the deviation characteristic value to obtain deviation analysis data; Dividing the time period based on the deviation analysis data, segmenting and labeling the deviation data with periodic fluctuations to obtain deviation trend characteristics; Comparing the deviation trend characteristics with real-time data and analyzing the fluctuation range to obtain a deviation amplitude record; Comparing the deviation amplitude record and the deviation analysis report with a preset deviation amplitude threshold, and performing normalization processing on the data exceeding the preset deviation amplitude threshold to obtain a deviation value matrix.

[0034] It should be noted that the deviation characteristic value such as trajectory deviation angle and speed change rate in the deviation analysis report is obtained; the deviation characteristic value is compared with the current environmental conditions such as temperature or humidity to obtain the fluctuation of the deviation characteristic value under different conditions to obtain deviation analysis data. Assuming that the deviation angle in the historical data is 0.2 degrees and the angle under the current environmental condition is 0.25 degrees, this difference may reflect the influence of environmental conditions on the deviation.

[0035] It is worth noting that the deviation analysis data is divided into time periods, combined with historical data and environmental conditions, to analyze the periodic trend characteristics of the deviation analysis data, and the time periods with periodic trend characteristics are marked to obtain the deviation trend characteristics. For example, in a certain period of time, the deviation value shows a significant rise and fall every 30 seconds. This periodic fluctuation can be determined by segmenting the segmented deviation trend set.

[0036] In one possible implementation, the deviation distribution in each time period is matched with the real-time amplitude to obtain the fluctuation range under the current condition. Assuming that the average value of the deviation distribution in a certain time period is 0.15 degrees, and the real-time amplitude is 0.18 degrees, the system will record this fluctuation range to form a deviation amplitude record. The deviation amplitude record and the deviation analysis report are compared with a preset deviation amplitude threshold, and data exceeding the preset deviation amplitude threshold is normalized to obtain a deviation value matrix. Assuming that the preset deviation threshold is 0.1 degrees, and the current deviation amplitude is 0.13 degrees, the data exceeding the range will be normalized to generate a quantized deviation value matrix.

[0037] In step S104, the spatial displacement and angular offset data are extracted from the deviation value matrix to obtain the basic adjustment amount, including: The offset parameter and the rotation angle in the deviation value matrix are extracted, and the offset parameter and the rotation angle are classified to obtain a classified deviation record. The classified deviation record is subjected to coordinate transformation processing, and the data exceeding the preset deviation threshold is subjected to boundary correction processing to obtain a displacement angle set. The spatial displacement data and the angular offset data in the displacement angle set are extracted, and the spatial displacement data and the angular offset data are subjected to data mapping processing to obtain a trajectory deviation distribution record. The trajectory deviation distribution record is subjected to motion trajectory correction and the error-prone point parameters are adjusted to obtain the basic adjustment amount.

[0038] It should be noted that the processing of the deviation decomposition result extracts the related data of the offset parameter and the rotation angle from the deviation value matrix, and classifies the offset parameter and the rotation angle to generate independent data sets. For example, assuming that the range of the offset parameter is between 0.1 and 0.3 mm, and the range of the rotation angle is between 0.2 and 0.5 degrees, two independent data sets are formed after classification, which facilitates subsequent targeted analysis to obtain the classified deviation record and determine the core parameter range.

[0039] It is worth noting that for the processing of classification bias records, preliminary processing of spatial displacement and angle offset is required, and the range is constrained by setting a preset threshold. Assuming that the displacement threshold is 0.2 mm and the angle threshold is 0.3 degrees, if the processing result shows that the displacement of a certain point is 0.25 mm, which exceeds the threshold, boundary correction is required, and the adjusted displacement angle set meets the requirements; compare the spatial displacement and angle offset with the motion trajectory, and analyze the deviation data of each trajectory point. Assuming that in a certain trajectory, the displacement deviation of a certain point is 0.15 mm and the angle deviation is 0.4 degrees, by comparing the historical trajectory data, it is determined that it is an abnormal point, and is recorded in the trajectory deviation distribution record.

