On-orbit calibration method for installation error of laser terminal

Abstract

The invention relates to the technical field of laser terminal pointing control, and discloses a laser terminal installation error on-orbit calibration method which comprises the following steps: S1, defining multiple coordinate systems including an ideal installation coordinate system of a laser terminal during ground calibration and an actual installation coordinate system caused by stress release or deformation after emission, and a real-time pointing coordinate system of in-orbit operation of the laser terminal. And S2, establishing a laser terminal installation error model, wherein the model defines an installation error angle through a direction relation of a coordinate system before and after terminal emission. According to the method, the installation error angle is dynamically estimated through the recursive least square method, high-precision dynamic correction of the laser terminal pointing error is achieved in combination with the rotation matrix pointing correction model, meanwhile, error feedback and time sequence control are introduced, and the problems of nonlinear error accumulation, insufficient fixed gain response and long-term error accumulation in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of laser terminal pointing control, and specifically to an on-orbit calibration method for the installation error of a laser terminal. Background Art

[0002] In the field of laser terminal pointing control, the influence of installation error on the system pointing accuracy has always been an important issue. In the prior art, technical means such as off-line calibration, static compensation, or simple feedback control are adopted to reduce the influence of installation error. The basic idea of these methods is to pre-estimate the installation error angle through calibration experiments or mathematical models and use it as a fixed parameter for subsequent error correction. However, in a dynamic environment, such as the drastic change of satellite attitude, vibration interference, and non-linear error drift caused by environmental factors, the fixed error model of the prior art often fails to meet the requirements of high-precision and dynamic compensation.

[0003] The structure of the prior art relies on static or linearized error models, and estimates and corrects the error angle by simply fitting the error relationship. Although this linear assumption is computationally simple, it has insufficient adaptability to non-linear dynamic error changes. Especially in the case of large error drift or drastic dynamic changes, the model error will rapidly accumulate, resulting in a significant decrease in system accuracy. In addition, most feedback control methods adopt fixed gain design and lack the ability of adaptive adjustment to changes in operating conditions. This makes the system prone to problems such as insufficient correction or overcorrection in a high-dynamic environment, and even affects the stability of pointing control. At the same time, the prior art often only focuses on the correction of instantaneous errors and ignores the cumulative effect of historical errors, resulting in the gradual amplification of errors during long-term operation, which limits the long-term reliability and pointing accuracy of the system. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides an on-orbit calibration method for the installation error of a laser terminal, which solves the problems in the prior art that it is impossible to accurately estimate and efficiently correct the installation error in a dynamic environment in real time, resulting in insufficient pointing accuracy and system stability.

[0005] To achieve the above object, the present invention is realized through the following technical solutions: An on-orbit calibration method for the installation error of a laser terminal, comprising the following steps: S1. Define multiple coordinate systems, including the ideal installation coordinate system of the laser terminal during ground calibration, the actual installation coordinate system caused by stress release or deformation after launch, and the real-time pointing coordinate system of the laser terminal during on-orbit operation; S2. Establish a laser terminal installation error model. The model defines the installation error angle through the directional relationship between the coordinate systems before and after the terminal emits. The installation error model is described by three-axis error angles. The coordinate system relationship before and after the laser terminal emits is established by the installation error angle. The installation error angle is approximated by small angles and simplified into angular deviations in three directions. The installation error model is as follows: The coordinate system transformation matrix before and after emission is a unit matrix superimposed with a rotation matrix composed of three-axis installation error angles; S3. After the laser terminal scans, captures, and establishes a communication link, obtain the real-time pointing angle and satellite attitude information of the laser terminal, calculate the theoretical pointing vector and the actual pointing vector, and construct an error observation quantity based on the difference between the two; S4. Based on the error observation quantity, use the recursive least squares method to estimate the installation error angle, and obtain the installation error angle of the laser terminal through iterative convergence; S5. Apply the installation error angle to the pointing correction of the laser terminal; S6. Multi-source information fusion and dynamic adaptation. Through multi-source information fusion technology, combine sensor data such as the temperature of the laser terminal and the satellite attitude to adjust the error correction strategy in real time; S7. Error feedback and optimization gain adjustment. Optimize the gain adjustment according to the error feedback mechanism and historical error data.

[0006] Preferably, the error observation quantity is determined by the difference between the real-time pointing vector and the theoretical pointing vector after the laser terminal scans, captures, and establishes a link. The real-time pointing vector is calculated through the elevation angle and azimuth angle of the laser terminal. The theoretical pointing vector is determined by the ground calibration result of the laser terminal and the target position.

[0007] Preferably, the theoretical pointing vector is calculated based on the following information: 1) The position vector of the target satellite or ground station in the inertial coordinate system; 2) The conversion relationship between the satellite coordinate system and the inertial coordinate system described by the satellite attitude information; 3) The initial installation position and attitude information of the laser terminal during ground calibration.

