Low earth orbit satellite orbit determination method, orbit determination equipment, readable storage medium and product

Abstract

The invention provides a low earth orbit satellite orbit determination method, orbit determination equipment, a readable storage medium and a product, and relates to the technical field of communication, the method comprises the following steps: collecting multi-source data, the multi-source data comprising low earth orbit satellite self-loading data, real-time observation data collected by a ground continuous operation reference station (CORS) and environmental data collected by an auxiliary sensor; carrying out multi-source data fusion on the low-orbit satellite self-loading data, the real-time observation data and the environment data to obtain comprehensive positioning data; inputting the comprehensive positioning data into a low-orbit satellite state space model containing position, speed and acceleration, and performing state estimation on the comprehensive positioning data through the low-orbit satellite state space model to obtain orbit state parameters of the low-orbit satellite, the orbit state parameters comprise at least one of the position, the speed and the acceleration of the low-orbit satellite; according to the method, the orbit determination precision of the low-orbit satellite can be improved, errors are reduced, and the robustness and real-time performance of the system are improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method, equipment, readable storage medium, and product for determining the orbit of a low-Earth orbit satellite. Background Technology

[0002] Currently, the main technologies used for orbit determination of low-Earth orbit satellites are as follows:

[0003] Satellite-borne Global Navigation Satellite System (GNSS) positioning: Low-Earth orbit satellites carry GNSS receivers. By receiving signals from navigation satellites such as the Global Positioning System (GPS), BeiDou, and GLONASS, they obtain their own position and velocity information using point positioning, pseudorange measurement, and carrier phase observation. This method can provide continuous data, but it relies solely on the satellite's onboard GNSS data and is limited in positioning accuracy by factors such as atmospheric delay, ionospheric refraction, and multipath effects.

[0004] Continuously Operating Reference Stations (CORS) differential positioning technology utilizes a nationwide or global network of CORS stations to collect high-precision GNSS observation data and performs real-time differential correction using known static baselines. This technology has been widely applied in geodesy and static ground positioning. However, when directly applied to low-Earth orbit (LEO) satellite dynamic orbit determination, it fails to consider the high-speed motion characteristics of LEO satellites, often resulting in difficulties meeting high-precision requirements due to data reporting delays, inaccurate time synchronization, and a lack of multi-source data fusion mechanisms.

[0005] Traditional filtering algorithms: The Extended Kalman Filter (EKF) is commonly used. It achieves state estimation and orbit prediction by establishing a satellite motion state space model. However, due to fixed parameters, large model errors, and a lack of adaptive capability, this method performs poorly in the complex and rapidly changing dynamic environment of low-Earth orbit satellites.

[0006] Existing solutions often focus on a single data source, such as relying solely on satellite-borne data or data from a single CORS station, failing to fully leverage the complementary advantages of multi-source data in error elimination and real-time dynamic correction. Furthermore, current solutions largely remain at the level of fixed parameters or static models, which cannot meet the orbit determination requirements of low-Earth orbit satellites. Summary of the Invention

[0007] The purpose of this application is to provide a method, device, readable storage medium, and product for determining the orbit of a low-Earth orbit satellite, in order to solve the problem that current satellite orbit determination schemes cannot meet the orbit determination requirements of low-Earth orbit satellites.

[0008] To address the aforementioned problems, this application provides a method for determining the orbit of a low-Earth orbit satellite, the method comprising:

[0009] Collect multi-source data, including: low-orbit satellite onboard data, real-time observation data collected by the continuously operating ground reference station CORS, and environmental data collected by auxiliary sensors;

[0010] Multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data;

[0011] The integrated positioning data is input into a low-Earth orbit (LEO) satellite state space model that includes position, velocity, and acceleration. The LEO satellite state space model is used to estimate the state of the integrated positioning data to obtain the LEO satellite's orbital state parameters. The orbital state parameters include at least one of the LEO satellite's position, velocity, and acceleration.

[0012] In some embodiments, the orbital state parameters of the low-Earth orbit satellite are obtained by performing state estimation on the integrated positioning data using the low-Earth orbit satellite state space model, including:

[0013] The state of the integrated positioning data is estimated using the target filtering method through the state space model of the low-Earth orbit satellite, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite.

[0014] The target filtering method includes at least one of the following: extended Kalman filtering; unscented Kalman filtering; particle filtering.

[0015] In some embodiments, the method further includes:

[0016] Based on the dynamic error model, obtain the real-time observation error;

[0017] Based on the real-time observation error, the filtering parameters of the target filtering method are dynamically adjusted, and the filtering parameters include: filtering gain and / or covariance matrix.

[0018] In some embodiments, before fusing the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data, the method further includes:

[0019] The time of the low-orbit satellite's onboard data, the real-time observation data, and the environmental data are calibrated using an atomic clock or a GPS clock.

[0020] Based on the observation data and the environmental data, the low-Earth orbit satellite self-load data is corrected to obtain the corrected low-Earth orbit satellite self-load data.

