A master-slave axis coupling control method, system, terminal device and medium

Abstract

The present invention discloses a master-slave axis coupling control method, system, terminal device and medium. The method includes: obtaining sample data of a master axis and a slave axis, and preprocessing the sample data using a filtering method to obtain observation data of the master axis and the slave axis; performing parameter identification based on the observation data to update system parameters of a control quantity model; calculating a motion distance control quantity based on the control quantity model and the observation data, and performing coupling control on the master-slave axis based on the motion distance control quantity. By performing real-time optimization on sampled data and combining parameter identification technology to dynamically adjust the control strategy, the present invention realizes gradual optimization and dynamic adjustment during the motion control process, and solves the problems of high noise sensitivity, insufficient parameter identification and lack of process optimization in the existing profiling control technology.

Description

Technical Field

[0001] The present invention relates to the technical field of master-slave axis coupling control, and particularly to a master-slave axis coupling control method, system, terminal device and medium for surface precision machining. Background Art

[0002] In the field of motion control, master-slave axis coupled motion is widely used in precision manufacturing equipment such as numerically controlled machine tools and industrial robots. In order to achieve high-precision master-slave axis motion cooperation, a contouring control method is usually adopted, that is, the slave axis is driven for synchronous control through the position, speed or acceleration signal of the master axis. However, traditional contouring control methods are often difficult to ensure the robustness and motion accuracy of the system when facing complex environments (such as signal noise and uncertain parameter changes).

[0003] In the prior art, contouring control depends on real-time signals. Once noise is introduced by sensors or the environment, it is easy to cause an increase in following error. Moreover, traditional methods mainly focus on the matching of output trajectories, rather than the data optimization of the motion process itself, and fail to effectively improve the overall performance of the system. Further, in a complex system, the kinematic parameters of the master-slave axis may change over time, and existing methods are difficult to identify parameters in real time and adjust control strategies. Finally, data-driven technologies have not been organically combined with filtering algorithms and parameter identification methods, and it is difficult to achieve global optimization of complex dynamic coupling processes. Specifically, in the prior art, traditional contouring control based on PID realizes synchronization between the master and slave axes through simple proportional, integral and derivative controls. However, due to the single model, it is difficult to adapt to complex non-linear motions. And the improved method based on feedforward compensation adds feedforward compensation in contouring control, which can improve the system response speed to a certain extent, but for complex coupled dynamic systems, the compensation effect is limited, and the parameters are difficult to accurately identify. Therefore, the prior art still needs to be improved. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a master-slave axis coupling control method, system, terminal device and medium aiming at the above-mentioned defects of the prior art, aiming to solve the problems of high noise sensitivity, insufficient parameter identification and lack of process optimization in the existing contouring control technology.

[0005] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0006] In a first aspect, the present invention provides a master-slave axis coupling control method, the method comprising:

[0007] Obtain sample data of the master axis and the slave axis, and preprocess the sample data using a filtering method to obtain observed data of the master axis and the slave axis;

[0008] Perform parameter identification based on the observed data to update the system parameters of the control quantity model;

[0009] Based on the control quantity model and the observed data, calculate the motion distance control quantity, and perform coupled control on the main shaft and the slave shaft based on the motion distance control quantity.

[0010] In one implementation, the obtaining of the sample data of the main shaft and the slave shaft includes:

[0011] Collect the encoder position data and sensor data of the main shaft and the slave shaft during an interruption;

[0012] Store the encoder position data and sensor data into the sample buffer in a first-in-first-out manner to obtain the sample data of the main shaft and the slave shaft.

[0013] In one implementation, the preprocessing of the sample data using a filtering method to obtain the observed data of the main shaft and the slave shaft includes:

[0014] The filtering method is the extended Kalman filter;

[0015] Use the extended Kalman filter to perform state estimation, error correction, and smoothing processing on the sample data to obtain the observed data of the main shaft and the slave shaft.

[0016] In one implementation, the performing of parameter identification based on the observed data to update the system parameters of the control quantity model includes:

[0017] The system parameters include the input-output gain coefficient and the bias coefficient;

[0018] Input the observed data into the parameter estimator of the pre-trained deep neural network to update the input-output gain coefficient and the bias coefficient of the control quantity model.

