A Multi-Level Co-phase Control Method for Distributed Space Interferometer Detector Loads

By using a multi-stage co-phase control method to coordinate the compensation sequence of coarse and fine-stage fast-reflection mirrors and optical delay lines, the problems of mismatched capture range of actuators and excessive control input in distributed spatial interferometer detectors are solved, achieving stable beam pointing and optical path difference compensation, and improving the stability and feasibility of control.

CN123113328APending Publication Date: 2026-11-17CHINA ACAD OF AEROSPACE SCI & TECH INNOVATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611236412.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-14
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the existing technology, distributed spatial interferometer detectors have problems such as mismatch in the capture range of the actuator, unreasonable start-up timing, control input that may exceed physical limits, and excessive changes in adjacent control commands during the beam pointing and optical path difference compensation process, which lead to control instability or actuator saturation.

Method used

A multi-stage common-phase control method is adopted. By setting coarse and fine fast-reflecting mirrors and optical delay lines in the beam propagation direction, a continuous-time state-space model is established and discretized using the zero-order hold method. The reference displacement and velocity are calculated periodically, the actuators are started in sequence, the predicted state sequence and objective function are constructed, and the optimal control input sequence is applied to ensure input amplitude constraints and smoothness.

Benefits of technology

It achieves coordinated compensation between coarse and fine actuators, avoids premature saturation of actuators, meets input capacity constraints, reduces micro-vibrations, and improves control stability and engineering feasibility.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113328A_ABST
    Figure CN123113328A_ABST
Patent Text Reader

Abstract

This invention discloses a multi-level co-phase control method for a distributed space interferometric detector payload. The method acquires residual pose deviation, residual velocity deviation, two-dimensional beam pointing error, optical path difference, and state feedback values ​​of each actuator. It calculates the geometric reference quantities required by each actuator to compensate for the current optical error, and applies a sampling-period rate limit to these geometric reference quantities. Following the order of coarse-level pointing compensation, fine-level pointing compensation, coarse-level optical path difference compensation, and fine-level optical path difference compensation, the method determines the current co-phase control stage and the set of actuators to be activated. For each actuator in the set, a state prediction within a finite prediction time domain is established. An objective function is constructed that simultaneously includes state tracking error, control input amplitude, and control input increment. Under the constraint of actuator input amplitude, the optimal control input sequence is solved until the final two-dimensional beam pointing error and optical path difference both meet the preset tolerance. This invention improves the smoothness of the switching process between coarse and fine-level actuators.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of optical co-phase control technology, specifically relating to a multi-stage co-phase control method for a distributed space interferometer detector payload that utilizes coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines to coordinate and compensate for beam pointing error and optical path difference at each stage. Background Technology

[0002] Distributed space interferometers (DIAs) use multiple space-separated optical collectors to receive target beams and then coherently combine them at a beam combiner to achieve an equivalent large aperture and high angular resolution through aperture synthesis. Unlike ground-based distributed interferometers, which rely on stable ground structures, have relatively fixed baselines, and operate at a single interference frequency, space-based distributed systems are built upon freely flying spacecraft formations. This dynamic, time-varying baseline environment, accompanied by complex space perturbations and relative motions, places extremely high demands on system stability. To achieve stable interferometry, the incident beams must maintain strict consistency in their propagation directions and keep their optical path differences within a specified range. Therefore, beam pointing and optical path difference are the core of space payload co-phase control.

[0003] In practical space engineering, before the optical payload performs co-phase control, the detector formation level has already completed platform-level coarse alignment using star sensors, guide cameras, and other means. This formation-level control can maintain the nominal spatial configuration between the collector and the combiner, ensuring that the relative pose residuals between the devices converge to a small error range, thus effectively constraining the initial co-phase state within the local neighborhood corresponding to the main interference fringes. However, platform-level control alone is insufficient to meet the payload-level accuracy required for interferometric synthesis; therefore, actuators must be introduced within the optical payload for further high-precision compensation.

[0004] For high-precision compensation within a local neighborhood, conventional methods often fail to strictly distinguish the order of error compensation. If optical path difference compensation is initiated before the beam pointing is stable, the residual pointing error will alter the effective optical path and optical path difference sensitivity, easily leading to control instability or even premature saturation of the actuator. Furthermore, single-stage actuators in space payloads face stringent physical limitations in terms of stroke, resolution, and input capability. If the control law does not explicitly consider the upper limit of the actuator input, the resulting driving force or torque may exceed physical limits. Simultaneously, if the control process focuses solely on eliminating tracking errors without smoothing abrupt changes in adjacent control commands, it can easily induce micro-vibrations in the load, compromising the optical stability required for interference. Therefore, a multi-stage co-phase control method for loads that can adapt to the characteristics of dynamic spatial baselines and uniformly address the sequential initiation of actuators, input amplitude constraints, and the smoothness of control commands is urgently needed. Summary of the Invention

[0005] The technical problem to be solved by this invention is: to address the issues of mismatched capture ranges of coarse and fine co-phase actuators, unreasonable timing of pointing compensation and optical path difference compensation, control inputs potentially exceeding physical limits, and excessive changes in adjacent control commands, a multi-level co-phase control method for distributed space interferometer payloads is provided to achieve sequential coordination of coarse pointing compensation, fine pointing compensation, coarse optical path difference compensation, and fine optical path difference compensation, and to obtain actuator control commands that satisfy input amplitude constraints and change smoothly through finite time-domain incremental model predictive control.