[0021] In some embodiments, the low-Earth orbit satellite self-loaded data is corrected based on the observation data and the environmental data to obtain corrected low-Earth orbit satellite self-loaded data, including:

[0022] The real-time observation data and the low-Earth orbit satellite's onboard data are input into a real-time differential correction model. The real-time differential correction model calculates the errors of atmospheric delay and multipath effects, and performs differential correction on the low-Earth orbit satellite's onboard data based on the errors of atmospheric delay and multipath effects to obtain the first data. The real-time differential correction model is constructed based on the baseline information of ground CORS and historical observation data collected by ground CORS.

[0023] Based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-orbit satellite self-loaded data.

[0024] In some embodiments, after collecting multi-source data, the method further includes:

[0025] The multi-source data is cleaned, outliers are removed, and normalization is performed.

[0026] In some embodiments, multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data, including:

[0027] The low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data are weighted and fused to obtain the comprehensive positioning data based on the real-time quality indicators of the low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data.

[0028] In some embodiments, the method further includes:

[0029] The dynamic error model is established based on historical information of the orbital state parameters of the low-orbit satellite and real-time environmental data collected by the auxiliary sensors.

[0030] In some embodiments, the method further includes:

[0031] The orbital state parameters of the low-orbit satellite are fed back to the satellite control center. These orbital state parameters are used to assist the satellite control center in adjusting the multi-source data fusion strategy and the filtering parameters of the target filtering method.

[0032] This application embodiment also provides an orbit determination device, including a processor and a transceiver. The transceiver receives and transmits data under the control of the processor, and the processor is used to perform the following operations:

[0033] Collect multi-source data, including: low-orbit satellite onboard data, real-time observation data collected by the continuously operating ground reference station CORS, and environmental data collected by auxiliary sensors;

[0034] Multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data;

[0035] The integrated positioning data is input into a low-Earth orbit (LEO) satellite state space model that includes position, velocity, and acceleration. The LEO satellite state space model is used to estimate the state of the integrated positioning data to obtain the LEO satellite's orbital state parameters. The orbital state parameters include at least one of the LEO satellite's position, velocity, and acceleration.

[0036] In some embodiments, the processor is further configured to perform the following operations:

[0037] The state of the integrated positioning data is estimated using the target filtering method through the state space model of the low-Earth orbit satellite, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite.

[0038] The target filtering method includes at least one of the following: extended Kalman filtering; unscented Kalman filtering; particle filtering.

[0039] In some embodiments, the processor is further configured to perform the following operations:

[0040] Based on the dynamic error model, obtain the real-time observation error;

[0041] Based on the real-time observation error, the filtering parameters of the target filtering method are dynamically adjusted, and the filtering parameters include: filtering gain and / or covariance matrix.

[0042] In some embodiments, the processor is further configured to perform the following operations:

[0043] The time of the low-orbit satellite's onboard data, the real-time observation data, and the environmental data are calibrated using an atomic clock or a GPS clock.

[0044] Based on the observation data and the environmental data, the low-Earth orbit satellite self-load data is corrected to obtain the corrected low-Earth orbit satellite self-load data.

[0045] In some embodiments, the processor is further configured to perform the following operations:

[0046] The real-time observation data and the low-Earth orbit satellite's onboard data are input into a real-time differential correction model. The real-time differential correction model calculates the errors of atmospheric delay and multipath effects, and performs differential correction on the low-Earth orbit satellite's onboard data based on the errors of atmospheric delay and multipath effects to obtain the first data. The real-time differential correction model is constructed based on the baseline information of ground CORS and historical observation data collected by ground CORS.

[0047] Based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-orbit satellite self-loaded data.

[0048] In some embodiments, the processor is further configured to perform the following operations:

[0049] The multi-source data is cleaned, outliers are removed, and normalization is performed.

[0050] In some embodiments, the processor is further configured to perform the following operations:

[0051] The low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data are weighted and fused to obtain the comprehensive positioning data based on the real-time quality indicators of the low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data.

[0052] In some embodiments, the processor is further configured to perform the following operations:

[0053] The dynamic error model is established based on historical information of the orbital state parameters of the low-orbit satellite and real-time environmental data collected by the auxiliary sensors.

[0054] In some embodiments, the processor is further configured to perform the following operations:

[0055] The orbital state parameters of the low-orbit satellite are fed back to the satellite control center. These orbital state parameters are used to assist the satellite control center in adjusting the multi-source data fusion strategy and the filtering parameters of the target filtering method.

[0056] This application also provides an orbit determination device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the low-Earth orbit satellite orbit determination method as described above.

[0057] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps in the low-Earth orbit satellite orbit determination method described above.

[0058] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the low-Earth orbit satellite orbit determination method described above.