[0019] In one implementation, the calculating of the motion distance control quantity based on the control quantity model and the observed data includes:

[0020] The observed data includes the current positions of the main shaft and the slave shaft;

[0021] Input the observed data into the control quantity model to calculate the expected positions of the main shaft and the slave shaft;

[0022] Based on the deviation between the expected position and the current position, calculate the control quantity of the motion distance.

[0023] In one implementation, the calculating of the control quantity of the motion distance based on the deviation between the expected position and the current position includes:

[0024] Calculate the sliding surface parameters based on the deviation between the desired position and the current position to obtain an adaptive sliding surface.

[0025] Calculate the motion distance control quantity based on the adaptive sliding surface.

[0026] In one implementation, the method further includes:

[0027] Add an additional sensor at the actual machining position, and the sensor is used to collect error data in real time after the system input.

[0028] Perform secondary parameter identification on the error data to obtain the input-output gain coefficient and bias coefficient of the compensation quantity model of the motion distance control quantity.

[0029] Calculate a compensation value based on the compensation quantity model and the observation data.

[0030] Perform error compensation on the motion distance control quantity based on the compensation value.

[0031] In a second aspect, an embodiment of the present invention further provides a master-slave axis coupling control system, and the system includes:

[0032] An observation data acquisition module, configured to acquire sample data of the main axis and the slave axis, and preprocess the sample data using a filtering method to obtain the observation data of the main axis and the slave axis.

[0033] A control quantity model acquisition module, configured to perform parameter identification based on the observation data and update the system parameters of the control quantity model.

[0034] A control quantity acquisition module, configured to calculate the motion distance control quantity based on the control quantity model and the observation data, and perform coupling control on the main axis and the slave axis based on the motion distance control quantity.

[0035] In a third aspect, an embodiment of the present invention further provides a terminal device, and the terminal device includes a memory, a processor, and a master-slave axis coupling control program stored in the memory and executable on the processor. When the processor executes the master-slave axis coupling control program, the steps of the master-slave axis coupling control method in any one of the above solutions are implemented.

[0036] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a master-slave axis coupling control program is stored. When the master-slave axis coupling control program is executed by a processor, the steps of the master-slave axis coupling control method in any one of the above solutions are implemented.

[0037] Beneficial effects: The present invention discloses a master-slave axis coupling control method, system, terminal device and medium. Compared with the prior art, the method first obtains the sample data of the master axis and the slave axis, and preprocesses the sample data using a filtering method to obtain the observed data of the master axis and the slave axis; then, based on the observed data, parameter identification is performed to update the system parameters of the control quantity model; finally, based on the control quantity model and the observed data, the motion distance control quantity is calculated, and based on the motion distance control quantity, the master-slave axes are coupled and controlled. By optimizing the sampling data in real time, combining parameter identification technology to dynamically adjust the control strategy, and using the first-in-first-out motion controller buffer management algorithm to realize the real-time incremental optimization of sample parameters during the motion control process, the present invention realizes the coupling control of the master-slave axis motion control quantity and the target state value of the actuator, realizes high-precision profiling motion control, and solves the problems of high noise sensitivity, insufficient parameter identification and lack of process optimization in the existing profiling control technology. Description of the Drawings

[0038] Figure 1 It is a flowchart of the specific implementation manner of the master-slave axis coupling control method provided by the embodiment of the present invention.

[0039] Figure 2 It is a motion control flowchart provided by the embodiment of the present invention.

[0040] Figure 3 It is a flowchart of sample data processing, adaptive controller generation and control quantity calculation provided by the embodiment of the present invention.

[0041] Figure 4 It is a flowchart of the master-slave axis data coupling and the first-in-first-out algorithm to realize real-time incremental optimization provided by the embodiment of the present invention.

[0042] Figure 5 It is a principle block diagram of the master-slave axis coupling control device provided by the embodiment of the present invention.

[0043] Figure 6 It is an internal structure principle block diagram of the intelligent terminal provided by the embodiment of the present invention. Detailed Description of the Invention

[0044] To make the purpose, technical solution and effects of the present invention clearer and more definite, the following further describes the present invention in detail with reference to the accompanying drawings and by way of examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0045] The flowcharts shown in the accompanying drawings are only illustrative examples and do not necessarily include all contents, operations or steps, nor do they have to be executed in the described order. For example, some operations or steps can be decomposed, combined or partially merged, so the actual execution order may change according to the actual situation.

