Iterative learning control method, system and equipment for controlled system and medium

Abstract

The invention relates to the technical field of iterative learning control, and discloses a controlled system iterative learning control method, system and device and a medium, and the method comprises the steps: obtaining a historical input signal, a current error sequence and a historical error sequence of a controlled system; under the condition that data loss exists in the current input signal, determining a target round input signal in the historical input signals; setting a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence; obtaining an updated input signal according to the high-order feedback iterative learning control law, and inputting the updated input signal to the controlled system; and iterating the step of obtaining the updated input signal until an end condition is reached. The method has the beneficial effects that the influence of data loss on input signal updating is reduced, so that the output of the controlled system can still quickly approach to expected output under the condition of data loss, and the accurate control of the controlled system is realized.

Description

Technical Field

[0001] The present application relates to the field of iterative learning control technology, and in particular to an iterative learning control method, system, device and medium for a controlled system. Background Art

[0002] In today's era of rapid technological development, controlled systems and controllers have been able to transcend geographical restrictions and achieve remote collaborative work. Through the efficient communication medium of wireless networks, even if the controlled system and controller are in different locations, real-time transmission of information and accurate execution of instructions can be ensured, thereby controlling the normal operation of the controlled system and meeting various application requirements. This long-distance wireless control method not only improves work efficiency, but also greatly broadens the application scope of controlled systems and controllers.

[0003] When data is transmitted between the controlled system and the controller, the huge amount of data not only easily leads to delays in system response, but also often comes with the risk of data loss. Therefore, in the context of large data transmission volume and delayed system response, how to reduce the impact of data loss and achieve precise control of the controlled system has become a technical problem. Summary of the invention

[0004] The present application provides an iterative learning control method, system, device and medium for a controlled system, which solves the technical problem of input signal stopping updating due to data loss, thereby reducing the impact of data loss on input signal updating, so that in the case of data loss, the output signal of the controlled system can still quickly approach the expected output signal, thereby achieving the technical effect of precise control of the controlled system.

[0005] In order to achieve the above objectives, the main technical solutions adopted in this application include:

[0006] In a first aspect, an embodiment of the present application provides an iterative learning control method for a controlled system, the method comprising:

[0007] Acquire a historical input signal of a controlled system, a current error sequence of the controlled system at a current moment, and a historical error sequence of the controlled system at a moment before the current moment; wherein the historical input signal at least includes an input signal of a previous iteration round, the current error sequence is determined based on a difference between an expected output sequence of the controlled system and a current output sequence in a current iteration round, and the current output sequence is generated by controlling the controlled system using the input signal of the previous iteration round as a current input signal;

[0008] In the case where data is lost in the current input signal, determining a target round input signal before the previous iteration round in the historical input signal;

[0009] Setting a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence;

[0010] Obtaining an updated input signal according to the high-order feedback iterative learning control law, and inputting the updated input signal into the controlled system;

[0011] The above steps of obtaining the updated input signal are iterated until the end condition is reached.

[0012] The iterative learning control method for the controlled system proposed in the embodiment of the present application, when data loss occurs in the output sequence of the controlled system, updates the input signal according to the historical error sequence and the target round input signal in the historical input signal on the basis of the conventional control law by setting a high-order feedback iterative learning control law. This method solves the technical problem of stopping the update of the input signal due to data loss, reduces the impact of data loss on the update of the input signal, so that in the case of data loss, the output of the controlled system can still quickly approach the expected output, thereby realizing precise control of the controlled system.

[0013] Optionally, the high-order feedback iterative learning control law includes a high-order learning term and a feedback update term; and setting the high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence includes:

[0014] Based on the target round input signal, the current error sequence and the data loss characterization parameter, the high-order learning item is obtained to perform data compensation on the current input signal;

[0015] The feedback update term is obtained based on the historical error sequence and the data loss characterization parameter to supplement the target round input signal when the target round input signal does not meet the preset iteration requirement.

[0016] Optionally, the high-order learning item includes an input learning item and an error learning item; the obtaining the high-order learning item based on the target round input signal, the current error sequence and the data loss characterization parameter to perform data compensation on the current input signal includes:

[0017] Learning the target round input signal according to the input learning gain to obtain the input learning item;

[0018] The current error sequence is learned according to the error learning gain and the data loss characterization parameter to determine the error learning item.

[0019] Optionally, the high-order learning item is in the following form:

[0020]

[0021] Among them, u f,k+1 (n) represents the high-order learning item, represents the input learning item, represents the error learning term; u k-j+1 (n) is the input signal at the nth moment in the k-j+1th iteration round; γ k-j+1 (n+1) is the data loss parameter at the n+1th moment in the k-j+1th iteration round, γ k-j+1 (n+1)=1 means no data loss occurs, γ k-j+1 (n+1)=0 means data loss has occurred; is the current error sequence at the n+1th moment; P j is the learning gain of the jth input; L j is the jth error learning gain; N is the number of input signals in the target round.

[0022] Optionally, the input learning gain and the error learning gain satisfy the following convergence condition:

[0023]

[0024] Among them, γ j (n+1) is the j-th data loss characterization parameter at the n+1th time, γ j (n+1)=1 means that there is no data missing in the historical output sequence corresponding to the j-th target round input signal, γ j (n+1)=0 means that data is missing in the historical output sequence corresponding to the j-th target round input signal; ρ j is the convergence coefficient corresponding to the jth historical input signal; C and B are related to the controlled system.