[0059] The above-mentioned technical solution of this application has at least the following beneficial effects:

[0060] In the low-Earth orbit satellite orbit determination method, orbit determination equipment, readable storage medium, and product of this application, on the one hand, multi-source data fusion is performed using low-Earth orbit satellite onboard data, real-time observation data collected by ground CORS, and environmental data collected by auxiliary sensors to obtain comprehensive positioning data. This method fully utilizes multi-source data to achieve complementary advantages between different types of data. On the other hand, the state of the above-mentioned comprehensive positioning data is estimated through the low-Earth orbit satellite state space model to obtain the orbital state parameters of the low-Earth orbit satellite, thereby achieving high-precision real-time estimation of the satellite orbit, significantly improving orbit determination accuracy, reducing errors, and enhancing system robustness and real-time performance. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the steps of the low-Earth orbit satellite orbit determination method provided in the embodiments of this application;

[0062] Figure 2 This is a schematic diagram of the structure of an application system for the orbit determination method for orbit determination satellites provided in the embodiments of this application;

[0063] Figure 3 A flowchart illustrating an application system for the positioning satellite orbit determination method provided in this application embodiment;

[0064] Figure 4 This is a schematic diagram showing the structure of the orbit determination device provided in the embodiments of this application. Detailed Implementation

[0065] To make the technical problems, technical solutions and advantages of this application clearer, a detailed description will be provided below in conjunction with the accompanying drawings and specific embodiments.

[0066] like Figure 1As shown in the embodiment of this application, a method for determining the orbit of a low-Earth orbit (LEO) satellite is provided. This method is applicable to precise orbit determination, orbit maintenance, and dynamic correction of LEO satellites in fields such as communication, remote sensing, navigation, and space science experiments. The method includes:

[0067] Step 101: Collect multi-source data, including: low-orbit satellite self-loaded data, real-time observation data collected by the ground continuously operating reference station CORS, and environmental data collected by auxiliary sensors;

[0068] In this step, the low-Earth orbit satellite carries a high-precision GNSS receiver to continuously collect information such as pseudorange, carrier phase, and signal strength. The instrument integrates a high-precision clock, ensuring a time accuracy of 10ns and a daily average accuracy of <5E⁻¹³, thus guaranteeing accurate sampling time.

[0069] In this step, CORS stations located in different geographical locations capture observation data of low-orbit satellites in real time through high-precision GNSS receivers, and a distributed acquisition system is used to ensure data continuity.

[0070] In this step, auxiliary sensors such as atmospheric sounders and meteorological satellites provide additional environmental data to help build atmospheric delay models and improve the accuracy of the algorithm.

[0071] Optionally, all of the above data acquisition devices use high-precision measuring instruments and are equipped with atomic clocks or GPS clocks for high-precision time synchronization to ensure the consistency of data timestamps.

[0072] It should be noted that, in addition to traditional satellite-borne and CORS station data, the embodiments of this application can also introduce data sources such as satellite laser ranging, ground mobile stations, meteorological satellites and UAV observations to form multimodal data fusion, thereby achieving optimal orbit determination in different environments.

[0073] Step 102: Perform multi-source data fusion on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data;

[0074] Optionally, this step introduces an adaptive multi-source data fusion strategy or algorithm to weight and fuse data from satellites, CORS stations, and auxiliary sensors according to real-time quality indicators to form comprehensive positioning data.

[0075] In one implementation, machine learning algorithms (such as deep learning algorithms or reinforcement learning algorithms) are used to dynamically evaluate the real-time quality indicators of each data in the multi-source data, and the weights of each data are dynamically adjusted according to the real-time quality indicators of each data. Thus, the data are weighted and fused according to their respective weights to obtain the aforementioned comprehensive positioning data.

[0076] Step 103: Input the integrated positioning data into a low-Earth orbit satellite state space model that includes position, velocity, and acceleration. Perform state estimation on the integrated positioning data through the low-Earth orbit satellite state space model to obtain the orbital state parameters of the low-Earth orbit satellite. The orbital state parameters include at least one of the low-Earth orbit satellite's position, velocity, and acceleration.

[0077] In this step, a state space model of a low-Earth orbit (LEO) satellite is established, including state variables such as position, velocity, and acceleration. This LEO satellite state space model is used to estimate the state of the fused integrated positioning data, and the orbital state parameters of the LEO satellite are output in real time. Based on these orbital state parameters, the operational orbit information of the LEO satellite can be determined. For example, the operational orbit information of the LEO satellite can be determined based on its position, velocity, and acceleration, thereby achieving accurate orbit determination of the LEO satellite.

[0078] Optionally, step 103 includes:

[0079] The state of the integrated positioning data is estimated using the target filtering method through the state space model of the low-Earth orbit satellite, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite.

[0080] The target filtering method includes at least one of the following: extended Kalman filtering; unscented Kalman filtering; particle filtering.

[0081] In one implementation, the extended Kalman filter mentioned above is specifically an adaptive extended Kalman filter (AEKF).

[0082] The embodiments of this application may also employ unscented Kalman filtering (UKF), particle filtering (PF), adaptive extended Kalman filtering (AEKF), or hybrid filtering (hybrid filtering can be understood as a combination of at least two of the above extended Kalman filtering, unscented Kalman filtering, and particle filtering) to perform state estimation on the integrated positioning data, thereby achieving more robust state estimation in nonlinear or non-Gaussian noise environments.