[0006] The technical solution adopted in this invention is: a multi-level co-phase control method for a distributed space interferometer detector payload, comprising the following steps: In each controlled optical path, coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines are sequentially set along the beam propagation direction, and are respectively denoted as coarse and fine pointing actuators and coarse and fine optical path difference actuators; the residual pose deviation of the platform, the residual velocity deviation of the platform, the two-dimensional pointing error of the beam, the optical path difference, and the status feedback values ​​of each actuator are obtained; Establish continuous-time state-space models of coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines, and discretize them using the zero-order preservation method to obtain the corresponding discrete-time models. Based on the current residual pose deviation, residual velocity deviation, state feedback values ​​of each actuator, and optical path geometry, calculate the original reference displacement required by each actuator to compensate for the current optical error. The original reference displacement is then subjected to a rate limit on a sampling cycle basis to obtain a rate-limited reference displacement. and reference speed ; Based on the relationship between the two-dimensional pointing error or displacement tracking error of the beam output by the preceding actuator and the corresponding switching threshold, the following steps are taken in the order of coarse-level pointing compensation, fine-level pointing compensation, coarse-level optical path difference compensation, and fine-level optical path difference compensation: Common-phase control phase at each sampling time and the set of initiating execution mechanisms ; Based on the aforementioned discrete-time model, the set of initiating actuators is... Each actuator in the process establishes a predicted state sequence within a finite prediction time domain. Construct an objective function that simultaneously includes the predicted state tracking error, the predicted control input magnitude, and the predicted control input increment. And solve the optimal control input sequence under the actuator input amplitude constraint. ; Apply only the optimal control input sequence The first control input in At the next sampling time, the state feedback values ​​of each actuator are reacquired, and the original reference displacement is regenerated. Speed ​​limit reference displacement and reference speed Reassess the common phase control stage and the set of initiating execution mechanisms The optimization problem is solved again until the two-dimensional pointing error of the terminal beam is found. Optical path difference with terminal All meet the preset tolerance.

[0007] Furthermore, the distributed spatial interferometer detector includes a beam combiner and at least two optical collectors, each optical collector forming an optical path entering the beam combiner; each optical path has the same coarse and fine stage actuator configuration, and the multi-stage co-phase control method can be executed independently according to the optical path, or it can be executed in parallel under the same sampling timing.

[0008] Furthermore, the continuous-time state-space model of the coarse and fine fast-reflection mirrors and the coarse and fine optical delay lines is established and discretized using the zero-order preservation method, specifically as follows: The first Each degree of freedom of the actuator is represented as a local generalized displacement near the nominal operating point. Its continuous-time second-order mechanical dynamics model is: In the formula, For the first The degree-of-freedom index of each actuator, for a two-axis fast-reflection mirror, , For optical delay lines, corresponding to the two axes, h =1; For local generalized displacement relative to the nominal operating point, where the fast-reflection mirror corresponds to the mirror angular displacement and the optical delay line corresponds to the linear displacement; , and These are the equivalent mass or moment of inertia, damping, and stiffness for the corresponding degrees of freedom; For driving force or driving torque; Definition of the first The first executive agency The continuous-time state vector with degrees of freedom is: The continuous-time state-space model for this degree of freedom can then be written as: For a two-axis fast-reflection mirror, the continuous state matrix and continuous input matrix Each model is composed of two single-axis models arranged diagonally in blocks; for optical delay lines, the corresponding single-degree-of-freedom continuous-time state-space model is directly used. Let the sampling period be The zero-order preservation method is used for the first... The continuous-time state-space model of each actuator is precisely discretized to obtain: In the formula, and The first The implementing agency in the The state vector and control input at each sampling time; and These are the continuous state matrix and continuous input matrix, respectively, assembled based on the parameters of each degree of freedom of the actuator. and These are the discrete state matrix and discrete input matrix obtained by zero-order preserved discretization, respectively.

[0009] Furthermore, the original reference displacement The residual pose deviation of the current platform, the residual velocity deviation of the platform, the nominal mirror position, the nominal mirror normal, and the first Status feedback value of each actuator The original reference displacement of the fast-reflection mirror is determined jointly. The original reference displacement of the optical delay line is the mirror rotation angle required to make the output beam point in the target direction. To compensate for the current optical path difference, the required linear displacement; Rate limiting of the original reference displacement on a sampling cycle basis includes: In the formula, For the first The sampling time is calculated by the geometric optical path. The original reference displacement of each actuator, and The first The and the first The velocity-limiting reference displacement at each sampling time. For the first The maximum allowable change in reference displacement for each actuator within a single sampling period. By amplitude A function for limiting amplitude; No. The implementing agency in the Reference velocity at each sampling time Calculate according to the following formula: In the formula, The sampling period is For the first The maximum reference speed allowed by each actuator.