[0025] Optionally, the feedback update item is in the following form:

[0026]

[0027] Among them, u b,k+1 (n) represents the feedback update term; γ k+1 (n) is the data loss characterization parameter at the nth moment, γ k+1 (n) = 1 means that no data loss occurs in the current output sequence. k+1 (n) = 0 means that data loss has occurred in the current output sequence; is the historical error sequence of the kth iteration round; Γ is the feedback constraint coefficient.

[0028] Optionally, the controlled system and the controller are connected via a codec network, the codec network includes an encoder, a quantizer, a network path and a decoder, and the quantizer is a unit uniform quantizer.

[0029] In a second aspect, an embodiment of the present application provides an iterative learning control system for a controlled system, the system comprising:

[0030] A data acquisition module, used to acquire historical input signals of a controlled system, a current error sequence of the controlled system at a current moment, and a historical error sequence of the controlled system at a moment before the current moment; wherein the historical input signals at least include input signals of a previous iteration round, the current error sequence is determined based on the difference between an expected output sequence of the controlled system and a current output sequence of a current iteration round, and the current output sequence is generated by taking the input signal of the previous iteration round as a current input signal and controlling the controlled system;

[0031] a target round determination module, configured to determine, in the case where data loss exists in the current input signal, a target round input signal before the previous iteration round in the historical input signal;

[0032] A learning control module, used for setting a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence;

[0033] A signal updating module, used for obtaining an updated input signal according to the high-order feedback iterative learning control law, and inputting the updated input signal to the controlled system;

[0034] The iterative calculation module is used to iteratively execute the module for obtaining the updated input signal until the end condition is reached.

[0035] In a third aspect, an embodiment of the present application provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the iterative learning control method for the controlled system described in any one of the above embodiments by executing the computer instructions.

[0036] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the iterative learning control method for a controlled system described in any one of the above embodiments.

[0037] In a fifth aspect, an embodiment of the present application provides a computer program product, including computer instructions, wherein the computer instructions are used to enable a computer to execute the iterative learning control method for a controlled system described in any one of the above embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A step diagram of an iterative learning control method for a controlled system provided in an embodiment of the present application;

[0040] Figure 2 It is an iterative learning control block diagram in an embodiment of the present application;

[0041] Figure 3 A diagram showing the steps for setting a high-order feedback iterative learning control law in an embodiment of the present application;

[0042] Figure 4 This is an iterative learning control block diagram of a codec network included in an embodiment of the present application;

[0043] Figure 5 It is a diagram showing the expected output trajectory of the permanent magnet linear motor and the system output at different iteration times in the embodiment of the present application;

[0044] Figure 6 This is a comparison diagram of the output errors of the permanent magnet linear motor under different data loss rates in the embodiment of the present application;

[0045] Figure 7 This is a comparison diagram of the output error of the permanent magnet linear motor in the embodiment of the present application when the data loss rate is 20% and the output error of the traditional iterative learning control;

[0046] Figure 8 A module diagram of an iterative learning control system for a controlled system provided in an embodiment of the present application;

[0047] Fig. 9 A schematic diagram of the structure of a computer device provided in an embodiment of the present application.

[0048] Among them, the figure numbers of the drawings in the specification are as follows: 100. Data acquisition module, 200. Target round determination module, 300. Learning control module, 400. Signal update module, 500. Iterative calculation module. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0050] In some engineering applications that require extremely high control accuracy, not only does the controlled system need to be able to continuously repeat the same control task within a specific range, but it must also ensure that the output of the controlled system can accurately and without deviation track the expected output trajectory set above. Such requirements are crucial to ensuring the stability and performance of the system and are the key to achieving high-precision control. Conventional controlled systems include motors and other systems that require feedback updates through errors. In today's era of rapid technological development, motors and controllers have been able to transcend geographical limitations and achieve remote collaborative work. Through wireless networks, an efficient communication medium, even if the motor and controller are in different locations, real-time transmission of information and accurate execution of instructions can be ensured, thereby controlling the normal operation of the motor and meeting various application requirements. This long-distance wireless control method not only improves work efficiency, but also greatly broadens the application scope of motors and controllers.

[0051] When the motor system is transmitting data, the huge amount of data may not only cause delays in system response, but also often be accompanied by the risk of data loss. This delay and data loss problem will cause time lag after the control input, especially for high-speed motors, which undoubtedly increases the complexity and difficulty of control. What is more difficult is that if the model information of the motor system cannot be accurately obtained, the mainstream control algorithms that mainly rely on time domain information will be stretched when pursuing high-speed and accurate tracking control.

[0052] The connection methods between the motor and the control system include communication interface and communication protocol, fiber optic communication, power line communication and codec network, etc. Different connection methods may cause the loss of transmitted data due to signal interference, network congestion, hardware failure or protocol error. In the motor control process, data loss will cause the motor output signal received in the controller to be incomplete, and during the iteration process, the input signal output by the controller to the motor will not be updated, resulting in the motor output signal being unable to approach the expected output signal.

[0053] Therefore, in the context of large data transmission volume and delayed system response, how to achieve precise control of high-speed motors has become a technical challenge.