[0040] In one possible implementation, when the motion trajectory is locally corrected, the displacement and angle range corresponding to the abnormal point can be focused on for parameter adjustment. Assuming that the displacement deviation of an abnormal point is 0.18 mm and the angle deviation is 0.35 degrees, the basic adjustment amount is obtained. By adjusting the parameters, it is corrected to close to the ideal value, and finally the corrected trajectory parameter combination is generated.

[0041] In step S105, the initial parameter adjustment scheme of each axis is calculated according to the basic adjustment amount and the constraint conditions of multi-axis cooperative control, and the parameter optimization result is obtained, including: obtain real-time motion state data of each axis; compare the real-time motion state data of each axis with the constraint conditions of multi-axis cooperative control to obtain a state data set; analyze the coupling relationship between the axes based on the state data set, generate an association data table, and extract key influence indicators from the association data table; independent control parameter calculation is performed on the key influence indicators to obtain the parameter optimization result.

[0042] It should be noted that the real-time motion state data of each axis, including the speed, position and other information of each axis, is obtained, and the real-time motion state data of each axis is compared with the constraint conditions of multi-axis cooperative control to obtain a state data set. For example, in the multi-axis cooperative control scene of an industrial robot arm, assuming that the upper limit of the running speed of a certain axis is 50 mm / s, and the real-time data shows that its speed reaches 55 mm / s, which exceeds the threshold of 5 mm, the exceeding part needs to be preliminarily limited, and the state data set that meets the constraints is obtained after adjustment.

[0043] It is worth noting that the constraint conditions of multi-axis cooperative control include the preset upper limit of the motion speed of each axis, the position tolerance range and the dynamic response time threshold, which are used as the basis for constraint judgment and parameter adjustment of the real-time motion state of each axis.

[0044] In a possible implementation, for processing of the initial state data set and analyzing the coupling relationship between the axes, the mechanical influence between the axes can be focused on. For example, acceleration of an axis can cause vibration of an adjacent axis, and a key influence index between the axes can be vibration frequency or torque transmission ratio, so as to determine the mutual influence range of the axes.

[0045] It should be noted that the key influence index includes the control parameter of each axis, and the control parameter is matched and adjusted according to the requirement of dynamic balance, to obtain an independent control parameter set suitable for each axis, and to ensure that the independent control parameter set meets the control precision standard; assuming that a certain axis needs to maintain stability in high-speed motion, the control parameter of the axis can need to reduce the upper limit of acceleration to 40 mm / s, and the control parameter of another axis can need to increase the response speed to 60 mm / s, to finally form an independent control parameter set suitable for each axis.

[0046] It should be noted that the independent control parameter set is used for local trajectory optimization of the overall motion trajectory, to obtain a final parameter optimization data set, and to ensure that the parameter optimization data set can adapt to the target requirement of multi-axis cooperative control; the local trajectory optimization can focus on parameter fine-tuning of a deviation point in the trajectory, for example, in the overall motion trajectory, the actual position of a local point deviates from the planned position by 0.2 mm, and the deviation is reduced to within 0.05 mm by fine-tuning the control parameter of the related axis, to finally obtain an optimized parameter data set. The local trajectory optimization can ensure the overall consistency of the trajectory, adapt to the target requirement of multi-axis cooperative control, and provide guarantee for execution of a precision task.

[0047] In step S106, the parameter optimization result needs to be fused with a compensation factor, and a compensation factor is recalculated for a parameter whose parameter adjustment result exceeds a preset accuracy range, to obtain a final parameter adjustment scheme, including: According to the parameter optimization result, mechanical wear history information of the motion component is extracted and compared, to obtain mechanical wear influence data; The mechanical wear influence data is used for compensation factor calculation, and compared with a preset compensation factor threshold value; when the threshold value is exceeded, the compensation factor is recalculated, to obtain a compensation factor data set; According to the compensation factor data set, the parameter optimization result is dynamically compensated, to obtain a final parameter adjustment scheme.

