A method, apparatus, medium, and electronic equipment for predicting satellite attitude in orbit.

Abstract

This application provides a method, apparatus, medium, and electronic device for on-orbit prediction of satellite attitude. The method includes: inputting the current attitude, current angular velocity, flywheel speed change, environmental disturbance parameters, and observed attitude into an on-orbit satellite attitude extrapolation model. The on-orbit satellite attitude extrapolation model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and thus perform on-orbit satellite attitude prediction; obtaining the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model; and providing the extrapolated satellite attitude to an attitude control system to achieve attitude control and adjustment of the satellite. The method described in this application is applicable to non-steady-state satellite attitude extrapolation and can also adapt to interference with satellite attitude caused by environmental changes (environmental magnetic disturbance, solar illumination, atmospheric drag), thereby making the attitude extrapolation more accurate.

Description

Technical Field

[0001] This application relates to the field of satellite attitude control, and more specifically, the embodiments of this application relate to a method, apparatus, medium, and electronic equipment for predicting satellite attitude in orbit. Background Technology

[0002] In traditional attitude measurement systems, the gyroscope is the inertial reference, providing the most basic attitude and angular velocity information for the satellite. However, defects such as gyroscope constant drift, correlation drift, and measurement noise affect the measurement accuracy.

[0003] In recent years, with the continuous improvement of satellite technology, various space missions have placed increasingly higher demands on the precision of satellite attitude control. The pointing accuracy of Earth observation satellites has increased from 1° in the 1970s to 0.001° at the beginning of this century.

[0004] To meet the design requirements of high accuracy and high reliability for satellite attitude determination, combined attitude determination methods based on Kalman filters have been widely studied and applied. One major method is star-sensor and gyroscope combined attitude determination based on Kalman filters, such as... Figure 1 As shown: The satellite attitude is extrapolated using the gyro angular velocity and the attitude of the previous period. When the star sensor is unavailable, the satellite attitude is the extrapolated value. When the star sensor is available, the attitude deviation is calculated using the observed attitude based on the star sensor and the attitude extrapolated prediction value. The attitude deviation is then used as a state variable to correct the attitude and gyro drift.

[0005] The technical drawbacks of this approach are: existing attitude extrapolation methods are applicable when the satellite is in a steady state; if the extrapolated attitude remains uncorrected, attitude deviations will accumulate. Existing integral attitude extrapolation methods do not consider environmental disturbance torques, resulting in inaccurate attitude predictions. Summary of the Invention

[0006] The purpose of this application is to provide a method, apparatus, medium, and electronic device for on-orbit prediction of satellite attitude. The method described in this application can adapt to the interference on satellite attitude caused by environmental changes (environmental magnetic disturbance, sunlight, atmospheric drag), thereby making attitude extrapolation more accurate.

[0007] In a first aspect, this application provides a method for predicting satellite attitude in orbit. The method includes: inputting the current attitude, current angular velocity, flywheel speed change, environmental disturbance parameter values, and observed attitude into an on-orbit satellite attitude extrapolation model, wherein the on-orbit satellite attitude extrapolation model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and then perform on-orbit satellite attitude prediction; obtaining the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model; and providing the extrapolated satellite attitude to the attitude control system to achieve attitude control adjustment of the satellite.

[0008] Some embodiments of this application introduce parameters related to environmental changes (i.e., environmental disturbance parameters, such as environmental magnetic disturbance, sunlight, and atmospheric drag), thereby enabling the corresponding model to adapt to the interference on satellite attitude caused by environmental changes (environmental magnetic disturbance, sunlight, and atmospheric drag), thus making attitude extrapolation more accurate.

[0009] In some embodiments, obtaining the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model includes: if it is determined that the observed attitude is unavailable, then using the predicted attitude obtained through the on-orbit satellite attitude extrapolation model as the extrapolated satellite attitude.

[0010] In some embodiments of this application, when it is confirmed that the observed attitude obtained through the star sensor is unavailable, the predicted value is used as the current attitude data to adjust and control the satellite attitude, thereby improving the versatility of the technical solution.

