Multi-track fusion real-time precise positioning method and system

Abstract

This invention provides a real-time precise positioning method and system based on multi-orbit fusion, relating to the field of satellite positioning technology. The method includes: obtaining corresponding orbit fusion weight sets based on historical orbit accuracy assessment samples from various analysis centers; collecting raw ground observation data, broadcast ephemeris data, and real-time orbit corrections from various analysis centers, and performing weighted fusion to obtain fused real-time orbit corrections; further correcting and superimposing the broadcast ephemeris data to generate real-time precise orbits; calculating clock errors based on these precise orbits and raw ground observation data to obtain precise satellite clock errors; and finally generating and broadcasting a real-time precise positioning correction data stream. The user terminal uses this data to perform positioning corrections and obtain the target positioning result. This method suppresses anomalies from a single analysis center by weighted fusion of real-time orbit corrections and achieves high self-consistency between orbits and clock errors by recalculating clock errors based on the fused orbits, thereby improving the accuracy and continuity of real-time precise positioning.

Description

Technical Field

[0001] This invention belongs to the field of satellite positioning technology, specifically relating to a real-time precise positioning method and system with multi-orbit fusion. Background Technology

[0002] Real-time precise positioning services of the Global Navigation Satellite System rely on high-precision orbit and satellite clock correction information. Currently, to improve service availability, related technologies typically acquire real-time orbit and clock correction products broadcast by multiple analysis centers and directly weight and combine or simply stitch together the correction information from each source at the server or user end to generate a fused correction data stream for use by the positioning terminal. However, due to differences in data processing strategies, reference definitions, and dynamic models among analysis centers, their orbit and clock correction products implicitly contain different spatial and temporal references. Existing technologies fuse orbit corrections and clock corrections as independent products separately or combine them directly at the user end, resulting in a lack of geometric and physical consistency between the fused orbit and clock corrections. This easily leads to clock reference fluctuations and the accumulation of systematic deviations, thereby restricting positioning accuracy and service continuity. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a real-time precise positioning method and system based on multi-track fusion to solve the aforementioned problems. This method suppresses anomalies of a single analysis center by weighted fusion of real-time track corrections and achieves high self-consistency between track and clock error by recalculating the clock error based on the fused track, thereby improving the accuracy and continuity of real-time precise positioning.

[0004] To address the aforementioned technical problems, this invention provides a real-time precise positioning method using multi-track fusion, comprising the following steps: Obtain historical samples of orbit accuracy assessment from several analysis centers, and obtain the orbit fusion weight set corresponding to each analysis center based on the historical samples of orbit accuracy assessment. Acquire raw ground observation data, broadcast ephemeris, and real-time orbital corrections from several analysis centers; Based on the track fusion weight set corresponding to each analysis center, the real-time track corrections of each analysis center are weighted and fused to obtain the fused real-time track corrections. Based on the fused real-time orbit corrections, the broadcast ephemeris is corrected and superimposed to obtain a real-time precise orbit. Based on the real-time precise orbit and ground raw observation data, an observation equation is established, and based on the observation equation, the clock error is calculated to obtain the precise satellite clock error; Based on the precise satellite clock bias, the fused real-time orbit correction data, and the broadcast ephemeris, a real-time precise positioning correction data stream is obtained and broadcast, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result.

[0005] In the above scheme, historical samples of orbit accuracy assessment from several analysis centers are obtained, and an orbit fusion weight set is acquired. Based on the orbit fusion weight set, the real-time orbit corrections of each analysis center are weighted and fused to obtain a fused real-time orbit correction, thereby suppressing the impact of anomalies or accuracy fluctuations in a single analysis center. Based on the fused real-time orbit correction, the broadcast ephemeris is corrected and superimposed to obtain a real-time precise orbit. Then, based on the real-time precise orbit and the original ground observation data, an observation equation is established to calculate the clock error. The precise satellite clock error is obtained by recalculating. Since the clock error calculation is directly re-estimated using the fused real-time precise orbit as the geometric reference, rather than directly combining multi-source clock error products, the inconsistency of multi-source clock error references and the error of reference jump are fundamentally avoided, ensuring high self-consistency between orbit and clock error. Finally, based on the precise satellite clock error, the fused real-time orbit correction, and the broadcast ephemeris, a real-time precise positioning correction data stream is obtained and broadcast. The user terminal performs positioning correction based on the real-time precise positioning correction data stream, thus obtaining a target positioning result with higher accuracy and stronger continuity.

