Multi-axis servo turntable system and control method

Abstract

The application discloses a multi-axis servo turntable system and a control method, and belongs to the technical field of mechanical control. The multi-axis servo turntable system comprises a driving mechanism, a control unit and a turntable body. The driving mechanism comprises a plurality of servo motors and is used for driving the turntable body to rotate and axially move. The control unit controls each servo motor by using a fuzzy self-adaptive sliding mode iterative learning control algorithm. By introducing fuzzy logic to dynamically adjust the sliding mode parameters and combining the compensation of periodic disturbance by iterative learning, the system chattering is significantly reduced, the dynamic response speed is improved, and high-precision control of multi-axis cooperation is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mechanical control, and more particularly to a multi-axis servo turntable system and a control method. BACKGROUND

[0002] At present, the demand for target monitoring in optical systems is increasing, especially in the fields of space exploration, remote sensing, military reconnaissance, etc. In the above target monitoring, a multi-axis servo turntable capable of realizing free rotation is usually used to realize the monitoring of the target. However, the existing multi-axis servo turntable mostly adopts analog signal control, and has some defects, such as low monitoring precision, slow reaction speed, etc., and it is difficult to meet the requirements of high-precision position control and speed control.

[0003] Therefore, how to improve the control precision of the multi-axis servo turntable system is a problem to be solved by those skilled in the art. SUMMARY

[0004] Therefore, the present application provides a multi-axis servo turntable system and a control method.

[0005] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0006] On the one hand, the present application discloses a multi-axis servo turntable system, comprising a driving mechanism, a control unit and a turntable body, wherein the driving mechanism comprises a plurality of servo motors for driving the turntable body to rotate and move axially; the control unit controls each servo motor by using a fuzzy self-adaptive sliding mode iterative learning control algorithm.

[0007] Preferably, the plurality of servo motors drive the turntable body to rotate and move axially, and specifically comprising the following steps:

[0008] The servo controller of each servo motor receives feedback information from the corresponding encoder, and controls each axis according to the fuzzy self-adaptive sliding mode iterative learning control algorithm preset by the control unit, and the encoder is arranged on each axis.

[0009] Preferably, each axis is controlled according to the fuzzy self-adaptive sliding mode iterative learning control algorithm preset by the control unit, and specifically comprising the following steps:

[0010] An adaptive sliding mode iterative learning controller is constructed, and the sliding film control rate of the adaptive sliding mode iterative learning controller is taken as a state variable of the control unit of each servo motor, and the state variable comprises the rotating speed of the servo motor;

[0011] The switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller are dynamically adjusted by using a fuzzy logic module, and an optimized switching gain and adaptive sliding mode iterative learning controller are obtained.

[0012] The optimized adaptive iterative learning sliding mode controller model is analyzed for stability and error convergence by Lyapunov method;

[0013] The adaptive iterative learning sliding mode controller model meeting the stability and error convergence conditions is applied to the control unit of each servo motor.

[0014] Preferably, in the step of constructing the adaptive sliding mode iterative learning controller, the adaptive sliding mode iterative learning controller is represented by the following formula:

[0015]

[0016] Wherein, u k (k) is the output of the adaptive sliding mode iterative learning controller at the kth time; b -1 represents the reciprocal of constant b; c is an integral coefficient, e k (k) is the tracking error of the output speed at the kth time; represents the expected target speed; B(x k , t) is a known friction torque function; represents the expected iterative control component for learning unknown periodic functions; v k (k) represents the system state variable at the kth iteration; represents the disturbance estimation of the system at the kth time.

[0017] Preferably, the disturbance estimation of the system is Based on the following adaptive law:

[0018] ;

[0019] Wherein γ is the learning gain of the adaptive control law, and γ > 0; s k (k) represents the integral sliding surface at the kth iteration.