[0008] Preferably, the recursive least squares method is implemented through the following steps: 1) Construct a state equation for the installation error, assuming that the installation error angle is a constant; 2) Establish an error observation equation, and the observation quantity is the error vector; 3) Calculate the gain matrix for recursive estimation, and update the installation error angle according to the gain matrix; 4) Optimize the estimation accuracy of subsequent observation quantities by updating the error covariance matrix; 5) Repeat the iteration until the error change amount meets the termination condition.

[0009] Preferably, the gain matrix is calculated from an error covariance matrix, an observation matrix, and an observation noise covariance matrix. The initial error covariance matrix of the recursive least squares method is valued according to the laser terminal error range and is dynamically adjusted during the iteration process.

[0010] Preferably, the iteration termination condition of the method is that the change in the installation error angle in two consecutive iterations is less than a preset threshold, and the preset threshold is determined according to the laser terminal pointing accuracy requirement.

[0011] Preferably, the installation error angle is a set of three-axis installation error angles, and the three-axis installation error angles are calculated by multi-vector attitude determination. The accuracy of the multi-vector attitude determination is determined by the included angle of the reference vectors, and the included angle of the reference vectors is 90 degrees.

[0012] Preferably, the installation error angle after the S5 calibration is applied to the scanning capture and link tracking processes of the laser terminal. By substituting the installation error angle into the pointing control algorithm of the laser terminal, the pointing deviation of the laser terminal is corrected to shorten the link establishment time.

[0013] Preferably, the method is applicable to the on-orbit calibration of batch laser communication terminal devices, can avoid adding additional high-cost attitude sensors, and can adapt to the deformation characteristics of multiple satellite platforms at the same time.

[0014] Preferably, the multi-source information fusion technology includes a Kalman filter or a particle filter algorithm for fusing satellite attitude, laser terminal temperature, and acceleration data.

[0015] The present invention provides a method for on-orbit calibration of laser terminal installation errors. It has the following beneficial effects: 1. By adopting the recursive least squares method to dynamically estimate the installation error angle, the present invention achieves the effect of real-time tracking of the installation error. Compared with the prior art technical solutions that only rely on offline calibration or static calculation, the problem that the error cannot be accurately captured when the error changes dynamically is solved, and the flexibility and adaptability of the estimation are significantly improved.

[0016] 2. By introducing a pointing correction model based on the rotation matrix, the present invention realizes the high-precision dynamic correction of the laser terminal pointing error. Different from the common simple linear correction methods in the prior art, this solution effectively solves the problem of non-linear error accumulation under complex attitude changes, enabling the system to maintain stable pointing accuracy in a variety of dynamic environments.

[0017] 3. The present invention combines an error feedback mechanism and an optimized gain adjustment strategy, achieving the effects of fast response and error self-adaptive adjustment. In the prior art, fixed gain parameters are often used, which tend to cause overcorrection or response lag. However, this solution solves the problem of insufficient robustness of the system under different operating conditions through dynamic optimization.

[0018] 4. By adopting the time series integration control technology, the present invention significantly reduces the cumulative error in the error correction process. Compared with the technical solutions in the traditional methods that ignore the historical error data, this solution effectively compensates for the long-term deviation problem caused by insufficient instantaneous correction, enabling the system to maintain high reliability and pointing stability during continuous operation.

[0019] 5. The present invention adopts a multi-source information fusion technology to fuse the data of different sensors in real time and optimize the error correction process. In this way, the system can automatically adapt in a complex dynamic environment, reducing the negative impact of environmental changes on the system performance, enhancing the stability and robustness of the system. The adjustment mechanism of multi-level feedback and historical error effectively slows down the accumulation of errors, ensuring the stability of the laser terminal during long-term operation. The system can continuously optimize the error correction in multiple iteration cycles, improving the reliability of the system in long-term tasks. Description of the Drawings

[0020] Figure 1 It is a schematic flow chart of the method of the present invention. Detailed Embodiment

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0022] Please refer to the attached Figure 1 , the embodiment of the present invention provides a method for on-orbit calibration of the installation error of a laser terminal, including the following steps: S1. Define multiple coordinate systems, including the ideal installation coordinate system of the laser terminal during ground calibration, the actual installation coordinate system caused by stress release or deformation after launch, and the real-time pointing coordinate system of the laser terminal during on-orbit operation; Definition of Coordinate Systems and Modeling of Installation Errors In the present invention, in order to accurately describe the installation errors of the laser terminal during on-orbit operation and provide a reliable mathematical basis for subsequent calibration and pointing correction, it is first necessary to establish a coordinate system and an installation error model related to the laser terminal. These coordinate systems are interrelated and are respectively used to describe the ideal installation state of the laser terminal during ground calibration, the actual offset state after launch, and the pointing characteristics during on-orbit operation. By establishing the installation error model, the deviation angles generated by the laser terminal due to deformation or stress release can be clearly characterized, providing the necessary conditions for subsequent error estimation and correction.