[0010] Furthermore, the determination of the first Common-phase control phase at each sampling time and the set of initiating execution mechanisms Specifically: All implementing agencies and the set of initiating execution agencies Represented as: In the formula, , , and These represent coarse-grade and fine-grade fast-reflecting mirrors and coarse-grade and fine-grade optical delay lines, respectively. For the stage The corresponding set of initiating actuators, coarse-level fast-reaction mirror It is activated from the start of the co-phase control to eliminate the main incident pointing error; When the coarse-grained fast-reflective mirror Two-dimensional beam pointing error at the output end When the following formula is satisfied, the precision fast-reflection mirror is activated. : When the precision fast-reflective mirror Two-dimensional beam pointing error at the output end When the following formula is satisfied, the coarse-order optical delay line is activated. : The reference displacement of the coarse-order optical delay line is determined by the optical path difference to be compensated. When the displacement tracking error of the coarse-order optical delay line... When the following formula is satisfied, it indicates that the coarse-level optical path difference compensation has reached the corresponding accuracy. At this time, the fine-level optical delay line is activated to perform fine compensation for the remaining tracking deviation of the coarse-level optical delay line: In the formula, , and These are the start-up thresholds for precision fast-reflecting mirrors, coarse-level optical delay lines, and precision-level optical delay lines, respectively.

[0011] Furthermore, the set of initiating actuators Each actuator in the process establishes a predicted state sequence within a finite prediction time domain. Specifically: For the set of initiating execution agencies The first in Each implementing agency provides a feedback value based on its current status. As the initial value for prediction, in a length of Established within the prediction time domain: In the formula, In the first The sampling time for the first sampling time The sampling time of the first sampling moment Predicted values ​​for the status of each actuator. The corresponding control input to be optimized; the predicted state sequence Predictive control input sequence and reference state sequence They are written as: Therefore, predict the state sequence. Written as: In the formula, For the discrete state matrix The constructed state prediction matrix, For the discrete state matrix and discrete input matrix The constructed control prediction matrix; Reference state sequence By the Speed-limiting reference displacement of each actuator and reference speed Expanding along the prediction time domain, for a fast-reflection mirror, the reference state is obtained. It consists of a two-axis speed-limiting reference angle and its reference angular velocity; for an optical delay line, the reference state... It consists of a speed-limiting reference displacement and its reference linear velocity.

[0012] Furthermore, the objective function : No. Incremental model prediction control objective function for each actuator for: In the formula, To predict the control input increment sequence, The prediction state tracking error weight matrix is ​​as follows: To control the input magnitude weighting matrix, To control the input increment weight matrix; and It is a positive semi-definite matrix. It is a positive definite matrix; predicts the control input increment sequence. Defined as: In the formula, For the first The implementing agency in the The actual control input applied at each sampling time; The first item is The remaining terms represent the differences between adjacent predictive control inputs within the prediction time domain; the control input increment term is used to account for changes in adjacent soft-constraint control inputs, and is configured in conjunction with other parameters. , and This allows for a trade-off between tracking speed, input amplitude, and input smoothness.

[0013] Furthermore, the incremental model predictive control optimization problem satisfies the following input magnitude constraints: In the formula, and The first The minimum and maximum control inputs allowed for each actuator. In the first The sampling time for the first sampling time The predictive control input is obtained by solving at each sampling time point.

[0014] Furthermore, when solving the optimization problem again, for those belonging to For the actuators that have not yet started, apply the corresponding control inputs; for the actuators that have not yet started, apply zero control inputs, and the actuators that have already started continue to operate in a closed loop.

[0015] Furthermore, the two-dimensional pointing error of the terminal beam Optical path difference with terminal They are defined as follows: In the formula, and For the first At each sampling moment, the angle between the terminal beam pointing in two orthogonal directions is... and For the corresponding expected angle, and The first The terminal optical path difference and the desired optical path difference at each sampling time; the conditions for co-phase control to be completed are: In the formula, and These are the terminal beam pointing error tolerance and the terminal optical path difference tolerance, respectively.

[0016] The advantages of this invention compared to the prior art are: (1) Set sequential start conditions based on the capture range of the subsequent actuator, coordinate the large stroke coarse compensation and high resolution fine compensation in stages, avoid premature saturation of the fine actuator, and ensure that pointing compensation precedes optical path difference compensation.

[0017] (2) Explicitly apply input amplitude constraints of driving force or driving torque within the finite prediction time domain so that the optimized control command meets the physical input capability of the actuator and improves the engineering feasibility of the control method.

[0018] (3) In addition to the state tracking error and the control input amplitude, a control input increment regularization term is introduced to suppress abrupt changes in adjacent control commands, reduce jitter induced by the actuator, and improve the smoothness of the switching process between coarse and fine actuators.