[0020] Preferably, the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller are dynamically adjusted by using a fuzzy logic module to obtain an optimized switching gain and adaptive sliding mode iterative learning controller, which specifically includes the following steps:

[0021] Define the integral sliding surface s k (k) at the kth iteration and its rate of change as input variables of the fuzzy logic module;

[0022] The input variables are mapped to corresponding fuzzy sets by membership functions;

[0023] The fuzzy set output is converted into parameter adjustment amount Δβ1 and Δc by using the barycenter method, wherein Δβ1 is a switching gain adjustment amount, and Δc is an integral coefficient adjustment amount;

[0024] The optimized switching gain is obtained according to the switching gain adjustment amount Δβ1:

[0025] β1(k) = β1(k-1) + Δβ1;

[0026] The optimized integral coefficient is obtained according to the integral coefficient adjustment amount Δc:

[0027] c(k) = c(k-1) + Δc;

[0028] The optimized integral coefficient is applied to the adaptive sliding mode iterative learning controller to obtain an optimized adaptive sliding mode iterative learning controller.

[0029] Preferably, the stability analysis and error convergence analysis of the optimized adaptive iterative learning sliding mode controller model are performed by using the Lyapunov method, and the method specifically includes the following steps:

[0030] A Lyapunov function at the kth iteration is established:

[0031]

[0032] According to the Lyapunov function, a plurality of Lyapunov sub-functions 、 、 、 are calculated:

[0033]

[0034]

[0035]

[0036]

[0037] In the formula, η1 and η2 represent system parameter constants; s k (t) represents an integral sliding surface at the kth iteration; represents a learning error at the kth iteration; q represents a constant; represents a disturbance estimation error at the kth iteration; γ represents a learning gain of the adaptive control law;

[0038] The difference between each Lyapunov sub-function at the kth iteration and the (k-1)th iteration is calculated , i = 1, 2, 3, 4;

[0039] The obtained , i = 1, 2, 3, 4, the total difference AV of the Lyapunov function between the kth iteration and the k-1th iteration is obtained by summation k (t) is less than or equal to 0, the adaptive iterative learning sliding mode controller model satisfies the stability and error convergence conditions.

[0040] The total difference AV of the Lyapunov function between the kth iteration and the k-1th iteration is obtained by summation k (t) is less than or equal to 0, the adaptive iterative learning sliding mode controller model satisfies the stability and error convergence conditions.

[0041] Preferably, the multi-axis servo turntable system further comprises a monitoring unit for monitoring the state and position information of each axis in real time, and realizing alarm in abnormal conditions.

[0042] Preferably, the multi-axis servo turntable system is applied to target monitoring of various optical systems.

[0043] In another aspect, the application further discloses a multi-axis servo turntable system control method, comprising the following steps:

[0044] comprising the following steps:

[0045] The servo motor speed is selected as the system state variable, and parameter perturbation and external disturbance are considered to establish a servo motor dynamic system;

[0046] An adaptive sliding mode iterative learning controller is constructed, and the sliding mode control rate of the adaptive sliding mode iterative learning controller is taken as the state variable of each servo motor control unit;

[0047] The switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller are dynamically adjusted by using a fuzzy logic module to obtain an optimized switching gain and adaptive sliding mode iterative learning controller;

[0048] The stability and error convergence of the optimized adaptive iterative learning sliding mode controller model are analyzed by Lyapunov method;

[0049] The adaptive iterative learning sliding mode controller model satisfying the stability and error convergence conditions is applied to the control unit of each servo motor.

[0050] Further, in the step of constructing the adaptive sliding mode iterative learning controller, the adaptive sliding mode iterative learning controller is represented by the following formula:

[0051]

[0052] Wherein, u k (t) is the output of the adaptive sliding mode iterative learning controller in the kth iteration; b -1represents the reciprocal of constant b; c is an integral coefficient, e k (t) is the tracking error of the kth output rotation speed; represents the expected target rotation speed; B(x k (t) is a known friction torque function; represents the expected iterative control component for learning the unknown periodic function; v k (t) represents the system state variable at the kth iteration; represents the disturbance estimation of the kth iteration of the system.

