Iterative learning control method based on incomplete error information utilization

By designing the learning gain matrix and noise model, an iterative learning control model for the utilization of non-complete error information is constructed, which solves the problem of the non-zero Markov parameter matrix dissatisfied with rank and noise in observing satellites, and realizes high-precision full path tracking and system reliance Great.

CN119087818APending Publication Date: 2024-12-06BEIHANG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411331450.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-24
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the repeated tracking task of observing satellites, the non-zero Markov parameter matrix is ​​not satisfied with the rank of the rank, resulting in the iterative learning control algorithm complex, and random noise and path length changes affect the tracking accuracy, making it difficult to achieve high-precision full path tracking.

Method used

Based on incomplete error information, learning gain matrix regulation parameters are designed, noise model and path tracking model are established, noise is used to describe noise using martingale differential sequence, and it is constructed an iterative learning control model for the utilization of incomplete error information to overcome the influence of noise and adapt to path length changes.

Benefits of technology

High-precision full path tracking of ground observation satellites under random interference and path length changes is achieved, improving control performance and system robustness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119087818A_ABST
    Figure CN119087818A_ABST
Patent Text Reader

Abstract

The invention relates to an iterative learning control method based on incomplete error information utilization, and belongs to the technical field of iterative learning control, and the method comprises the steps: building a path tracking model in a full tracking state, building a yoke difference sequence of a noise set based on a drainage basin, and obtaining a noise model; a learning gain matrix regulation and control parameter is designed again; obtaining a learning gain matrix based on the learning gain matrix regulation and control parameters; designing an iterative learning control model for incomplete error information utilization based on the learning gain matrix, the noise model and the path tracking model; finally, the target position of the earth observation satellite is preset; controlling the earth observation satellite to track a preset target position based on an iterative learning control model utilized by the incomplete error information; the problem of high-precision complete path tracking control of the observation satellite under the influence of random interference and path length iterative change is solved, and the purposes of improving the control performance of the observation satellite and enhancing the robustness and reliability of the system are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of iterative learning control, and in particular to an iterative learning control method based on the utilization of incomplete error information. Background Art

[0002] At present, the research on the observation frequency and resolution of earth observation satellites has achieved remarkable results, and its observation frequency and resolution have been significantly improved. As a result, the demand for scheduling and optimizing repetitive tasks of earth observation satellites has become more urgent. In the prior art, some optimization methods are used to improve the observation coverage and data quality, such as mixed integer programming, heuristic algorithms, etc. Therefore, the research on repeated tracking tasks of observation satellites is of great significance. In many practical application scenarios, observation satellites need to achieve complete path tracking tasks within a limited time, that is, to achieve complete tracking of the target path from beginning to end. The iterative learning control method is a control method that uses the historical iterative input and output data generated during the operation of the system to achieve complete tracking of the target path. However, when iterative learning control is applied to repeated observation tasks of observation satellites, there are the following difficulties that need to be overcome: (1) The non-zero Markov parameter matrix corresponding to the observation satellite usually does not satisfy the full rank condition. In this case, the design of the iterative learning control algorithm often requires all the historical iterative output data, which makes the iterative learning control algorithm design complex and increases the data storage requirements; (2) Due to the existence of sensor measurement noise and environmental interference, the observation satellite will inevitably be affected by random noise; (3) Due to the output limitation, the path length of the observation satellite in each iteration is different, so the iterative learning control algorithm is not suitable for the observation satellite. Figure 1 For example, the path length is 8 in the initial iteration, 10 in the first iteration, and 6 in the third iteration.