[0006] Furthermore, the step of obtaining the orbit fusion weight set corresponding to each analysis center based on historical samples of orbit accuracy assessment includes: For each analysis center, historical real-time orbit data and corresponding post-event precise orbit data are obtained based on the historical samples of orbit accuracy assessment corresponding to that analysis center. A historical orbit deviation dataset is obtained based on the aforementioned historical real-time orbit data and post-event precise orbit data. Based on the historical orbital deviation dataset, orbital error direction decomposition is performed to obtain historical radial error data, historical tangential error data, and historical normal error data; The weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data, resulting in a track fusion weight set.

[0007] In the above scheme, firstly, for each analysis center, historical real-time orbit data and corresponding post-hoc precise orbit data are obtained based on historical samples of orbit accuracy assessment. By introducing the post-hoc precise orbit data as a truth reference, an objective benchmark is provided for orbit accuracy assessment. Then, based on the historical real-time orbit data and the post-hoc precise orbit data, a historical orbit deviation dataset is obtained, and the orbit error is extracted in a quantitative form, providing a data foundation for subsequent error analysis. Next, based on the historical orbit deviation dataset, orbit error direction decomposition is performed to obtain historical radial error data, historical tangential error data, and historical normal error data. The three-dimensional orbit error is decomposed into three orthogonal directions, allowing for more refined weight determination by handling different error directions separately. Finally, based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data, weights are determined to obtain an orbit fusion weight set. Utilizing the inverse distance weighting principle, analysis centers with better historical accuracy performance receive higher weights, ensuring that the fusion result tends towards optimal accuracy in all directions.

[0008] Furthermore, the weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data to obtain a track fusion weight set. The preset inverse distance weighting algorithm is specifically as follows: Based on the historical radial error data, historical tangential error data, and historical normal error data, the corresponding root mean square values ​​of radial error, tangential error, and normal error are obtained respectively. The corresponding radial weight, tangential weight, and normal weight are determined based on the root mean square value of the radial error, the root mean square value of the tangential error, and the root mean square value of the normal error, respectively. The orbit fusion weight set is obtained based on the radial weight, tangential weight, and normal weight.

[0009] In the above scheme, firstly, based on the historical radial error data, historical tangential error data, and historical normal error data, the corresponding root mean square (RMS) values ​​of radial error, tangential error, and normal error are obtained respectively. The historical error sequences in each direction are then synthesized into statistics by calculating the RMS values, providing input for weight calculation. Next, based on the RMS values ​​of radial error, tangential error, and normal error, the corresponding radial weight, tangential weight, and normal weight are determined respectively, achieving weighting by direction. This ensures that the accuracy fluctuation of a certain analysis center in one direction does not affect the weights in other directions, guaranteeing independent optimization of the fusion results in each direction. Finally, based on the radial weight, tangential weight, and normal weight, the orbital fusion weight set is obtained, integrating the weights of the three directions into a complete weight set. This allows subsequent weighted fusion to apply the corresponding weights to each component of the orbital correction, achieving refined multi-directional weighted fusion, thereby ensuring that the fused real-time orbital correction achieves optimal accuracy in each direction.

[0010] Further, the weighted fusion of the real-time track corrections of each analysis center based on the track fusion weight set corresponding to each analysis center to obtain the fused real-time track corrections includes: Based on the real-time orbit corrections of each analysis center, the radial orbit correction component, tangential orbit correction component, and normal orbit correction component corresponding to each analysis center are obtained respectively. Based on the orbit fusion weight set corresponding to each analysis center, the radial orbit correction component, tangential orbit correction component, and normal orbit correction component of each analysis center are weighted and summed to obtain the corresponding fused radial orbit correction component, fused tangential orbit correction component, and fused normal orbit correction component. The fused real-time orbit correction number is obtained based on the fused radial orbit correction component, the fused tangential orbit correction component, and the fused normal orbit correction component.