[0053] Further, the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller are dynamically adjusted by using a fuzzy logic module, to obtain an optimized switching gain and adaptive sliding mode iterative learning controller, and the specific steps include the following steps:

[0054] the integral sliding surface s k (t) and its rate of change as input variables of the fuzzy logic module;

[0055] The input variables are mapped to the corresponding fuzzy sets by the membership function;

[0056] The fuzzy set output is converted into parameter adjustment amounts Δβ1 and Δc by using the barycenter method, wherein Δβ1 is the switching gain adjustment amount, and Δc is the integral coefficient adjustment amount;

[0057] The optimized switching gain is obtained according to the switching gain adjustment amount Δβ1:

[0058] β1(k)=β1(k-1)+Δβ1;

[0059] The optimized integral coefficient is obtained according to the integral coefficient adjustment amount Δc:

[0060] c(k)=c(k-1)+Δc ;

[0061] The optimized integral coefficient is applied to the adaptive sliding mode iterative learning controller to obtain an optimized adaptive sliding mode iterative learning controller.

[0062] Further, the nonlinear disturbance state of the system is estimated in real time by using a neural network.

[0063] According to the above technical solution, compared with the prior art, the application provides a multi-axis servo turntable system and a control method, which have the following beneficial effects:

[0064] The application significantly reduces system chattering and improves dynamic response speed by introducing fuzzy logic to dynamically adjust sliding mode parameters and combining iterative learning to compensate for periodic disturbances, thereby achieving high-precision control of multi-axis cooperation. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0066] Figure 1 This is a schematic structural diagram of the multi-axis servo turntable system provided by the present invention.

[0067] Figure 2 This is a schematic diagram of the overall flow of the multi-axis servo turntable system control method provided by the present invention.

[0068] Figure 3 : This is a fuzzy sliding mode control architecture diagram provided by the present invention, used to demonstrate the integrated relationship between the fuzzy reasoning module and the sliding mode controller;

[0069] Figure 4 :This is the neural network online identification flow chart provided by the present invention, which is used to describe the neural network input / output and weight update process DETAILED DESCRIPTION

[0070] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0071] The embodiment of the present invention first discloses a multi-axis servo turntable system for an optical system, such as Figure 1 The system includes: a driving mechanism, a control unit and a turntable body, wherein the driving mechanism includes multiple servo motors for driving the turntable body to rotate and move axially; the control unit uses an adaptive sliding mode iterative learning control algorithm to control each servo motor.

[0072] The turntable body of the multi-axis servo turntable system of this invention utilizes high-precision bearings to ensure stability and accuracy. The drive mechanism includes a servo motor and a matching reducer to drive the turntable body. The control unit, driven by the servo motor, precisely controls the position and speed of the turntable body using an adaptive sliding mode iterative learning control algorithm.

[0073] The multi-axis servo turntable of the application adopts high-precision encoders for position feedback, and can achieve nanometer-level positioning accuracy. Through precise control of the control system, the turntable can be accurately parked at any position.

[0074] The multi-axis servo turntable of the application adopts high-precision bearings and mechanical structures, and an advanced control system, which can effectively reduce jitter and errors and improve the stability of the turntable.

[0075] The multi-axis servo turntable of the application can be controlled through a simple operation interface without the need for professional technicians to operate and maintain. At the same time, the control system can also monitor various parameters of the turntable in real time to facilitate timely discovery and solution of problems.

[0076] Based on the above system, the embodiment of the application also discloses a multi-axis servo turntable system control method for an optical system, as shown in the whole method Figure 2 The specific implementation steps of the multi-axis servo turntable system control method are described below:

[0077] The application adopts an adaptive sliding mode iterative learning control algorithm: in order to facilitate the analysis of the stability of the adaptive sliding mode iterative learning algorithm (SMC-ILC), the motor speed is selected as the system state variable, and the parameter perturbation and external disturbance are considered, a permanent magnet synchronous motor dynamic system is established, and a single-input single-output system is taken as an example:

[0078] (1)

[0079] For example, the specific implementation mode of the adaptive sliding mode iterative learning control strategy is described, wherein x(t)=ω is a state variable related to system input (the system input can be motor speed), u(t)=i qref is a control variable input (the control variable can be a current reference value), y(t)=ω is a system output (the system output can be motor speed), f(x,t) is a to-be-learned function related to the state variable, b is a known non-zero constant, r(t) is parameter variation and external disturbance of the system, and B(x,t) is a known friction torque function (which can represent the relationship between system friction torque and input speed).

