High-response encoder position follow-up control method and device

Abstract

The invention relates to the technical field of encoder control, and particularly discloses a high-response encoder position follow-up control method and device, which directly connects an encoder signal to an IO port of a controller, and adopts a more efficient software algorithm to process the encoder signal, thereby reducing the delay of an intermediate link and improving the accuracy of the encoder position follow-up control. And the response speed and the action continuity of the system are improved. That is to say, as the signal line of the encoder is directly connected to the controller, the controller detects the forward rotation or reverse rotation of the encoder (roller) and the number of pulses generated by the encoder in unit time so as to calculate the target position and the target speed at which the manipulator needs to operate, and the target position and the target speed are sent to the servo driver. And the moving position is consistent with the rolling number of the rollers, and the position is accurate, so that the operation effect of high response between the rollers and the controller is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of encoder control, and more specifically, to a high-response encoder position follow-up control method and device. Background Art

[0002] In a machine tool manipulator control system, the teach pendant (host computer) and the controller (slave computer) communicate with each other through a serial cable. The user issues corresponding instructions to the controller by operating the teach pendant. The controller then detects the type of the instruction. If it is a servo instruction, after parsing, the controller calculates the target position and speed that the servo driver needs to execute, and then sends this data to the servo driver to achieve the purpose of the manipulator operation. Among all the instructions, there is a type of servo instruction for the operation of the roller (encoder). When the user scrolls the roller, the teach pendant detects the rotation of the roller and calculates the distance to be moved, and then sends this distance to the controller. However, due to the inherent delay characteristics of serial communication itself, plus the time required for the controller to parse the instruction, calculate the target position, and send the instruction to the servo driver, significant delays may occur during the entire process. Due to the various processing times and communication delays mentioned above, when the user continuously scrolls the roller, the servo driver cannot immediately respond to the new position instruction, but shows a stop-and-go phenomenon. This means that the real-time performance and response speed of the system are limited, and a smooth operation experience cannot be provided.

[0003] In addition, since a series of calculation and transmission steps are required to update the position of the servo driver after each roller operation, during continuous scrolling, the operation of the servo driver appears not smooth enough, affecting the overall operation effect. Especially in application scenarios with high-precision requirements, such behavior may lead to task failure or product quality degradation. As the roller continuously scrolls, if the controller fails to receive the latest position information in time, it will cause the servo driver to work based on outdated data, further exacerbating the problem of inconsistent operation and slow response.

[0004] Therefore, an optimized encoder position follow-up control scheme is desired. Summary of the Invention

[0005] The present application provides a high-response encoder position follow-up control method and device, which directly connects the encoder signal to the IO port of the controller and uses a more efficient software algorithm to process the encoder signal, thereby reducing the delay in the intermediate link and improving the response speed and action continuity of the system. In other words, since the signal line of the encoder is directly connected to the controller, the controller detects the forward or reverse rotation of the encoder (roller) and the number of pulses generated by the encoder per unit time, thereby calculating the target position and target speed that the manipulator needs to run, and sends the target position and target speed to the servo driver, thereby achieving the effect of the roller stopping and starting, and the moving position is consistent with the number of times the roller rolls, and the position is accurate, so as to achieve a high-response operation effect between the roller and the controller.

[0006] In a first aspect, a high-response encoder position following control method is provided, comprising:

[0007] Obtain the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder;

[0008] Transmitting the A-phase pulse signal and the B-phase pulse signal to the controller through the IO port;

[0009] The controller determines the movement update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal;

[0010] Determining the target position coordinates of the manipulator based on the movement update distance and the current position coordinates of the manipulator;

[0011] Determine the number of rotations of the roller encoder based on the A-phase pulse signal and the B-phase pulse signal, and determine the target speed sent to the servo driver based on the number of rotations;

[0012] The target position coordinates and the target speed of the robot are sent to a servo driver for driving the servo driver.

[0013] In a second aspect, a high-response encoder position follow-up control device is provided, comprising:

[0014] The pulse signal acquisition module is used to acquire the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder;

[0015] A pulse signal transmission module, used for transmitting the A-phase pulse signal and the B-phase pulse signal to a controller through an IO port;

[0016] A moving update distance determination module, used for the controller to determine the moving update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal;

[0017] A target position coordinate determination module, used to determine the target position coordinates of the manipulator based on the movement update distance and the current position coordinates of the manipulator;

[0018] A target speed determination module, used to determine the number of rotations of the roller encoder based on the A-phase pulse signal and the B-phase pulse signal, and determine the target speed sent to the servo driver for operation based on the number of rotations;

[0019] The servo drive instruction generating module is used to send the target position coordinates and the target speed of the manipulator to the servo drive for driving the servo drive.

[0020] The present application provides a high-response encoder position follow-up control method and device, which directly connects the encoder signal to the IO port of the controller and uses a more efficient software algorithm to process the encoder signal, thereby reducing the delay in the intermediate link and improving the response speed and action continuity of the system. In other words, since the signal line of the encoder is directly connected to the controller, the controller detects the forward or reverse rotation of the encoder (roller) and the number of pulses generated by the encoder per unit time, thereby calculating the target position and target speed that the manipulator needs to run, and sends the target position and target speed to the servo driver, thereby achieving the effect of the roller stopping and starting, and the moving position is consistent with the number of times the roller rolls, and the position is accurate, so as to achieve a high-response operation effect between the roller and the controller. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0022] Figure 1 Schematic diagram of the connection between the teaching pendant and the controller according to an embodiment of the present application.

[0023] Figure 2 This is a schematic diagram of the hardware circuit principle of an embodiment of the present application.

[0024] Figure 3 A schematic flow chart of a high-response encoder position following control method according to an embodiment of the present application.

[0025] Figure 4 This is a schematic flow chart of step S3 in the high-response encoder position following control method according to an embodiment of the present application.