[0054] Based on the above problems, the present application provides an iterative learning control method, system, device and medium for a controlled system, wherein the historical input signal, current error sequence and historical error sequence of the controlled system are obtained; when there is data loss in the current input signal, the target round input signal is determined in the historical input signal; according to the target round input signal, the current error sequence and the historical error sequence, a high-order feedback iterative learning control law is set; according to the high-order feedback iterative learning control law, an updated input signal is obtained, and the updated input signal is input to the controlled system; the above steps of obtaining the updated input signal are iterated until the end condition is reached. The beneficial effect of this method is that the influence of data loss on the update of the input signal is reduced, so that in the case of data loss, the output of the controlled system can still quickly approach the expected output, thereby realizing precise control of the controlled system.

[0055] The controlled system iterative learning control method provided in this specification can be applied to a controlled system-controller that is communicatively connected via a codec network, wherein the controller is also connected to a memory; the controlled system may include a permanent magnet linear motor, and the method can also be used for other types of motors after adaptive modification. It is understandable that the method can also be applied to other communication connection methods in which data loss may occur during the communication connection process. Exemplarily, the communication connection method between the controlled system and the controller can also include other communication connection methods, including but not limited to communication interfaces and communication protocols, optical fiber communication, and power line communication.

[0056] According to an embodiment of the present application, an embodiment of an iterative learning control method for a controlled system is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0057] In this embodiment, a controlled system iterative learning control method is provided, which can be used for the above-mentioned controlled system-controller which is connected for communication via a codec network. Figure 1 The step diagram of the iterative learning control method for the controlled system provided in the embodiment of the present application, as shown in the figure, the method includes:

[0058] S100. Obtain the historical input signal of the controlled system, the current error sequence of the controlled system at the current moment, and the historical error sequence of the controlled system at the moment before the current moment; wherein the historical input signal at least includes the input signal of the previous iteration round, the current error sequence is determined based on the difference between the expected output sequence of the controlled system and the current output sequence in the current iteration round, and the current output sequence is generated by taking the input signal of the previous iteration round as the current input signal and controlling the controlled system.

[0059] S200. In the case where there is data loss in the current input signal, determine a target round input signal before the previous iteration round in the historical input signal.

[0060] S300. Set a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence.

[0061] S400. Obtain an updated input signal according to a high-order feedback iterative learning control law, and input the updated input signal to the controlled system.

[0062] S500. Iterate the above steps of obtaining the updated input signal until the end condition is reached.

[0063] Specifically, the current output sequence is obtained by collecting the current output trajectory of the controlled system in the current iteration round at a certain sampling period. Before step S100, it also includes setting the sampling period of the controlled system. Suppose the sampling period T S For Δ, the sampling discrete sequence can be obtained as follows:

[0064]

[0065] Among them, T d is the motor running time. Accordingly, the sampling time series corresponding to the sampling discrete sequence is in the following form:

[0066]

[0067] It is understood that the moments in the expected output sequence are the same as the sampling time sequence. During the iterative calculation process, the expected output sequence can be manually set by the user and input into the controller.

[0068] Accordingly, the initial input signal when the controlled system starts working is in the following form:

[0069]

[0070] Wherein, u0(t) represents the initial input signal.

[0071] Reference Figure 2 As shown, Figure 2 This is an iterative learning control block diagram in an embodiment of the present application. As shown in the figure, the controlled system is connected to the controller through the first transmission network, and transmits the current output sequence to the controller through the first transmission network. There is random data loss in the first transmission network, resulting in a certain amount of data loss when the current output sequence reaches the controller after passing through the first transmission network, that is, the current output sequence transmitted to the controller is the current output sequence y where data loss occurs k(t+1). If data loss occurs in the current output sequence, the controller cannot update the input signal according to the current input signal, causing the input signal to stop updating, resulting in the output sequence of the controlled system being unable to approach the expected output sequence.

[0072] Accordingly, the controller is connected to the controlled system through the second transmission network, and transmits the updated input signal to the controlled system through the second transmission network. There is also random data loss in the second transmission network, resulting in a certain amount of data loss when the updated input signal reaches the controlled system after passing through the second transmission network, that is, the updated input signal transmitted to the controlled system The updated input signal u is the one where data loss occurs. k+1 (t). It should be noted that the present application mainly aims at improving the data loss caused by the first transmission network, and as for the data loss caused by the second transmission network, the impact of the data loss caused by the second transmission network on the controlled system can be reduced by adaptively modifying the controlled system. The present application does not make any specific limitation on this method.

[0073] Furthermore, the controller is connected to a memory, and the memory is used to store the initial input signal of the controlled system and multiple updated input signals obtained by the controller during the iteration process. Since the data loss in the first transmission network occurs randomly, if there is no data loss in the current output sequence transmitted to the controller, a conventional iterative learning control law can be used, that is, the current input signal u is updated according to the learning gain. k (t) and the current error sequence are learned to obtain the updated input signal. The conventional iterative learning control law here can be selected according to the actual scenario needs without specific limitation.