[0080] To achieve the control target, before designing the adaptive sliding mode iterative learning algorithm, ω ref is set as the expected speed signal of the system, and the speed loop control algorithm should ensure that the actual running speed of the servo system can track the given expected speed ω ref , e(t) is the tracking error of the output speed, and e(t)=ω ref(t) -ω(t), represents the difference between the actual speed and the desired speed. In any time period [0, T], the initial state of the system output speed error satisfies:

[0081] The initial error is bounded: the initial error e(0) is finite, that is, the speed error at the initial time of the system will not be too large.

[0082] The error change rate is bounded: the change rate de / dt of the initial error is also limited, that is, the change of the speed error at the initial time of the system will not be too fast.

[0083] The integral of the error and its derivative is bounded: the integral of the initial error and its derivative is limited, that is, the accumulation of the speed error in the past period of time will not be too large.

[0084] The second-order integral of the error and its derivative is bounded: the second-order integral of the initial error and its derivative is limited, that is, the accumulated change of the speed error in the past period of time will not be too large.

[0085] The integral sliding mode surface s(t) is selected as follows:

[0086] (2)

[0087] In the formula, c is the integral coefficient of the sliding mode surface (used to adjust the sliding mode surface to the speed), and c > 0. The difference between the two sides of the formula is taken with respect to time t:

[0088] (3)

[0089] Combining the above two formulas, we get:

[0090] (4)

[0091] The above is the dynamic equation of the sliding mode surface. Generally, the control input of the servo system is to make the system run towards the sliding mode surface, and the asymptotic rate needs to satisfy the time derivative σ> 0 of the sliding mode surface to ensure the stability of the dynamic system. The role of the controller is to make the system state trajectory reach the sliding mode surface in a limited time, where ; s(t) is the sliding mode surface, is the time derivative of the sliding mode surface.

[0092] The input control quantity u(t) needs to satisfy the sliding mode surface reaching condition, at this time the state variable v t is expressed by the sliding mode control rate, and the sliding mode control rate is selected as follows:

[0093] (5)

[0094] where β1>0 is the switching gain of the sign function, sgn(s)=1 when s>0, sgn(s)=0 when s=0, and sgn(s)=-1 when s<0. β2>0 is the linear gain.

[0095] Substituting equation (5) into equation (4), the adaptive sliding mode iterative learning controller at the kth iteration is designed as follows: (6)

[0096] where u k (t) is the output of the adaptive sliding mode iterative learning controller at the kth iteration; b -1 represents the inverse of the constant b; c is the integral coefficient; e k (t) is the tracking error of the kth output speed; represents the desired target speed; B(x k ,t) is a known friction torque function; represents the iterative control component (iterative control part) for learning the unknown periodic function, which can be used to learn the unknown periodic function; v k (t) represents the system state variable at the kth iteration; k is the iteration number of the controller, thus, the learning function f(x k ,t) about the state variable is learned, and the iterative learning control law is designed as follows:

[0097] (7)

[0098] where q, η1, η2 are all positive constants, s k (t) is the sliding mode surface dynamic at the kth iteration, and in order to improve the robustness of the iterative learning strategy, the concept of sliding mode control is combined in the design of the control law.

[0099] Combining equations (4) and (6), the simplified form of the sliding mode surface dynamic equation is obtained:

[0100] (8)

[0101] The above equation shows that if the learning function f(x k ,t) can be accurately learned, the sliding mode control rate v k (t) can effectively suppress the disturbance r k (t), and the sliding mode surface will tend to zero.