[0011] In the above scheme, firstly, the radial, tangential, and normal orbital correction components corresponding to each analysis center are obtained based on the real-time orbital corrections of each analysis center. The real-time orbital corrections are decomposed into three orthogonal components in the orbital coordinate system, allowing each component to be independently weighted. Then, based on the orbital fusion weight set corresponding to each analysis center, the radial, tangential, and normal orbital correction components of each analysis center are weighted and summed to obtain the corresponding fused radial, tangential, and normal orbital correction components. For each direction, the corresponding weights are applied independently for fusion, allowing the correction components of high-precision analysis centers to play a role in that direction, while the components of low-precision analysis centers are reasonably suppressed, ensuring that all directional components reach optimal accuracy. Finally, based on the fused radial, tangential, and normal orbital correction components, the fused real-time orbital corrections are obtained. The three independently optimized directional components are then recombine into a complete fused real-time orbital correction, ensuring that the fusion result maintains high accuracy in three-dimensional space.

[0012] Furthermore, the step of correcting and superimposing the broadcast ephemeris based on the fused real-time orbit corrections to obtain the real-time precise orbit includes: Based on the preset reference time and the fused real-time track correction number, the reference time track correction component and the reference time track correction velocity component are obtained; The current observation epoch, the satellite position vector at the current epoch, and the satellite velocity vector at the current epoch are obtained based on the broadcast ephemeris. The current orbit correction component is obtained based on the current observation epoch, the preset reference time, the orbit correction component at the reference time, and the orbit correction velocity component at the reference time. Based on the current epoch satellite position vector and current epoch satellite velocity vector, perform geocentric-ground-fixed coordinate transformation on the current orbit correction component to obtain the ground-fixed orbit correction component; The broadcast ephemeris is corrected based on the Earth-fixed orbit correction component to obtain the real-time precise orbit.

[0013] In the above scheme, firstly, the orbital correction component and the orbital correction velocity component at the reference time are obtained based on a preset reference time and the fused real-time orbital correction data, providing initial parameters for calculating the orbital correction component from the reference time to the current observation epoch. Then, the current observation epoch, the satellite position vector, and the satellite velocity vector at the current epoch are obtained based on the broadcast ephemeris. The current orbital correction component is then obtained based on the current observation epoch, the preset reference time, the orbital correction component at the reference time, and the orbital velocity component at the reference time, achieving accurate calculation of the orbital correction component at the current observation epoch and avoiding deviations introduced by epoch inconsistencies. Next, a geocentric-to-ground coordinate transformation is performed on the current orbital correction component based on the current epoch satellite position vector and the current epoch satellite velocity vector to obtain the geocentric-to-ground orbital correction component, unifying the coordinate reference of the orbital correction component with the broadcast ephemeris and ensuring the superpositionability of subsequent corrections. Finally, the broadcast ephemeris is corrected based on the ground-fixed orbit correction component to obtain the real-time precise orbit. The fused high-precision orbit correction is then directly applied to the broadcast ephemeris to obtain a real-time precise orbit with the same accuracy as the fused real-time orbit correction.

[0014] Furthermore, the step of establishing observation equations based on the real-time precise orbit and raw ground observation data, and calculating clock bias based on the observation equations to obtain precise satellite clock bias, includes: The current epoch observation equation set is constructed based on the preset set of parameters to be estimated, the real-time precise orbit, and the raw ground observation data. Obtain the optimal state estimate of the historical epoch, and make state prediction based on the optimal state estimate of the historical epoch and the preset random walk model to obtain the current epoch state prediction value and the prediction covariance matrix. The precise satellite clock bias is obtained by estimating parameters based on the current epoch state prediction, the forecast covariance matrix, and the current epoch observation equations.

[0015] In the above scheme, firstly, an observation equation set for the current epoch is constructed based on a preset set of parameters to be estimated, the real-time precise orbit, and the original ground observation data. The fused real-time precise orbit is used as a known geometric reference and substituted into the observation equations, which reduces the coupling effect of orbital errors on clock error estimation and improves the accuracy of clock error calculation. Then, the optimal state estimate for historical epochs is obtained, and state prediction is performed based on the optimal state estimate for historical epochs and a preset random walk model to obtain the predicted state value and forecast covariance matrix for the current epoch. The Kalman filter prediction mechanism is used to transfer the historical optimal estimate to the current epoch, ensuring that the clock error parameters remain continuous in time and avoiding instability caused by abrupt changes between epochs. Finally, parameter estimation is performed based on the predicted state value, forecast covariance matrix, and the observation equation set for the current epoch to obtain the precise satellite clock error. The predicted information is fused with the current observation information for solution. Because the geometric reference is a high-precision fused orbit and the state prediction ensures temporal continuity, the obtained precise satellite clock error has high accuracy and high stability.