[0023] Generally, this step first starts from the definition of the coordinate system to clarify the reference frame of the laser terminal in different states. On this basis, the specific method of using the direction cosine matrix to describe the installation error is further introduced. As an option, the error model is appropriately simplified through the small-angle assumption to reduce the computational complexity while ensuring the computational accuracy. In different embodiments, the installation error model can be flexibly adjusted to meet the design requirements of specific satellite platforms.

[0024] Specifically, in this embodiment, the definition of the coordinate system and the installation error model includes the following parts: Definition of the J2000 inertial coordinate system Generally, the J2000 inertial coordinate system is a reference coordinate system widely used in astronomy and space technology. Its origin is set at the Earth's center of mass, and the following three axes are defined: : Points in the direction of the vernal equinox; : Is parallel to the direction of the Earth's axis of rotation; : Is perpendicular to and and together satisfy the right-hand rule.

[0025] In the present invention, the J2000 coordinate system is mainly used to describe the global position of the target satellite or ground station, as well as the attitude information of the satellite in inertial space.

[0026] Definition of the satellite body coordinate system In a possible implementation, the satellite body coordinate system (SAT) is a reference coordinate system with the satellite's center of mass as the origin, used to describe the installation state of the laser terminal. The axis directions of this coordinate system are usually set according to the satellite's shape and flight direction, and its definition is as follows: : Along the longitudinal axis of the satellite, usually pointing in the satellite's flight direction; : Perpendicular to the satellite's symmetric plane; : Is perpendicular to and Vertical, satisfying the right-hand rule.

[0027] As an option, the measurement result of the satellite attitude can provide the conversion relationship between the satellite body coordinate system and the J2000 coordinate system, providing support for the subsequent laser terminal pointing modeling.

[0028] Definition of the coordinate system before the terminal prism is launched Specifically, the coordinate system before the terminal prism is launched (LCT-Before) describes the ideal installation state of the laser terminal during ground calibration. The definition of this coordinate system is as follows: The origin is located at the geometric center of the laser terminal prism; : The ideal light-emitting direction at the CPA zero position; : Perpendicular to the installation joint surface of the laser terminal; : Determined according to the right-hand rule.

[0029] In some embodiments, the coordinate system before the terminal prism is launched can be determined through ground experiments, and the conversion relationship with the satellite body coordinate system can be pre-calibrated before launch.

[0030] Definition of the coordinate system after the terminal prism is launched After launch, due to reasons such as stress release and deformation, the actual installation state of the laser terminal may deviate from the ground calibration state. Therefore, the coordinate system after the terminal prism is launched (LCT-After) is defined, and its relationship with LCT-Before is represented by the installation error matrix Described. Generally, the installation error matrix is represented by a direction cosine matrix.

[0031] As an option, the definition of the installation error matrix is as follows: ; Where: : The direction cosine matrix of rotation around the axis; ; : The direction cosine matrix of rotation around the axis; ; : The direction cosine matrix of rotation around the axis; ; In a possible implementation, assuming that the installation error angles are small, the above matrix can be approximately simplified as: ; Where: : Identity matrix; : Represent the three-axis deflection angles of the installation error respectively.

[0032] This simplified form can reduce the computational complexity and is applicable to most engineering application scenarios.

[0033] Definition of the CPA zero coordinate system The CPA zero coordinate system (CPA-Zero) is the initial reference frame for laser terminal scanning and capture. Generally, when both the pitch angle and azimuth angle of the CPA are zero, its coordinate axis directions are consistent with the LCT-Before coordinate system. The specific definitions are as follows: Origin: The geometric center of the CPA light output port; : Direction of the ideal pointing vector; : Perpendicular to the laser terminal joint surface; : Satisfy the right-hand rule.

[0034] In actual operation, when the pitch angle and azimuth angle of the CPA deviate from the zero position, a CPA instantaneous coordinate system (CPA-Instant) is formed, and the relationship between the two can be described by the real-time angles of the CPA.

[0035] This step is the basis of the whole method. Through the definition of the above coordinate system and the establishment of the installation error model, the working environment and state changes of the laser terminal can be completely described, providing an accurate and reliable mathematical model for subsequent error observation and parameter estimation. This method is applicable to different satellite platforms and has strong versatility and scalability.

[0036] S2. Establish a laser terminal installation error model, and the model defines the installation error angle through the direction relationship of the coordinate systems before and after the terminal emits light; Scanning capture and error observation quantity To realize the in-orbit calibration of the laser terminal installation error, after completing the coordinate system definition and error modeling in step 1, the present invention further establishes a link with the target through scanning capture, obtains the real-time pointing data of the laser terminal, and constructs an error observation quantity based on the difference between the theoretical pointing vector and the actual pointing vector. Through these real-time observation data, the installation error of the laser terminal during in-orbit operation can be directly reflected, providing the necessary data input for subsequent error estimation.