[0102] According to Lyapunov stability theorem, the adaptive sliding mode control method needs to satisfy the condition of the existence of sliding surface, and can effectively suppress system disturbance. Therefore, a simple adaptive algorithm is designed here to realize online estimation of system disturbance r(t), and its estimated value is used to compensate the control law, and the design of adaptive sliding mode iterative learning controller in the kth iteration is as follows:

[0103] (9)

[0104] In the formula, represents the disturbance estimation of the system, which is mainly based on the following adaptive law:

[0105] (10)

[0106] Here γ is the learning gain of the adaptive control law, and γ > 0.

[0107] The estimation error is defined as Where, represents the disturbance estimation, r k (t) represents the true disturbance value.

[0108] Equations (5), (7), (9), (10) define the adaptive sliding mode iterative learning control technology, and equations (4), (9) can be obtained:

[0109] (11)

[0110] The above formula shows that if can accurately learn the function to be learned f(x k ,t), the sliding mode control rate v k (t) can effectively control the disturbance error, and the sliding surface will tend to zero.

[0111] In order to estimate the sliding surface dynamics and the convergence of the speed response error, and analyze the stability of the control law designed above.

[0112] First, the Lyapunov function at the kth iteration is defined as:

[0113] (12)

[0114] The Lyapunov function defined in this application includes multiple Lyapunov sub-functions 、 、 、 The specific calculation formula of each sub-function is as follows:

[0115] ;

[0116] ;

[0117] ;

[0118] ;

[0119] where η1, η2 represent system parameter constants; s k (t) represents the integral sliding surface at the kth iteration; represents the learning error at the kth iteration, which can be represented and obtained by q represents a constant; represents the disturbance estimation error at the kth iteration; γ represents the learning gain of the adaptive control law.

[0120] Secondly, the difference between each Lyapunov sub-function in formula (12) at the kth iteration and the (k-1)th iteration is analyzed, and the specific steps are as follows:

[0121] The difference between the function V k 1 (t) at the kth iteration and the (k-1)th iteration is:

[0122] (13)

[0123] Combining formula (5), (11) can be obtained:

[0124] (14)

[0125] The difference between the function V k 2 (t) at the kth iteration and the (k-1)th iteration is:

[0126] (15)

[0127] Combining formula (5), (11) can be obtained:

[0128] (16)

[0129] The difference between the function V k 3 (t) at the kth iteration and the (k-1)th iteration is:

[0130] (17)

[0131] Since there is a relationship (a-b) T (a-b)- (a-c) T (a-c)= (c-b)T If (2(a-b) - (b-c)) > 0, then the above formula can be simplified as follows:

[0132] (18)

[0133] Combining formula (7), we have:

[0134] (19)

[0135] Similarly, we have:

[0136] (20)

[0137] Analysis function V k 4 The difference between the kth iteration and the (k-1)th iteration is:

[0138] (21)

[0139] Since the external disturbance is slower than the system state, combining formula (10), we have:

[0140] (22)

[0141] Combining formula (14), (16), (20), (21), we have the difference between the kth iteration and the (k-1)th iteration of the Lyapunov function:

[0142] (23)

[0143] Let the switching gain β1 satisfy the condition β1≥ |r|, then the above formula can be simplified as:

[0144] ΔV k (t)≤0 (24)

[0145] According to the Lyapunov stability theorem, the Lyapunov function V k (t) defined above is convergent, and the sliding surface s k (t) tends to zero and satisfies the reachable condition, and the speed loop tracking error of the servo system can also asymptotically tend to zero. The adaptive law added in the system can reduce the switching gain, ensure the robustness, and also reduce the chattering of the servo system.

[0146] On the basis of the above, the invention introduces a fuzzy logic module, which dynamically adjusts the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller, to obtain an optimized switching gain and adaptive sliding mode iterative learning controller.

[0147] Figure 3 The integration of fuzzy logic module and sliding mode controller is shown; the fuzzy logic module includes:

[0148] Fuzzy input module, for input variable sliding surface error s k (t) and its rate of change .