[0011] In some embodiments, obtaining the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model includes: if it is determined that the observed attitude is available, then using the observed attitude as the extrapolated satellite attitude; obtaining the difference between the observed attitude and the predicted attitude obtained using the on-orbit satellite attitude extrapolation model to obtain the prediction error; and confirming that the prediction error is greater than a threshold, then continuing on-orbit training of the on-orbit satellite attitude extrapolation model to correct the correlation coefficient.

[0012] In some embodiments of this application, when the observation value is confirmed to be available, the model will be trained on-orbit based on the observation value and the predicted attitude obtained by the current model. This can be applied to the impact of environmental changes on attitude prediction and effectively avoid the technical defect of attitude deviation accumulation caused by continuous uncorrected extrapolated attitude.

[0013] In some embodiments, the on-orbit satellite attitude extrapolation model is a model trained through model initialization.

[0014] Some embodiments of this application divide the training process of the satellite attitude acquisition model into two stages: initialization training and on-orbit operation training. This can effectively improve the model's adaptability to the environment and enhance the accuracy of the predicted attitude.

[0015] In some embodiments, the method further includes: collecting on-orbit data to obtain a training dataset; if it is determined that the prediction error is less than a set threshold, then continuing to perform linear regression training on the satellite attitude acquisition model until the on-orbit satellite attitude extrapolation model is obtained.

[0016] Some embodiments of this application provide a method for pre-training a model, which can be performed using a small amount of data.

[0017] In some embodiments, the environmental disturbance parameter values ​​include at least one of the following: environmental magnetic field value, solar radiation value, and atmospheric pressure value.

[0018] In some embodiments, the on-orbit satellite attitude extrapolation model includes: a regression attitude extrapolation module configured to obtain the predicted attitude based on all coefficients in the dependent function, the current attitude, the current angular velocity, the flywheel speed change, and the environmental disturbance parameter value; and a regression prediction module configured to provide all the coefficients to the regression attitude extrapolation module, wherein all the coefficients are obtained by training the regression prediction model on-orbit using the difference between historical prediction errors and historical observed attitudes.

[0019] Some embodiments of this application adjust the parameters of the model using effective observation data, thereby improving the accuracy of the predicted attitude obtained from the model and effectively avoiding the accumulation of attitude errors.

[0020] Secondly, some embodiments of this application provide an apparatus for predicting satellite attitude in orbit. The apparatus includes: an input module configured to input the current attitude, current angular velocity, flywheel speed change, environmental disturbance parameter values, and observed attitude into an on-orbit satellite attitude extrapolation model, wherein the on-orbit satellite attitude extrapolation model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and then perform on-orbit satellite attitude prediction; a prediction module configured to obtain the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model; and an output module configured to provide the extrapolated satellite attitude to an attitude control system to achieve attitude control adjustment of the satellite.

[0021] Thirdly, some embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the method described in any embodiment of the first aspect.

[0022] Fourthly, some embodiments of this application provide an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can implement the method as described in any embodiment of the first aspect. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 An architecture diagram of a system for predicting satellite attitude for related technologies;

[0025] Figure 2 This is one of the architecture diagrams of an on-orbit satellite attitude prediction system provided in an embodiment of this application;

[0026] Figure 3 One of the flowcharts for an on-orbit satellite attitude prediction method provided in the embodiments of this application;

[0027] Figure 4 A second architectural diagram of a system for predicting satellite attitude in orbit, provided in an embodiment of this application;

[0028] Figure 5 A second flowchart illustrating the method for predicting satellite attitude in orbit provided in this application embodiment;

[0029] Figure 6 A block diagram illustrating the composition of an on-orbit satellite attitude prediction device provided in an embodiment of this application;

[0030] Figure 7 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0031] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0032] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0033] To improve the accuracy of attitude prediction, some embodiments of this application establish a regression task model for attitude extrapolation (i.e., an on-orbit satellite attitude extrapolation model). This model combines the actual on-orbit operating conditions of the satellite with current attitude, current angular velocity, flywheel speed change, ambient magnetic field strength, solar illumination, and atmospheric pressure as input attitude-influencing variables. A linear regression machine learning algorithm is used to fit the satellite attitude dependent function, thereby achieving on-orbit satellite attitude prediction. Furthermore, this on-orbit satellite attitude extrapolation model possesses continuous on-orbit learning capabilities and can be adapted to the impact of environmental changes on attitude prediction.