[0019] (4) The coarse and fine fast reflection mirrors and coarse and fine optical delay lines adopt a unified discrete state model, prediction matrix construction and optimization process. By setting the corresponding model parameters, weight matrix, reference displacement change limit and input upper and lower limits, they can be adapted to distributed space interferometer detectors with different actuator parameters and different optical path numbers. Attached Figure Description

[0020] Figure 1 The following is a flowchart of the overall process of the multi-stage co-phase control method for load of the present invention, and is included as an abstract figure; Figure 2 This is a schematic diagram of the optical path vector and actuator configuration of the payload of the distributed spatial interferometer detector to which this invention applies; Figure 3 This is a schematic diagram of the sequential start-up state machine for coarse-level pointing, fine-level pointing, coarse-level optical path difference, and fine-level optical path difference of the present invention. Figure 4 This is a flowchart illustrating the incremental model predictive control workflow for a single actuator in this invention. Detailed Implementation

[0021] This invention proposes a multi-level co-phase control method for the payload of a distributed space interferometer detector, such as... Figure 1 As shown, it includes the following steps: Step 1: In each controlled optical path, coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines are sequentially set along the beam propagation direction, and are respectively denoted as coarse and fine pointing actuators and coarse and fine optical path difference actuators; obtain the platform residual pose deviation, platform residual velocity deviation, beam two-dimensional pointing error, optical path difference, and the status feedback value of each actuator; Step 2: Establish continuous-time state-space models of coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines, and discretize them using the zero-order preservation method to obtain the corresponding discrete-time models; Step 3: Based on the current residual pose deviation of the platform, the residual velocity deviation of the platform, the state feedback values ​​of each actuator, and the optical path geometry, calculate the original reference displacement required by each stage of the actuator to compensate for the current optical error. The original reference displacement is then subjected to a rate limit on a sampling cycle basis to obtain a rate-limited reference displacement. and reference speed ; Step 4: Based on the relationship between the two-dimensional pointing error or displacement tracking error of the beam output by the preceding actuator and the corresponding switching threshold, determine the next step in the order of coarse pointing compensation, fine pointing compensation, coarse optical path difference compensation, and fine optical path difference compensation. Common-phase control phase at each sampling time and the set of initiating execution mechanisms ; Step 5: Based on the discrete-time model, start the set of actuators. Each actuator in the process establishes a predicted state sequence within a finite prediction time domain. Construct an objective function that simultaneously includes the predicted state tracking error, the predicted control input magnitude, and the predicted control input increment. And solve the optimal control input sequence under the actuator input amplitude constraint. ; Step 6: Apply only the optimal control input sequence The first control input in At the next sampling time, the state feedback values ​​of each actuator are reacquired, and the original reference displacement is regenerated. Speed ​​limit reference displacement and reference speed Reassess the common phase control stage and the set of initiating execution mechanisms The optimization problem is solved again until the two-dimensional pointing error of the terminal beam is found. Optical path difference with terminal All meet the preset tolerance.

[0022] In step 2, the continuous-time state-space model of the coarse and fine fast-reflecting mirrors and the coarse and fine optical delay lines is established and discretized using the zero-order preservation method, specifically as follows: The first Each degree of freedom of the actuator is represented as a local generalized displacement near the nominal operating point. Its continuous-time second-order mechanical dynamics model is: In the formula, For the first The degree-of-freedom index of each actuator, for a two-axis fast-reflection mirror, , For optical delay lines, corresponding to the two axes, h =1; For local generalized displacement relative to the nominal operating point, where the fast-reflection mirror corresponds to the mirror angular displacement and the optical delay line corresponds to the linear displacement; , and These are the equivalent mass or moment of inertia, damping, and stiffness for the corresponding degrees of freedom; For driving force or driving torque; Definition of the first The first executive agency The continuous-time state vector with degrees of freedom is: The continuous-time state-space model for this degree of freedom can then be written as: For a two-axis fast-reflection mirror, the continuous state matrix and continuous input matrix Each model is composed of two single-axis models arranged diagonally in blocks; for optical delay lines, the corresponding single-degree-of-freedom continuous-time state-space model is directly used. Let the sampling period be The zero-order preservation method is used for the first... The continuous-time state-space model of each actuator is precisely discretized to obtain: In the formula, and The first The implementing agency in the The state vector and control input at each sampling time; and These are the continuous state matrix and continuous input matrix, respectively, assembled based on the parameters of each degree of freedom of the actuator. and These are the discrete state matrix and discrete input matrix obtained by zero-order preserved discretization, respectively.

[0023] In step 3, the original reference displacement of the fast-reflection mirror The original reference displacement of the optical delay line is the mirror rotation angle required to make the output beam point in the target direction. To compensate for the current optical path difference, the required linear displacement; Rate limiting of the original reference displacement on a sampling cycle basis includes: In the formula, For the first The sampling time is calculated by the geometric optical path. The original reference displacement of each actuator, and The first The and the first The velocity-limiting reference displacement at each sampling time. For the first The maximum allowable change in reference displacement for each actuator within a single sampling period. By amplitude A function for limiting amplitude; No. The implementing agency in the Reference velocity at each sampling time Calculate according to the following formula: In the formula, The sampling period is For the first The maximum reference speed allowed by each actuator.