[0149] Fuzzy module, which maps input variables into fuzzy sets (such as "negative large", "zero", "positive large") through triangular / trapezoidal membership functions.

[0150] Rule base, for storing representative rules (example):

[0151] If s k is "negative large" and is "positive large", then Δβ1= +0.2.

[0152] Defuzzification module, which converts fuzzy output into parameter adjustment amount Δβ1 and Δc using the center of gravity method.

[0153] Parameter dynamic adjustment module, for parameter update adjustment, the parameter update formula is as follows:

[0154] β1(k)=β1(k-1)+Δβ1,c(k)=c(k-1)+Δc.

[0155] The parameter dynamic adjustment module arrow points to the original sliding mode controller, indicating real-time feedback of parameters.

[0156] The fuzzy logic module is connected in series with the adaptive sliding mode iterative learning controller (SMC-ILC), and the optimized control amount u k (t) is output.

[0157] As another improved embodiment, the application also introduces a neural network online identification system, Figure 4 is a neural network online identification flowchart, which describes the input / output of the neural network and the weight update process, and the weight update path aims to "Lyapunov stability guarantee". The input layer inputs the system state variable x k (t). The hidden layer uses RBF (radial basis function) neural network; the base function uses Gaussian function:

[0158] ;

[0159] The number of hidden layer nodes is 5;

[0160] The output layer outputs "compensation of nonlinear disturbance", and the specific output of the output layer is the nonlinear function estimate value , the formula is:

[0161] ;

[0162] Weight update from the sliding surface error s k (t), feedback to the weight matrix W k (t), the update law is:

[0163] ;

[0164] Figure 4 The arrow marked "online learning" indicates the direction.

[0165] Finally, through compensation control output and input to the controller, used to offset the nonlinear term of the system.

[0166] Multi-axis servo turntable can be applied to various optical systems for target monitoring. Through sensors, optical sensors, current and voltage sensors, etc. can be used to monitor the working state according to the corresponding signal feedback. For example, using current and voltage sensors to detect the current of the servo motor, and then analyzing whether the turntable is running normally. Or using a camera to monitor the target can realize visual feedback control, which is used to monitor the position and attitude of the turntable in real time, and adjust the control to run at the correct position. This method is very effective for attitude recognition and position control of rotating objects. In addition, machine vision technology can be used to monitor the target position of the optical system in real time, process image data through algorithms, extract key information, and then evaluate the motion state of the turntable. The acquired data can also be processed and analyzed by software to realize the motion control of the turntable and the fault diagnosis of the optical system.

[0167] Specifically, a camera is installed on the rotating body of the turntable to collect multi-directional images of the target object. The mechanical connection between the optical system and the servo motor exists inside the device. Image data is usually collected by cameras or other imaging devices installed around the optical system. These devices will capture the position of the target and the changes in the environment in real time. The collected image data will contain visual information such as the position, shape, color of the target object, and the motion of the turntable.

[0168] The collected image data will be analyzed and processed by image processing algorithms. These algorithms include but are not limited to image preprocessing (such as denoising, enhancement), target detection, edge detection, morphological processing, etc. During processing, the algorithm extracts key information such as the position and shape of the target. Key information mainly includes the current position, motion trajectory, rotation speed, and acceleration of the target. These information can reflect the motion state of the target in the optical system. The extracted key information will be compared with the motion parameters of the turntable. Through comparison, it can be evaluated whether the current motion state of the turntable meets the expectations. For example, if the actual position of the target deviates from the expected position, the system can correct the deviation by adjusting the motion parameters of the turntable.

[0169] The motion state of the turntable directly affects the target position of the optical system. Image data provides real-time feedback of the target position, helping to adjust the motion of the turntable to achieve the best working state. Through this closed-loop control, it can ensure that the optical system is always in the best alignment state, improving the working efficiency and accuracy of the system. In summary, the processing of image data and the extraction of key information are achieved through real-time monitoring and analysis of the target position, and then the motion state of the turntable is adjusted to enable the optical system to accurately track and locate the target. Thus the present application has wide application prospects.

