Method for keeping surface type of primary mirror of spliced telescope based on multi-agent cooperative control

Abstract

The invention relates to the technical field of space telescopes, in particular to a method for keeping the surface type of a primary mirror of a spliced telescope based on multi-agent cooperative control, and the method comprises the steps: setting an agent on each sub-mirror in the primary mirror, and obtaining the current attitude information of each sub-mirror; each sub-mirror is connected with other sub-mirrors by utilizing an intelligent agent through wireless communication, and the current attitude information of each sub-mirror is shared to the connected sub-mirror connected with the sub-mirror in real time, so that the current shared attitude information of each sub-mirror is obtained; according to the current attitude information and the current shared attitude information of each sub-mirror, determining a current kinetic model of each sub-mirror, and further obtaining the attitude information of each sub-mirror at the next moment; and the above contents are repeated until the current attitude information of each sub-mirror converges to the ideal attitude information of each sub-mirror, so that information sharing between the sub-mirrors and collaborative decision between the connected sub-mirrors are realized, the dynamic adaptive capacity is improved, and the requirements of adaptive real-time adjustment of the sub-mirror attitude and the main mirror surface shape are met.

Description

Technical Field

[0001] The present invention belongs to the technical field of space telescopes, and in particular relates to a method for maintaining the surface shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control. Background Art

[0002] The main mirror of the spliced ​​telescope is composed of multiple independent sub-mirror units, which increases the aperture of the telescope, improves the resolution and light collection ability, but also brings technical challenges to maintain the surface shape. Existing sub-mirror cooperative control methods include centralized control, decentralized control, distributed control, feedback control, adaptive control, fuzzy control, H-infinity-based control, predictive control and intelligent control. Centralized control has the risk of single point failure, heavy computational burden and poor scalability; decentralized control lacks global coordination and is difficult to achieve global optimization of the overall surface shape; although distributed control improves coordination, it requires complex communication networks and algorithms; feedback control is not responsive enough in the face of rapidly changing environments; adaptive control design is complex and computationally expensive; fuzzy control rule formulation and optimization require a lot of experience and data, and performance is limited; H-infinity-based control design is complex and computationally burdensome; predictive control has a heavy computational burden and depends on model accuracy; intelligent control algorithms are complex, training and optimization require a lot of data and computing resources, and the control decision process is not transparent.

[0003] The Chinese patent publication number is CN117130172A, and the publication date is November 28, 2023. The patent name is "A global adjustment method, device, equipment and medium for a spliced ​​space telescope". By constructing the first-order linear relationship between different types of wave aberration coefficients and each misalignment, different misalignments are obtained in stages and adjusted accordingly, which is conducive to the development and application of large-aperture spliced ​​space telescopes. However, the global adjustment used in this patent only involves the debugging of the mechanical system without designing collaborative control, and cannot guarantee the dynamic performance of the system.

[0004] The collaborative control scheme based on multi-agents overcomes the shortcomings of existing technologies by introducing collaboration and information sharing between agents, improves the overall performance and robustness of the system, and can face the uncertainty of the use environment and maintain high performance when the primary mirror is disturbed. The collaboration and information exchange between agents realize the optimal control of the global surface shape and avoid the local optimal problem. The real-time response capability of the system is improved through distributed computing and real-time information exchange. The modular design of multi-agent collaborative control makes the design and implementation of the control algorithm more flexible and simple, providing reliable technical support for the high-precision and high-robust surface shape maintenance of spliced ​​telescopes in scientific research. Summary of the invention

[0005] In view of this, the present invention aims to provide a method for maintaining the surface shape of a spliced ​​telescope primary mirror based on multi-agent collaborative control, so as to set an intelligent agent on each sub-mirror of the spliced ​​primary mirror, share information between the sub-mirrors through the intelligent agent, and thus achieve global coordination, which not only improves the overall performance of the system, but also enhances the robustness of the system to external disturbances and uncertainties.

[0006] To achieve the above object, the technical solution created by the present invention is implemented as follows: A method for maintaining the surface shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control, comprising the following steps: S1: An intelligent agent is set on each sub-mirror in the main mirror to set the initial posture information of each sub-mirror, the ideal posture information of each sub-mirror and the dynamic model of each sub-mirror; S2: Obtain the current posture information of each sub-mirror, and make a difference with the ideal posture information of each sub-mirror to obtain the current posture information error of each sub-mirror; S3: Using an intelligent agent to connect each sub-mirror with other sub-mirrors through wireless communication, so that the current posture information of each sub-mirror is shared in real time with the connected sub-mirrors, thereby obtaining the current shared posture information received by each sub-mirror; S4: according to the current posture information of each sub-mirror and its current shared posture information, determine the current control input of each sub-mirror and the current dynamic model of each sub-mirror, and then obtain the posture information of each sub-mirror at the next moment; S5: Repeat steps S2 to S4 until the current posture information of each sub-mirror converges to the ideal posture information of each sub-mirror.