[0034] Please refer to Figure 1 , Figure 1 A system for predicting satellite attitude is provided for related technologies. This system can implement a combined attitude determination method based on Kalman filters, which is based on a combination of star-sensor and gyroscope attitude determination using Kalman filters. Figure 1In the architecture: angular velocity measured by gyroscope, and satellite attitude extrapolated from the previous period attitude q. k When the star sensor measures the observed attitude q H When unavailable, the satellite attitude is extrapolated; when the star sensor measurement is available, the star-sensor-based observation attitude q is used. H The attitude deviation is calculated by comparing the attitude extrapolation prediction with the attitude deviation, and the attitude deviation is used as a state variable to correct the attitude and gyroscope drift.

[0035] It is easy to understand that the technical problem with this architecture is that it cannot react to changes in the environment, and the accuracy of posture prediction drops significantly when the environment changes.

[0036] and Figure 1 The difference is that some embodiments of this application have systems for predicting satellite attitude in orbit, such as... Figure 2 As shown, the input parameters of this system include environmental disturbance parameters (e.g., Figure 2 The environmental magnetic field strength, solar radiation intensity, atmospheric pressure), wheel speed change, current angular velocity measured by the gyroscope, and attitude measured by the star sensor are all considered. It is understandable that due to the input... Figure 2 The on-orbit satellite attitude extrapolation model has a richer variety of input parameters, therefore compared to Figure 1 Traditional gyroscope integral extrapolation algorithms, in some embodiments of this application, incorporate factors affecting satellite attitude, such as changes in flywheel speed and environmental disturbances. This makes them applicable not only to attitude extrapolation when the satellite is in a steady state but also to attitude extrapolation during attitude adjustment. The method in some embodiments of this application can adapt to interference with satellite attitude caused by environmental changes (environmental magnetic disturbances, sunlight, atmospheric drag), thereby making attitude extrapolation more accurate.

[0037] The following combination Figure 3 This application provides an exemplary method for predicting satellite attitude in orbit, based on some embodiments thereof.

[0038] like Figure 3 As shown in the figure, an embodiment of this application provides a method for predicting satellite attitude in orbit, the method comprising:

[0039] S101 inputs the current attitude, current angular velocity, flywheel speed change, environmental disturbance parameter values, and observed attitude into the on-orbit satellite attitude extrapolation model.

[0040] It should be noted that the on-orbit satellite attitude extrapolation model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and then perform on-orbit satellite attitude prediction to obtain the predicted attitude.

[0041] S102, the extrapolated satellite attitude is obtained through the on-orbit satellite attitude extrapolation model.

[0042] S103, the extrapolated satellite attitude is provided to the attitude control system to achieve attitude control adjustment of the satellite.

[0043] Some embodiments of this application introduce parameters related to environmental changes (i.e., environmental disturbance parameters, such as environmental magnetic disturbance, sunlight, and atmospheric drag), thereby enabling the corresponding model to adapt to the interference on satellite attitude caused by environmental changes (environmental magnetic disturbance, sunlight, and atmospheric drag), thus making attitude extrapolation more accurate.

[0044] The implementation process of the above steps is illustrated below.

[0045] In some embodiments of this application, the environmental interference parameter value mentioned in S101 includes at least one of: ambient magnetic field value, solar radiation value, and atmospheric pressure value. For example, in some embodiments of this application, the ambient magnetic field value, solar radiation value, and atmospheric pressure value are all used as input parameters; in some embodiments of this application, the solar radiation value and atmospheric pressure value are used as input parameters; and in some embodiments of this application, the ambient magnetic field value and solar radiation value are used as input parameters.

[0046] like Figure 4 As shown, with Figure 2 The difference is that this figure presents an architecture for an on-orbit satellite attitude extrapolation model. Figure 4 The embodiments shown in this application include an on-orbit satellite attitude extrapolation model comprising: a regression attitude extrapolation module configured to obtain the current predicted attitude based on all coefficients in the dependent function, the current attitude, the current angular acceleration, the flywheel speed change, and environmental disturbance parameters; and a regression prediction module configured to provide the regression attitude extrapolation module with all coefficients required for the current calculation, wherein all coefficients are obtained by on-orbit training of the corresponding regression prediction model using the difference between historical prediction errors and historical observed attitudes.