[0037] In general, a laser terminal needs to establish a communication link with a target through a scanning and capturing method. During the link establishment process, the pointing vector of the laser terminal will be affected by installation errors, resulting in an extended link capture time or an unstable communication link. To solve this problem, the present invention constructs an observable quantity of the installation error based on the real-time pointing angle of the laser terminal and the ground calibration data, and accurately quantifies the influence of the installation error. In some embodiments, by observing the pointing data in real time, the pointing correction strategy of the laser terminal can be dynamically optimized, further improving the efficiency of link establishment.

[0038] Specifically, in this embodiment, the construction of the scanning capture and the error observable quantity includes the following: Obtaining the real-time pointing vector In this embodiment, the real-time pointing vector of the laser terminal is calculated based on the pitch angle and azimuth angle of the CPA (Coupled Positioning Assembly). The CPA is a key component of the laser terminal, which is composed of two servo motors and is used to adjust the pitch angle and azimuth angle of the laser, so as to achieve dynamic tracking of the target.

[0039] In a possible implementation manner, the real-time pointing vector of the laser terminal is determined by the following steps: First, according to the pitch angle of the CPA and the azimuth angle , a rotation relationship between the CPA zero-position coordinate system (CPA-Zero) and the CPA instantaneous coordinate system (CPA-Instant) is established.

[0040] The direction cosine matrix of the CPA instantaneous coordinate system is expressed as: ; Where: ; Combining the above rotation matrix and the initial pointing vector of the CPA zero-position coordinate system, the real-time pointing vector can be obtained: ; In general, the direction of is

[0041] which represents the ideal initial light-emitting direction. In some embodiments, the real-time pointing vector

[0042] is directly measured by the angle sensor of the CPA, or can be further calibrated in combination with satellite attitude data. In this embodiment, the theoretical pointing vector It is calculated based on the target position, satellite attitude, and ground calibration results. The theoretical pointing vector represents the light-emitting direction of the laser terminal in an ideal state without installation errors.

[0043] Specifically, the calculation steps of the theoretical pointing vector include the following: First, according to the J2000 coordinate system position of the target satellite or ground station , combined with the attitude transformation matrix of the satellite body coordinate system (SAT) , determine the projection of the target position in the satellite body coordinate system: ; Combined with the initial installation direction of the laser terminal during ground calibration , project the target position onto the CPA zero-position coordinate system: ; In a possible implementation, can be calculated from satellite orbit data and the fixed position of the ground station, while is provided in real time by the satellite's attitude sensors.

[0044] Construction of the error observation quantity Generally, the error observation quantity is defined as the difference between the theoretical pointing vector and the actual pointing vector, and its mathematical expression is: ; As an option, the error observation quantity can be decomposed into components in three directions: : Error in the light-emitting direction of the laser terminal; : Horizontal error perpendicular to the light-emitting direction; : Vertical error perpendicular to the light-emitting direction.

[0045] Specifically, The component of can be directly calculated through matrix operations. As a possible implementation, when the amplitude of the error component is small, it can be directly approximated in combination with the servo motor angle of the CPA to further simplify the calculation process of error observation.

[0046] In some embodiments, The magnitude of directly reflects the influence degree of the laser terminal installation error on link establishment. By analyzing the error observation quantity, the pointing angle of the laser terminal can be dynamically adjusted to improve the stability of link establishment.

[0047] The completion of this step provides a direct observable input for the recursive estimation of subsequent installation errors. Combining the precise calculation of the real-time pointing vector and the theoretical pointing vector ensures the accuracy and reliability of the error model. This method of constructing error observables based on differences is characterized by simple implementation and high computational efficiency, and is applicable to various on-orbit application scenarios of laser communication terminals.

[0048] S3. After the laser terminal scans, captures, and establishes a communication link, obtain the real-time pointing angle and satellite attitude information of the laser terminal, calculate the theoretical pointing vector and the actual pointing vector, and construct an error observable based on the difference between the two. Parameterization of the error model In step 2, a communication link was established through scanning and capture, and the real-time pointing vector and the theoretical pointing vector of the laser terminal were obtained. An error observable was constructed based on the difference between the two, providing raw data support for the calibration of the installation error of the laser terminal. On this basis, the present invention proposes a parameterization method based on the error model. Using the mathematical model and observables of the installation error, a direct relationship between the installation error angle and the error observable is established. Through the parameterization of this error model, the installation error angle can be expressed in a mathematical form, facilitating the subsequent implementation of error estimation and calibration.

[0049] Generally, the installation error mainly manifests as the deviation between the actual installation state and the ideal installation state of the laser terminal. This deviation can be described by the direction cosine matrix and the rotation angle. As an option, by combining the construction method of the error observable, the installation error model can be converted into an observable mathematical relationship, enabling the accurate expression of the error parameters. In some embodiments, the parameterization of the error model can be adapted according to different error distribution characteristics to improve the applicability of the calibration process.