[0046] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0047] It should be understood that in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. For example, the first control information and the second control information are only used to distinguish different control information and do not limit their sequence.

[0048] Those skilled in the art can understand that terms such as "first" and "second" do not limit the quantity and execution order, and "first", "second" and the like do not necessarily mean different.

[0049] It should also be understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0050] In the prior art, contouring control depends on real-time signals. Once noise is introduced by sensors or the environment, it is easy to cause an increase in following error. Moreover, traditional methods mainly focus on the matching of the output trajectory rather than the data optimization of the motion process itself, and fail to effectively improve the overall performance of the system. Further, in a complex system, the kinematic parameters of the master and slave axes may change over time, and it is difficult for existing methods to identify the parameters in real time and adjust the control strategy. Finally, data-driven technologies have not been organically combined with filtering algorithms and parameter identification methods, making it difficult to achieve global optimization of complex dynamic coupling processes. Specifically, in the prior art, traditional contouring control based on PID realizes synchronization between the master and slave axes through simple proportional, integral and differential controls. However, due to the single model, it is difficult to adapt to complex non-linear motions. And the improved method based on feed-forward compensation adds feed-forward compensation in contouring control, which can improve the system response speed to a certain extent, but for complex coupled dynamic systems, the compensation effect is limited and the parameters are difficult to accurately identify.

[0051] To solve the problems of the existing technology, the present invention provides a master-slave axis coupling control method. By optimizing the sampled data in real time, combining parameter identification technology to dynamically adjust the control strategy, and using the first-in-first-out motion controller buffer management algorithm in the motion control process to achieve real-time incremental optimization of sample parameters, the coupling control of the motion control amounts of the master-slave axes and the target state values of the actuator is realized, high-precision contouring motion control is achieved, and the problems of high noise sensitivity, insufficient parameter identification, and lack of process optimization in the existing contouring control technology are solved. In other words, the present invention can significantly improve the collaborative motion accuracy of the master-slave axes and the robustness of the system, especially showing superiority in high-noise and complex dynamic environments. Specifically, the enhanced robustness can effectively suppress the influence of sensor noise on the motion accuracy; by real-time parameter identification and dynamically adjusting the control strategy to adapt to various operation scenarios, the system obtains parameter self-adaptation ability; by combining data-driven and filtering technologies, the observed data is optimized while being read to ensure the continuity and smoothness of the motion process.

[0052] In the technical solution of the present invention, the specific motion control process is as Figure 2 shown. Among them, first, sample data processing is performed, then an adaptive sliding mode controller is generated, and then the parameters of the adaptive controller are calculated in real time. If the conditions are not met, it returns for re-calculation. After the conditions are met, the adaptive controller calculates the control amount. If the conditions are not met, it also needs to return for re-calculation. Then, the master-slave axis data is coupled, and the control amount is stored in the first-in-first-out buffer. During this process, it is judged whether the buffer is full. If it is not full, it continues to store. If it is full, the cam motion mode is started. Finally, when the cam is moving, it is judged whether the motion is in place. If it is not in place, the cam motion mode is restarted. If it is in place, the motion control process ends.

[0053] A master-slave axis coupling control method provided by this embodiment, as Figure 1 shown, specifically includes the following steps:

[0054] Step S100, obtain the sample data of the main axis and the slave axis, and preprocess the sample data using a filtering method to obtain the observed data of the main axis and the slave axis.

[0055] In this embodiment, relevant operation data of the spindle and the slave axis in the profiling control system are obtained by sampling. The operation data obtained by sampling are sample data, which may include position data, speed data, acceleration data, force or torque data, etc. Specifically, the position data represents the real-time position information of the spindle and the slave axis during movement, and is used to determine their current states and movement trajectories; the speed data represents the movement speeds of the spindle and the slave axis, and is used to analyze the dynamic characteristics of the movement; the acceleration data represents the acceleration information of the spindle and the slave axis, which helps to evaluate the smoothness and response speed of the movement; the force or torque data represents the force or torque borne by the spindle and the slave axis, and is used to analyze the load conditions and control effects. After obtaining the sample data, a filtering method is used to preprocess the above data to achieve data optimization and smoothing control. The filtering method may include low-pass filtering, high-pass filtering, band-pass filtering, Kalman filtering, moving average filtering, etc., and may also include advanced filtering methods such as particle filtering or nonlinear observers. The purpose of the filtering method is to remove high-frequency, low-frequency, random noises and interferences and smooth the data. After preprocessing the sample data, the observed data of the spindle and the slave axis are obtained, and the observed data are the preprocessed sample data.