[0007] Furthermore, in step S1, the dynamic model of each sub-mirror for: ; in, represents the posture information of the i-th sub-mirror, represents the control input of the ith sub-mirror, A represents the state matrix, B represents the control matrix.

[0008] Furthermore, in step S3, the current shared posture information of each sub-mirror is determined by the following formula: for: ; in, represents the number of connected sub-mirrors connected to the i-th sub-mirror, Represents the total number of connected sub-mirrors connected to the i-th sub-mirror.

[0009] Further, in step S4, the current control input of each sub-mirror is determined by the following formula: for: ; in, K represents the control gain matrix, represents the information weighting factor of the connected sub-mirror of the i-th sub-mirror, represents the current shared posture information received by the nth connected sub-mirror connected to the i-th sub-mirror at the jth moment; Then combined with the dynamic model of each sub-mirror Get the current dynamic model of each sub-mirror for: ; The next moment attitude information of each sub-mirror is determined by the following formula: for: ; in, Represents the time step.

[0010] Compared with the prior art, the invention can achieve the following beneficial effects: (1) The method for maintaining the shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control created by the present invention realizes information sharing between sub-mirrors and collaborative decision-making with connected sub-mirrors by setting an agent on each sub-mirror of the spliced ​​primary mirror, thereby achieving more flexible dynamic adaptability. Unlike the traditional H-infinity control scheme that mainly focuses on global performance optimization, the collaborative control can respond more quickly to changes in the surface shape error of the primary mirror, meet the needs of real-time adjustment of the sub-mirror posture and the surface shape of the primary mirror, and each sub-mirror is not only optimized according to its own state, but also coordinated with the help of the state information of the connected sub-mirrors, thereby improving the overall control accuracy and the real-time response capability of the system. (2) The method for maintaining the shape of the primary mirror of a spliced ​​telescope based on multi-agent collaborative control created by the present invention has excellent scalability: as the number of sub-mirrors increases, the system can be expanded by adding more agents without significantly modifying the overall control algorithm or hardware configuration. This feature enables it to effectively avoid the computational burden of the central controller when facing a large spliced ​​telescope, thereby improving the overall efficiency and flexibility of the system; (3) In the method for maintaining the surface shape of the primary mirror of a spliced ​​telescope based on multi-agent collaborative control described in the present invention, each sub-mirror independently calculates its control input, which improves the computing processing capability and reliability. Even if some sub-mirrors or related modules fail, the overall normal operation can still be maintained. This method significantly improves the robustness and fault tolerance of the system; at the same time, it finds a balance between the local and the global, ensuring that the spliced ​​telescope can maintain a high-precision primary mirror surface shape under various complex conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings constituting part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments and descriptions of the present invention are used to explain the present invention and do not constitute an improper limitation on the present invention. In the drawings: Figure 1 A flow chart of a method for maintaining the surface shape of a spliced ​​telescope primary mirror based on multi-agent collaborative control described in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solution and advantages of the invention more clear, the invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the invention and do not constitute a limitation of the invention.

[0013] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0014] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0015] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the invention can be understood according to specific circumstances.

[0016] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0017] like Figure 1 As shown, the method for maintaining the surface shape of the primary mirror of a spliced ​​telescope based on multi-agent collaborative control described in the embodiment of the present invention comprises the following steps: S1: An intelligent agent is set on each sub-mirror in the main mirror, and the initial posture information of each sub-mirror, the ideal posture information of each sub-mirror and the dynamic model of each sub-mirror are set.

[0018] Specifically, the base coordinate system is established with the main mirror, and the initial attitude information of each sub-mirror is Including the initial position coordinates of each sub-mirror in the base coordinate system , and the initial rotation angles of each sub-mirror around the x-axis, y-axis, and z-axis of the base coordinate system , i represents the number of sub-mirrors, that is, the initial posture information of each sub-mirror The ideal posture information of each sub-mirror Includes the ideal position coordinates of each sub-mirror in the base coordinate system , and the ideal rotation angles of each sub-mirror around the x-axis, y-axis, and z-axis of the base coordinate system , that is, the ideal posture information of each sub-mirror The initial posture information of each sub-mirror and the ideal posture information of each sub-mirror are obtained through measurement or preset configuration.