[0083] In addition to extended Kalman filtering, this application embodiment can also employ unscented Kalman filtering (UKF), particle filtering (PF), or hybrid filtering strategies to provide more robust state estimation in nonlinear and non-Gaussian noise environments, thus avoiding the limitations of a single filtering algorithm.

[0084] In the data processing of this application embodiment, an extended Kalman filter, unscented Kalman filter (UKF), particle filter (PF), or hybrid filter strategy are combined with a parallel automatic selection mechanism to take the state estimation result with the smallest error as the orbital state parameter of the low-Earth orbit satellite, so as to adapt to the state estimation requirements under nonlinear and non-Gaussian noise environment.

[0085] In this embodiment, on the one hand, multi-source data fusion is performed using data carried by the low-orbit satellite, real-time observation data collected by ground CORS, and environmental data collected by auxiliary sensors to obtain comprehensive positioning data. This method makes full use of multi-source data to achieve complementary advantages between different types of data. On the other hand, the state of the above comprehensive positioning data is estimated by using an extended Kalman filter through the low-orbit satellite state space model, thereby obtaining the orbital state parameters of the low-orbit satellite. This achieves high-precision real-time estimation of the satellite orbit, thereby significantly improving orbit determination accuracy, reducing errors, and enhancing system robustness and real-time performance.

[0086] This low-Earth orbit (LEO) satellite orbit determination method is based on multi-source data fusion and real-time differential correction to enhance LEO satellite orbit determination accuracy. It comprehensively solves key technical problems in existing LEO satellite orbit determination methods, such as insufficient data timeliness, large differential correction errors, inadequate multi-source data fusion, and poor filter adaptability. Through a series of techniques including high-precision clock synchronization, real-time data acquisition, differential correction, multi-frequency joint correction, adaptive extended Kalman filtering, and intelligent closed-loop feedback, it achieves high-precision real-time estimation of satellite orbit status, significantly improving the orbit determination accuracy and system robustness of LEO satellites. This can substantially reduce satellite operation and maintenance costs and increase satellite mission success rates.

[0087] In some embodiments of this application, the method further includes:

[0088] Based on the dynamic error model, obtain the real-time observation error;

[0089] Based on the real-time observation error, the filtering parameters of the target filtering method are dynamically adjusted, and the filtering parameters include: filtering gain and / or covariance matrix.

[0090] Optionally, the method further includes:

[0091] The method further includes:

[0092] Based on historical information about the orbital state parameters of low-Earth orbit satellites and real-time environmental data collected by the auxiliary sensors, a dynamic error model is established. This dynamic error model comprehensively considers the influence of multiple external factors such as the atmosphere, ionosphere, multipath propagation, and instrument noise. Furthermore, the real-time observation error output by this dynamic error model is used to compensate for the filtering parameters during the filtering process, effectively addressing the problem of insufficient accuracy in state estimation for low-Earth orbit satellites in high-speed, dynamic, and complex environments.

[0093] In some embodiments of this application, prior to step 102, the method further includes:

[0094] Using an atomic clock or a GPS clock, the time of the low-orbit satellite's onboard data, the real-time observation data, and the environmental data are respectively calibrated to achieve microsecond-level alignment of the multi-source data.

[0095] Based on the observation data and the environmental data, the low-Earth orbit satellite self-load data is corrected to obtain the corrected low-Earth orbit satellite self-load data.

[0096] Optionally, embodiments of this application use atomic clocks or GPS clocks for time calibration. Through high-precision NTP protocol and dedicated synchronization module, timestamps are applied at the data link layer (rather than the business application layer) to avoid operating system scheduling delays, achieve microsecond-level alignment of each data source, and ensure that data fusion is free of time delay deviations.

[0097] This application uses atomic clocks or GPS clocks for microsecond-level time synchronization and utilizes multi-channel data transmission and redundant backup technology to ensure high-precision synchronization between different data sources, reducing the impact of data latency and packet loss on differential correction and data fusion.

[0098] Optionally, the low-Earth orbit satellite's onboard data is corrected based on the observation data and the environmental data to obtain corrected low-Earth orbit satellite onboard data, including:

[0099] The real-time observation data and the low-Earth orbit satellite's onboard data are input into a real-time differential correction model. The real-time differential correction model calculates the errors of atmospheric delay and multipath effects, and performs differential correction on the low-Earth orbit satellite's onboard data based on the errors of atmospheric delay and multipath effects to obtain the first data. The real-time differential correction model is constructed based on the baseline information of ground CORS and historical observation data collected by ground CORS.

[0100] Based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-orbit satellite self-loaded data.

[0101] Optionally, the baseline information of the ground CORS includes at least one of the following: geometric parameters such as length or elevation difference, coordinate results, accuracy indicators, and calculation methods.