[0003] Therefore, the partial error information stochastic iterative learning control method for variable path length tracking tasks is a key and difficult problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above problems, the present invention proposes an iterative learning control method based on the utilization of incomplete error information. First, a path tracking model of the full tracking state is established. Secondly, a martingale difference sequence of a noise set is established based on a watershed to obtain a noise model. Thirdly, a learning gain matrix control parameter is designed. Based on the learning gain matrix control parameter, a learning gain matrix is ​​obtained. Based on the learning gain matrix, an iterative learning control model using incomplete error information is designed. Finally, a target position of an earth observation satellite is preset. Based on the iterative learning control model using incomplete error information, the earth observation satellite is controlled to track the preset target position. The problem of high-precision complete path tracking control of the observation satellite under the influence of random interference and iterative changes in path length is solved, so as to improve the control performance of the observation satellite and enhance the robustness and reliability of the system.

[0005] The present invention provides an iterative learning control method based on the utilization of incomplete error information, comprising:

[0006] Step S1, determining the scenario of the path tracking task;

[0007] Acquire multiple paths of the earth observation satellite performing the path tracking task in the scenario of the path tracking task; set the tracking state of each path to obtain multiple updated paths;

[0008] establishing a path tracking model using the multiple updated paths;

[0009] Preferably, the earth observation satellite is a multiple-input multiple-output time-varying earth observation satellite;

[0010] Preferably, the expression of the earth observation satellite is:

[0011]

[0012] Among them, x i (t) is the path tracking state of the Earth observation satellite in the i-th iteration at time t, i = 0, 1, 2, 3...N, N is the total number of preset iterations; t = 0, 1, 2, 3...N i , N i represents the total running time of the iterative change, u i (t) is the input of the earth observation satellite at the time t for the i-th iteration, y i (t) is the position of the earth observation satellite at the time t, the i-th iteration; A(t) is the first time-varying system matrix corresponding to the earth observation satellite at the time t, B(t) is the second time-varying system matrix corresponding to the earth observation satellite at the time t, and C(t) is the third time-varying system matrix corresponding to the earth observation satellite at the time t; v i (t) is the noise of the earth observation satellite at the i-th iteration at time t.

[0013] Preferably, the path tracking model is a finite-state Markov chain;

[0014] Preferably, the path with the longest length N H Set the path of the Earth observation satellite to complete tracking status.

[0015] Further, the complete tracking state of each path length is returned normally.

[0016] Preferably, the scenario in which the earth observation satellite performs a tracking mission includes:

[0017] In high-resolution observation missions, due to the change in terrain height, the path length of the satellite must be dynamically adjusted to maintain a stable observation trajectory;

[0018] For dynamic target monitoring tasks, such as tracking ships or vehicles, the target's moving speed and direction changes will cause the satellite to constantly adjust its path length to maintain continuous observation of the target;

[0019] Environmental interference, such as weather changes, affects the quality of observations, requiring satellites to adjust path lengths to avoid or reduce interference effects; when satellites switch between different missions, such as switching from large-scale terrain scanning to high-precision local observations, path lengths need to be adjusted according to mission requirements.

[0020] Step S2, obtaining the noise existing when the earth observation satellite performs the mission on a repeated path, and obtaining a corresponding noise set; and establishing a watershed corresponding to the noise set;

[0021] Based on the watershed, the martingale difference sequence of the noise set is established to obtain the noise model;

[0022] The expression of the watershed is:

[0023]

[0024] Among them, Z i is the watershed of the earth observation satellite in the i-th iteration; represents the element in the flow domain of the i-th iteration, σ is algebraic, v i (t) is the noise of the earth observation satellite at the time t, Represents the set {0,1,2,…,N}.

[0025] The expression of the martingale difference sequence is:

[0026]

[0027] Among them, v i+1(t) is the noise of the Earth observation satellite at the time t, the i+1th iteration, is the noise v of the Earth observation satellite at the time t, the i+1th iteration i+1 (t) relative to the basin conditional expectation.

[0028] Preferably, the noise is the noise randomly generated when the earth observation satellite repeatedly performs the path tracking task;

[0029] Furthermore, the noise includes sensor noise, atmospheric disturbance, and noise caused by manufacturing and assembly errors;

[0030] Exemplarily, the sensor noise is generated by measurement errors of a gyroscope, an accelerometer and a star sensor;

[0031] The atmospheric disturbance is caused by the uneven atmospheric density and short-term changes in atmospheric density caused by the Earth's orbit;

[0032] The manufacturing and assembly errors are generated during the satellite manufacturing and assembly process and are accumulated over a long period of time during orbital operation.