[0050] Specifically, in this embodiment, the parameterization of the error model includes the following: Establishment of the installation error model In this embodiment, the installation error of the laser terminal is described by the direction cosine matrix relationship between the coordinate system before the terminal prism emits (LCT - Before) and the coordinate system after emission (LCT - After). The installation error matrix is defined as: ; where: represents the rotation matrix around the axis: ; represents the rotation matrix around the axis: ; represents the rotation matrix about the axis: ; In a possible implementation, through the small-angle assumption, the above matrix can be simplified to: ; where: represents the identity matrix; respectively represent the deflection angles of the installation error in the direction.

[0051] This simplified form has higher computational efficiency in practical applications and can meet the accuracy requirements for error description in most scenarios.

[0052] Mathematical expression of the error observation quantity Generally, the error observation quantity of the laser terminal is defined as the difference between the actual pointing vector and the theoretical pointing vector : ; As an option, through the installation error model, the actual pointing vector can be expressed as the result of the transformation of the theoretical pointing vector through the error matrix: ; Combining the above expressions, the error observation quantity can be further written as: ; In some embodiments, by introducing the error matrix with small-angle simplification, this formula can be extended to: ; Relationship between the installation error angle and the observation quantity Specifically, in this embodiment, the relationship between the installation error angle and the error observation quantity can be expressed as: ; where is the observation matrix, and its form depends on the theoretical pointing vector is the installation error angle, which is the calibrated target parameter.

[0053] In a possible implementation, the specific form of the observation matrix is: ; Wherein: They are respectively the three components of the theoretical pointing vector.

[0054] In this step, by parameterizing the error observation quantity and the installation error angle, a complete mathematical basis is provided for the implementation of the subsequent recursive least squares method. The parameterization of the error model is applicable to both the simple scenario of a small laser terminal and can be extended to the application scenarios of complex satellite platforms, providing flexibility for the actual operation of error calibration.

[0055] S4. Based on the error observation quantity, use the recursive least squares method to estimate the installation error angle, and obtain the installation error angle of the laser terminal through iterative convergence; In step 3, a mathematical model between the installation error and the observation quantity has been established, and the basic relationship of error observation has been clarified. Based on this model, in step 4, the installation error angle is dynamically estimated by the recursive least squares method (RLS) to achieve real-time update and optimization of the error parameters. The recursive least squares method can effectively handle the noise in the observation data and gradually optimize the error estimation, enabling the dynamic characteristics of the installation error angle to be accurately described and providing the necessary input conditions for the error correction in the subsequent steps.

[0056] Generally, the recursive least squares method is applicable to dynamic systems with unknown error characteristics or gradually changing over time. In this embodiment, the installation error angle is defined as a constant variable, and the influence of the observation residual on the error angle is dynamically adjusted by the least squares method to ensure the convergence and accuracy of the estimated value. As an option, the gain matrix and the noise covariance matrix in the recursive process can be dynamically adjusted according to the error characteristics to improve the adaptability to different system conditions.

[0057] Establishment of the state equation In this embodiment, the state equation of the installation error angle is assumed to be a constant model, and the specific expression is: ; Wherein: represents the installation error angle at the th observation; represents the installation error angle at the th observation.

[0058] This assumption is based on the fact that the change of the installation error angle can be ignored within a short time, thus simplifying the complexity of state update and making the estimation process more robust.

[0059] Establishment of the observation equation In this embodiment, the observation equation is used to describe the error observation quantity and the installation error angle The linear relationship between them has the following mathematical form: ; Where: is the observation matrix for the th observation; represents the observation noise, whose mean is assumed to be 0 and the covariance matrix is .

[0060] Generally, the observation matrix is directly related to the theoretical steering vector , and its form can be specifically expressed as: ; Where: are the components of the theoretical steering vector on the , , axes respectively.

[0061] The error observation quantity , whose components are the projections of the error vector on the three axes, results from the deviation between the actual observation result and the theoretical steering result.

[0062] The core formula of the recursive least squares method In this embodiment, the specific implementation of the recursive least squares method includes the following key steps: State update The state update formula in the recursive least squares method is: ; Where: is the gain matrix for the th update; is the difference between the observation quantity and the predicted quantity, called the residual.

[0063] Calculation of the gain matrix The calculation formula of the gain matrix is as follows: ; Where: is the value of the error covariance matrix at the previous moment; represents the covariance matrix of the observation noise.

[0064] Update of the error covariance matrix The update formula of the error covariance matrix is: ; Wherein: is the identity matrix; is the error covariance matrix after the

[0065] Setting of Initial Conditions In a possible implementation, the setting of the initial conditions for the recursive least squares method includes: Initial state value , assuming that the initial installation error angle is zero; Initial error covariance matrix is usually set as a diagonal matrix, and its diagonal elements are multiples of the possible range of the installation error angle; Noise covariance matrix The initial value of

[0066] is estimated according to the actual sensor noise characteristics. In some embodiments, in this embodiment, the gain matrix and the noise covariance matrix are dynamically adjusted to adapt to different system operating conditions. For example: When the observation noise is large, the value of can be appropriately increased to reduce the influence of the observation noise on the state estimation; After the error residual is stable, the value of can be gradually reduced to improve the stability and accuracy of the estimation.