[0026] Figure 5 This is a schematic flowchart of step S32 in the high-response encoder position following control method according to an embodiment of the present application.

[0027] Figure 6 This is a schematic flowchart of step S33 in the high-response encoder position following control method according to an embodiment of the present application.

[0028] Figure 7 This is a schematic flowchart of step S331 in the high-response encoder position following control method according to an embodiment of the present application.

[0029] Figure 8 It is a schematic block diagram of a high-response encoder position following control device according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0031] In response to the above technical problems, a high-response encoder position follow-up control method is proposed in the technical solution of the present application. In terms of hardware, the A and B phase signal lines of the roller encoder originally connected to the teach pendant are modified to directly connect the two signal lines to the controller for connection, and the A and B phase signal lines of the encoder are connected to the controller as IO port signals input by the controller. This reduces the delay of the interactive communication between the teach pendant and the controller, enhances the stability and anti-interference ability of the signal transmission, and enables the entire system to operate stably at a higher frequency, providing a better user experience. The improved connection method between the teach pendant and the controller is as follows: Figure 1 shown.

[0032] Correspondingly, the problems encountered in the hardware after the improvement are: 1. The original direct connection line between the teach pendant and the controller has a length range of 5 meters to 12 meters according to customer requirements. Since the roller signal line on the teach pendant is connected to the controller, these two lines will be connected to the connection line between the controller and the teach pendant, which will cause communication interference problems caused by long-distance transmission. 2. Since the recommended operating voltage and current of the encoder are 5V and 10mA, but the input IO port on the controller requires a working voltage of 24V, directly connecting the encoder signal line on the teach pendant to the input IO port of the controller will cause strong electrical interference to the encoder. Therefore, an improvement to the hardware circuit is proposed. Taking into account the recommended operating voltage of the encoder of 5V and current of 10mA, and the need for long-distance transmission, it is necessary to increase the driving capability and anti-interference capability. Interference isolation is performed through optical coupling, and MOS tubes are added to convert 5V to a working voltage of 24V, and the current driving capability is increased to reach a driving current of 500mA. The specific hardware circuit principle diagram is shown as follows Figure 2 shown.

[0033] Furthermore, in the above-mentioned high-response encoder position follow-up control method, by directly connecting the encoder signal to the IO port of the controller and using a more efficient software algorithm to process the encoder signal, the delay in the intermediate link is reduced, and the response speed and action continuity of the system are improved. That is to say, since the signal line of the encoder is directly connected to the controller, the controller detects the forward or reverse rotation of the encoder (roller) and the number of pulses generated by the encoder per unit time, thereby calculating the target position and target speed that the manipulator needs to run, and sending the target position and target speed to the servo driver, thereby achieving the effect of the roller stopping and starting, and the moving position is consistent with the number of times the roller rolls, and the position is accurate, so as to achieve a high-response operation effect between the roller and the controller.

[0034] Specifically, in the technical solution of this application, if Figure 3 As shown, the high-response encoder position follow-up control method includes: S1, obtaining the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder; S2, transmitting the A-phase pulse signal and the B-phase pulse signal to the controller through the IO port; S3, the controller determines the movement update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal; S4, based on the movement update distance and the current position coordinates of the manipulator, determines the target position coordinates of the manipulator; S5, determines the number of rotations of the roller encoder based on the A-phase pulse signal and the B-phase pulse signal, and determines the target speed sent to the servo driver for operation based on the number of rotations; S6, sends the target position coordinates and the target speed of the manipulator to the servo driver for driving the servo driver.

[0035] In step S1 and step S2, the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder are obtained, and the A-phase pulse signal and the B-phase pulse signal are transmitted to the controller through the IO port. It should be understood that since each rotation of the roller will generate a series of A-phase and B-phase pulse signals, these signals directly reflect the direction and speed of the roller rotation. By quickly and accurately parsing these pulse signals, the controller can calculate the distance and direction that the manipulator needs to move in almost real time, thereby immediately updating the target position coordinates. This ensures the real-time responsiveness of the system and makes the movement of the manipulator smoother and smoother. In addition, the A-phase and B-phase pulse signals contain detailed information about the rotation of the roller. By accurately processing these two signals, it can be ensured that each roller rotation can be correctly recorded and converted into the corresponding manipulator movement. This high-precision position detection is crucial to achieve precise control, especially in application scenarios with extremely high precision requirements, such as precision assembly, micromachining, etc.

[0036] Based on this, in the process of determining the mobile update distance of the manipulator, the technical concept of the present application is to analyze the A-phase pulse signal and the B-phase pulse signal by using a signal processing and analysis algorithm based on artificial intelligence and deep learning, so as to capture the information representing the interactive response of the time-frequency characteristics of the A-phase pulse signal and the B-phase pulse signal, thereby helping to determine the direction and number of rotations of the roller encoder, so as to decode the mobile update distance of the manipulator. In this way, the artificial intelligence analysis algorithm can be used to automatically analyze the A-phase and B-phase pulse signals generated each time to more effectively obtain the mobile update distance of the manipulator, providing support for the position follow-up control of the high-response encoder.

[0037] In one embodiment, Figure 4 As shown, in step S3, the controller determines the moving update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal, including: S31, performing S transformation on the A-phase pulse signal and the B-phase pulse signal to obtain a two-dimensional time-frequency diagram of the A-phase pulse signal and a two-dimensional time-frequency diagram of the B-phase pulse signal; S32, performing signal time-frequency semantic feature extraction on the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal to obtain the time-frequency feature of the A-phase pulse signal and the time-frequency feature of the B-phase pulse signal; S33, performing interactive response analysis based on feature positive and negative attention on the time-frequency feature of the A-phase pulse signal and the time-frequency feature of the B-phase pulse signal to obtain the interactive response coding feature of the A-phase-B-phase pulse signal time-frequency feature; S34, determining the moving update distance of the manipulator based on the interactive response coding feature of the A-phase-B-phase pulse signal time-frequency feature.