[0033] The random noise of the present application is modeled as a martingale difference sequence, which can more accurately describe various random noise sources and improve the applicability and accuracy of the model.

[0034] Step S3, designing a learning gain matrix control parameter; obtaining a learning gain matrix based on the learning gain matrix control parameter;

[0035] Design an iterative learning control model that utilizes incomplete error information based on learning gain matrix, noise model and path tracking model;

[0036] Preferably, the learning gain matrix control parameters include a noise attenuation factor, an iterative convergence learning parameter and a traceable learning parameter;

[0037] The present invention utilizes random noise attenuation factors to overcome the problems caused by random noise, utilizes iterative convergent learning parameters to solve the traceability problem of earth observation satellites, and utilizes traceable learning parameters to ensure the convergence of iterative learning control models utilizing incomplete error information.

[0038] Furthermore, the learning gain matrix control parameters also include a random noise attenuation factor that satisfies the condition, expressed as Among them, α i is the noise attenuation factor of the i-th iteration of the earth observation satellite;

[0039] The expression for the iterative convergence learning parameter to meet the conditions is:

[0040] Re(λ i(W(t)C(t+1)B(t)))>0

[0041] Among them, Re(·) is the eigenvalue function, W(t)C(t+1)B(t) is the algorithm condition matrix, Re(λ k (W(t)C(t+1)B(t)))>0 means that the eigenvalues ​​of the algorithm condition matrix W(t)C(t+1)B(t) all have positive real parts, W(t) is the learning gain matrix at time t, C(t+1) is the third time-varying system matrix corresponding to the observed satellite at time t+1, B(t) is the second time-varying system matrix corresponding to the observed satellite at time t, λ i is the parameter for the i-th iteration.

[0042] Step S4, presetting a target position of the earth observation satellite; and controlling the earth observation satellite to track the preset target position based on an iterative learning control model utilizing incomplete error information.

[0043] Preferably, the specific steps of controlling the tracking of a preset target position by an earth observation satellite based on an iterative learning control model using incomplete error information include:

[0044] When t = 1, it is the initial time of the path; when i = 1, it is the first iteration at time t;

[0045] Preset the target position of the earth observation satellite at time t; predict the target path and obtain the position at time t;

[0046] Determine an input of an i-th iteration of an earth observation satellite at time t; obtain a position of an i-th iteration of an earth observation satellite at time t based on the input of the i-th iteration of the earth observation satellite at time t and the noise of the i-th iteration at time t;

[0047] Based on the target position of the earth observation satellite at the i-th iteration at time t, the position of the i-th iteration at time t is corrected to obtain the updated position of the i-th iteration at time t and the tracking error of the i-th iteration at time t;

[0048] The tracking error of the i-th iteration at time t is updated based on the learning gain matrix to obtain the tracking error of the i-th iteration at time t;

[0049] Combine the input of the i-th iteration of the earth observation satellite at time t with the i-th iteration tracking error of the time t to obtain the updated input of the earth observation satellite at time t; and use the updated input at time t as the input of the next iteration;

[0050] Determine whether i is greater than or equal to N, where N is the preset total number of iterations. If so, stop the calculation, obtain the updated input of the earth observation satellite at time t, obtain the position of the earth observation satellite at time t based on the input of the earth observation satellite at time t, and record it, complete the tracking of the target position at time t, and proceed to the next step;

[0051] Determine whether t is greater than or equal to N i , N i Represents the total running time of the iterative change. If yes, stop the calculation, get the position of the earth observation satellite at each time, and complete the tracking of the target position by the earth observation satellite at each time.

[0052] Furthermore, the updated position expression of the earth observation satellite at time t for the i-th iteration is:

[0053]

[0054] in, is the updated position of the Earth observation satellite at time t for the i-th iteration, N t,i represents the path length of the i-th iteration at time t, N H is the longest path, y d (t) is the target position of the observation satellite at time t.