[0048] It should be noted that the mechanical wear history information of the moving parts needs to be extracted from the equipment operation record first, and the history information is compared with the preset wear influence range to obtain mechanical wear influence data; for example, the wear data in the long-term operation of the equipment is obtained through the sensor and the log system, such as the change of the friction coefficient of the bearing or the deformation record of the transmission part. Assuming that the friction coefficient of the bearing of a shaft increases from 0.02 to 0.05 in the past 6 months, which exceeds the preset wear influence range 0.03, the specific influence on the parameters needs to be further analyzed to generate the corresponding influence data set.

[0049] It should be noted that the potential interference of the mechanical wear influence data on the control accuracy is analyzed to obtain the compensation factor, and the compensation factor is compared with the preset compensation factor threshold. When the threshold is exceeded, the compensation factor is recalculated to obtain the compensation factor data set; assuming that the response delay of a shaft increases by 0.1 seconds due to wear, and the preset threshold is 0.05 seconds, the calculation logic needs to be adjusted after the range is exceeded to determine the compensation factor again. The updated compensation factor data set may reduce the acceleration parameters of the shaft by 10% to reduce further wear.

[0050] In a possible implementation, when the preliminary parameter adjustment scheme is dynamically compensated based on the updated compensation factor data set, the adjustment amplitude of each shaft can be matched with the accuracy requirement. Assuming that the adjustment amplitude of a shaft needs to be controlled within 5%, and the compensation factor suggests a reduction of 8%, the accuracy and compensation demand need to be further balanced to finally determine the adjusted parameter data set.

[0051] In step S107, the motion parameters of each shaft need to be adjusted synchronously according to the final parameter adjustment scheme, including: Obtaining real-time motion trajectory data of each shaft, performing deviation analysis on the real-time motion trajectory data of each shaft to obtain error detection results; Classifying the error detection results, adjusting the parameter scheme for error data exceeding the preset error threshold, and obtaining and updating the motion instructions of each shaft in real time; Adjusting the motion parameters of each shaft synchronously according to the motion instructions of each shaft, monitoring the motion state of each shaft and comparing it with the operation standard to obtain the control output result.

[0052] It should be noted that the real-time motion trajectory data of each shaft needs to be obtained from the running state of each shaft through high-precision sensors. For example, when an industrial robot arm performs a welding task, the sensor detects that the actual trajectory of a shaft deviates from the preset path by 0.2 millimeters, and the preset trajectory planning range is 0.1 millimeters. The deviation needs to be further analyzed after the threshold is exceeded.

[0053] It is worth noting that for error detection results, the deviation data needs to be classified and arranged. The deviation can be classified into systematic deviation and random deviation according to types. Assuming that the systematic deviation of a certain shaft accounts for 70% of the deviation data, which is mainly caused by the loosening of components caused by long-term operation, and the random deviation accounts for 30%, which may be related to external environmental interference.

[0054] In a possible implementation, when determining the parameter synchronization scheme, the classified deviation data needs to be matched with the inter-shaft coordination requirement. Assuming that the systematic deviation of a certain shaft causes a 0.3-second delay in the coordinated motion with other shafts, and the coordination requirement is within 0.1 second, the motion parameters of the shaft need to be adjusted, such as reducing the running speed or increasing the pre-compensation time; the motion instructions of each shaft are updated in real time based on the parameter synchronization scheme, and the motion state of each shaft is monitored in real time according to the adjusted instruction data set, the monitoring data is compared with the running stability standard to determine whether the trajectory deviation correction requirement is met, and the final control output result is obtained. For example, the monitoring data shows that a certain shaft still has slight jitter after adjustment, but the jitter amplitude is controlled within 0.02 millimeters, which meets the running stability standard, so it can be judged that it meets the correction requirement.