[0038] Exemplarily, in step S31, the A-phase pulse signal and the B-phase pulse signal are subjected to S transformation to obtain a two-dimensional time-frequency diagram of the A-phase pulse signal and a two-dimensional time-frequency diagram of the B-phase pulse signal. It should be understood that the S-transform is a time-frequency analysis tool that combines the characteristics of the short-time Fourier transform (STFT) and the wavelet transform. Compared with the traditional wavelet transform, the S-transform can provide a time-frequency distribution, in which each point represents the energy of a specific frequency at a specific moment, and provides the frequency component information of the A-phase pulse signal and the B-phase pulse signal at different time points. This ability makes the S-transform very suitable for analyzing signals whose frequency components change over time, that is, non-stationary signals. Therefore, by performing S transformation on the A-phase pulse signal and the B-phase pulse signal respectively, the time-frequency feature information related to the encoder position follow-up control task can be better identified in a complex background. In addition, the S-transform also has good energy concentration, which can effectively reduce the influence of noise and improve the readability of the signal.

[0039] In one embodiment, Figure 5 As shown, in step S32, signal time-frequency semantic feature extraction is performed on the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal to obtain the time-frequency feature diagram of the A-phase pulse signal and the time-frequency feature diagram of the B-phase pulse signal, including: S321, the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal are respectively passed through a signal time-frequency semantic feature extractor based on DCNN to obtain the time-frequency feature diagram of the A-phase pulse signal and the time-frequency feature diagram of the B-phase pulse signal; S322, the time-frequency feature diagram of the A-phase pulse signal and the time-frequency feature diagram of the B-phase pulse signal are respectively passed through a signal time-frequency global domain sensor based on a fully connected layer to obtain the time-frequency feature vector of the A-phase pulse signal as the time-frequency feature of the A-phase pulse signal and the time-frequency feature vector of the B-phase pulse signal as the time-frequency feature of the B-phase pulse signal.

[0040] Exemplarily, in step S321, the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal are respectively passed through a signal time-frequency semantic feature extractor based on DCNN to obtain a time-frequency feature diagram of the A-phase pulse signal and a time-frequency feature diagram of the B-phase pulse signal. It should be understood that since the dilated convolution (Dilated Convolutional Neural Network, DCNN) is a technology for increasing the field of view of the convolution kernel while keeping the size of the receptive field unchanged. By inserting holes (i.e., zero values) between the weights of the convolution kernel, the receptive field of the convolution can be expanded without reducing the resolution or increasing the amount of parameters. The introduction of DCNN can expand the scope of feature extraction, so that the network can capture the contextual information of the time-frequency features of the A-phase pulse signal and the time-frequency features of the B-phase pulse signal in a larger range, which is very important for understanding the temporal dynamics of the signal. By adjusting the expansion rate, DCNN can capture features at different scales. Therefore, by introducing a signal time-frequency semantic feature extractor based on DCNN to perform feature extraction of the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal, signals with different frequency components can be effectively processed.

[0041] Exemplarily, in step S322, the A-phase pulse signal time-frequency feature diagram and the B-phase pulse signal time-frequency feature diagram are respectively passed through a signal time-frequency global sensor based on a fully connected layer to obtain an A-phase pulse signal time-frequency feature vector as the A-phase pulse signal time-frequency feature and a B-phase pulse signal time-frequency feature vector as the B-phase pulse signal time-frequency feature. It should be understood that since the A-phase pulse signal time-frequency feature diagram and the B-phase pulse signal time-frequency feature diagram respectively contain local time-frequency features of different scales captured by the A-phase pulse signal and the B-phase pulse signal through the expansion convolution. There is a correlation effect between these local time-frequency features. In order to be able to more comprehensively perceive the time-frequency global information in the A-phase pulse signal and the B-phase pulse signal to form a more abstract and more representative feature vector, in the technical solution of the present application, the A-phase pulse signal time-frequency feature diagram and the B-phase pulse signal time-frequency feature diagram are further passed through a signal time-frequency global sensor based on a fully connected layer to obtain an A-phase pulse signal time-frequency feature vector and a B-phase pulse signal time-frequency feature vector. Through the processing of the signal time-frequency global perceptron based on the fully connected layer, not only the amount of data for subsequent processing can be reduced, but also the model can be helped to better learn the essential characteristics of the data and avoid overfitting.

[0042] Exemplarily, in step S33, the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature are subjected to interactive response analysis based on feature positive and negative attention to obtain the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature. It should be understood that since the A-phase pulse signal time-frequency feature vector and the B-phase pulse signal time-frequency feature vector respectively contain the time-frequency global perception feature information about the A-phase pulse signal and the time-frequency global perception feature information about the B-phase pulse signal, and since the A-phase pulse signal time-frequency semantics and the B-phase pulse signal time-frequency semantics have mutual synergy and response characteristics, this feature and effect are crucial to the subsequent manipulator movement distance control. Therefore, in order to be able to more deeply understand the interaction between the A-phase and B-phase pulse signals, so as to more intelligently predict and respond to the user's input to control the movement distance of the manipulator and achieve an instant and smooth operation experience, in the technical solution of the present application, the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature are further subjected to interactive response analysis based on feature positive and negative attention to obtain the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature. In particular, the interactive response analysis based on feature positive and negative attention can efficiently capture and quantify the mutual influence between the time-frequency characteristics of the A-phase pulse signal and the time-frequency characteristics of the B-phase pulse signal by introducing homography projection transformation, bidirectional attention mechanism and interactive response encoding, so as to gain a deep insight into the interactive response relationship between the time-frequency characteristics of the A-phase and B-phase pulse signals. For example, in the case of continuous rolling of the wheel, it can help ensure that each tiny wheel rotation can be correctly identified and quickly converted into the corresponding manipulator movement without the phenomenon of stopping and starting.