[0074] If data loss occurs in the current output sequence transmitted to the controller, the high-order feedback iterative learning control law set in the method is used at this time, and the target round input signal of the controlled system before the previous iteration round, the current error sequence and the historical error sequence of the controlled system at the previous moment of the current moment are output to the controller through the memory, so as to obtain the updated input signal according to the high-order feedback iterative learning control law, so that the input signal of the controlled system is updated in each iteration round, and can approach the expected input signal more quickly, and the controlled system can obtain the expected output signal, that is, the expected output sequence, through the expected input signal. Among them, the target round is a plurality of iteration rounds arranged in sequence before the previous iteration round, and the number of target rounds can be determined by the number of data loss in the historical output sequence before the previous iteration round in the memory. The more data loss occurs in the historical output sequence, the more target rounds there are; the fewer data loss occurs in the historical output sequence, the fewer target rounds there are; therefore, it can be understood that the number of target rounds is proportional to the number of data loss in the historical output sequence.

[0075] Furthermore, the obtained updated input signal is transmitted to the controlled system, so that the controlled system works according to the updated input signal, that is, one iteration round is completed. By repeating the steps of iteratively obtaining the updated input signal, the output sequence of the controlled system gradually approaches the expected output sequence, thereby achieving precise control of the controlled system. Exemplarily, the end condition includes an error condition and an iteration number condition, wherein the error condition is that the current error sequence is within the allowable range, indicating that the output sequence of the controlled system is close enough to the expected output sequence at this time, meeting the precise control requirements; the iteration number condition is that the number of iterative calculations in the controller reaches the preset number.

[0076] In some embodiments, the controlled system is a permanent magnet linear motor, which can be represented by the following model:

[0077]

[0078] Among them, x k (t) is the motor position; v k (t) is the motor speed; y k (n) is the output trajectory of the motor; R is the stator resistance; m is the rotor mass; ψ f is the magnetic flux; k1 and k2 are the magnetic pole pitches, k1 = π / τ, k2 = 1.5π / τ, and τ = 0.031m. For this model, there is an expected output trajectory y d (t).

[0079] By discretizing the above model, we can get a discrete model in the following form:

[0080]

[0081] in, C =

[01] ; is the output sequence. Accordingly, according to the expected output trajectory, the sampling time sequence is sampled to obtain the discretized expected output sequence

[0082] According to the expected output sequence and the current output sequence in the current iteration round, the current error sequence of the model can be obtained, which is in the following form:

[0083]

[0084] According to the current error sequence and the current input signal for generating the current output sequence, the control input signal of the permanent magnet linear motor can be updated to obtain an updated input signal.

[0085] The iterative learning control method for the controlled system provided in this embodiment, when data loss occurs in the output sequence of the controlled system, updates the input signal according to the historical error sequence and the target round input signal in the historical input signal on the basis of the conventional control law by setting a high-order feedback iterative learning control law. This method solves the technical problem of stopping the update of the input signal due to data loss, reduces the impact of data loss on the update of the input signal, so that in the case of data loss, the output of the controlled system can still quickly approach the expected output, thereby realizing precise control of the controlled system.

[0086] Reference Figure 3 As shown, Figure 3 This is a step diagram for setting a high-order feedback iterative learning control law in an embodiment of the present application. As shown in the figure, as an embodiment of the present application, the high-order feedback iterative learning control law includes a high-order learning term and a feedback update term; according to the target round input signal, the current error sequence and the historical error sequence, the high-order feedback iterative learning control law is set, including:

[0087] S310. Based on the target round input signal, the current error sequence and the data loss characterization parameter, a high-order learning item is obtained to perform data compensation on the current input signal.

[0088] S320. Based on the historical error sequence and the data loss characterization parameter, a feedback update term is obtained to supplement the target round input signal when the target round input signal does not meet the preset iteration requirement.

[0089] Specifically, the high-order learning item is an improvement on the conventional iterative learning control law. According to the target round input signal, the current error sequence and the data loss characterization parameter, the high-order learning item is obtained to perform data compensation on the current input signal. Among them, the current error sequence and the data loss characterization parameter are used to feedback the deviation between the current output sequence and the expected output sequence to the controller, thereby reducing the deviation through iterative calculation, reducing the instability of the controlled system, and improving the control accuracy of the controlled system. It should be noted that the data loss characterization parameter in the high-order learning item corresponds to the target round, indicating whether the output sequence in the corresponding target round has data loss. By using the high-order learning item obtained by the target round input signal, the current error sequence and the data loss characterization parameter to compensate for the current input signal, the influence of data loss on the input signal update is reduced, so that in the case of data loss, the output of the controlled system can still quickly approach the expected output, thereby achieving precise control of the controlled system.

[0090] Furthermore, if the number of target rounds is set lower than the preset iteration requirement, the effect of data compensation of the current input signal only through the high-order learning term does not meet the requirement, resulting in a decrease in the speed at which the updated input signal approaches the expected input signal. The feedback update term is used to supplement the input signal of the target round, that is, to update the input signal of the current iteration round using the historical error sequence of the previous iteration round, so that when the number of target rounds does not meet the preset iteration requirement, the input signal can also be updated, so that the output of the controlled system can quickly approach the expected output.

[0091] As an embodiment of the present application, the high-order learning item includes an input learning item and an error learning item; based on the target round input signal, the current error sequence and the data loss characterization parameter, the high-order learning item is obtained to perform data compensation on the current input signal, including:

[0092] S312. Learn the target round input signal according to the input learning gain to obtain an input learning item.

[0093] S314. Learn the current error sequence according to the error learning gain and the data loss characterization parameter to determine the error learning item.