[0056] In one implementation, the obtaining of the sample data of the spindle and the slave axis specifically includes the following steps:

[0057] Step S110: Collect the encoder position data and sensor data of the spindle and the slave axis in the interrupt;

[0058] Step S120: Store the encoder position data and sensor data into the sample buffer in a first-in-first-out manner to obtain the sample data of the spindle and the slave axis.

[0059] In this embodiment, as Figure 3 shown, first, the position data of the spindle and the slave axis are obtained through the encoder. The position data may include absolute position, relative position, and position increment. Then, through the sensor, speed data, acceleration data, force or torque data are obtained. Specifically, the above data can be obtained through actual values such as the voltage analog quantity of the pressure value read by the sensor or the voltage analog quantity of the distance. Immediately afterwards, the above encoder position data and sensor data are stored in the buffer in a first-in-first-out manner. Using the first-in-first-out manner to store the above data in the buffer is to ensure the timeliness and data consistency of the sampled data. The above data stored in the buffer are used as sample data for subsequent data processing. Further, it is judged whether the sample data meet the limitations of the profiling control. If they meet, the next step of data processing is entered. If there are abnormal data, the encoder and sensor data are read again.

[0060] In one implementation, the sample data is preprocessed using a filtering method to obtain the observed data of the main axis and the slave axis, which specifically includes the following steps:

[0061] Step S130: The filtering method is the extended Kalman filter;

[0062] Step S140: Use the extended Kalman filter to perform state estimation, error correction, and smoothing processing on the sample data to obtain the observed data of the main axis and the slave axis.

[0063] In this embodiment, the filtering method is specifically selected as the extended Kalman filter. As Figure 3 shown, the extended Kalman filter performs state estimation, error correction, and smoothing processing on the sample data. The extended Kalman filter performs real-time processing on the sampled data with large noise to ensure the smooth transition of the sampled data and avoid sudden jumps in motion caused by noise interference. Compared with traditional simple filtering methods, the extended Kalman filter has more advantages in noise suppression and data continuity guarantee. Among them, in the actual master-slave axis coupling control scenario, the system is often non-linear. For example, the motion of the master and slave axes may be affected by various non-linear factors, such as friction, load changes, etc. The extended Kalman filter can effectively perform state estimation on non-linear systems. It performs linearization approximation processing on the non-linear system and uses the framework of the Kalman filter to estimate the state of the system. Further, the sample data of the master and slave axes includes multi-source data such as encoder position data and sensor data. The extended Kalman filter can effectively fuse these different types of data. Considering the data from different sensors comprehensively, these data are weighted processed to obtain accurate observed data. At the same time, during the motion of the master and slave axes, due to the existence of various interference factors, there will be certain errors in the collected data. The extended Kalman filter can correct these errors in real time. Similarly, the sampled master-slave axis sample data may have noise, resulting in large data fluctuations. The extended Kalman filter can smooth these data. Through its filtering mechanism, while retaining the main features of the data, it reduces the influence of noise on the data. For example, for the high-frequency noise generated in the encoder position data due to mechanical vibration and other reasons, it can be effectively filtered out, making the obtained observed data smoother.

[0064] During the sampling process of the above data, traditional embedded systems solely rely on general-purpose processors to execute the sampling task. Limited by their clock frequency and architecture characteristics, the sampling frequency often fails to meet high-frequency requirements. Therefore, to further improve the system performance, dedicated hardware acceleration units, such as digital signal processors (DSPs) and field-programmable gate arrays (FPGAs), are added to the embedded system to accelerate the computational efficiency of the data sampling and filtering optimization process.

[0065] Step S200: Perform parameter identification based on the observed data to update the system parameters of the control quantity model.

[0066] In this embodiment, the control quantity model is used to obtain the expected positions of the master and slave axes. Specifically, the input of this model is the observed data, that is, the position data of the master and slave axes, etc., and the output is the expected positions of the master and slave axes. The system parameters are also included in this model. The system parameters are updated in real time through parameter identification, that is, the control quantity model is updated through parameter identification.