[0043] In one embodiment, Figure 6 As shown, in step S33, the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics are subjected to interactive response analysis based on feature positive and negative attention to obtain the A-phase-B-phase pulse signal time-frequency characteristics interactive response coding characteristics, including: S331, constructing a positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field between the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics; S332, based on the positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field, modulating the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics and then performing interactive response analysis to obtain the A-phase-B-phase pulse signal time-frequency characteristics interactive response coding characteristics.

[0044] In one embodiment, Figure 7As shown, in step S331, a positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field is constructed between the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature, including: S3311, performing homography projection transformation on the A-phase pulse signal time-frequency feature vector and the B-phase pulse signal time-frequency feature vector to obtain the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector; S3312, based on the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector, constructing the forward A-phase-B-phase pulse signal time-frequency attention score field and the reverse A-phase-B-phase pulse signal time-frequency attention score field; S3313, constructing a dual-perspective attention balance field between the forward A-phase-B-phase pulse signal time-frequency attention score field and the reverse A-phase-B-phase pulse signal time-frequency attention score field to obtain the positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field.

[0045] Exemplarily, in step S3311, the A-phase pulse signal time-frequency feature vector and the B-phase pulse signal time-frequency feature vector are subjected to homography projection transformation to obtain the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector. Specifically, the process can be expressed by the formula:

[0046] v′1=H1(v1)

[0047] v′2=H2(v2)

[0048] Among them, v1 is the time-frequency feature vector of the A-phase pulse signal, v2 is the time-frequency feature vector of the B-phase pulse signal, H1 is the time-frequency mapping homography matrix of the A-phase pulse signal, H2 is the time-frequency mapping homography matrix of the B-phase pulse signal, v'1 is the time-frequency feature homography projection coding vector of the A-phase pulse signal, and v'2 is the time-frequency feature homography projection coding vector of the B-phase pulse signal.

[0049] In one embodiment, in step S3312, based on the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector, a forward A-phase-B-phase pulse signal time-frequency attention score field and a reverse A-phase-B-phase pulse signal time-frequency attention score field are constructed, including: calculating the forward attention score field of the A-phase pulse signal time-frequency feature homography projection coding vector relative to the B-phase pulse signal time-frequency feature homography projection coding vector to obtain the forward A-phase-B-phase pulse signal time-frequency attention score field; calculating the reverse attention score field of the B-phase pulse signal time-frequency feature homography projection coding vector relative to the A-phase pulse signal time-frequency feature homography projection coding vector to obtain the reverse A-phase-B-phase pulse signal time-frequency attention score field. Specifically, the process can be expressed by the formula:

[0050]

[0051] Among them, v'2 T is the transposed vector of v'2, v'1 T is the transposed vector of v'1, is matrix multiplication, S1 is the length of v'2, S2 is the length of v'1, M z is the time-frequency attention score field of the positive A-phase-B-phase pulse signal, M f It is the time-frequency attention score field of the reverse A-phase-B-phase pulse signal.

[0052] That is, in one embodiment, the forward attention score field of the A-phase pulse signal time-frequency feature homography projection coding vector relative to the B-phase pulse signal time-frequency feature homography projection coding vector is calculated to obtain the forward A-phase-B-phase pulse signal time-frequency attention score field, including: calculating the product of the A-phase pulse signal time-frequency feature homography projection coding vector and the transposed vector of the B-phase pulse signal time-frequency feature homography projection coding vector, and then dividing the obtained A-phase-B-phase pulse signal time-frequency interaction matrix by the square root of the length of the B-phase pulse signal time-frequency feature homography projection coding vector by position to obtain the A-phase-B-phase pulse signal time-frequency interaction correlation matrix as the forward A-phase-B-phase pulse signal time-frequency attention score field.

[0053] Exemplarily, in step S3313, a dual-perspective attention balance field is constructed between the forward A-phase-B-phase pulse signal time-frequency attention score field and the reverse A-phase-B-phase pulse signal time-frequency attention score field to obtain the forward and reverse A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field. Specifically, the process can be expressed by the formula:

[0054] Ω=Conv 3×3 [cat(M z ,Nf )]

[0055] Among them, cat(·) is the feature concatenation operation, Conv 3×3 (·) is the convolution code with a convolution kernel of 3×3, and Ω is the time-frequency dual-perspective attention balance field of the positive and negative A-phase-B-phase pulse signals.

[0056] In one embodiment, in step S332, based on the positive and negative A-phase-B-phase pulse signal time-frequency dual-view attention balance field, the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature are modulated and then interactive response analysis is performed to obtain the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature, including: mapping the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector to the positive and negative A-phase-B-phase pulse signal time-frequency dual-view attention balance field respectively to obtain the A-phase pulse signal time-frequency feature homography projection attention modulation coding vector and the B-phase pulse signal time-frequency feature homography projection attention modulation coding vector. Specifically, the process can be expressed by the formula:

[0057]

[0058] Among them, v 1t is the time-frequency feature homography projection attention modulation encoding vector of the A-phase pulse signal, v 2t It is the homography projection attention modulation encoding vector of the time-frequency feature of the B-phase pulse signal.

[0059] The time-frequency feature homography projection attention modulation coding vector of the A-phase pulse signal and the time-frequency feature homography projection attention modulation coding vector of the B-phase pulse signal are calculated and divided by the position point to obtain the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector as the A-phase-B-phase pulse signal time-frequency feature interaction response coding feature. Specifically, the process can be expressed by the formula:

[0060]

[0061] Wherein, V is the time-frequency characteristic interaction response encoding vector of the A-phase-B-phase pulse signal.