[0019] The intelligent body of each sub-mirror is connected to the adjustment mechanism for adjusting the sub-mirror, so that the intelligent body can collect the posture information of the corresponding sub-mirror in real time and receive the posture information of all connected sub-mirrors connected to it. The intelligent body controls the adjustment mechanism according to the posture information of the corresponding sub-mirror and the connected sub-mirrors, and then completes the posture adjustment of the corresponding sub-mirror. In the embodiment of the present invention, a six-degree-of-freedom adjustment platform is used as the intelligent body, the output end of the six-degree-of-freedom adjustment platform is connected to the adjustment mechanism of the sub-mirror, and the control center of the six-degree-of-freedom adjustment platform completes the collection, transmission, reception and calculation processing of the sub-mirror data.

[0020] Dynamic model of each sub-mirror for: ; in, represents the posture information of the i-th sub-mirror, represents the control input of the ith sub-mirror, A represents the state matrix, B Represents the control matrix. The state matrix A is identified by a single sub-mirror system and is used to describe the state change of each sub-mirror. Each sub-mirror has a specific state matrix A. The control matrix B is used to describe the output control quantity to make the actuator actuate and control the posture and position of the sub-mirror. The control matrix B is also identified by a single sub-mirror system and each sub-mirror has a specific control matrix B. The control input of each sub-mirror Refers to the signal or action applied to the system to drive the system to achieve the desired behavior. Specifically, in the attitude adjustment of the sub-mirror of the spliced ​​telescope, the control input refers to the force, torque or electrical signal used to adjust the position and attitude of the sub-mirror. These inputs are generated by the controller in the intelligent agent and act on the adjustment mechanism of the sub-mirror to compensate or correct the error of the measured attitude information of the sub-mirror.

[0021] S2: Obtain the current posture information of each sub-mirror, and make a difference with the ideal posture information of each sub-mirror to obtain the current posture information error of each sub-mirror.

[0022] In step S2, the posture information of each sub-mirror at the first moment is the initial posture information of the sub-mirror. From the second moment on, the posture information of each sub-mirror is measured by the wavefront sensor and the interferometric measurement device. The current posture information error of each sub-mirror is obtained by the following formula: : ; Where j represents the time. The attitude information error data is usually expressed as a wavefront error distribution on the entire main mirror surface, usually a two-dimensional matrix or phase distribution diagram, which represents the deviation at each sampling point and decomposes it into each sub-mirror. The surface error of the main mirror can be directly mapped to each sub-mirror based on the position and geometric relationship of the sub-mirror. This method assumes that the error occurs locally and is expressed as follows: ; in, represents the surface error of the primary mirror at the jth moment, represents the area where the i-th sub-mirror is located, represents the area of ​​the ith sub-mirror, A represents the error amount, and the current attitude information error of each sub-mirror is the average error. The error value of each sub-mirror area can be obtained by integration. This method assumes that the attitude information error of each sub-mirror is the superposition of the local surface error of the main mirror, so the global error can be distributed to each sub-mirror more accurately.

[0023] S3: Each sub-mirror is connected to other sub-mirrors through wireless communication using an intelligent agent, and the current posture information of each sub-mirror is shared in real time with the connected sub-mirrors, thereby obtaining the current shared posture information received by each sub-mirror.

[0024] Each sub-mirror shares its own state and error information with other sub-mirrors in real time through an intelligent agent in the form of wireless communication or forming a local area network, so as to ensure the real-time and accuracy of the information. In an embodiment of the present invention, the wireless communication connection method of step S3 is: the intelligent agents of all sub-mirrors are connected to an information processing center such as a server or a host computer, and the information processing center collects the information of all sub-mirrors through the intelligent agents of each sub-mirror, and then distributes the information of all sub-mirrors to the intelligent agent of each sub-mirror, thereby realizing the sharing of sub-mirror posture information. In some other embodiments, the wireless communication connection method of step S3 can also be: the intelligent agent of each sub-mirror is directly connected to the intelligent agent of other sub-mirrors, that is, every two connected intelligent agents are directly connected to communicate to form wireless communication or a local area network, without the need for an information processing center such as a server or a host computer.

[0025] In step S3, the current shared posture information of each sub-mirror is determined by the following formula: for: ; in, represents the number of connected sub-mirrors connected to the i-th sub-mirror, Indicates the total number of connected sub-mirrors connected to the i-th sub-mirror, and the current shared posture information of each sub-mirror is the average posture information of all connected sub-mirrors connected to the i-th sub-mirror. Thus, through the information of connected sub-mirrors, each sub-mirror can better understand the surrounding environment and adjust the control strategy to achieve collaboration between sub-mirrors.

[0026] S4: According to the current posture information of each sub-mirror and its current shared posture information, determine the current control input of each sub-mirror and the current dynamic model of each sub-mirror, and then obtain the posture information of each sub-mirror at the next moment.