[0055] Furthermore, the tracking error expression is:

[0056]

[0057] in, is the tracking error of the earth observation satellite at the i-th iteration at time t.

[0058] Furthermore, the iterative learning control model using the incomplete error information is expressed as:

[0059]

[0060] Among them, u i+1 (t) represents the input of the Earth observation satellite at the time t, i+1th iteration, u i (t) represents the input of the Earth observation satellite at the time t for the i-th iteration, is the tracking error of the Earth observation satellite at the time t for the i-th iteration, is the learning gain matrix of the earth observation satellite at the i-th iteration at time t, α iis the noise attenuation factor of the earth observation satellite for the i-th iteration, which is used to overcome the influence of random noise. W(t) is the iterative convergence learning parameter of the iterative learning control model for controlling the utilization of incomplete error information at time t, which is used to ensure the convergence of the iterative learning control model for the utilization of incomplete error information. Z(t) is the trackable learning parameter of the earth observation satellite position and the target position at time t, which is used to overcome the trackability problem of the observation satellite system.

[0061] The present invention also includes step S5 of ensuring the precise tracking effect of the earth observation satellite by using an iterative learning control model using the incomplete error information;

[0062] Specifically, the specific steps to ensure the precise tracking effect of the observation satellite include:

[0063] By using Lyapunov function and stochastic approximation method, the convergence of iterative learning control model using incomplete error information is analyzed in the mean square sense and in the probability 1 sense.

[0064] Based on the convergence in the mean square sense and in the probability 1 sense, the stability and accuracy of the iterative learning control model using incomplete error information under random noise are obtained;

[0065] Based on the stability and accuracy under random noise, the precise tracking effect of the earth observation satellite is guaranteed.

[0066] Compared with the prior art, the present invention has at least the following beneficial effects:

[0067] (1) The iterative learning control method using incomplete error information of the present invention can effectively overcome the influence of noise on the tracking effect and realize the precise tracking of the specified path by the observation satellite;

[0068] (2) The iterative learning control method using incomplete error information of the present invention is applicable to the scenario where the system non-zero Markov parameter matrix does not satisfy the full rank condition, thereby improving the adaptability of satellite observation;

[0069] (3) The iterative learning control method using incomplete error information of the present invention is applicable to the scenario where the path length changes iteratively, thereby improving the practicality of random iterative learning control;

[0070] (4) The iterative learning control model using incomplete error information of the present invention can ensure that the earth observation satellite completely tracks the specified path from beginning to end within a period of time. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] The drawings are only for the purpose of illustrating particular embodiments and are not to be construed as limiting the invention.

[0072] Figure 1A schematic diagram of the iteratively changing path length of the present invention;

[0073] Figure 2 It is a schematic diagram of a full error information random iterative learning control block diagram for variable path length tracking tasks of the present invention;

[0074] Figure 3 It is a schematic diagram of a flow chart of a full error information stochastic iterative learning control algorithm for variable path length tracking tasks in an embodiment of the present invention;

[0075] Figure 4 Schematic diagram of a satellite path tracking simulation diagram in an embodiment of the present invention. DETAILED DESCRIPTION

[0076] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In addition, the present invention can also be implemented in other ways different from those described herein, and therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0077] A specific embodiment of the present invention, as Figure 1-4 , discloses a full error iterative learning control algorithm for repeated tracking tasks of observation satellites. In order to illustrate the effectiveness of the method proposed in the present invention, the above technical solution of the present invention is described in detail through a specific embodiment below. The specific implementation steps are as follows:

[0078] The present invention provides an iterative learning control method based on the utilization of incomplete error information, comprising:

[0079] Step S1, determining the scenario of the path tracking task;

[0080] Acquire multiple paths of the earth observation satellite performing the path tracking task in the scenario of the path tracking task; set the tracking state of each path to obtain multiple updated paths;