[0062] Specifically, the interactive response analysis process based on feature positive and negative attention can deeply explore and express the intrinsic connection between the time-frequency characteristics of the A-phase pulse signal and the time-frequency characteristics of the B-phase pulse signal. Compared with the traditional feature interaction architecture, the interactive response analysis process based on feature positive and negative attention can more carefully analyze the interaction between the time-frequency characteristics of the A-phase and B-phase pulse signals through the bidirectional attention mechanism, and improve the model's ability to understand the relationship between the complex time-frequency characteristics of the A-phase and B-phase pulse signals. In addition, it not only considers the influence from the time-frequency characteristics of the A-phase pulse signal to the time-frequency characteristics of the B-phase pulse signal, but also considers the influence in the opposite direction, ensuring that no important interaction details are missed, thereby enhancing the understanding of the complex dynamic relationship between the A-phase and B-phase pulse signals, and enhancing the real-time and robustness of the system, achieving smoother and more precise manipulator control.

[0063] Exemplarily, in step S34, the mobile update distance of the manipulator is determined based on the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature. In one embodiment, the mobile update distance of the manipulator is determined based on the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature, including: passing the A-phase-B-phase pulse signal time-frequency feature interactive response coding vector through a decoder-based manipulator mobile distance parser to obtain a recommended decoded value of the manipulator's mobile update distance. In other words, decoding regression is performed by utilizing the feature interactive response coding representation between the A-phase pulse signal time-frequency global perception feature and the B-phase pulse signal time-frequency global perception feature to perform a recommendation for the manipulator's mobile update distance. In this way, an artificial intelligence analysis algorithm can be used to automatically analyze the A-phase and B-phase pulse signals generated each time to more effectively obtain the manipulator's mobile update distance, providing support for achieving position follow-up control of a high-response encoder.

[0064] In one embodiment, the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector is passed through a decoder-based manipulator movement distance parser to obtain a recommended decoding value of the manipulator's movement update distance, including: using a decoder to decode and regress the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector using the following formula to obtain a recommended decoding value of the manipulator's movement update distance; wherein the formula is: Wherein, V is the time-frequency characteristic interactive response encoding vector of the A-phase-B-phase pulse signal, Y is the recommended decoding value of the mobile update distance of the manipulator, W is the weight matrix, Represents matrix multiplication.

[0065] Preferably, in one example, the A-phase-B-phase pulse signal time-frequency feature interactive response coding vector is passed through a decoder-based manipulator movement distance parser to obtain a recommended decoding value of the manipulator movement update distance, including:

[0066] Arranging the eigenvalues ​​of the A-phase-B-phase pulse signal time-frequency characteristic interaction response coding vector in ascending order to form an A-phase-B-phase pulse signal time-frequency characteristic interaction response sequential coding vector;

[0067] In response to the absolute value of the difference between the i-th eigenvalue and the i+1-th eigenvalue of the A-phase-B-phase pulse signal time-frequency feature interaction response sequence encoding vector being less than or equal to the distance difference hyperparameter ε, a weighted sum between the i-th eigenvalue and the i+1-th eigenvalue is calculated as the optimized i+1-th eigenvalue;

[0068] Calculate the square root of the sum of squares of all eigenvalues ​​of the A-phase-B-phase pulse signal time-frequency characteristic interaction response encoding vector:

[0069]

[0070] Among them, v i represents the i-th eigenvalue of the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector, L represents the length of the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector, and γ1 represents the square root of the sum of the squares of all eigenvalues ​​of the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector.

[0071] The square root of the sum of the squares of all eigenvalues ​​of the A-phase-B-phase pulse signal time-frequency characteristic interaction response coding vector is multiplied by 2 and then divided by the square of the length of the A-phase-B-phase pulse signal time-frequency characteristic interaction response coding vector to obtain the A-phase-B-phase pulse signal time-frequency characteristic interaction response space primitive value γ2=2×γ1 / L 2 ;

[0072] In response to the absolute value of the difference between the i-th eigenvalue and the i+1-th eigenvalue of the A-phase-B-phase pulse signal time-frequency feature interaction response sequential coding vector being greater than the distance difference hyperparameter ε, after multiplying the A-phase-B-phase pulse signal time-frequency feature interaction response space primitive value by the i-th eigenvalue, a weighted subtraction between the product and the i+1-th eigenvalue is calculated to obtain an optimized i+1-th eigenvalue;

[0073] On the basis of keeping the first eigenvalue of the A-phase-B-phase pulse signal time-frequency characteristic interaction response sequence coding vector unchanged, combining the optimized i+1th eigenvalue to obtain the optimized A-phase-B-phase pulse signal time-frequency characteristic interaction response coding vector;

[0074] The optimized A-phase-B-phase pulse signal time-frequency feature interaction response encoding vector is passed through a decoder-based manipulator movement distance parser to obtain a recommended decoding value of the manipulator's movement update distance.

[0075] Here, when the A-phase pulse signal time-frequency feature vector and the B-phase pulse signal time-frequency feature vector respectively represent the pulse signal time-frequency features of the A-phase pulse signal two-dimensional time-frequency diagram and the B-phase pulse signal two-dimensional time-frequency diagram determined by S-transformation, when performing an interactive response based on characteristic positive and negative attention fields, the complexity of the positive and negative attention field mechanism based on the significant difference in signal source domain content between the A-phase pulse signal and the B-phase pulse signal will lead to insufficient representation of the long-distance interactive response of the A-phase-B-phase pulse signal time-frequency feature interactive response encoding vector, thereby reducing the expression effect of the A-phase-B-phase pulse signal time-frequency feature interactive response encoding vector, affecting the accuracy of the recommended decoding value of the manipulator's movement update distance obtained by the decoder-based manipulator movement distance parser.