[0047] In other words, the on-orbit satellite attitude extrapolation model of some embodiments of this application obtains the predicted attitude q using the following formula. k+1 :

[0048] q k+1 =f(q) k ω k ΔN k B k S k , P k )

[0049] Where f is obtained by fitting using a linear regression algorithm, q k Representing the current attitude, ω k Characterizing the current angular acceleration, ΔN k B represents the change in flywheel speed. kCharacterizing environmental magnetic field value, s k Characterizing solar irradiance and P k It represents atmospheric pressure.

[0050] Some embodiments of this application employ regression prediction methods to fit the coefficients of the function in the model (from...). Figure 4 These coefficients are derived from the regression prediction model. For example, in some embodiments of this application, the on-orbit satellite attitude extrapolation model is the RP-EKF (Regression Predict EKF) attitude extrapolation model. The inputs of this model include: the current attitude q k Angular acceleration ω measured by gyroscope k Flywheel speed change ΔN k The output of this model is the predicted attitude value q extrapolated from the attitude. k+1 .

[0051] The following combination Figure 5 The process of S102 obtaining the predicted attitude through on-orbit training is illustrated by way of example.

[0052] The workflow of the linear regression prediction algorithm for the on-orbit satellite attitude extrapolation model in some embodiments of this application includes: on-orbit attitude extrapolation is divided into two stages: model initialization and on-orbit operation, and the workflow is as follows: Figure 5 As shown. For example, if the on-orbit satellite attitude extrapolation model is the RP-EKF model, only 300 data points are needed to complete the initial training, and the prediction accuracy is better than that of traditional attitude extrapolation algorithms. After the model training is completed, the iterative cycle of attitude extrapolation and model training begins (i.e., the on-orbit training phase of the on-orbit satellite attitude extrapolation model). This phase includes: if the observed attitude is unavailable, the satellite attitude is the predicted attitude; if the observed attitude is available, the satellite attitude is the observed attitude; if the prediction error is greater than the threshold, the model is trained on-orbit using this set of data to correct the correlation coefficient.

[0053] Some embodiments of this application divide the training process of the satellite attitude acquisition model into two stages: initialization training and on-orbit operation training. This can effectively improve the model's adaptability to the environment and enhance the accuracy of the predicted attitude.

[0054] For example, in some embodiments of this application, the on-orbit satellite attitude extrapolation model is a model trained through model initialization.

[0055] like Figure 5 As shown, in some embodiments of this application, the linear regression training (i.e., initial training) process includes, for example,:

[0056] The first step is on-orbit data acquisition, which involves collecting on-orbit data to obtain the training dataset.

[0057] The second step is data preprocessing.

[0058] The third step is to determine the relationship between the prediction error and the threshold.

[0059] If the prediction error is determined to be greater than or equal to the set threshold, then the satellite attitude acquisition model continues to be trained by linear regression (i.e., initial training) until the on-orbit satellite attitude extrapolation model is obtained.

[0060] If the prediction error is determined to be less than the set threshold, the initial training phase is stopped, and an in-orbit satellite attitude extrapolation model is obtained. Then, based on this model, subsequent linear regression predictions are performed, and in-orbit training is conducted in combination with the prediction results to correct the parameters of the model.

[0061] It should be noted that training the model during the on-orbit operation phase requires combining observation of the attitude.

[0062] For example, in some embodiments of this application, S102 includes, for instance, if it is determined that the observed attitude is unavailable, then the predicted attitude obtained through the on-orbit satellite attitude extrapolation model is used as the extrapolated satellite attitude. In other words, some embodiments of this application, when confirming that the observed attitude obtained through the star sensor is unavailable, use the value of the predicted attitude as the current attitude data to achieve adjustment and control of the satellite attitude, thereby improving the versatility of the technical solution.