[0067] In one implementation, the step of performing parameter identification based on the observed data to update the system parameters of the control quantity model specifically includes the following steps:

[0068] Step S210: The system parameters include input-output gain coefficients and bias coefficients;

[0069] Step S220: Input the observed data into the parameter estimator of the pre-trained deep neural network to update the input-output gain coefficients and bias coefficients of the control quantity model.

[0070] In this embodiment, as Figure 3 shown, parameter identification is performed based on the observed data. Parameter identification is used to obtain system parameters, and the system parameters include input-output gain coefficients and bias coefficients. Further, the system parameters may also include parameters such as inertia, damping coefficient, and delay. Among them, the output gain coefficient reflects the amplification factor relationship between the system input signal and the output signal. The gain coefficient quantifies the degree of enhancement or weakening of the output signal corresponding to the change of the input signal. If the input is a small position deviation signal of the main axis, after passing through the control system, the output is a relatively large displacement adjustment amount of the slave axis. The bias coefficient is the output offset that the system itself has when there is no input signal. Even if the position deviation feedback by the encoder is zero and no additional load is detected by the sensor, the motor or actuator may still have an initial output state due to inherent factors such as mechanical installation error and electromagnetic characteristics. The quantization value corresponding to this initial output state is the bias coefficient.

[0071] To obtain the above system parameters, a parameter identification method can be used, or a parameter estimator of a pre-trained deep neural network can be used. If a parameter estimator is used, by setting a loss function, the historical data of the master and slave axes covering different working conditions is used as training data to train and obtain the parameter estimator of the pre-trained deep neural network. The loss function can use the mean square error function to minimize the error between the predicted gain coefficient, bias coefficient and the actual value. The iterative training method can use the stochastic gradient descent algorithm. In each round of training, a batch of data is input into the network, the loss is calculated, and the network weights are updated by backpropagation to continuously optimize the network parameters until the loss on the validation set no longer decreases significantly and reaches a convergence state, obtaining the parameter estimator of the pre-trained deep neural network. After obtaining the parameter estimator of the pre-trained deep neural network, the observed data of the master and slave axes is input into the parameter estimator, and the real-time system parameters, that is, the real-time input-output gain coefficient and bias coefficient, can be obtained. Further, the system parameters are updated into the control quantity model.

[0072] Step S300: Based on the control quantity model and the observed data, calculate the motion distance control quantity, and based on the motion distance control quantity, perform coupled control on the main axis and the slave axis.

[0073] In this embodiment, after the system parameters of the control quantity model are updated in real time through the above steps, the control quantity model can be used to calculate the motion distance control quantity. The control quantity is used to design the parameters of a PD controller (Proportional-Derivative Controller) in real time, that is, the parameters of the actuator, to realize the coupled control of the motion control quantity of the master and slave axes and the target state value of the actuator, and realize high-precision contouring motion control. The target state value of the actuator includes the voltage analog quantity of the pressure sensor or the voltage analog quantity of the distance sensor.

[0074] In one implementation manner, the calculating the motion distance control quantity based on the control quantity model and the observed data specifically includes the following steps:

[0075] Step S310: The observed data includes the current positions of the main axis and the slave axis;

[0076] Step S320: Input the observed data into the control quantity model to calculate the expected positions of the main axis and the slave axis;

[0077] Step S330: Based on the deviation between the expected position and the current position, calculate the control quantity of the motion distance.

[0078] In this embodiment, the observed data includes the current positions of the master and slave axes. As Figure 3As shown, input the current position into the control quantity model, and predict and calculate the expected positions of the master and slave axes. Calculate the control quantity of the movement distance based on the deviation between the expected position and the current position. The control quantity is used for the coupled control of the subsequent actuator.

[0079] In one implementation manner, calculating the control quantity of the movement distance based on the deviation between the expected position and the current position of the master and slave axes specifically includes the following steps:

[0080] Step S331: Calculate the sliding surface parameters based on the deviation between the expected position and the current position to obtain an adaptive sliding surface;

[0081] Step S332: Calculate the control quantity of the movement distance based on the adaptive sliding surface.