[0102] This application embodiment utilizes known baseline information and historical observation data between ground CORS to construct a real-time differential correction model capable of calculating atmospheric delay errors and multipath effect errors in real time. Furthermore, it corrects ionospheric delay errors and tropospheric delay errors to obtain corrected low-Earth orbit satellite self-loaded data, thereby ensuring a higher accuracy correction effect.

[0103] In one implementation, this embodiment first utilizes known baseline information between ground CORSs and combines it with historical observation data to establish a real-time differential correction model. This real-time differential correction model calculates the errors of atmospheric delay and multipath effects based on the real-time observation data, and then performs real-time correction on the low-Earth orbit (LEO) satellite-borne data based on these errors to obtain first data. For example, based on the errors of atmospheric delay and multipath effects, a correction amount for the LEO satellite-borne data is determined (this correction amount can be positive or negative), and the LEO satellite-borne data is further summed with the aforementioned correction amount to obtain the first data; this first data can be understood as the LEO satellite-borne data after correcting for atmospheric delay and multipath effects.

[0104] In another implementation, based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-Earth orbit satellite self-borne data, including:

[0105] Based on environmental data from multiple dimensions (such as atmospheric temperature, humidity, air pressure, ionospheric parameters, etc.) and real-time observation data from multiple frequency bands, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-Earth orbit satellite self-loaded data. This enables the correction of ionospheric delay error and tropospheric delay error, ensuring a higher precision differential correction effect for the corrected low-Earth orbit satellite self-loaded data.

[0106] In at least one embodiment of this application, prior to step 102, the method further includes:

[0107] The multi-source data is cleaned, outliers are removed, and normalization is performed.

[0108] Specifically, to address issues such as noise, outliers, and missing data in the original multi-source data, statistical filtering (smoothing high-frequency noise and handling outliers), median filtering (removing impulse noise and isolated outliers), and interpolation methods (ensuring data integrity) are used sequentially to clean the original data and ensure its usability.

[0109] Specifically, for data from different data sources, due to inconsistencies in units, ranges, etc., Z-Score or Min-Max normalization is often used to ensure that all types of data have the same dimensions and scale. This makes data from different indicators comparable and results in a more even data distribution, facilitating subsequent comprehensive analysis.

[0110] In other words, before correcting multi-source data, the multi-source data undergoes cleaning, outlier removal, and normalization to ensure data quality. Furthermore, the low-Earth orbit satellite-borne data, ground-based CORS-acquired observation data, and environmental data acquired by auxiliary sensors mentioned in subsequent steps are all data that have undergone cleaning, outlier removal, and normalization; no specific limitations are imposed here.

[0111] In some embodiments of this application, step 102 includes:

[0112] The low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data are weighted and fused to obtain the comprehensive positioning data based on the real-time quality indicators of the low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data.

[0113] This application introduces an adaptive multi-source data fusion strategy or algorithm to weight and fuse data from satellites, ground CORS, and auxiliary sensors based on real-time quality indicators to form comprehensive positioning data. For example, based on a comprehensive evaluation of various dimensions of indicators such as real-time signal-to-noise ratio, observation accuracy, and data integrity of each data source, the data preprocessing, fusion, and filtering parameters are automatically adjusted, and an adaptive weighted average method is used to fuse the data. Furthermore, this application introduces machine learning algorithms (such as deep learning or reinforcement learning) to dynamically evaluate data quality and automatically adjust the weights of each data source.

[0114] This application embodiment utilizes deep learning or reinforcement learning methods to achieve real-time evaluation of data quality and environmental changes, and automatically adjusts data fusion weights and filtering parameters to further improve system adaptability and form an intelligent closed-loop adjustment system.

[0115] In summary, the embodiments of this application provide an adaptive data fusion algorithm based on machine learning for data fusion of multi-source data. This method can dynamically evaluate data quality, automatically adjust weighting coefficients, and make full use of multi-source input data such as satellite-borne data, ground CORS data, and auxiliary sensors to achieve complementary advantages between different types of data and significantly improve orbit determination accuracy.

[0116] In at least one embodiment of this application, the method further includes:

[0117] The orbital state parameters of the low-orbit satellite are fed back to the satellite control center. These orbital state parameters are used to assist the satellite control center in adjusting the multi-source data fusion strategy and the filtering parameters of the target filtering method.

[0118] In one implementation, the satellite control center compares the feedback orbital state parameters with the actual operating status of the low-orbit satellite, and adjusts the multi-source data fusion strategy and the filtering parameters of the target filtering method based on the comparison results.

[0119] The embodiments of this application adopt a distributed storage and computing architecture to realize real-time data acquisition, correction, fusion and state estimation; and feed back the high-precision orbital state parameters obtained after fusion to the satellite control center in real time, so that the satellite control center can realize dynamic orbit correction and attitude control.

[0120] Specifically, the satellite control center compares the feedback data with the actual operating status to form a closed-loop control. It uses an adaptive fault-tolerant mechanism to automatically adjust the data fusion strategy and filtering parameters to ensure that the system can maintain high-precision orbit determination even in the event of partial data source failure or sudden environmental changes.