[0067] Recursive Termination Condition In this embodiment, the termination condition of the recursive algorithm is usually set as the change amount of the installation error angle being less than the preset threshold , that is: ; Wherein, The specific value of

[0068] can be determined according to the pointing accuracy requirements of the laser terminal.

[0069] 5) Apply the installation error angle to the pointing correction of the laser terminal to optimize the efficiency and stability of subsequent scan capture and link tracking.

[0070] S5, Error Correction and Pointing Optimization In step 4, the dynamic estimation result of the installation error angle is obtained by the recursive least squares method. Based on this result, in step 5, the pointing error of the laser terminal is further corrected in real time, and the dynamic adjustment of the pointing parameters is realized by combining with the optimization model, so as to improve the pointing accuracy and robustness of the system. In this embodiment, a pointing correction model based on the rotation matrix is adopted, and through the real-time feedback and adaptive adjustment mechanism of the target point deviation, the problem of insufficient response to complex dynamic environments in the prior art is solved.

[0071] Generally, the pointing correction of the laser terminal needs to comprehensively consider the relationship between the installation error angle and the theoretical pointing vector, and at the same time combine the actual deviation of the target point to optimize the dynamic response of the system. In this embodiment, the error correction model takes the recursive estimation of the installation error angle as the core, and realizes the efficient compensation of the pointing error through feedback adjustment and integral control of historical errors. As an option, this embodiment also proposes a dynamic weight optimization strategy to adapt to the pointing requirements under various complex operating conditions.

[0072] Establishment of the Pointing Correction Model In this embodiment, the actual pointing vector of the laser terminal is calculated based on the correction relationship between the installation error angle and the theoretical pointing vector, and the specific expression is: ; Where: represents the actual pointing vector; represents the theoretical pointing vector; is the rotation matrix determined by the installation error angle

[0073] Generally, the rotation matrix represents the coordinate system transformation caused by the installation error angle, specifically: ; Where: , are the matrices of rotation around , , axes respectively, and the specific definitions are as follows: ; ; ; ​In a possible implementation, to improve the correction accuracy, the actual pointing vector is also combined with the platform's dynamic attitude information for real-time compensation to offset the secondary errors caused by platform vibration.

[0074] Deviation feedback and optimization model To further improve the pointing accuracy, an optimization model is designed based on the error feedback mechanism in this embodiment. Specifically, the target point deviation vector is defined as the difference between the actual target point position and the ideal target point position, and its expression is: ; Where: is the actual measured position of the target; is the ideal position of the target.

[0075] The optimization model dynamically adjusts the pointing vector through deviation feedback, and its calculation formula is: ; Where: is the optimized pointing vector; is the optimization gain matrix, and its size is dynamically adjusted according to the target deviation characteristics and the laser beam divergence angle.

[0076] Generally, the design of the optimization gain matrix needs to balance the correction accuracy and the response speed. For example, when the error is small, the gain can be appropriately reduced to avoid overcorrection, and when the deviation is large, the gain should be increased to respond quickly.

[0077] Time series error control In some embodiments, this embodiment further reduces the cumulative effect of long-term errors through integral control of time series. The historical error integral term The calculation formula is: ; The feedback control strategy based on historical errors can be expressed as: ; Where: is the finally corrected pointing vector; is the integral gain coefficient, which controls the influence weight of the integral term on the correction process.

[0078] As an option, the size of the integral gain can be dynamically adjusted according to the stability of historical errors to avoid system oscillation caused by excessive integration.

[0079] Initial conditions and boundary settings The initial condition settings in this embodiment include: Optimization gain matrix The initial value is set according to the target accuracy requirements of the laser terminal; Deviation vector The tolerance range is defined as 10% - 20% of the laser beam width; Integral gain The initial value range is from 0.1 to 0.5.

[0080] The setting of boundary conditions can further ensure the stability of the correction process. For example, when the historical error integral term exceeds a certain threshold, the system will trigger a dynamic amplitude limiting mechanism to avoid the negative impact of excessive adjustment on the pointing accuracy.

[0081] Dynamic adjustment mechanism In this embodiment, by introducing a dynamic adjustment mechanism, the system is ensured to have self - adaptive ability under different operating conditions. For example: When the target deviation is large, by increasing the optimization gain The correction speed is increased; When the noise level increases, the integral gain is reduced To reduce system jitter.

[0082] In some embodiments, the dynamic adjustment mechanism combines real - time monitoring data of the operating environment, such as platform attitude stability or link quality, thereby further optimizing the system performance.

[0083] In this embodiment, through the design of the pointing correction and optimization model, high - precision dynamic compensation for the laser terminal pointing error is achieved. Combining deviation feedback and historical error control, it effectively solves the problems of non - linear error accumulation, insufficient dynamic response, and long - term error accumulation in the prior art, ensuring the stability and reliability of the system in complex operating environments.