[0082] In this embodiment, as Figure 3 shown, adaptive sliding mode control is introduced when calculating the control quantity of the movement distance. Specifically, first construct a sliding mode surface, that is, a sliding surface. The sliding surface is a function of the system state variables, and it plans the ideal trajectory for the system to tend from any initial state to the expected steady state. First, through the previous steps, obtain the deviation between the expected position and the current position of the master and slave axes. The deviation is a vector or scalar value, reflecting the distance between the actual operating state of the system and the ideal target. In a second-order system, the first derivative (velocity deviation) and second derivative (acceleration deviation) of the deviation are combined with preset weight coefficients to construct a formula for calculating the sliding surface parameters, and the deviation is associated with the sliding surface parameters through the formula. Substitute the calculated sliding surface parameters into the sliding surface function expression to obtain the adaptive sliding surface. Through the adaptive sliding surface, obtain the control quantity of the movement distance. According to the target value of the slave axis state to be maintained (such as the voltage analog quantity of the pressure value or the voltage analog quantity of the distance), through the first-in-first-out sequential processing method, the sliding surface parameters of the sliding mode control are updated in real time to gradually approach the target value.

[0083] In one implementation manner, the method further includes the following steps:

[0084] Step S410: Additionally set a sensor at the actual machining position, and the sensor is used to collect the error data after the system input in real time;

[0085] Step S420: Perform secondary parameter identification on the error data to obtain the input-output gain coefficient and bias coefficient of the compensation quantity model of the control quantity of the movement distance;

[0086] Step S430: Calculate the compensation value based on the compensation quantity model and the observed data;

[0087] Step S440: Perform error compensation on the control quantity of the movement distance based on the compensation value.

[0088] In this embodiment, at the actual machining position, an additional sensor is deployed, and the sensor can be a high-precision laser displacement sensor. This sensor captures the deviation information between the machining position and the ideal model in real time, and uses this as the error data after system input. After the error data is collected, the system parameters of the compensation amount model of the control amount are obtained by using the method of parameter identification. The parameter identification method can select algorithms such as the least squares method. The system parameters of the compensation amount model specifically include the input-output gain coefficient and the bias coefficient. The observed data includes the current speed, acceleration, and position information of the master and slave axes. These data are multiplied and accumulated with the system parameters in the compensation amount model to calculate the compensation value. Specifically, the gain coefficient is multiplied by the error data and then added to the bias coefficient to obtain the compensation value. Based on the calculated compensation value, error compensation is implemented for the motion distance control amount.

[0089] In the face of a large-scale multi-axis control scenario, the traditional centralized master-slave axis control architecture has limitations. Therefore, distributing the master-slave axis control logic in multiple control nodes and using network communication to achieve data synchronization and distributed optimization can adapt to a larger-scale multi-axis control scenario.

[0090] In the technical solution of the present invention, the specific process of realizing real-time incremental optimization based on the first-in-first-out algorithm is as Figure 4 shown. Among them, first, the adaptive controller calculates the control amount, and then reads the position data of the spindle encoder. Then, a two-dimensional coordinate system of the master and slave axes is established, and based on this, a two-dimensional array of the master and slave axes is generated. The generated two-dimensional array is stored in the first-in-first-out buffer area, and then it is judged whether the buffer area is full. If it is not full, continue to read the position of the spindle encoder, continue to generate the two-dimensional array and store it in the buffer area. When the buffer area is full, the cam motion is executed, and the position of the spindle encoder is read again to judge whether it has moved to the next sampling position. If it has not reached, continue to execute the cam motion and read the position; if it has reached, delete the used array, and re-read the sensor sampling value and process the sample data in the buffer area, and then return to the step of the adaptive controller calculating the control amount, so as to realize real-time incremental optimization in a loop. The whole process ensures the effective processing of data and the stable operation of the system through the data coupling of the master and slave axes and the first-in-first-out algorithm.

[0091] In summary, under the technical solution of the above embodiment, by performing real-time optimization on the sampled data and combining parameter identification technology to dynamically adjust the control strategy, gradual optimization and dynamic adjustment are realized in the motion control process, so as to solve the problems of high noise sensitivity, insufficient parameter identification, and lack of process optimization in the existing profiling control technology.

[0092] As Figure 5As shown in the figure, an embodiment of the present invention provides a master-slave axis coupling control system, which includes: an observation data acquisition module 10, a control quantity model acquisition module 20, and a control quantity acquisition module 30.