[0121] The aforementioned intelligent adaptive adjustment of filtering parameters, hybrid filtering, and closed-loop fault-tolerant mechanism further expand the system's application capabilities in complex environments, achieving precise orbit determination for low-Earth orbit satellites. Based on the technical solution of this application, it can not only be widely used in satellite navigation, remote sensing, and scientific experiments, but also extended to other high-precision positioning fields such as autonomous driving, smart cities, and geodesy, without specific limitations here.

[0122] In one implementation, such as Figure 2 and Figure 3 As shown, the equipment or system capable of implementing the aforementioned low-Earth orbit satellite orbit determination method consists of five layers, arranged in the order of data processing: data acquisition layer, data transmission and synchronization layer, data processing and correction layer, data fusion and estimation layer, and data closed-loop feedback control layer. Among them:

[0123] The data acquisition layer includes a low-Earth orbit satellite-borne GNSS data acquisition unit, a ground CORS station data acquisition unit, and an auxiliary sensor data acquisition unit. Each data acquisition device employs high-precision measuring instruments and is equipped with an atomic clock or GPS clock for high-precision time synchronization, ensuring the consistency of data timestamps.

[0124] The data transmission and synchronization layer transmits data from various data sources to the data processing and correction layer in real time via dedicated communication links (such as fiber optics, satellite relay, and wireless transmission). It employs Network Time Protocol (NTP) and high-precision clock synchronization technology to achieve millisecond-level or even microsecond-level synchronization of multi-source data, eliminating data time deviations.

[0125] The data processing and correction layer cleanses and normalizes data from various data sources to ensure data quality. It also constructs a real-time differential correction model, using baseline data and historical observation data between CORS stations to calculate errors such as atmospheric delay and multipath effects, and applies the corrections to the satellite's onboard data to form a high-precision data stream after differential correction. Furthermore, multi-frequency joint correction further suppresses ionospheric and tropospheric delay errors, improving orbit determination accuracy.

[0126] The data fusion and estimation layer introduces an adaptive multi-source data fusion algorithm, which weights and fuses data from satellites, CORS stations, and auxiliary sensors according to real-time quality indicators to form comprehensive positioning data. An adaptive extended Kalman filter (AEKF) combined with a dynamic error model is then used to perform state estimation on the fused data, outputting high-precision satellite position, velocity, acceleration, and other state parameters in real time.

[0127] The data closed-loop feedback control layer feeds the estimation results back to the satellite control center in real time, enabling closed-loop monitoring and dynamic correction of the orbital state. It also constructs an intelligent fault-tolerance and self-healing mechanism, automatically switching to backup data sources and adjusting fusion weights in the event of data anomalies or partial data source failures, ensuring stable system operation.

[0128] Optionally, a simulation platform can be built in a laboratory environment to conduct simulation experiments on the multi-source data fusion and real-time differential correction algorithm using historical observation data and simulated environmental variables, verifying its performance under different scenarios. Test results show that the trajectory determination error is significantly reduced after processing by this system, improving accuracy by 20%-30% compared to traditional methods.

[0129] It should be noted that the various layers of the equipment or system that can realize the above-mentioned low-Earth orbit satellite orbit determination method can be set on one physical device or multiple physical devices, and no specific limitation is made here.

[0130] like Figure 4 As shown in the illustration, this application embodiment also provides an orbit determination device, including a processor 400 and a transceiver 410. The transceiver 410 receives and transmits data under the control of the processor 400, and the processor 400 is used to perform the following operations:

[0131] Collect multi-source data, including: low-orbit satellite onboard data, real-time observation data collected by the continuously operating ground reference station CORS, and environmental data collected by auxiliary sensors;

[0132] Multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data;

[0133] The integrated positioning data is input into a low-Earth orbit (LEO) satellite state space model that includes position, velocity, and acceleration. The LEO satellite state space model is used to estimate the state of the integrated positioning data to obtain the LEO satellite's orbital state parameters. The orbital state parameters include at least one of the LEO satellite's position, velocity, and acceleration.

[0134] In some embodiments of this application, the processor is also configured to perform the following operations:

[0135] The state of the integrated positioning data is estimated using the target filtering method through the state space model of the low-Earth orbit satellite, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite.

[0136] The target filtering method includes at least one of the following: extended Kalman filtering; unscented Kalman filtering; particle filtering.

[0137] In some embodiments of this application, the processor is also configured to perform the following operations:

[0138] Based on the dynamic error model, obtain the real-time observation error;

[0139] Based on the real-time observation error, the filtering parameters of the target filtering method are dynamically adjusted, and the filtering parameters include: filtering gain and / or covariance matrix.

[0140] In some embodiments of this application, the processor is also configured to perform the following operations:

[0141] The time of the low-orbit satellite's onboard data, the real-time observation data, and the environmental data are calibrated using an atomic clock or a GPS clock.