[0081] establishing a path tracking model using the multiple updated paths;

[0082] Preferably, the earth observation satellite is a multiple-input multiple-output time-varying earth observation satellite;

[0083] Preferably, the expression of the earth observation satellite is:

[0084]

[0085] Among them, x i(t) is the path tracking state of the Earth observation satellite in the i-th iteration at time t, i = 0, 1, 2, 3...N, N is the total number of preset iterations; t = 0, 1, 2, 3...N i , N i represents the total running time of the iterative change, u i (t) is the input of the earth observation satellite at the time t for the i-th iteration, y i (t) is the position of the earth observation satellite at the time t, the i-th iteration; A(t) is the first time-varying system matrix corresponding to the earth observation satellite at the time t, B(t) is the second time-varying system matrix corresponding to the earth observation satellite at the time t, and C(t) is the third time-varying system matrix corresponding to the earth observation satellite at the time t; v i (t) is the noise of the earth observation satellite at the i-th iteration at time t.

[0086] Preferably, the path tracking model is a finite-state Markov chain;

[0087] Preferably, the path with the longest length N H Set the path of the Earth observation satellite to complete tracking status.

[0088] Further, the complete tracking state of each path length is returned normally.

[0089] Preferably, the scenario in which the earth observation satellite performs a tracking mission includes:

[0090] In high-resolution observation missions, due to the change in terrain height, the path length of the satellite must be dynamically adjusted to maintain a stable observation trajectory;

[0091] For dynamic target monitoring tasks, such as tracking ships or vehicles, the target's moving speed and direction changes will cause the satellite to constantly adjust its path length to maintain continuous observation of the target;

[0092] Environmental interference, such as weather changes, affects the quality of observations, requiring satellites to adjust path lengths to avoid or reduce interference effects; when satellites switch between different missions, such as switching from large-scale terrain scanning to high-precision local observations, path lengths need to be adjusted according to mission requirements.

[0093] Step S2, obtaining the noise existing when the earth observation satellite performs the mission on a repeated path, and obtaining a corresponding noise set; and establishing a watershed corresponding to the noise set;

[0094] Based on the watershed, the martingale difference sequence of the noise set is established to obtain the noise model;

[0095] The expression of the watershed is:

[0096]

[0097] Among them, Z i is the watershed of the earth observation satellite in the i-th iteration; represents the element in the flow domain of the i-th iteration, σ is algebraic, v i (t) is the noise of the earth observation satellite at the time t, Represents the set {0,1,2,…,N}.

[0098] The expression of the martingale difference sequence is:

[0099]

[0100] Among them, v i+1 (t) is the noise of the Earth observation satellite at the time t, the i+1th iteration, is the noise v of the Earth observation satellite at the time t, the i+1th iteration i+1 (t) relative to the basin conditional expectation.

[0101] Preferably, the noise is the noise randomly generated when the earth observation satellite repeatedly performs the path tracking task;

[0102] Furthermore, the noise includes sensor noise, atmospheric disturbance, and noise caused by manufacturing and assembly errors;

[0103] Exemplarily, the sensor noise is generated by measurement errors of a gyroscope, an accelerometer and a star sensor;

[0104] The atmospheric disturbance is caused by the uneven atmospheric density and short-term changes in atmospheric density caused by the Earth's orbit;

[0105] The manufacturing and assembly errors are generated during the satellite manufacturing and assembly process and are accumulated over a long period of time during orbital operation.

[0106] The random noise of the present application is modeled as a martingale difference sequence, which can more accurately describe various random noise sources and improve the applicability and accuracy of the model.

[0107] Step S3, designing a learning gain matrix control parameter; obtaining a learning gain matrix based on the learning gain matrix control parameter;

[0108] Design an iterative learning control model that utilizes incomplete error information based on learning gain matrix, noise model and path tracking model;

[0109] Preferably, the learning gain matrix control parameters include a noise attenuation factor, an iterative convergence learning parameter and a traceable learning parameter;

[0110] The present invention utilizes random noise attenuation factors to overcome the problems caused by random noise, utilizes iterative convergent learning parameters to solve the traceability problem of earth observation satellites, and utilizes traceable learning parameters to ensure the convergence of iterative learning control models utilizing incomplete error information.