[0024] In step 4, the determination of the first Common-phase control phase at each sampling time and the set of initiating execution mechanisms Specifically: All implementing agencies and the set of initiating execution agencies Represented as: In the formula, , , and These represent coarse-grade and fine-grade fast-reflecting mirrors and coarse-grade and fine-grade optical delay lines, respectively. For the stage The corresponding set of initiating actuators, coarse-level fast-reaction mirror It is activated from the start of the co-phase control to eliminate the main incident pointing error; When the coarse-grained fast-reflective mirror Two-dimensional beam pointing error at the output end When the following formula is satisfied, the precision fast-reflection mirror is activated. : When the precision fast-reflective mirror Two-dimensional beam pointing error at the output end When the following formula is satisfied, the coarse-order optical delay line is activated. : The reference displacement of the coarse-order optical delay line is determined by the optical path difference to be compensated. When the displacement tracking error of the coarse-order optical delay line... When the following formula is satisfied, it indicates that the coarse-level optical path difference compensation has reached the corresponding accuracy. At this time, the fine-level optical delay line is activated to perform fine compensation for the remaining tracking deviation of the coarse-level optical delay line: In the formula, , and These are the start-up thresholds for the precision fast-reflection mirror, the coarse-level optical delay line, and the precision optical delay line, respectively. The sequential start-up states of each actuator are as follows: Figure 3 As shown.

[0025] In step 5, the set of start-up actuators is... Each actuator in the process establishes a predicted state sequence within a finite prediction time domain. Specifically: For the set of initiating execution agencies The first in Each implementing agency provides a feedback value based on its current status. As the initial value for prediction, in a length of Established within the prediction time domain: In the formula, In the first The sampling time for the first sampling time The sampling time of the first sampling moment Predicted values ​​for the status of each actuator. The corresponding control input to be optimized; the predicted state sequence Predictive control input sequence and reference state sequence They are written as: Therefore, predict the state sequence. Written as: In the formula, For the discrete state matrix The constructed state prediction matrix, For the discrete state matrix and discrete input matrix The constructed control prediction matrix; Reference state sequence By the Speed-limiting reference displacement of each actuator and reference speed Expanding along the prediction time domain, for a fast-reflection mirror, the reference state is obtained. It consists of a two-axis speed-limiting reference angle and its reference angular velocity; for an optical delay line, the reference state... It consists of a speed-limiting reference displacement and its reference linear velocity.

[0026] Establish the first Incremental model prediction control objective function for each actuator for: In the formula, To predict the control input increment sequence, The prediction state tracking error weight matrix is ​​as follows: To control the input magnitude weighting matrix, To control the input increment weight matrix; and It is a positive semi-definite matrix. It is a positive definite matrix; predicts the control input increment sequence. Defined as: In the formula, For the first The implementing agency in the The actual control input applied at each sampling time; The first item is The remaining terms represent the differences between adjacent predictive control inputs within the prediction time domain; the control input increment term is used to account for changes in adjacent soft-constraint control inputs, and is configured in conjunction with other parameters. , and This allows for a trade-off between tracking speed, input amplitude, and input smoothness.

[0027] The incremental model predictive control optimization problem satisfies the following input magnitude constraints: and The first The minimum and maximum control inputs allowed for each actuator. In the first The sampling time for the first sampling time The predictive control input is obtained by solving at each sampling time point.

[0028] In step 6, when solving the optimization problem again, for those belonging to For the actuators that have not yet started, apply the corresponding control inputs; for the actuators that have not yet started, apply zero control inputs, and the actuators that have already started continue to operate in a closed loop.

[0029] Two-dimensional pointing error of the terminal beam Optical path difference with terminal They are defined as follows: In the formula, and For the first At each sampling moment, the angle between the terminal beam pointing in two orthogonal directions is... and For the corresponding expected angle, and The first The terminal optical path difference and the desired optical path difference at each sampling time; the conditions for co-phase control to be completed are: In the formula, and These are the terminal beam pointing error tolerance and the terminal optical path difference tolerance, respectively. The incremental model predictive control rolling optimization process for a single actuator is as follows: Figure 4 As shown.

[0030] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0031] Example 1 Taking the simplest distributed space interferometer detector consisting of one beam combiner and two optical collectors as an example, the two optical paths have a symmetrical structure; the control process is explained below using one of the controlled optical paths, while the other optical path is executed independently using the same method or in parallel under the same sampling sequence.