[0063] For example, in some embodiments of this application, step S102 includes: if it is determined that the observed attitude is available, then the observed attitude is used as the extrapolated satellite attitude; the difference between the observed attitude and the predicted attitude obtained by using the on-orbit satellite attitude extrapolation model is obtained to obtain the prediction error; if it is confirmed that the prediction error is greater than a threshold, then the on-orbit satellite attitude extrapolation model is further trained on-orbit to correct the correlation coefficient.

[0064] For example, such as Figure 5 As shown in some embodiments of this application, the process of obtaining extrapolated attitude values ​​during the on-orbit operation phase includes:

[0065] The first step is to initiate linear regression prediction.

[0066] Soon Figure 4 All input parameters are input into the on-orbit satellite attitude extrapolation model obtained through linear regression training, and the predicted attitude for each iteration is obtained by this module.

[0067] The second step is to determine whether the observation attitude is available. If not, proceed to the fourth step; otherwise, proceed to the third step.

[0068] Third, if the observed attitude is available, proceed to step eight.

[0069] The fourth step is to use the observed attitude as the satellite attitude.

[0070] The fifth step is to calculate the predicted attitude deviation, which is the difference between the observed attitude and the predicted attitude. This is also known as the prediction error.

[0071] Step 6: Determine if the prediction error is greater than the threshold. If not, proceed to step 8; otherwise, proceed to step 7.

[0072] The seventh step is to train the on-orbit satellite attitude extrapolation model on-orbit until the training termination condition is met, so as to obtain the on-orbit satellite attitude extrapolation model used for the next attitude prediction. Finally, the observed attitude is output as the satellite attitude.

[0073] The eighth step is to directly use the predicted attitude obtained this time as the satellite attitude.

[0074] In other words, some embodiments of this application, when confirming the availability of observations, will train the model on-orbit based on the observations and the predicted attitude obtained by the current model. This can be applied to the impact of environmental changes on attitude prediction and effectively avoid the technical defect of attitude deviation accumulation caused by continuous uncorrected extrapolated attitude.

[0075] It is easy to understand that the on-orbit satellite attitude extrapolation model of some embodiments of this application has strong self-learning ability. The model is continuously trained and updated based on attitude measurement data, magnetometer measurement data, barometer measurement data, and solar sensor measurement data. It learns quickly and requires a small number of samples.

[0076] Some embodiments of this application utilize linear regression technology for satellite attitude extrapolation in orbit, enabling the satellite to possess self-learning and adaptive capabilities with high prediction accuracy. The models in some embodiments of this application incorporate attitude-influencing factors such as angular velocity, wheel speed changes, ambient magnetic field strength, sunlight intensity, and atmospheric pressure, making the attitude extrapolation capability adaptable to environmental variations.

[0077] Please refer to Figure 6 , Figure 6 The present application illustrates an apparatus for predicting satellite attitude in orbit, as provided in an embodiment. It should be understood that this apparatus is similar to the one described above. Figure 3 Corresponding to the method embodiments, it can execute the various steps involved in the above method embodiments. The specific functions of the device can be found in the description above. To avoid repetition, detailed descriptions are appropriately omitted here. The device includes at least one software function module that can be stored in a memory or embedded in the device's operating system in the form of software or firmware. The on-orbit satellite attitude prediction device includes: an input module 101, a prediction module 102, and an output module 103.

[0078] The input module is configured to input the current attitude, current angular velocity, flywheel speed change, environmental disturbance parameter values, and observed attitude into the on-orbit satellite attitude extrapolation model. The on-orbit satellite attitude extrapolation model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and then perform on-orbit satellite attitude prediction.

[0079] The prediction module is configured to obtain the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model.

[0080] The output module is configured to provide the extrapolated satellite attitude to the attitude control system in order to achieve attitude control adjustment of the satellite.

[0081] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the aforementioned method, and will not be elaborated further here.

[0082] Some embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the method described in any of the embodiments included in the above-described method for predicting satellite attitude.

[0083] like Figure 7 As shown, some embodiments of this application provide an electronic device 700, including a memory 710, a processor 720, and a computer program stored in the memory 710 and executable on the processor 720, wherein the processor 720 can implement the method described in any of the embodiments of the above-described method for predicting satellite attitude via a bus 730 when it reads and executes the program.

[0084] Processor 720 can process digital signals and can include various computing architectures. For example, it can be a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements multiple instruction set combinations. In some examples, processor 720 can be a microprocessor.