[0111] Furthermore, the learning gain matrix control parameters also include a random noise attenuation factor that satisfies the condition, expressed as Among them, α i is the noise attenuation factor of the i-th iteration of the earth observation satellite;

[0112] The expression for the iterative convergence learning parameter to meet the conditions is:

[0113] Re(λ i (W(t)C(t+1)B(t)))>0

[0114] Among them, Re(·) is the eigenvalue function, W(t)C(t+1)B(t) is the algorithm condition matrix, Re(λ k (W(t)C(t+1)B(t)))>0 means that the eigenvalues ​​of the algorithm condition matrix W(t)C(t+1)B(t) all have positive real parts, W(t) is the learning gain matrix at time t, C(t+1) is the third time-varying system matrix corresponding to the observed satellite at time t+1, B(t) is the second time-varying system matrix corresponding to the observed satellite at time t, λ i is the parameter for the i-th iteration.

[0115] Step S4, presetting a target position of the earth observation satellite; and controlling the earth observation satellite to track the preset target position based on an iterative learning control model utilizing incomplete error information.

[0116] Preferably, the specific steps of controlling the tracking of a preset target position by an earth observation satellite based on an iterative learning control model using incomplete error information include:

[0117] When t = 1, it is the initial time of the path; when i = 1, it is the first iteration at time t;

[0118] Preset the target position of the earth observation satellite at time t; predict the target path and obtain the position at time t;

[0119] Determine an input of an i-th iteration of an earth observation satellite at time t; obtain a position of an i-th iteration of an earth observation satellite at time t based on the input of the i-th iteration of the earth observation satellite at time t and the noise of the i-th iteration at time t;

[0120] Based on the target position of the earth observation satellite at the i-th iteration at time t, the position of the i-th iteration at time t is corrected to obtain the updated position of the i-th iteration at time t and the tracking error of the i-th iteration at time t;

[0121] The tracking error of the i-th iteration at time t is updated based on the learning gain matrix to obtain the tracking error of the i-th iteration at time t;

[0122] Combine the input of the i-th iteration of the earth observation satellite at time t with the i-th iteration tracking error of the time t to obtain the updated input of the earth observation satellite at time t; and use the updated input at time t as the input of the next iteration;

[0123] Determine whether i is greater than or equal to N, where N is the preset total number of iterations. If so, stop the calculation, obtain the updated input of the earth observation satellite at time t, obtain the position of the earth observation satellite at time t based on the input of the earth observation satellite at time t, and record it, complete the tracking of the target position at time t, and proceed to the next step;

[0124] Determine whether t is greater than or equal to N i , N i Represents the total running time of the iterative change. If yes, stop the calculation, get the position of the earth observation satellite at each time, and complete the tracking of the target position by the earth observation satellite at each time.

[0125] Furthermore, the updated position expression of the earth observation satellite at time t for the i-th iteration is:

[0126]

[0127] in, is the updated position of the Earth observation satellite at time t for the i-th iteration, N t,i represents the path length of the i-th iteration at time t, N H is the longest path, y d (t) is the target position of the observation satellite at time t.

[0128] Furthermore, the tracking error expression is:

[0129]

[0130] in, is the tracking error of the earth observation satellite at the i-th iteration at time t.

[0131] Furthermore, the iterative learning control model using the incomplete error information is expressed as:

[0132]

[0133] in, is the tracking error of the Earth observation satellite at the time t for the i-th iteration, is the learning gain matrix of the earth observation satellite at the i-th iteration at time t, α i is the noise attenuation factor of the earth observation satellite for the i-th iteration, which is used to overcome the influence of random noise. W(t) is the iterative convergence learning parameter of the iterative learning control model for controlling the utilization of incomplete error information at time t, which is used to ensure the convergence of the iterative learning control model for the utilization of incomplete error information. Z(t) is the trackable learning parameter of the earth observation satellite position and the target position at time t, which is used to overcome the trackability problem of the observation satellite system.