[0093] Specifically, the observation data acquisition module 10 is configured to acquire sample data of the main axis and the slave axis, and preprocess the sample data using a filtering method to obtain the observation data of the main axis and the slave axis; the control quantity model acquisition module 20 is configured to perform parameter identification based on the observation data and update the system parameters of the control quantity model; the control quantity acquisition module 30 is configured to calculate the motion distance control quantity based on the control quantity model and the observation data, and perform coupling control on the main axis and the slave axis based on the motion distance control quantity.

[0094] In one implementation, the observation data acquisition module 10 includes:

[0095] An encoder and sensor data acquisition unit, configured to acquire encoder position data and sensor data of the main axis and the slave axis during an interruption;

[0096] A sample data buffer unit, configured to store the encoder position data and sensor data into a sample buffer in a first-in-first-out manner to obtain the sample data of the main axis and the slave axis;

[0097] A sample data filtering unit, configured to perform state estimation, error correction, and smoothing processing on the sample data using the extended Kalman filter to obtain the observation data of the main axis and the slave axis, and the filtering method is the extended Kalman filter.

[0098] In one implementation, the control quantity model acquisition module 20 includes:

[0099] A system parameter acquisition unit, configured to input the observation data into a parameter estimator of a pre-trained deep neural network to update the input-output gain coefficient and bias coefficient of the control quantity model, and the system parameters include the input-output gain coefficient and bias coefficient.

[0100] In one implementation, the control quantity acquisition module 30 includes:

[0101] An expected position calculation unit, configured to input the observation data into the control quantity model to calculate the expected positions of the main axis and the slave axis, and the observation data includes the current positions of the main axis and the slave axis;

[0102] A control quantity calculation unit, configured to calculate the control quantity of the motion distance based on the deviation between the expected position and the current position.

[0103] In one implementation, the control quantity calculation unit includes:

[0104] An adaptive sliding surface acquisition subunit, configured to calculate sliding surface parameters based on the deviation between the desired position and the current position, so as to obtain an adaptive sliding surface;

[0105] A control quantity calculation subunit, configured to calculate a motion distance control quantity based on the adaptive sliding surface.

[0106] In one implementation, the system further includes:

[0107] An error data acquisition unit, configured to additionally set a sensor at the actual machining position, and the sensor is configured to collect error data in real time after the system input;

[0108] A compensation quantity model system parameter acquisition unit, configured to perform secondary parameter identification on the error data to obtain the input-output gain coefficient and bias coefficient of the compensation quantity model of the motion distance control quantity;

[0109] A compensation value acquisition unit, configured to calculate a compensation value based on the compensation quantity model and the observation data;

[0110] An error compensation unit, configured to perform error compensation on the motion distance control quantity based on the compensation value.

[0111] Based on the above embodiments, the present invention further provides an intelligent terminal, and its principle block diagram can be as Figure 6 shown. The intelligent terminal includes a processor, a memory, a network interface, a display screen, and a temperature sensor connected through a system bus. Among them, the processor of the intelligent terminal is configured to provide computing and control capabilities. The memory of the intelligent terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the intelligent terminal is configured to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a master-slave axis coupling control method. The display screen of the intelligent terminal can be a liquid crystal display screen or an electronic ink display screen, and the temperature sensor of the intelligent terminal is pre-set inside the intelligent terminal and is configured to detect the operating temperature of internal devices.

[0112] Those skilled in the art can understand that Figure 6 the principle block diagram shown in is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the intelligent terminal to which the solution of the present invention is applied. The specific intelligent terminal may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0113] In one embodiment, an intelligent terminal is provided, including a memory, and one or more programs, where the one or more programs are stored in the memory and are configured to be executed by one or more processors. The one or more programs include instructions for performing the following operations:

[0114] Obtain sample data of the main axis and the slave axis, and preprocess the sample data using a filtering method to obtain observation data of the main axis and the slave axis;

[0115] Perform parameter identification based on the observation data to update the system parameters of the control quantity model;

[0116] Calculate a motion distance control quantity based on the control quantity model and the observation data, and perform coupled control on the master and slave axes based on the motion distance control quantity.