[0094] Specifically, the input learning item is used to learn according to the target round input signal, thereby updating the input signal. It should be noted that in addition to learning the target round input signal, similar to the conventional iterative learning control law, the input learning item also learns the current input signal, that is, the input learning item learns the input signals of the current iteration round and the previous multiple iteration rounds to reduce the impact of data loss on the input signal update.

[0095] Furthermore, the error learning term is used to learn according to the current error sequence, and to feedback the deviation between the current output sequence and the expected output sequence, thereby reducing the deviation through iterative calculation, reducing the instability of the controlled system, and improving the control accuracy of the controlled system. It should be noted that the data loss characterization parameter in the error learning term corresponds to the target round, indicating whether data loss occurs in the output sequence in the corresponding target round.

[0096] As an embodiment of the present application, the form of the high-order learning item is as follows:

[0097]

[0098] Among them, u f,k+1 (n) represents a high-order learning item, represents the input learning item, represents the error learning term; u k-j+1 (n) is the input signal at the nth moment in the k-j+1th iteration round; γ k-j+1 (n+1) is the data loss parameter at the n+1th moment in the k-j+1th iteration round, γ k-j+1 (n+1)=1 means no data loss occurs, γ k-j+1 (n+1)=0 means data loss has occurred; is the current error sequence at the n+1th moment; P j is the learning gain of the jth input; L j is the jth error learning gain; N is the number of input signals in the target round.

[0099] It should be noted that any iteration round contains multiple moments, that is, the updated input signal obtained is a discrete sequence containing multiple moments. Figure 2 As shown, the error learning term is one moment different from the high-order learning term.

[0100] As an embodiment of the present application, the input learning gain and the error learning gain satisfy the following convergence conditions:

[0101]

[0102] Among them, γ j (n+1) is the j-th data loss characterization parameter at the n+1th time, γ j (n+1)=1 means that there is no data missing in the historical output sequence corresponding to the j-th target round input signal, γ j (n+1)=0 means that data is missing in the historical output sequence corresponding to the j-th target round input signal; ρ j is the convergence coefficient corresponding to the jth historical input signal; C and B are related to the controlled system.

[0103] Specifically, for each target round, there is a corresponding input learning gain and error learning gain, wherein the input learning gain is used to learn the input signal of the corresponding target round; the error learning gain is used to learn the error sequence of the corresponding target round, thereby constraining the error of the updated input signal.

[0104] Furthermore, for the convenience of explanation, the convergence conditions are referred to as the first convergence condition, the second convergence condition and the third convergence condition from top to bottom. Among them, the first convergence condition indicates that the updated input signal can converge after multiple iterations, reducing the error between the output sequence of the controlled system and the expected output sequence to zero, and keeping the input signal of the previous iteration round and the next iteration round the same after the iteration, thereby improving the stability of the controlled system. The second convergence condition indicates that the high-order learning items corresponding to any target round in the iteration process converge, that is, after learning for the target round, the error of the updated input signal is reduced and does not exceed the preset convergence coefficient, so that the updated input signal can converge after multiple iterations. The third convergence condition indicates that the sum of the convergence coefficients is not greater than 1, thereby improving the convergence speed in the iteration, so that the updated input signal can approach the expected input signal faster.

[0105] As an embodiment of the present application, the form of the feedback update item is as follows:

[0106]

[0107] Among them, u b,k+1 (n) represents the feedback update term; γ k+1 (n) is the data loss characterization parameter at the nth moment, γ k+1 (n) = 1 means that no data loss occurs in the current output sequence. k+1 (n) = 0 means that data loss has occurred in the current output sequence; is the historical error sequence of the kth iteration round; Γ is the feedback constraint coefficient.

[0108] Specifically, Figure 2 As shown, γ k+1 (n) corresponds to the current output sequence and is used to indicate whether data loss occurs in the current output sequence. The feedback constraint coefficient is used to constrain the feedback update term to avoid the problem of non-convergence of the updated input signal obtained by the feedback update term, thereby improving the convergence speed in the iteration and making the updated input signal approach the expected input signal faster.

[0109] Further, according to the high-order learning term and the feedback update term, the form of the high-order feedback iterative learning control law in this application can be obtained as follows:

[0110]

[0111] Among them, u k+1 (n) represents the updated input sequence. The updated input signal can be obtained according to the updated input sequence at different times, and its form is as follows:

[0112]

[0113] Among them, u k+1 (t) represents the updated input signal.

[0114] As an embodiment of the present application, the controlled system and the controller are connected via a codec network, the codec network includes an encoder, a quantizer, a network path and a decoder, and the quantizer is a unit uniform quantizer.

[0115] Reference Figure 4 As shown, Figure 4 It is an iterative learning control block diagram including a codec network in an embodiment of the present application. As shown in the figure, in this embodiment, the first transmission network and the second transmission network are both codec networks, and the output data of the controlled system is encoded by the first encoder E1 in the first transmission network, and the encoded data is transmitted through the first network path, and decoded by the first decoder D1, so as to be output to the controller. Correspondingly, the output data of the controller is encoded by the second encoder E2 in the second transmission network, and the encoded data is transmitted through the second network path, and decoded by the second decoder D2, so as to be output to the controlled system. It should be noted that both the first network path and the second network path have random data loss. Through the codec network, the analog data output by the controlled system or controller is quantized and data compressed, so as to be converted into digital data for transmission. Compared with the original data, the amount of information of the digital data will be greatly reduced, which can solve the problem of large data transmission volume.