[0142] Based on the observation data and the environmental data, the low-Earth orbit satellite self-load data is corrected to obtain the corrected low-Earth orbit satellite self-load data.

[0143] In some embodiments of this application, the processor is also configured to perform the following operations:

[0144] The real-time observation data and the low-Earth orbit satellite's onboard data are input into a real-time differential correction model. The real-time differential correction model calculates the errors of atmospheric delay and multipath effects, and performs differential correction on the low-Earth orbit satellite's onboard data based on the errors of atmospheric delay and multipath effects to obtain the first data. The real-time differential correction model is constructed based on the baseline information of ground CORS and historical observation data collected by ground CORS.

[0145] Based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-orbit satellite self-loaded data.

[0146] In some embodiments of this application, the processor is also configured to perform the following operations:

[0147] The multi-source data is cleaned, outliers are removed, and normalization is performed.

[0148] In some embodiments of this application, the processor is also configured to perform the following operations:

[0149] The low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data are weighted and fused to obtain the comprehensive positioning data based on the real-time quality indicators of the low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data.

[0150] In some embodiments of this application, the processor is also configured to perform the following operations:

[0151] The dynamic error model is established based on historical information of the orbital state parameters of the low-orbit satellite and real-time environmental data collected by the auxiliary sensors.

[0152] In some embodiments of this application, the processor is also configured to perform the following operations:

[0153] The state estimation of the integrated positioning data is performed using extended Kalman filtering, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite.

[0154] In some embodiments of this application, the processor is also configured to perform the following operations:

[0155] The orbital state parameters of the low-orbit satellite are fed back to the satellite control center. These orbital state parameters are used to assist the satellite control center in adjusting the multi-source data fusion strategy and the filtering parameters of the target filtering method.

[0156] This application effectively improves the orbit determination accuracy of low-Earth orbit satellites through multi-source data fusion and real-time differential correction, enabling satellites to obtain more accurate orbit parameters in missions such as communication, remote sensing, navigation, and scientific experiments, directly improving mission success rate and satellite lifespan; moreover, precise orbit determination can reduce attitude adjustment and orbit maintenance operations caused by orbital errors, reduce satellite control frequency and operation and maintenance costs, and improve long-term satellite operation efficiency.

[0157] It should be noted that the orbit determination device provided in this application is a device capable of performing the above-described low-Earth orbit satellite orbit determination method. Therefore, all embodiments of the above-described low-Earth orbit satellite orbit determination method are applicable to this device and can achieve the same or similar beneficial effects, which will not be repeated here.

[0158] This application also provides an orbit determination device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the various processes in the low-Earth orbit satellite orbit determination method embodiments described above and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0159] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this program implements the various processes described above in the low-Earth orbit satellite orbit determination method embodiment, and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0160] This application also provides a computer program product, including computer instructions. When executed by a processor, these computer instructions implement the various processes of the above-described low-Earth orbit satellite orbit determination method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0161] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0162] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 A device for one or more processes and / or the functions specified in one or more boxes.

[0163] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce a paper article including an instruction means, the instruction means being implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0164] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment, causing the computer or other programmable equipment to perform a series of operational steps to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0165] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining the orbit of a low-Earth orbit satellite, characterized in that, The method includes: Collect multi-source data, including: low-orbit satellite onboard data, real-time observation data collected by the continuously operating ground reference station CORS, and environmental data collected by auxiliary sensors; Multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data; The integrated positioning data is input into a low-Earth orbit (LEO) satellite state space model that includes position, velocity, and acceleration. The LEO satellite state space model is used to estimate the state of the integrated positioning data to obtain the LEO satellite's orbital state parameters. The orbital state parameters include at least one of the LEO satellite's position, velocity, and acceleration.

2. The method according to claim 1, characterized in that, The orbital state parameters of the low-Earth orbit satellite are obtained by performing state estimation on the integrated positioning data using the low-Earth orbit satellite state space model, including: The state of the integrated positioning data is estimated using the target filtering method through the state space model of the low-Earth orbit satellite, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite. The target filtering method includes at least one of the following: extended Kalman filtering; unscented Kalman filtering; particle filtering.

3. The method according to claim 2, characterized in that, The method further includes: Based on the dynamic error model, obtain the real-time observation error; Based on the real-time observation error, the filtering parameters of the target filtering method are dynamically adjusted, and the filtering parameters include: filtering gain and / or covariance matrix.

4. The method according to any one of claims 1-3, characterized in that, Before performing multi-source data fusion on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data, the method further includes: The time of the low-orbit satellite's onboard data, the real-time observation data, and the environmental data are calibrated using an atomic clock or a GPS clock. Based on the observation data and the environmental data, the low-Earth orbit satellite self-load data is corrected to obtain the corrected low-Earth orbit satellite self-load data.