[0134] The present invention also includes step S5 of ensuring the precise tracking effect of the earth observation satellite by using an iterative learning control model using the incomplete error information;

[0135] Specifically, the specific steps to ensure the precise tracking effect of the observation satellite include:

[0136] By using Lyapunov function and stochastic approximation method, the convergence of iterative learning control model using incomplete error information is analyzed in the mean square sense and in the probability 1 sense.

[0137] Based on the convergence in the mean square sense and in the probability 1 sense, the stability and accuracy of the iterative learning control model using incomplete error information under random noise are obtained;

[0138] Based on the stability and accuracy under random noise, the precise tracking effect of the earth observation satellite is guaranteed.

[0139] Embodiment 2

[0140] The earth observation satellite system was integrated and designed in Matlab, and simulation experiments were carried out. The main simulation process is as follows:

[0141] Parameter settings

[0142] Using linear dynamics for the Earth observation satellite system, the expression is:

[0143]

[0144] Among them, x i (t) is the path tracking state of the Earth observation satellite in the i-th iteration at time t, i = 1, 2, 3…N, N is the total number of preset iterations; t = 1, 2, 3…N i , N i represents the total running time of the iterative change, u i(t) is the input of the earth observation satellite at the time t for the i-th iteration, y i (t) is the position of the earth observation satellite at the time t, the i-th iteration; A(t) is the first time-varying system matrix corresponding to the earth observation satellite at the time t, B(t) is the second time-varying system matrix corresponding to the earth observation satellite at the time t, and C(t) is the third time-varying system matrix corresponding to the earth observation satellite at the time t; v i (t) is the noise of the earth observation satellite at the i-th iteration at time t.

[0145]

[0146] Assuming the total running time N=20, the rank of the Markov parameter matrix C(1)B(0) of the observation satellite system is 1, which does not satisfy the row full rank condition nor the column full rank condition.

[0147] The present invention selects the attenuation factor

[0148] Iterative convergence learning parameter Z(t) = 0.25I, I is the unit matrix, which can track the learning parameters

[0149] The expression of the target path is:

[0150] (2) Results Analysis

[0151] Figure 4 The dynamic evolution of the position of the earth observation satellite and the expected target path along the time axis after 27 and 50 iterations respectively is shown. The figure shows that the position output of the earth observation satellite system can track the expected target path in real time at any time, and has extremely high tracking accuracy. Therefore, the iterative learning control model using incomplete error information provided by the present invention can overcome the influence of random noise of the observation satellite system, so that any linear time-varying observation satellite system with a Markov parameter matrix of insufficient rank can track the trackable target path in real time at all times within a limited time period, and has extremely high tracking accuracy.

[0152] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. An iterative learning control method based on the use of incomplete error information, characterized in that: include: Step S1, determining the scenario of the path tracking task; Acquire multiple paths of an earth observation satellite performing a path tracking task in a scenario of the path tracking task; Set the tracking status of each path to obtain multiple update paths; establishing a path tracking model using the multiple updated paths; Step S2, obtaining the noise existing when the earth observation satellite performs the mission on a repeated path, and obtaining a corresponding noise set; Establishing a watershed corresponding to the noise set; Based on the watershed, the martingale difference sequence of the noise set is established to obtain the noise model; Step S3, designing the learning gain matrix control parameters; Obtaining a learning gain matrix based on the learning gain matrix control parameters; Design an iterative learning control model that utilizes incomplete error information based on learning gain matrix, noise model and path tracking model; Step S4, presetting the target position of the earth observation satellite; An iterative learning control model based on the use of incomplete error information controls the tracking of an earth observation satellite to a preset target position.

2. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The expression of the earth observation satellite is: Among them, x i (t) is the path tracking state of the Earth observation satellite in the i-th iteration at time t, i = 0, 1, 2, 3...N, N is the total number of preset iterations; t = 0, 1, 2, 3...N i , N i represents the total running time of the iterative change, u i (t) is the input of the earth observation satellite at the time t for the i-th iteration, y i (t) is the position of the earth observation satellite at the time t, the i-th iteration; A(t) is the first time-varying system matrix corresponding to the earth observation satellite at the time t, B(t) is the second time-varying system matrix corresponding to the earth observation satellite at the time t, and C(t) is the third time-varying system matrix corresponding to the earth observation satellite at the time t; v i (t) is the noise of the earth observation satellite at the i-th iteration at time t.

3. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The expression of the watershed is: Among them, Z i is the watershed of the earth observation satellite in the i-th iteration; represents the element in the flow domain of the i-th iteration, σ is algebraic, v i (t) is the noise of the earth observation satellite at the time t, Represents the set {0,1,2,…,N}.

4. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The noise is the noise randomly generated when the earth observation satellite repeatedly performs the path tracking task.

5. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The learning gain matrix control parameters include a noise attenuation factor, an iterative convergence learning parameter and a traceable learning parameter.

6. The iterative learning control method based on the use of incomplete error information according to claim 5, characterized in that: The learning gain matrix control parameters also include a random noise attenuation factor that satisfies the condition, expressed as Among them, α i is the noise attenuation factor of the earth observation satellite at the i-th iteration.

7. The iterative learning control method based on the use of incomplete error information according to claim 5, characterized in that: The expression for the iterative convergence learning parameter to meet the conditions is: Re(λ i (W(t)C(t+1)B(t)))>0 Among them, W(t) is the learning gain matrix at time t, C(t+1) is the third time-varying system matrix corresponding to the observation satellite at time t+1, and B(t) is the second time-varying system matrix corresponding to the observation satellite at time t.

8. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The specific steps of controlling the tracking of the preset target position by the iterative learning control model based on the use of incomplete error information include: When t = 1, it is the initial time of the path; when i = 1, it is the first iteration at time t; Preset the target position of the earth observation satellite at time t; predict the target path and obtain the position at time t; Determine an input of an i-th iteration of an earth observation satellite at time t; obtain a position of an i-th iteration of an earth observation satellite at time t based on the input of the i-th iteration of the earth observation satellite at time t and the noise of the i-th iteration at time t; Based on the target position of the earth observation satellite at the i-th iteration at time t, the position of the i-th iteration at time t is corrected to obtain the updated position of the i-th iteration at time t and the tracking error of the i-th iteration at time t; The tracking error of the i-th iteration at time t is updated based on the learning gain matrix to obtain the tracking error of the i-th iteration at time t; Combine the input of the i-th iteration of the earth observation satellite at time t with the i-th iteration tracking error of the time t to obtain the updated input of the earth observation satellite at time t; and use the updated input at time t as the input of the next iteration; Determine whether i is greater than or equal to N, where N is the preset total number of iterations. If so, obtain an updated input of the earth observation satellite at time t, obtain the position of the earth observation satellite at time t based on the input of the earth observation satellite at time t, and record it, complete the tracking of the target position at time t, and proceed to the next step; Determine whether t is greater than or equal to N i , N i It represents the total running time of the iterative change. If the position of the earth observation satellite at each moment is obtained, the earth observation satellite can track the target position at each moment.

9. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The iterative learning control model using the incomplete error information is expressed as: Among them, u i+1 (t) represents the input of the Earth observation satellite at the time t, i+1th iteration, u i (t) represents the input of the Earth observation satellite at the time t for the i-th iteration, is the tracking error of the Earth observation satellite at the time t for the i-th iteration, is the learning gain matrix of the earth observation satellite at the time t, the i-th iteration, N H Indicates the path with the longest length.

10. The iterative learning control method based on the use of incomplete error information according to claim 1, characterized in that: The method also includes step S5, which uses an iterative learning control model using incomplete error information to ensure accurate tracking of earth observation satellites.