[0116] In some embodiments, the first encoder E1 is of the form:

[0117]

[0118] Among them, y k+1 (t) is the input of the first encoder E1, i.e., the original output of the controlled system; is the output of the first encoder E1; is the internal state of the first encoder E1; is a scaling index of the first encoder E1, used to eliminate the influence of quantization error so that the decoded data is equal to the data before encoding; is the data loss characterization parameter of the first network path, Indicates that no data loss occurs after passing through the first network path. Indicates that data loss occurred after passing through the first network path.

[0119] Furthermore, a first quantizer is applied in the first encoder E1. The first quantizer is a unit uniform quantizer, and its form is as follows:

[0120]

[0121] Wherein, i is the output of the first quantizer, and z is an arbitrary value.

[0122] Accordingly, the first decoder D1 is of the following form:

[0123]

[0124] in, is the output of the first decoder D1, that is, the output of the controlled system after encoding and decoding. According to the form of the first encoder E1 and the first decoder D1, the relationship between the output of the controlled system before and after encoding and decoding can be obtained, and its form is as follows:

[0125]

[0126] The output of the controlled system after coding and decoding is a continuous signal. By sampling the output of the controlled system according to the sampling time sequence, the output sequence of the controlled system after coding and decoding can be obtained and transmitted to the controller.

[0127] In some embodiments, the second encoder E2 is of the form:

[0128]

[0129] Among them, u k+1 (t) is the input of the second encoder E2; is the output of the second encoder E2; is the internal state of the second encoder E2; is a scaling index of the second encoder E2, used to eliminate the influence of quantization error so that the decoded data is equal to the data before encoding; is the data loss characterization parameter of the second network path, Indicates that no data loss occurs after passing through the second network path. Indicates that data loss occurred after passing through the second network path.

[0130] Furthermore, a second quantizer is applied in the second encoder E2, and the second quantizer is a unit uniform quantizer, and its form is the same as the first quantizer. In other embodiments, different types of quantizers can also be selected according to actual scenarios.

[0131] Accordingly, the second decoder D2 is of the following form:

[0132]

[0133] in, is the output of the second decoder D2, that is, the controller output after encoding and decoding. According to the form of the second encoder E2 and the second decoder D2, the relationship between the controller output before and after encoding and decoding can be obtained, and its form is as follows:

[0134]

[0135] The output of the controller after encoding and decoding is a continuous signal. By sampling the controller output according to the sampling time sequence, the output sequence of the controller after encoding and decoding can be obtained and transmitted to the controlled system.

[0136] Based on the iterative learning control method for the controlled system provided in the above embodiment, a practical scenario for applying the method is provided below. In this practical scenario, a permanent magnet linear motor is used as the controlled system, and its model form is consistent with the aforementioned permanent magnet linear motor. The permanent magnet linear motor is provided with an expected output trajectory, and its form is as follows:

[0137] y d (t)=0.5-sin(0.05πt)-0.5cos(0.02t),t∈[0,0.5]

[0138] For this permanent magnet linear motor model, where x k (n) is the sampling value of the motor position. The initial motor position before iterative calculation is 0m; v k (n) is the sampling value of the motor speed. The initial motor speed before iterative calculation is 0 m / s; stator resistance R = 8.6 Ω; rotor mass m = 1.635 kg; magnetic flux ψ f =0.35Wb. Set the sampling period T S =Δ=10ms, motor running time T d =0.5s.

[0139] In the iterative calculation process, the number of target round input signals N=2 is selected, and for different target round input signals, the input learning gain P1=0.7, P2=0.3 is set; the error learning gain L1=7, L2=5 is set; among which, for 0<γ1(n)<1. In this actual scenario, the end condition is set to the error indicator function converges to the allowable range within a limited number of iterations. The error indicator function is obtained according to the current error sequence, and its form is as follows:

[0140]

[0141] Among them, y d(n) is the expected output sequence obtained according to the expected output trajectory; is the motor output sequence of the kth iteration.

[0142] According to the above conditions, set the corresponding high-order feedback iterative learning control law and perform iterative calculations. Figures 5 to 7 As shown, Figure 5 The expected output trajectory of the permanent magnet linear motor in the embodiment of the present application and the system output diagram of different iteration numbers are shown in the figure. Figure 5 The desired trajectory of the permanent magnet linear motor is shown (corresponding to y d (t)) and at the 5th iteration (corresponding to y5(t)), the 15th iteration (corresponding to y 15 (t)) and the 25th iteration (corresponding to y 25 (t)), where the horizontal axis is the motor running time and the vertical axis is the output tracking. Figure 5 It can be seen that according to the high-order feedback iterative learning control law, the motor output can converge to the desired trajectory within a finite number of iterations, and as the number of iterations increases, at the 25th iteration, the motor output almost converges to the desired trajectory, that is, the output trajectory of the permanent magnet linear motor almost coincides with the desired output trajectory, thereby achieving precise control of the permanent magnet linear motor.