[0084] S6. Multi - source information fusion and dynamic adaptation, through multi - source information fusion technology, combining sensor data such as laser terminal temperature and satellite attitude, to adjust the error correction strategy in real - time; In this embodiment, algorithms such as Kalman filtering and particle filtering are used for data fusion. Through these methods, the laser terminal can effectively combine multi - source information for dynamic error correction. Specifically, the Kalman filter uses the state equation and the observation equation to recursively update the estimated value, optimizing the accuracy of error estimation. During this process, the error covariance matrix is continuously updated, thereby improving the reliability of the estimation. The gain matrix is calculated by the following formula: ; Among them, represents the feature vector of the observed quantity, is the error covariance matrix at the previous moment, and the gain matrix affects the degree of error correction.

[0085] As an option, particle filtering can be used as a supplementary method to Kalman filtering. Particle filtering approximates the state distribution of the system by introducing a set of particles The weights of these particles are adjusted according to the observation results at each update to ensure that the error correction process remains efficient in the presence of non-linear and non-Gaussian noise. The update formula for this process is as follows: ; Among them, is the weight of the particle, is the likelihood function of the observed value.

[0086] Generally, Kalman filtering is applicable to linear systems, while particle filtering is suitable for dealing with non-linear or non-Gaussian systems. In some embodiments, according to the requirements of the actual system, Kalman filtering or particle filtering can be selected, or both can be used in combination to improve the flexibility and adaptability of error correction.

[0087] In a possible implementation, when combining the information of multiple sensors, the situation of information redundancy or sensor failure may be faced. In this case, using a multi-sensor fusion algorithm can effectively remove redundant information and process the data of failed sensors. For example, redundant sensors can be set to avoid the impact of a single sensor failure on the system performance.

[0088] In addition, sensor information fusion is not limited to the combination of error observed quantities, but can further perform comprehensive correction through the fusion of historical error data and real-time observation data. This method can quickly adjust the pointing error correction strategy of the laser terminal when monitoring environmental changes in real time, thereby improving the response ability of the system.

[0089] In some embodiments, deep learning algorithms can also be used to enhance the accuracy of information fusion. By training on historical error data and sensor information, a deep neural network can adaptively identify error patterns and optimize the error correction strategy according to new environmental information. At this time, the network input is the data of multiple sensors and historical errors, and the network output is the corrected error angle. In this way, the system can achieve dynamic adaptability and improve the accuracy of error correction.

[0090] Specifically, deep learning models such as convolutional neural networks (CNNs) or long short-term memory networks (LSTMs) can be used to process time-series error data, accurately predict future error trends, and adjust the correction strategy accordingly. By using a standard error correction dataset during training, the network can gradually optimize the error prediction process and achieve more efficient error correction.

[0091] Generally speaking, this step realizes the real-time correction and dynamic adjustment of the laser terminal pointing error by integrating multi-source information and advanced filtering algorithms, thus effectively improving the adaptability and stability of the system in various complex environments. This not only enhances the system performance but also provides a stable error correction ability for long-term operation.

[0092] S7. Error feedback and optimization gain adjustment. Optimize the gain adjustment according to the error feedback mechanism and historical error data; In this embodiment, the gain adjustment mechanism optimizes the gain setting by feeding back historical error data. In the previous step, the error was dynamically estimated by the recursive least squares (RLS) method. As time goes by, the gain matrix will be adjusted according to the changes in historical data to avoid over-correction or under-correction of errors caused by too large or too small gains. Specifically, the update formula for the gain matrix is: ; where is the error covariance matrix, is the observed value at the current moment, is the gain dynamically adjusted according to the current error and covariance matrix. In this way, the adjustment of the gain matrix can react to the changes in the system state and optimize the correction effect of the system.

[0093] As an option, the adjustment of the gain matrix can be combined with other feedback mechanisms such as genetic algorithms or particle swarm optimization to further improve the adjustment accuracy of the gain. These methods can optimize the gain adjustment based on the historical data of the feedback error, ensuring that the system can achieve the best correction performance in different working environments.

[0094] Specifically, the design of the feedback mechanism takes into account the delay of the feedback signal and the response ability of the system. In practical applications, there is often a certain delay in the feedback signal. Especially during long-term operation, the accumulation of errors may cause signal lag. To make up for this, adaptive filtering technology can be combined to adjust the feedback gain in real time, enabling the system to maintain high stability and correction ability even when the error signal is delayed.

[0095] In a possible implementation, the system can design a gain adjustment mechanism with multiple levels. For example, for errors that change rapidly in the short term, a higher gain is adopted; while for situations where long-term error accumulation is less, the gain is appropriately reduced. This dynamic adjustment method can not only improve the system response speed but also prevent the system from overcorrecting.

[0096] In some embodiments, different sensor information can also be combined and the gain can be adjusted using the weighted average method. The contribution degree of each sensor data to the gain is weighted according to its accuracy and real-time performance, so that different sensor data have different effects on the correction process when adjusting the gain. This can further enhance the flexibility of the system and enable it to automatically adjust the correction strategy according to different situations.