[0055] In summary, the present application discloses a three-axis motion component control method based on machine vision, comprising: obtaining a motion trajectory original data stream, and preprocessing to obtain a trajectory feature data set; analyzing the difference between the actual trajectory and the theoretical trajectory according to the trajectory feature data set, and obtaining a deviation distribution report; calculating the real-time change amplitude of the deviation according to the deviation analysis report, combining historical data and current environmental fluctuation conditions, and obtaining a deviation value matrix; extracting spatial displacement and angle offset data according to the deviation value matrix, and obtaining a basic adjustment amount; calculating the initial parameter adjustment scheme of each shaft according to the basic adjustment amount, combining the constraint conditions of multi-axis collaborative control, and obtaining a parameter optimization result; re-computing the compensation factor for the parameters whose parameter adjustment results exceed the preset accuracy range based on the parameter optimization result, and obtaining a final parameter adjustment scheme; and synchronously adjusting the motion parameters of each shaft according to the final parameter adjustment scheme.

[0056] The present application discloses a multi-axis motion component trajectory deviation dynamic correction method, which collects motion trajectory data in real time through a sensor network, identifies trajectory deviation using a deep learning model, constructs a deviation dynamic quantization mechanism to calculate real-time change amplitude, decomposes deviation components to extract related parameters, analyzes inter-shaft coupling relationship combining multi-axis collaborative control constraint conditions, and finally obtains a parameter adjustment scheme by integrating a compensation factor for mechanical wear influence.

[0057] The application updates the execution instruction of the multi-axis motion control system in real time, synchronously adjusts the motion parameters of each axis, completes dynamic correction of trajectory deviation, effectively improves the motion precision and stability of the multi-axis motion component, is suitable for the field of high-precision motion control, and can significantly improve the motion trajectory control effect under a complex environment.

[0058] Referring to Figure 2 The application provides a three-axis motion component control system based on machine vision, comprising: A data acquisition module is configured to acquire a motion trajectory original data stream of a multi-axis motion component in real time through a sensor network, wherein the original data stream contains position, speed and acceleration information. A data preprocessing module is configured to preprocess the acquired original data stream to obtain a trajectory feature data set. A deviation identification and analysis module is configured to compare the trajectory feature data set with a preset theoretical trajectory template, identify and analyze the difference between an actual trajectory and a theoretical trajectory, and determine the distribution and variation trend of the deviation. A deviation quantification module is configured to calculate the real-time variation amplitude of the deviation according to the deviation analysis report, combine historical data and current environmental fluctuation conditions, and obtain a deviation value matrix. A parameter extraction and correction module is configured to extract spatial displacement and angle offset data according to the deviation value matrix, and obtain a basic adjustment amount. A multi-axis coordination and optimization module is configured to combine the basic adjustment amount with the constraint conditions of multi-axis coordination control, calculate an initial parameter adjustment scheme of each axis, and obtain a parameter optimization result. A motion control instruction generation and execution module is configured to integrate a compensation factor into the parameter optimization result, recalculate the compensation factor for a parameter adjustment result that exceeds a preset precision range, obtain a final parameter adjustment scheme, and synchronously adjust the motion parameters of each axis according to the final parameter adjustment scheme.

[0059] It should be noted that the three-axis motion component control system based on machine vision provided by the embodiment of the application is used to execute all process steps of the three-axis motion component control method based on machine vision provided by the above embodiment, and the working principles and beneficial effects of the two are one-to-one corresponding, and thus will not be repeated.

[0060] The embodiment of the application further provides an electronic device. The electronic device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a three-axis motion component control program based on machine vision. The processor implements the steps in the above-mentioned three-axis motion component control method based on machine vision when executing the computer program, such as Figure 1The step S101 is shown. Alternatively, the processor implements the functions of each module / unit in the above-mentioned device embodiments when executing the computer program, such as the motion control instruction generation and execution module.

[0061] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device.