[0143] Figure 6 This is a comparison diagram of the output errors of the permanent magnet linear motor under different data loss rates in the embodiment of the present application, as shown in the figure, Figure 6 The figure shows the convergence speed of the iterative learning of the permanent magnet linear motor with the set high-order feedback iterative learning control law when the data loss rate is 20%, 50% and 70%, respectively. The horizontal axis is the iteration round and the vertical axis is the error index function. Figure 6 It can be seen that the greater the data loss rate, the slower the system convergence speed.

[0144] Figure 7 This is a comparison diagram of the output error of the permanent magnet linear motor in the embodiment of the present application when the data loss rate is 20% and the output error of the traditional iterative learning control, as shown in the figure, Figure 7The figure shows the convergence speed of the high-order feedback iterative learning control law set by the iterative learning control method for the controlled system provided by the present application, and the traditional iterative learning control law, wherein the horizontal axis is the iteration round and the vertical axis is the error index function, and both are iteratively calculated under the condition of 20% data loss rate. According to the above error index function, when the data loss rate is 20%, when the high-order feedback iterative learning control law is used, the error index function can converge to close to zero at the 30th iteration round; while the traditional iterative learning control law needs to converge to close to zero after the 80th iteration round. It can be seen that under certain data loss conditions, the high-order feedback iterative learning control law has a certain data compensation effect, which enables the controlled system to converge to the desired trajectory faster. The high-order feedback iterative learning control law realizes that when the output error is within an acceptable range, the number of iterations is greatly reduced by about 50 times, which fully proves that the iterative learning control method for the controlled system provided by the present application can significantly improve the convergence speed and greatly shorten the iterative calculation time while effectively reducing the tracking error, thereby effectively avoiding unnecessary energy consumption.

[0145] Accordingly, please refer to Figure 8 , the embodiment of the present application provides an iterative learning control system for a controlled system, the system comprising:

[0146] The data acquisition module 100 is used to obtain the historical input signal of the controlled system, the current error sequence of the controlled system at the current moment, and the historical error sequence of the controlled system at the previous moment; wherein the historical input signal at least includes the input signal of the previous iteration round, the current error sequence is determined based on the difference between the expected output sequence of the controlled system and the current output sequence in the current iteration round, and the current output sequence is generated by using the input signal of the previous iteration round as the current input signal and controlling the controlled system.

[0147] The target round determination module 200 is used to determine the target round input signal before the previous iteration round in the historical input signal when there is data loss in the current input signal.

[0148] The learning control module 300 is used to set a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence.

[0149] The signal updating module 400 is used to iteratively learn the control law according to the high-order feedback to obtain an updated input signal, and input the updated input signal to the controlled system.

[0150] The iterative calculation module 500 is used to iteratively execute the module for obtaining the updated input signal until the end condition is reached.

[0151] In some optional implementations, the high-order feedback iterative learning control law includes a high-order learning term and a feedback update term; the learning control module 300 includes:

[0152] A data compensation unit, used to obtain a high-order learning item based on a target round input signal, a current error sequence and a data loss characterization parameter to perform data compensation on a current input signal;

[0153] The signal supplement unit is used to obtain a feedback update item based on a historical error sequence and a data loss characterization parameter, so as to supplement the target round input signal when the target round input signal does not meet the preset iteration requirement.

[0154] In some optional implementations, the high-order learning item includes an input learning item and an error learning item; and the data compensation unit includes:

[0155] An input learning subunit, used for learning the target round input signal according to the input learning gain to obtain an input learning item;

[0156] The error learning subunit is used to learn the current error sequence according to the error learning gain and the data loss characterization parameter to determine the error learning item.

[0157] In some optional implementations, the form of the high-order learning item in the data compensation unit is as follows:

[0158]

[0159] Among them, u f,k+1 (n) represents a high-order learning item, represents the input learning item, represents the error learning term; u k-j+1 (n) is the input signal at the nth moment in the k-j+1th iteration round; γ k-j+1 (n+1) is the data loss parameter at the n+1th moment in the k-j+1th iteration round, γ k-j+1 (n+1)=1 means no data loss occurs, γ k-j+1 (n+1)=0 means data loss has occurred; is the current error sequence at the n+1th moment; P j is the learning gain of the jth input; L j is the jth error learning gain; N is the number of input signals in the target round.

[0160] In some optional implementations, the input learning gain and the error learning gain in the data compensation unit satisfy the following convergence conditions:

[0161]

[0162] Among them, γ j (n+1) is the j-th data loss characterization parameter at the n+1th time, γ j (n+1)=1 means that there is no data missing in the historical output sequence corresponding to the j-th target round input signal, γ j (n+1)=0 means that data is missing in the historical output sequence corresponding to the j-th target round input signal; ρ j is the convergence coefficient corresponding to the jth historical input signal; C and B are related to the controlled system.

[0163] In some optional implementations, the feedback update item in the signal supplement unit is in the following form:

[0164]

[0165] Among them, u b,k+1 (n) represents the feedback update term; γ k+1 (n) is the data loss characterization parameter at the nth moment, γ k+1 (n) = 1 means that no data loss occurs in the current output sequence. k+1 (n) = 0 means that data loss has occurred in the current output sequence; is the historical error sequence of the kth iteration round; Γ is the feedback constraint coefficient.

[0166] In some optional implementations, the controlled system and the controller are connected via a codec network, which includes an encoder, a quantizer, a network path and a decoder, and the quantizer is a unit uniform quantizer.