[0032] like Figure 2As shown, coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines are sequentially arranged along the beam propagation direction of a controlled optical path. The fast-reflecting mirrors are mainly used to compensate for beam propagation direction errors, while the optical delay lines are mainly used to compensate for optical path differences between different optical paths. These actuators together constitute the load-level co-phase error adjustment link. The coarse-stage fast-reflecting mirror has a large angular travel to achieve primary pointing capture; the fine-stage fast-reflecting mirror has high angular resolution to further reduce pointing residuals; the coarse-stage optical delay line has a large linear travel to eliminate primary optical path differences; and the fine-stage optical delay line has high displacement resolution to achieve fine adjustment of the terminal optical path difference.

[0033] The following control steps are performed in each sampling period: Step 1: Read the platform's position deviation, attitude deviation, translational velocity deviation, and angular velocity deviation relative to the nominal configuration; read the two-dimensional pointing error and optical path difference of the terminal beam; and obtain the position and velocity feedback values ​​of the four-stage actuators. The actuator status feedback values ​​are obtained from the position and velocity feedback devices of the corresponding actuators.

[0034] Step 2: Based on the current platform residual pose deviation, platform residual velocity deviation, nominal optical element position and attitude, and actuator status feedback values. Optical path geometry calculations were performed. The coordinate transformation, ray-mirror intersection, mirror reflection, and piecewise optical path tracing methods used were referenced from "Index Decomposition Method for Optical Co-phase Control Systems Based on Geometric Models" (Optics & Precision Engineering, 2026, accepted). For the fast-reflecting mirror, the original reference rotation angle required to make the output beam direction consistent with the desired direction was obtained. For optical delay lines, determine the original reference displacement required to compensate for the current optical path difference. The speed-limiting reference displacement is then obtained according to the aforementioned speed-limiting formula. and reference speed This avoids the actuator receiving a reference displacement similar to a step signal when switching from an inactive state to an active state.

[0035] Step 3, according to the first Common-phase control phase at each sampling time Determine the set of initiating execution agencies Initially, only the coarse-stage fast-reflection mirror is activated. When coarse pointing error The norm is less than the threshold. Then, activate the precision fast-reflective mirror. When the precision pointing error The norm is less than the threshold. Then, start the coarse-stage optical delay line. When the coarse-order optical delay line displacement tracking error The absolute value is less than the start threshold. Then, start the precision optical delay line. The initiated front-stage actuators operate continuously in a closed loop, forming a hierarchical coarse-to-fine coordinated control. Each subsequent actuator generates a reference displacement based on the tracking deviation of the preceding actuator. The coarse-stage actuators perform major error compensation as much as possible, while the fine-stage actuators further compensate for the remaining tracking deviations of the preceding actuators. The coarse and fine-stage actuators do not simultaneously perform repeated compensation for the same total error. The reference displacement of the coarse-stage optical delay line is determined by the optical path difference to be compensated. When its displacement tracking error enters a specified threshold, it indicates that the coarse-stage optical path difference compensation has reached the corresponding accuracy. At this point, the fine-stage optical delay line is initiated to perform fine compensation for the remaining tracking deviations of the coarse-stage optical delay line.

[0036] Step 4, for The first in Each actuator employs a discrete state matrix. and discrete input matrix In length of Establish a predicted state sequence within the prediction time domain Feedback value based on the current state As the initial value for prediction, the reference state sequence is expanded along the prediction time domain. Set the predicted state tracking error weight matrix according to the state tracking accuracy requirements. Set the control input amplitude weight matrix according to the actuator input amplitude requirements. Set the control input increment weight matrix according to the smoothness requirements of the control command. and the lower limit of input allowed by the actuator. and input limit Write the input amplitude constraint.

[0037] Step 5: Solve the quadratic programming problem under the resulting input magnitude constraints to obtain a length of... Optimal control input sequence Only The first control input in Apply to the For each actuator, the remaining predictive control inputs are not directly applied; for actuators that have not yet started, zero control input is applied, and the upstream actuators that have already started continue to operate in a closed loop.

[0038] Step Six: After the actuator status and optical output are updated, proceed to the next sampling cycle and reacquire the status feedback value. Regenerate the original reference displacement Speed ​​limit reference displacement and reference speed Reassess the common phase control stage and the set of initiating execution mechanisms And solve the optimization problem. When the two-dimensional pointing error of the terminal beam The norm is not greater than And the terminal optical path difference The absolute value is not greater than When the load reaches a common phase state, the controller continues to operate in the same manner to maintain the common phase.

[0039] Regarding parameter selection, the prediction time domain length According to the sampling period The bandwidth and online computing capabilities of the actuators are determined; , and The settings are configured according to the tracking speed, input capability, and control command smoothness requirements of each stage. and Take the first The permissible range of driving force or driving torque for each actuator; , and The range is determined based on the capture range of the corresponding downstream actuator, and a margin is allowed in conjunction with the measured noise.

[0040] This implementation uses the same incremental model predictive control structure for all four levels of actuators. The coarse-level actuator achieves rapid acquisition through a larger allowable input and appropriate state tracking error weights, while the fine-level actuator achieves smooth and fine adjustment through a smaller reference displacement change limit and a larger control input increment weight. Each actuator level does not require changing its control algorithm; only the corresponding model parameters, weight matrix, reference displacement change limit, and input upper and lower limits need to be set to achieve coordinated coarse and fine-level control.