[0097] Generally speaking, by optimizing the gain adjustment, the present invention can not only improve the correction accuracy of the laser terminal pointing error but also ensure that the system can continuously maintain an efficient and stable operating state in a dynamic environment. Through this feedback and gain optimization mechanism, the laser terminal can better adapt to complex and changing environments, improving the overall performance of the system and the reliability of long-term operation.

[0098] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for on-orbit calibration of laser terminal installation error, characterized in that: The following steps are involved: S1. Define multiple coordinate systems, including the ideal installation coordinate system of the laser terminal when it is calibrated on the ground, the actual installation coordinate system caused by stress release or deformation after launch, and the real-time pointing coordinate system of the laser terminal when it is running on orbit; S2. Establish a laser terminal installation error model. The model defines the installation error angle by the direction relationship of the coordinate system before and after the terminal launches. The installation error model is described by the three-axis error angle. The coordinate system relationship before and after the laser terminal launches is established by the installation error angle. The installation error angle is simplified to the angle deviation in three directions by small angle approximation. The installation error model is: The coordinate system transformation matrix before and after launch is a rotation matrix consisting of the unit matrix and the three-axis installation error angle; S3, after the laser terminal scans and captures and establishes a communication link, obtain the real-time pointing angle of the laser terminal and the satellite attitude information, calculate the theoretical pointing vector and the actual pointing vector, and construct the error observation based on the difference between the two; S4. Based on the error observation, the installation error angle is estimated by using the recursive least square method, and the installation error angle of the laser terminal is obtained through iterative convergence; S5, applying the installation error angle to the pointing correction of the laser terminal; S6, multi-source information fusion and dynamic adaptation, through multi-source information fusion technology, combined with the sensor data of laser terminal temperature and satellite attitude, real-time adjustment of error correction strategy; S7, error feedback and optimized gain adjustment, optimize gain adjustment according to the error feedback mechanism and historical error data.

2. According to claim 1, a method for on-orbit calibration of laser terminal installation error is characterized in that: The error observation amount is determined by the difference between the real-time pointing vector and the theoretical pointing vector after the laser terminal scans, captures and establishes a link. The real-time pointing vector is calculated by the pitch angle and azimuth angle of the laser terminal, and the theoretical pointing vector is determined by the ground calibration result of the laser terminal and the target position.

3. The on-orbit calibration method of a laser terminal installation error according to claim 1 is characterized in that: The theoretical pointing vector is calculated based on the following information: 1) The position vector of the target satellite or ground station in the inertial coordinate system; 2) The transformation relationship between the satellite coordinate system and the inertial coordinate system described by the satellite attitude information; 3) The initial installation position and attitude information of the laser terminal during ground calibration.

4. The on-orbit calibration method of a laser terminal installation error according to claim 1 is characterized in that: The recursive least squares method is implemented by the following steps: 1) Construct the state equation of the installation error, assuming that the installation error angle is a constant; 2) Establish the error observation equation, and the observed quantity is the error vector; 3) Calculate the recursive estimated gain matrix and update the installation error angle according to the gain matrix; 4) Optimize the estimation accuracy of subsequent observations by updating the error covariance matrix; 5) Repeat the iteration until the error change meets the termination condition.

5. The on-orbit calibration method of laser terminal installation error according to claim 4 is characterized in that: The gain matrix is ​​calculated by using the error covariance matrix, the observation matrix and the observation noise covariance matrix. The initial error covariance matrix of the recursive least squares method is taken according to the laser terminal error range and is dynamically adjusted during the iteration process.

6. The on-orbit calibration method of a laser terminal installation error according to claim 1, characterized in that: The iterative termination condition of the method is that the change in the installation error angle in two consecutive iterations is less than a preset threshold, and the preset threshold is determined according to the pointing accuracy requirement of the laser terminal.

7. The on-orbit calibration method of laser terminal installation error according to claim 1, characterized in that: The installation error angle is a collection of three-axis installation error angles, and the three-axis installation error angles are obtained by multi-vector attitude determination. The accuracy of the multi-vector attitude determination is determined by the angle of the reference vectors, and the reference vector angle is 90 degrees.

8. The on-orbit calibration method of laser terminal installation error according to claim 1, characterized in that: S5: After the calibration is completed, the installation error angle is applied to the scanning capture and link tracking process of the laser terminal. By substituting the installation error angle into the pointing control algorithm of the laser terminal, the pointing deviation of the laser terminal is corrected to shorten the link establishment time.

9. The on-orbit calibration method of laser terminal installation error according to claim 1, characterized in that: The method is suitable for on-orbit calibration of mass-produced laser communication terminal equipment, can avoid the addition of high-cost attitude sensors, and is adaptable to the deformation characteristics of various satellite platforms.

10. The on-orbit calibration method of laser terminal installation error according to claim 1, characterized in that: The multi-source information fusion technology includes Kalman filtering or particle filtering algorithm to fuse satellite attitude, laser terminal temperature and acceleration data.

Images