[0027] In the embodiment of the present invention, the current control input of each sub-mirror is calculated and determined by designing the H infinite control law. ,Right now: ; in, represents the current shared posture information received by the nth connected sub-mirror connected to the i-th sub-mirror at the jth moment; represents the information weighting factor of the connected sub-mirror of the i-th sub-mirror, K represents the control gain matrix.

[0028] The information weighting factor in the above formula is It is based on the feedback of system status or performance indicators that are adjusted dynamically over time, thereby changing the degree of dependence of each sub-mirror on the connected sub-mirror. For example, if the difference between the posture information of the current sub-mirror and the posture information of the connected sub-mirror is large, the information weighting factor can be increased. value, to adjust it more by relying on the attitude information of the connected sub-mirrors; when the error decreases, the information weighting factor is reduced A common form is to adjust the information weighting factor by the following formula: ; in, and They are the current posture information of the a-th sub-mirror and the b-th sub-mirror respectively. The a-th sub-mirror and the b-th sub-mirror are connected through intelligent body wireless communication.

[0029] In the above formula, the control gain matrix K The following linear matrix inequality needs to be transformed: ; Transformed into a convex optimization problem with H infinity, where P represents a symmetric positive definite matrix, , represents the upper bound of the infinity norm of H, , C and D represent the output matrix and the transfer matrix respectively, which are used to describe the relationship between the system output and the state and input. In the embodiment of the present invention, by changing the variables for substitution, the problem is transformed from a non-convex form to a convex optimization form, thereby solving the control gain matrix K , that is, introducing new variables and Using these new variables, the original inequality is transformed into a linear matrix inequality about variables W and Y. Now consider the state feedback problem. The control gain matrix in the original problem is K The solution can be achieved by the following formula: ; In the embodiment of the present invention, a convex optimization tool, such as LMI Solver or YALMIP in MATLAB, can be used to solve this type of linear matrix inequality problem. The convex optimization tool will solve the linear matrix inequality problem according to the defined objective function (i.e., the upper bound of the H infinity norm). Minimize) to solve the variables W and Y, and finally get the control gain matrix K .

[0030] Through the above content, the current control input of each sub-mirror is obtained Finally, combined with the dynamic model of each sub-mirror Get the current dynamic model of each sub-mirror for: ; The next moment attitude information of each sub-mirror is determined by the following formula: for: ; in, Represents the time step.

[0031] S5: Repeat steps S2 to S4 until the current posture information of each sub-mirror converges to the ideal posture information of each sub-mirror. The posture information error of each sub-mirror at the next moment can be expressed as: ; The control input of each sub-mirror at the next moment can be expressed as: .

[0032] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the disclosure of the present invention can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0033] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for maintaining the surface shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control, characterized in that: The following steps are involved: S1: An intelligent agent is set on each sub-mirror in the main mirror to set the initial posture information of each sub-mirror, the ideal posture information of each sub-mirror and the dynamic model of each sub-mirror; S2: Obtain the current posture information of each sub-mirror, and make a difference with the ideal posture information of each sub-mirror to obtain the current posture information error of each sub-mirror; S3: using the intelligent agent to wirelessly connect each sub-mirror with other sub-mirrors, so that the current posture information of each sub-mirror is shared with the connected sub-mirrors in real time, thereby obtaining the current shared posture information received by each sub-mirror; S4: according to the current posture information of each sub-mirror and its current shared posture information, determine the current control input of each sub-mirror and the current dynamic model of each sub-mirror, and then obtain the posture information of each sub-mirror at the next moment; S5: Repeat steps S2 to S4 until the current posture information of each sub-mirror converges to the ideal posture information of each sub-mirror.

2. The method for maintaining the surface shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control according to claim 1 is characterized in that: In step S1, the dynamic model of each sub-mirror for: ； in, represents the posture information of the i-th sub-mirror, represents the control input of the ith sub-mirror, A represents the state matrix, B represents the control matrix.

3. The method for maintaining the surface shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control according to claim 2 is characterized in that: In step S3, the current shared posture information of each sub-mirror is determined by the following formula: for: ； in, represents the number of connected sub-mirrors connected to the i-th sub-mirror, Represents the total number of connected sub-mirrors connected to the i-th sub-mirror.

4. The method for maintaining the surface shape of a primary mirror of a spliced ​​telescope based on multi-agent collaborative control according to claim 3 is characterized in that: In step S4, the current control input of each sub-mirror is determined by the following formula: for: ； in, K represents the control gain matrix, represents the information weighting factor of the connected sub-mirror of the i-th sub-mirror, represents the current shared posture information received by the nth connected sub-mirror connected to the i-th sub-mirror at the jth moment; Then combined with the dynamic model of each sub-mirror Get the current dynamic model of each sub-mirror for: ； The next moment attitude information of each sub-mirror is determined by the following formula: for: ； in, Represents the time step.

Images