[0167] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0168] The iterative learning control system of the controlled system in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0169] See also Fig. 9 , Fig. 9It is a structural diagram of a computer device provided by an embodiment of the present application. As shown in the figure, the computer device includes: one or more processors 10, a memory 20, and an interface for connecting various components, including a high-speed interface and a low-speed interface. Each component uses different buses to communicate with each other, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig. 9 A processor 10 is taken as an example.

[0170] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0171] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0172] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0173] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0174] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0175] The embodiment of the present application also provides a computer-readable storage medium. The above method according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0176] The embodiment of the present application provides a computer program product, which includes computer instructions, which are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of any embodiment of the present application.

[0177] Although the embodiments of the present application are described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

[0178] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0179] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0180] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0181] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0182] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0183] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0184] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0185] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0186] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

[0187] Although the embodiments of the present application have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A controlled system iterative learning control method, characterized in that: The method comprises: Acquire a historical input signal of a controlled system, a current error sequence of the controlled system at a current moment, and a historical error sequence of the controlled system at a moment before the current moment; wherein the historical input signal at least includes an input signal of a previous iteration round, the current error sequence is determined based on a difference between an expected output sequence of the controlled system and a current output sequence in a current iteration round, and the current output sequence is generated by controlling the controlled system using the input signal of the previous iteration round as a current input signal; In the case where data is lost in the current input signal, determining a target round input signal before the previous iteration round in the historical input signal; Setting a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence; Obtaining an updated input signal according to the high-order feedback iterative learning control law, and inputting the updated input signal into the controlled system; The above steps of obtaining the updated input signal are iterated until the end condition is reached.

2. The method according to claim 1, characterized in that The high-order feedback iterative learning control law includes a high-order learning term and a feedback update term; the high-order feedback iterative learning control law is set according to the target round input signal, the current error sequence and the historical error sequence, including: Based on the target round input signal, the current error sequence and the data loss characterization parameter, the high-order learning item is obtained to perform data compensation on the current input signal; The feedback update term is obtained based on the historical error sequence and the data loss characterization parameter to supplement the target round input signal when the target round input signal does not meet the preset iteration requirement.

3. The method according to claim 2, characterized in that The high-order learning item includes an input learning item and an error learning item; the high-order learning item is obtained based on the target round input signal, the current error sequence and the data loss characterization parameter to perform data compensation on the current input signal, including: Learning the target round input signal according to the input learning gain to obtain the input learning item; The current error sequence is learned according to the error learning gain and the data loss characterization parameter to determine the error learning item.

4. The method according to claim 3, characterized in that The form of the high-level learning item is as follows: Among them, u f,k+1 (n) represents the high-order learning item, represents the input learning item, represents the error learning term; u k-j+1 (n) is the input signal at the nth moment in the k-j+1th iteration round; γ k-j+1 (n+1) is the data loss parameter at the n+1th moment in the k-j+1th iteration round, γ k-j+1 (n+1)=1 means no data loss occurs, γ k-j+1 (n+1)=0 means data loss has occurred; is the current error sequence at the n+1th moment; P j is the learning gain of the jth input; L j is the jth error learning gain; N is the number of input signals in the target round.

5. The method according to claim 4, characterized in that The input learning gain and the error learning gain satisfy the following convergence conditions: Among them, γ j (n+1) is the j-th data loss characterization parameter at the n+1th time, γ j (n+1)=1 means that there is no data missing in the historical output sequence corresponding to the j-th target round input signal, γ j (n+1)=0 means that data is missing in the historical output sequence corresponding to the j-th target round input signal; ρ j is the convergence coefficient corresponding to the jth historical input signal; C and B are related to the controlled system.

6. The method according to claim 2, characterized in that The feedback update item is in the following form: Among them, u b,k+1 (n) represents the feedback update term; γ k+1 (n) is the data loss characterization parameter at the nth moment, γ k+1 (n) = 1 means that no data loss occurs in the current output sequence. k+1 (n) = 0 means that data loss has occurred in the current output sequence; is the historical error sequence of the kth iteration round; Γ is the feedback constraint coefficient.

7. The method according to any one of claims 1 to 6, characterized in that The controlled system and the controller are connected via a codec network, which includes an encoder, a quantizer, a network path and a decoder, and the quantizer is a unit uniform quantizer.

8. A controlled system iterative learning control system, characterized in that: The system comprises: A data acquisition module, used to acquire historical input signals of a controlled system, a current error sequence of the controlled system at a current moment, and a historical error sequence of the controlled system at a moment before the current moment; wherein the historical input signals at least include input signals of a previous iteration round, the current error sequence is determined based on the difference between an expected output sequence of the controlled system and a current output sequence of a current iteration round, and the current output sequence is generated by taking the input signal of the previous iteration round as a current input signal and controlling the controlled system; a target round determination module, configured to determine, in the case where data loss exists in the current input signal, a target round input signal before the previous iteration round in the historical input signal; A learning control module, used for setting a high-order feedback iterative learning control law according to the target round input signal, the current error sequence and the historical error sequence; A signal updating module, used for obtaining an updated input signal according to the high-order feedback iterative learning control law, and inputting the updated input signal to the controlled system; The iterative calculation module is used to iteratively execute the module for obtaining the updated input signal until the end condition is reached.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the iterative learning control method for the controlled system according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the iterative learning control method for a controlled system according to any one of claims 1 to 7.

Images