5. The method according to claim 4, characterized in that, Based on the observation data and the environmental data, the low-Earth orbit satellite's onboard data is corrected to obtain corrected low-Earth orbit satellite onboard data, including: The real-time observation data and the low-Earth orbit satellite's onboard data are input into a real-time differential correction model. The real-time differential correction model calculates the errors of atmospheric delay and multipath effects, and performs differential correction on the low-Earth orbit satellite's onboard data based on the errors of atmospheric delay and multipath effects to obtain the first data. The real-time differential correction model is constructed based on the baseline information of ground CORS and historical observation data collected by ground CORS. Based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-orbit satellite self-loaded data.

6. The method according to claim 1, characterized in that, After collecting multi-source data, the method further includes: The multi-source data is cleaned, outliers are removed, and normalization is performed.

7. The method according to claim 1, characterized in that, Multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data, including: The low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data are weighted and fused to obtain the comprehensive positioning data based on the real-time quality indicators of the low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data.

8. The method according to claim 3, characterized in that, The method further includes: The dynamic error model is established based on historical information of the orbital state parameters of the low-orbit satellite and real-time environmental data collected by the auxiliary sensors.

9. The method according to any one of claims 1-8, characterized in that, The method further includes: The orbital state parameters of the low-orbit satellite are fed back to the satellite control center. These orbital state parameters are used to assist the satellite control center in adjusting the multi-source data fusion strategy and the filtering parameters of the target filtering method.

10. An orbit determination device, comprising a processor and a transceiver, wherein the transceiver receives and transmits data under the control of the processor, characterized in that, The processor is used to perform the following operations: Collect multi-source data, including: low-orbit satellite onboard data, real-time observation data collected by the continuously operating ground reference station CORS, and environmental data collected by auxiliary sensors; Multi-source data fusion is performed on the low-orbit satellite's onboard data, the real-time observation data, and the environmental data to obtain comprehensive positioning data; The integrated positioning data is input into a low-Earth orbit (LEO) satellite state space model that includes position, velocity, and acceleration. The LEO satellite state space model is used to estimate the state of the integrated positioning data to obtain the LEO satellite's orbital state parameters. The orbital state parameters include at least one of the LEO satellite's position, velocity, and acceleration.

11. The device according to claim 10, characterized in that, The processor is also used to perform the following operations: The state of the integrated positioning data is estimated using the target filtering method through the state space model of the low-Earth orbit satellite, and the state estimation result with the smallest error is selected as the orbital state parameter of the low-Earth orbit satellite. The target filtering method includes at least one of the following: extended Kalman filtering; unscented Kalman filtering; particle filtering.

12. The device according to claim 11, characterized in that, The processor is also used to perform the following operations: Based on the dynamic error model, obtain the real-time observation error; Based on the real-time observation error, the filtering parameters of the target filtering method are dynamically adjusted, and the filtering parameters include: filtering gain and / or covariance matrix.

13. The device according to any one of claims 10-12, characterized in that, The processor is also used to perform the following operations: The time of the low-orbit satellite's onboard data, the real-time observation data, and the environmental data are calibrated using an atomic clock or a GPS clock. Based on the observation data and the environmental data, the low-Earth orbit satellite self-load data is corrected to obtain the corrected low-Earth orbit satellite self-load data.

14. The device according to claim 13, characterized in that, The processor is also used to perform the following operations: The real-time observation data and the low-Earth orbit satellite's onboard data are input into a real-time differential correction model. The real-time differential correction model calculates the errors of atmospheric delay and multipath effects, and performs differential correction on the low-Earth orbit satellite's onboard data based on the errors of atmospheric delay and multipath effects to obtain the first data. The real-time differential correction model is constructed based on the baseline information of ground CORS and historical observation data collected by ground CORS. Based on the environmental data and the real-time observation data, the ionospheric delay error and tropospheric delay error of the first data are corrected to obtain the corrected low-orbit satellite self-loaded data.

15. The device according to claim 10, characterized in that, The processor is also used to perform the following operations: The multi-source data is cleaned, outliers are removed, and normalization is performed.

16. The device according to claim 10, characterized in that, The processor is also used to perform the following operations: The low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data are weighted and fused to obtain the comprehensive positioning data based on the real-time quality indicators of the low-Earth orbit satellite self-loaded data, the real-time observation data, and the environmental data.

17. The device according to claim 12, characterized in that, The processor is also used to perform the following operations: The dynamic error model is established based on historical information of the orbital state parameters of the low-orbit satellite and real-time environmental data collected by the auxiliary sensors.

18. The device according to any one of claims 10-17, characterized in that, The processor is also used to perform the following operations: The orbital state parameters of the low-orbit satellite are fed back to the satellite control center. These orbital state parameters are used to assist the satellite control center in adjusting the multi-source data fusion strategy and the filtering parameters of the target filtering method.

19. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the steps in the low-Earth orbit satellite orbit determination method as described in any one of claims 1-9.

20. A computer program product, characterized in that, It includes computer instructions, which, when executed by a processor, implement the steps of the low-Earth orbit satellite orbit determination method as described in any one of claims 1-9.

Images