[0041] For a distributed spatial interferometer detector with more than two optical collectors, the four-level actuators can be set up in each controlled optical path and the same control process can be executed; each optical path uses the two-dimensional pointing error and optical path difference at the beam combiner as the terminal phase criterion.

[0042] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

[0043] The contents not described in detail in this specification are common knowledge to those skilled in the art.

Claims

1. A method for multi-level co-phase control of a distributed space interferometric detector payload, characterized in that, Includes the following steps: In each controlled optical path, coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines are sequentially set along the beam propagation direction, and are respectively denoted as coarse and fine pointing actuators and coarse and fine optical path difference actuators; the residual pose deviation of the platform, the residual velocity deviation of the platform, the two-dimensional pointing error of the beam, the optical path difference, and the status feedback values ​​of each actuator are obtained; Establish continuous-time state-space models of coarse and fine fast-reflecting mirrors and coarse and fine optical delay lines, and discretize them using the zero-order preservation method to obtain the corresponding discrete-time models. Based on the current residual pose deviation, residual velocity deviation, state feedback values ​​of each actuator, and optical path geometry, calculate the original reference displacement required by each actuator to compensate for the current optical error. The original reference displacement is then subjected to a rate limit on a sampling cycle basis to obtain a rate-limited reference displacement. and reference speed ; Based on the relationship between the two-dimensional pointing error or displacement tracking error of the beam output by the preceding actuator and the corresponding switching threshold, the following steps are taken in the order of coarse-level pointing compensation, fine-level pointing compensation, coarse-level optical path difference compensation, and fine-level optical path difference compensation: Common-phase control phase at each sampling time and the set of initiating execution mechanisms ; Based on the aforementioned discrete-time model, the set of initiating actuators is... Each executing agency in the process establishes a predicted state sequence within a finite prediction time domain. Construct an objective function that simultaneously includes the predicted state tracking error, the predicted control input magnitude, and the predicted control input increment. And solve the optimal control input sequence under the actuator input amplitude constraint. ; Apply only the optimal control input sequence The first control input in ; At the next sampling time, the state feedback values ​​of each actuator are reacquired, and the original reference displacement is regenerated. Speed ​​limit reference displacement and reference speed Reassess the common phase control stage and the set of initiating execution mechanisms The optimization problem is solved again until the two-dimensional pointing error of the terminal beam is found. Optical path difference with terminal All meet the preset tolerance.

2. The method for multi-level phase control of a distributed space interferometer detector payload according to claim 1, characterized in that, The distributed space interferometer detector includes a beam combiner and at least two optical collectors, each optical collector forming an optical path into the beam combiner; each optical path has the same coarse and fine stage actuator configuration, and the multi-stage co-phase control method can be executed independently according to the optical path, or it can be executed in parallel under the same sampling time sequence.

3. The method for multi-level co-phase control of a distributed space interferometer detector payload according to claim 1, characterized in that, The continuous-time state-space model of the coarse and fine fast-reflection mirrors and the coarse and fine optical delay lines is established and discretized using the zero-order preservation method, specifically as follows: The first Each degree of freedom of the actuator is represented as a local generalized displacement near the nominal operating point. Its continuous-time second-order mechanical dynamics model is: In the formula, For the first The degree-of-freedom index of each actuator, for a two-axis fast-reflection mirror, , For optical delay lines, corresponding to the two axes, h =1; For local generalized displacement relative to the nominal operating point, where the fast-reflection mirror corresponds to the mirror angular displacement and the optical delay line corresponds to the linear displacement; , and These are the equivalent mass or moment of inertia, damping, and stiffness for the corresponding degrees of freedom; For driving force or driving torque; Definition of the first The first executive agency The continuous-time state vector with degrees of freedom is: The continuous-time state-space model for this degree of freedom can then be written as: For a two-axis fast-reflection mirror, the continuous state matrix and continuous input matrix Each model is composed of two single-axis models arranged diagonally in blocks; for optical delay lines, the corresponding single-degree-of-freedom continuous-time state-space model is directly used. Let the sampling period be The zero-order preservation method is used for the first... The continuous-time state-space model of each actuator is precisely discretized to obtain: In the formula, and The first The implementing agency in the The state vector and control input at each sampling time; and These are the continuous state matrix and continuous input matrix, respectively, assembled based on the parameters of each degree of freedom of the actuator. and These are the discrete state matrix and discrete input matrix obtained by zero-order preserved discretization, respectively.