[0016] Further, the parameter estimation based on the current epoch state prediction value, the forecast covariance matrix, and the current epoch observation equations to obtain the precise satellite clock bias includes: Construct the normal equation matrix and the right normal equation matrix based on the current epoch observation equation set; Based on the current epoch state prediction value, the forecast covariance matrix, the normal equation matrix, and the right normal equation matrix, a joint parameter estimation equation is constructed. The precise satellite clock bias is obtained based on the joint parameter estimation equation.

[0017] In the above scheme, firstly, a normal equation matrix and a right-hand matrix are constructed based on the current epoch observation equations, transforming the observation equations into normal equation form, enabling efficient solution of multi-parameter joint estimation through matrix operations. Then, a joint parameter estimation equation is constructed based on the current epoch state prediction, forecast covariance matrix, normal equation matrix, and right-hand matrix, incorporating prediction information into the normal equations in the form of a covariance matrix, forming a joint estimation framework that integrates prediction and observation. This ensures the estimation results are simultaneously constrained by historical continuity and current observation information. Finally, the precise satellite clock bias is obtained based on the joint parameter estimation equation. By solving the joint parameter estimation equation, the optimal parameter estimate under prediction constraints is obtained. The obtained precise satellite clock bias, due to the dual guarantees of geometric constraints from the high-precision fused orbit and temporal constraints from the random walk model, possesses higher accuracy and stronger stability.

[0018] Furthermore, the step of obtaining the real-time precise positioning correction data stream based on the precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris includes: The receiver clock bias is obtained based on the joint parameter estimation equation. The pseudorange post-test residual is obtained based on the precision satellite clock error, receiver clock error, and raw ground observation data. The spread code bias is obtained by fusing the post-pseudorange residual and the pre-acquired true inter-code bias of the satellite. The real-time precise positioning correction data stream is obtained based on the extended code deviation, precise satellite clock error, fused real-time orbit corrections, and broadcast ephemeris.

[0019] In the above scheme, the receiver clock bias is first obtained based on the joint parameter estimation equation, ensuring that the receiver clock bias and the precise satellite clock bias are obtained under the same estimation framework, guaranteeing their consistency. Then, the pseudorange post-hoc residual is obtained based on the precise satellite clock bias, receiver clock bias, and raw ground observation data. The optimal clock bias estimate is substituted back into the observation equation to calculate the residual, providing a basis for subsequent bias extraction. Next, bias fusion is performed based on the pseudorange post-hoc residual and the pre-acquired true inter-code bias of the satellites to obtain the spreading code bias. This spreading code bias includes both hardware delay and server-side systematic bias, eliminating the need for the user to process multiple bias sources separately. Finally, a real-time precise positioning correction data stream is obtained based on the spreading code bias, precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris. The orbit correction, clock bias correction, and code bias correction are integrated into a unified data stream, allowing the user to obtain complete precise positioning correction information through a single data source, simplifying the processing flow and ensuring consistency among the correction items.

[0020] Furthermore, the step of acquiring and broadcasting a real-time precise positioning correction data stream based on the precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result, includes: Based on the broadcast ephemeris, obtain the broadcast ephemeris orbital parameters, broadcast ephemeris clock error parameters, and group delay parameters; Based on the fused real-time orbit corrections, the broadcast ephemeris orbit parameters are corrected to obtain real-time precise orbit correction parameters; Real-time precision clock error correction parameters are determined based on the precision satellite clock error and broadcast ephemeris clock error parameters; The real-time code deviation correction parameters are determined based on the aforementioned extended code deviation. Based on the real-time precise track correction parameters, real-time precise clock error correction parameters and real-time code deviation correction parameters, binary data stream encoding is performed to obtain the real-time precise positioning correction data stream. The real-time precise positioning correction data stream is continuously broadcast externally through a preset differential data network transmission protocol so that the user terminal can receive the real-time precise positioning correction data stream; Based on the real-time precise orbit correction parameters and real-time precise clock error correction parameters in the real-time precise positioning correction data stream, the original phase observation value of the user terminal is corrected to obtain the corrected phase observation value. Based on the real-time precise orbit correction parameters, real-time precise clock error correction parameters, real-time code deviation correction parameters and the group delay parameters in the real-time precise positioning correction data stream, the original pseudorange observation value of the user terminal is corrected to obtain the corrected pseudorange observation value. The target positioning result is obtained by performing positioning calculations based on the corrected phase observations and corrected pseudorange observations.