[0076] Therefore, in view of the problem that the feature set of the A-phase-B phase pulse signal time-frequency feature interaction response coding vector is insufficient in the global interactive response representation capability due to the long distance exceeding the predetermined local distribution interval threshold under the predetermined feature value sequence distribution, the high-dimensional feature space primitive representation based on the self-inner product fusion of the A-phase-B phase pulse signal time-frequency feature interaction response coding vector is used to capture the complex structure of the global network interaction of its feature value, so as to reconstruct the interactive response relationship between the feature values ​​of the A-phase-B phase pulse signal time-frequency feature interaction response coding vector by simulating the scale-based high-dimensional feature space potential primitive, so as to realize the coding reconstruction of the real sequence distribution behavior of the A-phase-B phase pulse signal time-frequency feature interaction response coding vector under the long distance, improve the interactive response expression effect of the A-phase-B phase pulse signal time-frequency feature interaction response coding vector, and improve the accuracy of the recommended decoding value of the manipulator's mobile update distance obtained by the manipulator mobile distance parser based on the decoder. In this way, the artificial intelligence analysis algorithm can be used to automatically analyze the A-phase and B-phase pulse signals generated each time to more effectively obtain the mobile update distance of the manipulator, and provide support for the position follow-up control of the high-response encoder.

[0077] In another embodiment, the controller detects the level change of phase A of the roller encoder at a very high frequency (for example, once every 100 microseconds). Specifically, it checks whether there is a falling edge signal change from high level to low level in phase A, and this change needs to last for 5 times to be confirmed as a valid falling edge. Once such a change is detected, the controller will immediately determine the state of the phase B signal at this time. If phase B is high (for 5 times), it is considered that the forward rotation count of the roller is increased by one; if phase B is low (for 5 times), it is considered that the reverse rotation count of the roller is increased by one. The controller will detect again every 5 milliseconds whether the forward and reverse rotation count of the roller changes. If the roller does rotate during this period, that is, the forward and reverse rotation count is greater than zero, then the total change of the roller in these 5 milliseconds is calculated, that is, [RollerSum] = [RollerForeNum] - [RollerBackNum]. Here, [RollerSum] represents the net number of roller rotations, [RollerForeNum] represents the number of times the roller rotates forward, and [RollerBackNum] represents the number of times the roller rotates backward. According to the set roller unit length (such as 0.01mm, 0.10mm, etc.), the controller will use the formula [Dist] = [RollerSum] * [unit] to calculate the actual distance the robot should move. Here, [Dist] represents the distance the robot should move, [RollerSum] is the net number of roller rotations, and [unit] is the physical distance unit corresponding to each grid of the roller.

[0078] Exemplarily, in step S4, the target position coordinates of the manipulator are determined based on the movement update distance and the current position coordinates of the manipulator. That is, after determining the movement update distance, the target position coordinates are determined in combination with the current position coordinates of the manipulator to ensure that the manipulator can accurately respond to the user's operating instructions and achieve precise movement from the current position to the new position. This ensures that each time the user scrolls the wheel, the manipulator can move smoothly and accurately to the expected position, avoiding motion errors or delays caused by inaccurate position calculations. It should be understood that the controller needs to perform calculations in combination with the current coordinate position of the manipulator. This is because each movement of the manipulator is relative to its current position. Only by knowing "where it is now" can we correctly calculate "where it should move to".

[0079] In one embodiment, specifically, after obtaining the moving update distance [Dist], the target position coordinates of the manipulator are determined based on the moving update distance and the current position coordinates of the manipulator. If this is the first scrolling, the initial position [CurrentPos] is used as the current position, and the initial position [CurrentPos] is added to the moving distance [Dist] to obtain the target position coordinates [TargetPos]; if it is not the first scrolling, the previous target position [PreTargetPos] is used as the current position, and the previous target position [PreTargetPos] is added to the moving distance [Dist] this time.

[0080] Exemplarily, in step S5, the number of rotations of the roller encoder is determined based on the A-phase pulse signal and the B-phase pulse signal, and the target speed sent to the servo driver is determined based on the number of rotations. Specifically, in order to determine the number of rotations of the roller encoder, the controller detects whether there is a change from a high level to a low level in the roller A phase in each interrupt cycle, and determines whether the roller is forward or reversed in combination with the state of the B phase at this time. For example, if it is detected that the A phase has a falling edge and the B phase is at a high level at this time, it is considered that the roller forward count is increased by one; if the B phase is at a low level, it is considered that the roller reverse count is increased by one. In this way, the controller can cumulatively calculate the net number of rollers during this period, and the net number of roller rotations [RollerSum] is equal to the number of roller forward rotations [RollerForeNum] minus the number of roller reverse rotations [RollerBackNum]. Next, in order to determine the target speed sent to the servo driver, the controller uses a linear function to map the number of rotations to the target speed. This function is defined by the minimum speed [SpeedMin], the maximum speed [SpeedMax], and the speed slope [SpeedSlope], where the speed slope is equal to the maximum speed minus the minimum speed. The specific formula is as follows: [TargetSpeed] = [SpeedSlope] * [RollerSum] + [SpeedMin], where [TargetSpeed] represents the target speed, [RollerSum] represents the net number of roller rotations, [SpeedSlope] determines the rate of speed change, and [SpeedMin] is the set minimum speed. The advantage of this design is that the manipulator can maintain a certain minimum speed when stationary to prevent impact when suddenly starting, and as the number of roller rotations increases, the speed of the manipulator will increase accordingly until the set maximum speed is reached.

[0081] Exemplarily, in step S6, the target position coordinates and the target speed of the manipulator are sent to a servo driver to drive the servo driver. That is, the calculated target position and target speed are updated to the servo driver, and the servo driver starts to move at the latest position and speed.

[0082] In summary, according to the embodiment of the present application, a high-response encoder position follow-up control method is explained, which directly connects the encoder signal to the IO port of the controller and uses a more efficient software algorithm to process the encoder signal, thereby reducing the delay in the intermediate links and improving the response speed and action continuity of the system. In other words, since the signal line of the encoder is directly connected to the controller, the controller detects the forward or reverse rotation of the encoder (roller) and the number of pulses generated by the encoder per unit time, thereby calculating the target position and target speed that the manipulator needs to run, and sending the target position and target speed to the servo driver, thereby achieving the effect of the roller stopping and starting, and the moving position is consistent with the number of times the roller rolls, and the position is accurate, so as to achieve a high-response operation effect between the roller and the controller.