[0085] The memory 710 can be used to store instructions executed by the processor 720 or data related to the execution of instructions. These instructions and / or data may include code used to implement some or all of the functions of one or more modules described in the embodiments of this application. The processor 720 of the embodiments of this disclosure can be used to execute the instructions in the memory 710 to implement… Figure 3 The method shown. Memory 710 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memory well known to those skilled in the art.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0087] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0088] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0089] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0090] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0091] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for predicting satellite attitude in orbit, characterized in that, The method includes: The current attitude, current angular acceleration, flywheel speed change, environmental disturbance parameters, and observed attitude are input into the on-orbit satellite attitude extrapolation model. This model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and thus predict the satellite attitude in orbit. The training process for the satellite attitude acquisition model is divided into two stages: initialization training and on-orbit operational training. The on-orbit satellite attitude extrapolation model includes: The regression attitude extrapolation module is configured to obtain the predicted attitude based on all coefficients in the dependent function, the current attitude, the current angular acceleration, the flywheel speed change, and the environmental disturbance parameter value. The regression prediction module is configured to provide all the coefficients to the regression attitude extrapolation module, wherein all the coefficients are obtained by training the regression prediction module on-orbit using the prediction error calculated in the previous iteration; The extrapolated satellite attitude is obtained through the on-orbit satellite attitude extrapolation model; The extrapolated satellite attitude is provided to the attitude control system to achieve attitude control and adjustment of the satellite; The process of obtaining the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model includes: If the observed attitude is determined to be unavailable, the predicted attitude obtained through the on-orbit satellite attitude extrapolation model will be used as the extrapolated satellite attitude. If the observation attitude is determined to be available, then the observation attitude is used as the extrapolated satellite attitude.

2. The method as described in claim 1, characterized in that, The process of obtaining the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model includes: The difference between the observed attitude and the predicted attitude obtained using the on-orbit satellite attitude extrapolation model is used to obtain the prediction error. If the prediction error is confirmed to be greater than the threshold, the on-orbit satellite attitude extrapolation model will continue to be trained in orbit to correct the correlation coefficient.

3. The method as described in claim 2, characterized in that, The on-orbit satellite attitude extrapolation model is a model trained through model initialization.

4. The method as described in claim 3, characterized in that, The method further includes: Collect on-orbit data to obtain the training dataset; If the prediction error is determined to be greater than or equal to the set threshold, then the satellite attitude acquisition model continues to be trained by linear regression until the on-orbit satellite attitude acquisition model is obtained.

5. The method as described in claim 1, characterized in that, The environmental disturbance parameters include at least one of the following: environmental magnetic field value, solar radiation value, and atmospheric pressure value.

6. A device for predicting satellite attitude in orbit, characterized in that, The device includes: The input module is configured to input the current attitude, current angular acceleration, flywheel speed change, environmental disturbance parameters, and observed attitude into the on-orbit satellite attitude extrapolation model. The on-orbit satellite attitude extrapolation model is configured to use a linear regression machine learning algorithm to fit the satellite attitude dependent function and then perform on-orbit satellite attitude prediction. The on-orbit satellite attitude extrapolation model includes: The regression attitude extrapolation module is configured to obtain the predicted attitude based on all coefficients in the dependent function, the current attitude, the current angular acceleration, the flywheel speed change, and the environmental disturbance parameter value. The regression prediction module is configured to provide all the coefficients to the regression attitude extrapolation module, wherein all the coefficients are obtained by training the regression prediction module on-orbit using the prediction error calculated in the previous iteration; The prediction module, configured to obtain the extrapolated satellite attitude through the on-orbit satellite attitude extrapolation model, includes: If the observed attitude is determined to be unavailable, the predicted attitude obtained through the on-orbit satellite attitude extrapolation model will be used as the extrapolated satellite attitude. If the observation attitude is determined to be available, then the observation attitude is used as the extrapolated satellite attitude; The output module is configured to provide the extrapolated satellite attitude to the attitude control system in order to achieve attitude control adjustment of the satellite.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it can implement the method described in any one of claims 1-5.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, When the processor executes the program, it can implement the method described in any one of claims 1-5.

Images