4. The method for multi-level co-phase control of a distributed space interferometric detector payload according to claim 1, characterized in that, The original reference displacement The residual pose deviation of the current platform, the residual velocity deviation of the platform, the nominal mirror position, the nominal mirror normal, and the first Status feedback value of each actuator The original reference displacement of the fast-reflection mirror is determined jointly. The original reference displacement of the optical delay line is the mirror rotation angle required to make the output beam point in the target direction. To compensate for the current optical path difference, the required linear displacement; Rate limiting of the original reference displacement on a sampling cycle basis includes: In the formula, For the first The sampling time is calculated by the geometric optical path. The original reference displacement of each actuator, and The first The and the first The velocity-limiting reference displacement at each sampling time. For the first The maximum allowable change in reference displacement for each actuator within a single sampling period. By amplitude A function for limiting amplitude; No. The implementing agency in the Reference velocity at each sampling time Calculate according to the following formula: In the formula, The sampling period is For the first The maximum reference speed allowed by each actuator.

5. The method for multi-level co-phase control of a distributed space interferometer detector payload according to claim 1, characterized in that, The determination of the first Common-phase control phase at each sampling time and the set of initiating execution mechanisms Specifically: All implementing agencies and the set of initiating execution agencies Represented as: In the formula, , , and These represent coarse-grade and fine-grade fast-reflecting mirrors and coarse-grade and fine-grade optical delay lines, respectively. For the stage The corresponding set of initiating actuators, coarse-level fast-reaction mirror It is activated from the start of the co-phase control to eliminate the main incident pointing error; When the coarse-grained fast-reflective mirror Two-dimensional beam pointing error at the output end When the following formula is satisfied, the precision fast-reflection mirror is activated. : When the precision fast-reflective mirror Two-dimensional beam pointing error at the output end When the following formula is satisfied, the coarse-order optical delay line is activated. : The reference displacement of the coarse-order optical delay line is determined by the optical path difference to be compensated. When the displacement tracking error of the coarse-order optical delay line... When the following formula is satisfied, it indicates that the coarse-level optical path difference compensation has reached the corresponding accuracy. At this time, the fine-level optical delay line is activated to perform fine compensation for the remaining tracking deviation of the coarse-level optical delay line: In the formula, , and These are the start-up thresholds for precision fast-reflecting mirrors, coarse-level optical delay lines, and precision-level optical delay lines, respectively.

6. The method for multi-level co-phase control of a distributed space interferometer detector payload according to claim 3, characterized in that, The set of initiating actuators Each executing agency in the process establishes a predicted state sequence within a finite prediction time domain. Specifically: For the set of initiating execution agencies The first in Each implementing agency provides a feedback value based on its current status. As the initial value for prediction, in a length of Established within the prediction time domain: In the formula, In the first The sampling time for the first sampling time The sampling time of the first sampling moment Predicted values ​​for the status of each actuator. The corresponding control input to be optimized; the predicted state sequence Predictive control input sequence and reference state sequence Write them as follows: Therefore, predict the state sequence. Written as: In the formula, For the discrete state matrix The constructed state prediction matrix, For the discrete state matrix and discrete input matrix The constructed control prediction matrix; Reference state sequence By the Speed-limiting reference displacement of each actuator and reference speed Expanding along the prediction time domain, for a fast-reflection mirror, the reference state is obtained. It consists of a two-axis speed-limiting reference angle and its reference angular velocity; for an optical delay line, the reference state... It consists of a speed-limiting reference displacement and its reference linear velocity.

7. The method for multi-level phase control of a distributed space interferometer detector payload according to claim 6, characterized in that, The objective function : No. Incremental model prediction control objective function for each actuator for: In the formula, To predict the control input increment sequence, The prediction state tracking error weight matrix is ​​as follows: To control the input magnitude weighting matrix, To control the input increment weight matrix; and It is a positive semi-definite matrix. It is a positive definite matrix; predicts the control input increment sequence. Defined as: In the formula, For the first The implementing agency in the The actual control input applied at each sampling time; The first item is The remaining terms represent the differences between adjacent predictive control inputs within the prediction time domain; the control input increment term is used to account for changes in adjacent soft-constraint control inputs, and is configured in conjunction with other settings. , and This allows for a trade-off between tracking speed, input amplitude, and input smoothness.

8. The method for multi-level phase control of a distributed space interferometer detector payload according to claim 7, characterized in that, The incremental model predictive control optimization problem satisfies the following input magnitude constraints: In the formula, and The first The minimum and maximum control inputs allowed for each actuator. In the first The sampling time for the first sampling time The predictive control input is obtained by solving at each sampling time point.

9. The method for multi-level co-phase control of a distributed space interferometer detector payload according to claim 7, characterized in that, When solving the optimization problem again, for those belonging to For the actuators that have not yet started, apply the corresponding control inputs; for the actuators that have not yet started, apply zero control inputs, and the actuators that have already started continue to operate in a closed loop.

10. The method for multi-level phase control of a distributed space interferometer detector payload according to claim 1, characterized in that, Two-dimensional pointing error of the terminal beam Optical path difference with terminal They are defined as follows: In the formula, and For the first At each sampling moment, the angle between the terminal beam pointing in two orthogonal directions is... and For the corresponding expected angle, and The first The terminal optical path difference and the desired optical path difference at each sampling time; the conditions for co-phase control to be completed are: In the formula, and These are the terminal beam pointing error tolerance and the terminal optical path difference tolerance, respectively.