[0170] Multi-axis servo turntable can realize high-precision target positioning and rotation through multi-axis cooperation, and the tracking accuracy can reach 30urad. Each axis is provided with a high-precision encoder to accurately and reliably feedback position information. In addition, the monitoring system monitors the state and position information of each axis in real time, improving the safety and stability of the system.

[0171] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0172] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-axis servo turntable system, characterized in that: include: A drive mechanism, a control unit, and a turntable body, wherein the drive mechanism includes multiple servo motors for driving the turntable body to rotate and move axially; the control unit uses a fuzzy adaptive sliding mode iterative learning control algorithm to control each servo motor, specifically including: Constructing an adaptive sliding mode iterative learning controller, and using the sliding mode control rate of the adaptive sliding mode iterative learning controller as a state variable of each servo motor control unit, wherein the state variable includes the rotational speed of the servo motor; The fuzzy logic module is used to dynamically adjust the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller to obtain the optimized switching gain and the adaptive sliding mode iterative learning controller; The stability analysis and error convergence analysis of the optimized adaptive iterative learning sliding mode controller model are performed using the Lyapunov method. An adaptive iterative learning sliding mode controller model that satisfies stability and error convergence conditions is applied to the control unit of each servo motor; It also includes the use of neural networks to estimate the nonlinear disturbance state of the system in real time. Specifically, Introduce the neural network online identification system, and input the system state variable x into the input layer k (t), the hidden layer uses RBF neural network, the number of hidden layer nodes is 5; the basis function φ(x k ,t) is a Gaussian function, and the nonlinear function estimate output by the output layer is The weight update is based on the sliding surface error s k (t), fed back to the weight matrix W k (t), the update law is: Finally, the output is controlled by compensation And input to the controller to offset the nonlinear terms of the system.

2. The multi-axis servo turntable system according to claim 1, characterized in that: Multiple servo motors drive the turntable body to rotate and move axially, specifically including the following steps: The servo controller of each servo motor receives feedback information from a corresponding encoder and controls each axis according to a fuzzy adaptive sliding mode iterative learning control algorithm preset by the control unit. The encoder is set on each axis.

3. The multi-axis servo turntable system according to claim 2, characterized in that: In the step of constructing an adaptive sliding mode iterative learning controller, the adaptive sliding mode iterative learning controller is expressed by the following formula: Among them, u k (t) is the k-th output of the adaptive sliding mode iterative learning controller; b -1 represents the reciprocal of the constant b; c is the integral coefficient, e k (t) is the tracking error of the k-th output speed; represents the desired target speed; B(x k ,t) is the known friction torque function; represents the desired iterative control component for learning the unknown periodic function; v k (t) represents the system state variable at the kth iteration; Represents the k-th disturbance estimate of the system.

4. The multi-axis servo turntable system according to claim 1, characterized in that: The fuzzy logic module is used to dynamically adjust the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller to obtain the optimized switching gain and the adaptive sliding mode iterative learning controller, which specifically includes the following steps: Define the integral sliding surface s at the kth iteration k (t) and its rate of change As input variables of fuzzy logic modules; Map the input variables to corresponding fuzzy sets through membership functions; The center of gravity method is used to convert the fuzzy set output into parameter adjustment values ​​Δβ1 and Δc, where Δβ1 is the switching gain adjustment value and Δc is the integral coefficient adjustment value. The optimized switching gain is obtained according to the switching gain adjustment amount Δβ1: β1(k)=β1(k-1)+Δβ1 The optimized integral coefficient is obtained according to the integral coefficient adjustment Δc: c(k)=c(k-1)+Δc The optimized integral coefficient is applied to the adaptive sliding mode iterative learning controller to obtain the optimized adaptive sliding mode iterative learning controller.