[0083] Figure 8 Schematic block diagram of a high-response encoder position follow-up control device according to an embodiment of the present application. Figure 8 As shown, the high-response encoder position follow-up control device 100 includes: a pulse signal acquisition module 110, used to acquire the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder; a pulse signal transmission module 120, used to transmit the A-phase pulse signal and the B-phase pulse signal to the controller through the IO port; a movement update distance determination module 130, used for the controller to determine the movement update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal; a target position coordinate determination module 140, used to determine the target position coordinates of the manipulator based on the movement update distance and the current position coordinates of the manipulator; a target speed determination module 150, used to determine the number of rotations of the roller encoder based on the A-phase pulse signal and the B-phase pulse signal, and determine the target speed sent to the servo driver based on the number of rotations; a servo drive instruction generation module 160, used to send the target position coordinates and the target speed of the manipulator to the servo driver for driving the servo driver.

[0084] In one embodiment, the moving update distance determination module includes: a pulse signal conversion unit, which is used to perform S transformation on the A-phase pulse signal and the B-phase pulse signal to obtain a two-dimensional time-frequency diagram of the A-phase pulse signal and a two-dimensional time-frequency diagram of the B-phase pulse signal; a pulse signal time-frequency feature extraction unit, which is used to extract signal time-frequency semantic features from the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal to obtain the time-frequency features of the A-phase pulse signal and the time-frequency features of the B-phase pulse signal; a pulse signal interactive response analysis unit, which is used to perform interactive response analysis based on feature positive and negative attention on the time-frequency features of the A-phase pulse signal and the time-frequency features of the B-phase pulse signal to obtain the A-phase-B-phase pulse signal time-frequency feature interactive response coding features, wherein, The pulse signal interactive response analysis unit includes: a dual-perspective attention balance field construction subunit, which is used to construct a positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field between the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature; a pulse signal time-frequency feature interactive response coding subunit, which is used to modulate the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature based on the positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field, and then perform interactive response analysis to obtain the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature; a manipulator movement update distance determination unit, which is used to determine the manipulator movement update distance based on the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature.

[0085] Here, those skilled in the art will appreciate that the specific operations of the various modules and units in the above-mentioned high-response encoder position follow-up control device have been described in the above reference. Figures 1 to 6 The high-response encoder position following control method has been described in detail, and therefore, its repeated description will be omitted.

[0086] The prefixes such as "first" and "second" used in the embodiments of the present application are only used to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of prefixes such as ordinal numbers used to distinguish description objects in the embodiments of the present application does not constitute a limitation on the described objects. For the description of the described objects, please refer to the description of the context in the embodiments, and the use of such prefixes should not constitute an unnecessary limitation.

[0087] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0088] In the various embodiments of the present application, unless otherwise specified or logically conflicting, the terms and / or descriptions between the various embodiments are consistent and can be referenced to each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0089] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0091] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A high-response encoder position following control method, characterized in that: include: Obtain the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder; Transmitting the A-phase pulse signal and the B-phase pulse signal to the controller through the IO port; The controller determines the movement update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal; Determining the target position coordinates of the manipulator based on the movement update distance and the current position coordinates of the manipulator; Determine the number of rotations of the roller encoder based on the A-phase pulse signal and the B-phase pulse signal, and determine the target speed sent to the servo driver based on the number of rotations; The target position coordinates and the target speed of the robot are sent to a servo driver for driving the servo driver.

2. The high-response encoder position following control method according to claim 1, characterized in that: The controller determines the movement update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal, including: Performing S transformation on the A-phase pulse signal and the B-phase pulse signal respectively to obtain a two-dimensional time-frequency diagram of the A-phase pulse signal and a two-dimensional time-frequency diagram of the B-phase pulse signal; Extracting signal time-frequency semantic features from the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal to obtain the time-frequency features of the A-phase pulse signal and the time-frequency features of the B-phase pulse signal; The A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics are subjected to interactive response analysis based on feature positive and negative attention to obtain A-phase-B-phase pulse signal time-frequency characteristics interactive response coding characteristics, including: constructing a positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field between the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics; based on the positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field, modulating the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics and then performing interactive response analysis to obtain the A-phase-B-phase pulse signal time-frequency characteristics interactive response coding characteristics; Based on the time-frequency characteristic interactive response coding characteristics of the A-phase-B-phase pulse signal, the movement update distance of the manipulator is determined.

3. The high-response encoder position following control method according to claim 2, characterized in that: The signal time-frequency semantic features are extracted from the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal respectively to obtain the time-frequency feature diagram of the A-phase pulse signal and the time-frequency feature diagram of the B-phase pulse signal, including: The two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal are respectively passed through a signal time-frequency semantic feature extractor based on DCNN to obtain a time-frequency feature diagram of the A-phase pulse signal and a time-frequency feature diagram of the B-phase pulse signal; The A-phase pulse signal time-frequency feature diagram and the B-phase pulse signal time-frequency feature diagram are respectively passed through a signal time-frequency global domain sensor based on a fully connected layer to obtain an A-phase pulse signal time-frequency feature vector as the A-phase pulse signal time-frequency feature and a B-phase pulse signal time-frequency feature vector as the B-phase pulse signal time-frequency feature.

4. The high-response encoder position following control method according to claim 3, characterized in that: Constructing a positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field between the A-phase pulse signal time-frequency feature and the B-phase pulse signal time-frequency feature, including: Performing homography projection transformation on the A-phase pulse signal time-frequency feature vector and the B-phase pulse signal time-frequency feature vector to obtain an A-phase pulse signal time-frequency feature homography projection coding vector and a B-phase pulse signal time-frequency feature homography projection coding vector; Based on the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector, construct a forward A-phase-B-phase pulse signal time-frequency attention score field and a reverse A-phase-B-phase pulse signal time-frequency attention score field; A dual-perspective attention balance field is constructed between the forward A-phase-B-phase pulse signal time-frequency attention score field and the reverse A-phase-B-phase pulse signal time-frequency attention score field to obtain the forward and reverse A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field.