[0117] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided by the present invention can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0118] In summary, the present invention provides a master-slave axis coupling control method, system, terminal device and medium. Compared with the prior art, the method first obtains the sample data of the master axis and the slave axis, and preprocesses the sample data using a filtering method to obtain the observed data of the master axis and the slave axis; then, based on the observed data, parameter identification is performed to update the system parameters of the control quantity model; finally, based on the control quantity model and the observed data, the motion distance control quantity is calculated, and based on the motion distance control quantity, the master-slave axes are coupledly controlled. By optimizing the sampling data in real time, combining parameter identification technology to dynamically adjust the control strategy, and using the first-in-first-out motion controller buffer management algorithm to achieve real-time incremental optimization of sample parameters during the motion control process, the present invention realizes the coupled control of the master-slave axis motion control quantity and the target state value of the actuator, realizes high-precision contouring motion control, and solves the problems of high noise sensitivity, insufficient parameter identification and lack of process optimization in the existing contouring control technology.

[0119] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0120] The above embodiments only represent several implementation manners of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A master-slave axis coupling control method, characterized in that: Applied in profiling control, the method comprises: Acquire sample data of the main axis and the slave axis, and pre-process the sample data using a filtering method to obtain observation data of the main axis and the slave axis; Perform parameter identification based on the observed data and update system parameters of the control quantity model; Calculating a motion distance control amount based on the control amount model and the observation data, and performing coupling control on the master axis and the slave axis based on the motion distance control amount; The step of calculating the motion distance control amount based on the control amount model and the observation data includes: The observation data includes the current positions of the master axis and the slave axis; Inputting the observed data into the control quantity model to calculate the desired positions of the main axis and the slave axis; Based on the deviation between the desired position and the current position, a control amount of the movement distance is calculated.

2. The master-slave axis coupling control method according to claim 1, characterized in that: The step of obtaining sample data of the master axis and the slave axis includes: Collect encoder position data and sensor data of the master and slave axes during interruption; The encoder position data and the sensor data are stored in a sample buffer in a first-in-first-out manner to obtain sample data of the master axis and the slave axis.

3. The master-slave axis coupling control method according to claim 1, characterized in that: The method of preprocessing the sample data using a filtering method to obtain observation data of the main axis and the slave axis includes: The filtering method is extended Kalman filtering; The extended Kalman filter is used to perform state estimation, error correction and smoothing on the sample data to obtain observation data of the main axis and the slave axis.

4. The master-slave axis coupling control method according to claim 1, characterized in that: The performing parameter identification based on the observed data and updating the system parameters of the control quantity model includes: The system parameters include input and output gain coefficients and bias coefficients; The observation data is input into a parameter estimator of a pre-trained deep neural network to update the input-output gain coefficient and bias coefficient of the control quantity model.

5. The master-slave axis coupling control method according to claim 1, characterized in that: The step of calculating the control amount of the movement distance based on the deviation between the desired position and the current position includes: Calculating sliding surface parameters based on the deviation between the desired position and the current position to obtain an adaptive sliding surface; Based on the adaptive sliding surface, a movement distance control amount is calculated.

6. The master-slave axis coupling control method according to claim 1, characterized in that: The method further comprises: An additional sensor is provided at the actual processing position, and the sensor is used to collect error data after the system input in real time; Performing secondary parameter identification on the error data to obtain input and output gain coefficients and bias coefficients of a compensation model of the motion distance control amount; Calculating a compensation value based on the compensation amount model and the observation data; Based on the compensation value, error compensation is performed on the movement distance control amount.

7. A master-slave axis coupling control system, characterized in that: The system comprises: An observation data acquisition module is used to acquire sample data of the main axis and the slave axis, and pre-process the sample data using a filtering method to obtain observation data of the main axis and the slave axis; A control quantity model acquisition module, used for performing parameter identification based on the observation data and updating system parameters of the control quantity model; A control quantity acquisition module, used for calculating the motion distance control quantity based on the control quantity model and the observation data, and performing coupling control on the main axis and the slave axis based on the motion distance control quantity; The control amount acquisition module includes: An expected position calculation unit, used for inputting the observation data into the control quantity model to calculate the expected positions of the main axis and the slave axis, wherein the observation data includes the current positions of the main axis and the slave axis; A control amount calculation unit is used to calculate the control amount of the movement distance based on the deviation between the expected position and the current position.

8. A terminal device, characterized in that: The terminal device includes a memory, a processor, and a master-slave axis coupling control program stored in the memory and executable on the processor. When the processor executes the master-slave axis coupling control program, the steps of the master-slave axis coupling control method as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a master-slave axis coupling control program, and when the master-slave axis coupling control program is executed by the processor, the steps of the master-slave axis coupling control method according to any one of claims 1 to 6 are implemented.

Images