5. The multi-axis servo turntable system according to claim 1, characterized in that: The stability analysis and error convergence analysis of the optimized adaptive iterative learning sliding mode controller model using the Lyapunov method specifically include the following steps: Establish the Lyapunov function at the kth iteration: Calculate multiple Lyapunov subfunctions respectively according to the Lyapunov function Where η1 and η2 represent system parameter constants; s k (t) represents the integrated synovial surface at the kth iteration; represents the learning error of the kth iteration; q represents a constant; represents the disturbance estimation error of the kth iteration; γ represents the learning gain of the adaptive control law; Compute the difference between the kth iteration and the k-1th iteration for each Lyapunov subfunction i=1,2,3,4; Will get The sum of i=1,2,3,4 gives the total difference ΔV between the kth iteration and the k-1th iteration of the Lyapunov function. k (t); Let the total difference ΔV between the kth iteration and the k-1th iteration of the Lyapunov function be k (t) is less than or equal to 0, then the adaptive iterative learning sliding mode controller model meets the stability and error convergence conditions.

6. A multi-axis servo turntable system control method, characterized in that: The following steps are involved: The servo motor speed is selected as the system state variable, and the parameter perturbation and external disturbance are considered to establish the servo motor dynamic system; Constructing an adaptive sliding mode iterative learning controller, and using the sliding film control rate of the adaptive sliding mode iterative learning controller as a state variable of each servo motor control unit; The fuzzy logic module is used to dynamically adjust the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller to obtain the optimized switching gain and the adaptive sliding mode iterative learning controller; The stability analysis and error convergence analysis of the optimized adaptive iterative learning sliding mode controller model are performed using the Lyapunov method. An adaptive iterative learning sliding mode controller model that satisfies stability and error convergence conditions is applied to the control unit of each servo motor; It also includes the use of neural networks to estimate the nonlinear disturbance state of the system in real time. Specifically, Introduce the neural network online identification system, and input the system state variable x into the input layer k (t), the hidden layer uses RBF neural network, the number of hidden layer nodes is 5; the basis function φ(x k , t) is a Gaussian function, and the nonlinear function estimate output by the output layer is The weight update is based on the sliding surface error s k (t), fed back to the weight matrix W k (t), the update law is: Finally, the output is controlled by compensation And input to the controller to offset the nonlinear terms of the system.

7. The multi-axis servo turntable system control method according to claim 6, characterized in that: In the step of constructing an adaptive sliding mode iterative learning controller, the adaptive sliding mode iterative learning controller is expressed by the following formula: Among them, u k (t) is the k-th output of the adaptive sliding mode iterative learning controller; b -1 represents the reciprocal of the constant b; c is the integral coefficient, e k (t) is the tracking error of the k-th output speed; represents the desired target speed; B(x k ,t) is the known friction torque function; represents the desired iterative control component for learning the unknown periodic function; v k (t) represents the system state variable at the kth iteration; Represents the k-th disturbance estimate of the system.

8. The multi-axis servo turntable system control method according to claim 6, characterized in that: The fuzzy logic module is used to dynamically adjust the switching gain of the sliding mode control rate and the integral coefficient of the adaptive sliding mode iterative learning controller to obtain the optimized switching gain and the adaptive sliding mode iterative learning controller, which specifically includes the following steps: Define the integral sliding surface s at the kth iteration k (t) and its rate of change As input variables of fuzzy logic modules; Map the input variables to corresponding fuzzy sets through membership functions; The center of gravity method is used to convert the fuzzy set output into parameter adjustment values ​​Δβ1 and Δc, where Δβ1 is the switching gain adjustment value and Δc is the integral coefficient adjustment value. The optimized switching gain is obtained according to the switching gain adjustment amount Δβ1: β1(k)=β1(k-1)+Δβ1 The optimized integral coefficient is obtained according to the integral coefficient adjustment Δc: c(k)=c(k-1)+Δc The optimized integral coefficient is applied to the adaptive sliding mode iterative learning controller to obtain the optimized adaptive sliding mode iterative learning controller.

Images