5. The high-response encoder position following control method according to claim 4, characterized in that: Based on the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector, constructing a forward A-phase-B-phase pulse signal time-frequency attention score field and a reverse A-phase-B-phase pulse signal time-frequency attention score field, including: Calculating the forward attention score field of the homography projection coding vector of the time-frequency feature of the A-phase pulse signal relative to the homography projection coding vector of the time-frequency feature of the B-phase pulse signal to obtain the forward A-phase-B-phase pulse signal time-frequency attention score field; The reverse attention score field of the B-phase pulse signal time-frequency feature homography projection coding vector relative to the A-phase pulse signal time-frequency feature homography projection coding vector is calculated to obtain the reverse A-phase-B-phase pulse signal time-frequency attention score field.

6. The high-response encoder position following control method according to claim 5, characterized in that: Calculate the forward attention score field of the A-phase pulse signal time-frequency feature homography projection coding vector relative to the B-phase pulse signal time-frequency feature homography projection coding vector to obtain the forward A-phase-B-phase pulse signal time-frequency attention score field, including: calculating the product of the A-phase pulse signal time-frequency feature homography projection coding vector and the transposed vector of the B-phase pulse signal time-frequency feature homography projection coding vector, and then dividing the obtained A-phase-B-phase pulse signal time-frequency interaction matrix by the square root of the length of the B-phase pulse signal time-frequency feature homography projection coding vector by position to obtain the A-phase-B-phase pulse signal time-frequency interaction correlation matrix as the forward A-phase-B-phase pulse signal time-frequency attention score field.

7. The high-response encoder position following control method according to claim 6, characterized in that: Based on the time-frequency dual-perspective attention balance field of the positive and negative A-phase-B-phase pulse signal, the time-frequency characteristics of the A-phase pulse signal and the time-frequency characteristics of the B-phase pulse signal are modulated and then interactive response analysis is performed to obtain the time-frequency characteristics interactive response coding characteristics of the A-phase-B-phase pulse signal, including: Mapping the A-phase pulse signal time-frequency feature homography projection coding vector and the B-phase pulse signal time-frequency feature homography projection coding vector to the positive and negative A-phase-B-phase pulse signal time-frequency dual-view attention balance field respectively to obtain the A-phase pulse signal time-frequency feature homography projection attention modulation coding vector and the B-phase pulse signal time-frequency feature homography projection attention modulation coding vector; Calculate the point-by-point division between the homography projection attention modulation coding vector of the time-frequency feature of the A-phase pulse signal and the homography projection attention modulation coding vector of the time-frequency feature of the B-phase pulse signal to obtain the A-phase-B-phase pulse signal time-frequency feature interaction response coding vector as the A-phase-B-phase pulse signal time-frequency feature interaction response coding feature.

8. The high-response encoder position following control method according to claim 7, characterized in that: Based on the A-phase-B-phase pulse signal time-frequency characteristic interaction response coding feature, the movement update distance of the manipulator is determined, including: passing the A-phase-B-phase pulse signal time-frequency characteristic interaction response coding vector through a decoder-based manipulator movement distance parser to obtain a recommended decoding value of the manipulator movement update distance.

9. A high-response encoder position follow-up control device, characterized in that: include: The pulse signal acquisition module is used to acquire the A-phase pulse signal and the B-phase pulse signal sent by the user through the roller encoder; A pulse signal transmission module, used for transmitting the A-phase pulse signal and the B-phase pulse signal to a controller through an IO port; A moving update distance determination module, used for the controller to determine the moving update distance of the manipulator based on the A-phase pulse signal and the B-phase pulse signal; A target position coordinate determination module, used to determine the target position coordinates of the manipulator based on the movement update distance and the current position coordinates of the manipulator; A target speed determination module, used to determine the number of rotations of the roller encoder based on the A-phase pulse signal and the B-phase pulse signal, and determine the target speed sent to the servo driver for operation based on the number of rotations; The servo drive instruction generation module is used to send the target position coordinates and the target speed of the manipulator to the servo drive to drive the servo drive.

10. The high-response encoder position follow-up control device according to claim 9, characterized in that: The moving update distance determination module comprises: A pulse signal conversion unit, used for performing S-conversion on the A-phase pulse signal and the B-phase pulse signal to obtain a two-dimensional time-frequency diagram of the A-phase pulse signal and a two-dimensional time-frequency diagram of the B-phase pulse signal; A pulse signal time-frequency feature extraction unit is used to extract signal time-frequency semantic features from the two-dimensional time-frequency diagram of the A-phase pulse signal and the two-dimensional time-frequency diagram of the B-phase pulse signal to obtain the time-frequency features of the A-phase pulse signal and the time-frequency features of the B-phase pulse signal; A pulse signal interactive response analysis unit, used for performing interactive response analysis on the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics based on feature positive and negative attention to obtain A-phase-B-phase pulse signal time-frequency characteristics interactive response coding characteristics, wherein the pulse signal interactive response analysis unit includes: a dual-perspective attention balance field construction subunit, used for constructing a positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field between the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics; a pulse signal time-frequency characteristics interactive response coding subunit, used for modulating the A-phase pulse signal time-frequency characteristics and the B-phase pulse signal time-frequency characteristics based on the positive and negative A-phase-B-phase pulse signal time-frequency dual-perspective attention balance field, and then performing interactive response analysis to obtain the A-phase-B-phase pulse signal time-frequency characteristics interactive response coding characteristics; The manipulator movement update distance determination unit is used to determine the manipulator movement update distance based on the A-phase-B-phase pulse signal time-frequency feature interactive response coding feature.

Images