[0062] The electronic device can be a desktop computer, a notebook computer, a palm computer, a smart tablet and the like. The electronic device can include, but is not limited to, a processor, a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device, and can include more or less components than the above, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus and the like.

[0063] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, which connects all parts of the electronic device through various interfaces and lines.

[0064] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer program and / or modules stored in the memory, and calling data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), and the like. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0065] The modules / units integrated in the electronic device, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can realize the steps of the above-mentioned various method embodiments when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the contents included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0066] It should be noted that the apparatus embodiments described above are merely illustrative, and the units described as separate units can or can not be physically separate, and the units displayed as units can or can not be physical units, i.e. can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. In addition, the connection relationship between the modules in the apparatus embodiment provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0067] The above specific embodiments further illustrate the purpose, technical scheme and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A three-axis motion component control method based on machine vision, characterized in that: Executed by a computer, including: Obtain the original data stream of the motion trajectory and preprocess it to obtain the trajectory feature data set; Analyze the difference between the actual trajectory and the theoretical trajectory according to the trajectory feature data set to obtain a deviation distribution report; According to the deviation analysis report, combined with historical data and current environmental fluctuation conditions, the real-time change range of the deviation is calculated to obtain a deviation value matrix; Extracting spatial displacement and angular offset data according to the deviation value matrix to obtain a basic adjustment amount; Based on the basic adjustment amount and in combination with the constraints of multi-axis coordinated control, an initial parameter adjustment scheme for each axis is calculated to obtain a parameter optimization result; Incorporating compensation factors into the parameter optimization results, recalculating compensation factors for parameters whose parameter adjustment results exceed a preset accuracy range, and obtaining a final parameter adjustment solution; According to the final parameter adjustment scheme, the motion parameters of each axis are adjusted synchronously.

2. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: The process of obtaining the original data stream of the motion trajectory and preprocessing it to obtain a trajectory feature data set includes: Obtain the original data stream of the motion trajectory containing position, velocity and acceleration information; Performing denoising on the original data stream to obtain a first data stream; Dividing the first data stream into time periods, marking time periods in which data fluctuations exceed a preset fluctuation threshold, and obtaining a second data stream; Performing motion trajectory simulation on the second data stream to obtain a trajectory feature set; The trajectory feature set is corrected using a pre-established trajectory feature template to obtain a trajectory feature data set.

3. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: Based on the trajectory feature dataset, the difference between the actual trajectory and the theoretical trajectory is analyzed to obtain a deviation distribution report, including: Obtaining trajectory feature values ​​in the trajectory feature dataset; Comparing the trajectory characteristic value with a pre-established trajectory template to obtain a deviation distribution record table; Dividing the deviation distribution record table into time series, performing statistical analysis on the deviation fluctuation range within each time series, and obtaining a deviation trend data set; performing correction processing on the deviation trend data set to obtain a deviation-corrected data set; Performing a future trajectory deviation simulation on the deviation-corrected data set to obtain a deviation distribution report.

4. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: According to the deviation analysis report, combined with historical data and current environmental fluctuation conditions, the real-time variation of the deviation is calculated to obtain the deviation value matrix, including: Obtaining the deviation characteristic value in the deviation analysis report; Statistically analyzing the fluctuation of the deviation distribution of the deviation characteristic value to obtain deviation analysis data; Divide the deviation analysis data into time periods, mark the deviation data with periodic fluctuations in sections, and obtain deviation trend characteristics; Comparing the deviation trend characteristics with real-time data, and performing fluctuation range analysis to obtain deviation amplitude records; The deviation amplitude record and the deviation analysis report are compared with a preset deviation amplitude threshold, and data exceeding the preset deviation amplitude threshold is normalized to obtain a deviation value matrix.

5. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: According to the deviation value matrix, the spatial displacement and angle offset data are extracted to obtain the basic adjustment amount, including: extracting the offset parameters and the rotation angles from the deviation value matrix, and classifying the offset parameters and the rotation angles to obtain classified deviation records; Performing coordinate transformation processing on the classified deviation records, performing boundary correction processing on data exceeding a preset deviation threshold, and obtaining a displacement angle set; extracting spatial displacement data and angular offset data from the displacement angle set, performing data mapping processing on the spatial displacement data and the angular offset data, and obtaining a trajectory deviation distribution record; The trajectory deviation distribution record is corrected and the error-prone point parameters are adjusted to obtain a basic adjustment amount.

6. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: Based on the basic adjustment amount and combined with the constraints of multi-axis coordinated control, the initial parameter adjustment scheme for each axis is calculated to obtain the parameter optimization results, including: Get real-time motion status data of each axis; Comparing the real-time motion state data of each axis with the constraint conditions of the multi-axis coordinated control to obtain a state data group; Analyzing the inter-axis coupling relationship of the state data group, generating a correlation data table, and extracting key impact indicators from the correlation data table; Independent control parameter calculation is performed on the key influencing indicators to obtain parameter optimization results.

7. The three-axis motion component control method based on machine vision according to claim 6, characterized in that: The constraints of the multi-axis coordinated control include: The constraints of the multi-axis collaborative control include preset upper limits on the motion speed of each axis, position tolerance ranges, and dynamic response time thresholds, which are used to perform constraint judgment and parameter adjustment on the real-time motion state of each axis.

8. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: Incorporating compensation factors into the parameter optimization results, recalculating compensation factors for parameters whose parameter adjustment results exceed the preset accuracy range, and obtaining a final parameter adjustment solution, including: Extracting mechanical wear history information of moving parts according to the parameter optimization results and performing comparative processing to obtain mechanical wear impact data; Calculating a compensation factor for the mechanical wear impact data and comparing it with a preset compensation factor threshold; recalculating the compensation factor when the threshold is exceeded to obtain a compensation factor data set; The parameter optimization result is dynamically compensated according to the compensation factor data set to obtain a final parameter adjustment solution.

9. The three-axis motion component control method based on machine vision according to claim 1, characterized in that: According to the final parameter adjustment scheme, the motion parameters of each axis are adjusted synchronously, including: Acquire real-time motion trajectory data of each axis, perform deviation analysis on the real-time motion trajectory data of each axis, and obtain error detection results; Classify and process the error detection results, adjust the parameter scheme for the error data exceeding the preset error threshold, and obtain and update the motion instructions of each axis in real time; The motion parameters of each axis are synchronously adjusted according to the motion instructions of each axis, the motion state of each axis is monitored and compared with the operation standard to obtain a control output result.

10. A three-axis motion component control system based on machine vision, characterized in that: include: A data acquisition module is used to collect the original data stream of the motion trajectory of the multi-axis motion component in real time through a sensor network, wherein the original data stream includes position, velocity and acceleration information; The data preprocessing module is used to preprocess the collected raw data stream to obtain the trajectory feature data set; Deviation identification and analysis module, used to compare the trajectory feature data set with the preset theoretical trajectory template, identify and analyze the differences between the actual trajectory and the theoretical trajectory, and determine the distribution and change trend of the deviation; A deviation quantification module is used to calculate the real-time variation of the deviation based on the deviation analysis report, combined with historical data and current environmental fluctuation conditions, to obtain a deviation value matrix; A parameter extraction and correction module is used to extract spatial displacement and angular offset data according to the deviation value matrix to obtain a basic adjustment value; The multi-axis coordination and optimization module is used for basic adjustment. It combines the constraints of multi-axis coordinated control to calculate the initial parameter adjustment plan for each axis and obtain the parameter optimization results. The motion control instruction generation and execution module is used to incorporate the compensation factor into the parameter optimization result, recalculate the compensation factor for the parameters whose parameter adjustment results exceed the preset accuracy range, obtain the final parameter adjustment plan, and synchronously adjust the motion parameters of each axis according to the final parameter adjustment plan.

Images