[0021] In the above scheme, firstly, broadcast ephemeris orbit parameters, broadcast ephemeris clock bias parameters, and group delay parameters are obtained based on the broadcast ephemeris, providing a reference for subsequent generation of correction parameters. Then, the broadcast ephemeris orbit parameters are corrected based on the fused real-time orbit corrections to obtain real-time precise orbit correction parameters. Real-time precise clock bias correction parameters are determined based on the precise satellite clock bias and broadcast ephemeris clock bias parameters, and real-time code bias correction parameters are determined based on the spreading code bias, ensuring the correction parameters are of a moderate magnitude for easy data compression and transmission. Next, binary data stream encoding is performed based on the real-time precise orbit correction parameters, real-time precise clock bias correction parameters, and real-time code bias correction parameters to obtain a real-time precise positioning correction data stream, achieving compact encoding to improve transmission efficiency. The real-time precise positioning correction data stream is continuously broadcast through a preset differential data network transmission protocol for user reception. The user terminal corrects the original phase observations based on the real-time precise orbit correction parameters and the real-time precise clock error correction parameters to obtain corrected phase observations. It then corrects the original pseudorange observations based on the same parameters, resulting in corrected pseudorange observations. This differentiated correction ensures high accuracy of the phase observations, unaffected by code error, and provides complete system error correction for the pseudorange observations. Finally, positioning calculations are performed based on the corrected phase and pseudorange observations to obtain high-precision, highly continuous target positioning results.

[0022] The present invention also provides a real-time precision positioning system with multi-track fusion, comprising: The weight determination module is used to obtain historical samples of orbit accuracy assessment from several analysis centers, and to obtain the orbit fusion weight set corresponding to each analysis center based on the historical samples of orbit accuracy assessment. The data acquisition module is used to acquire raw ground observation data, broadcast ephemeris data, and real-time orbit corrections from several analysis centers; The track weighted fusion module is used to perform weighted fusion of the real-time track corrections of each analysis center based on the track fusion weight set corresponding to each analysis center, so as to obtain the fused real-time track corrections. The orbit correction module is used to correct and superimpose the broadcast ephemeris based on the fused real-time orbit correction data to obtain a real-time precise orbit. The clock error calculation module is used to establish observation equations based on the real-time precise orbit and raw ground observation data, and to calculate the clock error based on the observation equations to obtain the precise satellite clock error. The data broadcasting module is used to acquire and broadcast real-time precise positioning correction data streams based on the precise satellite clock error, fused real-time orbit correction data, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data streams and obtain the target positioning result.

[0023] In the above scheme, the weight determination module obtains an orbit fusion weight set based on historical samples of each orbit's accuracy assessment. It achieves differentiated weighting across multiple analysis centers through accuracy analysis of historical data, providing a weighting basis for subsequent fusion. The data acquisition module acquires raw ground observation data, broadcast ephemeris data, and real-time orbit corrections from each analysis center, responsible for multi-source data collection and aggregation, providing raw input to the system. The orbit weighted fusion module performs weighted fusion of the real-time orbit corrections from each analysis center based on the orbit fusion weight set, obtaining a fused real-time orbit correction, achieving robust fusion of multi-source orbit products and suppressing the impact of anomalies from a single analysis center. The orbit correction module corrects and superimposes the broadcast ephemeris based on the fused real-time orbit correction, obtaining a real-time precise orbit. The fused high-precision correction is then applied to the broadcast ephemeris to improve orbit accuracy. The clock error calculation module establishes observation equations based on the real-time precise orbit and raw ground observation data and performs clock error calculation to obtain precise satellite clock errors. The clock errors are recalculated using the fused orbit as the geometric reference, avoiding inconsistencies in multi-source clock error references. The data broadcasting module acquires and broadcasts real-time precise positioning correction data streams based on precise satellite clock bias, real-time orbit correction data, and broadcast ephemeris data. This enables users to perform positioning corrections to obtain target positioning results. The server can uniformly generate and broadcast high-precision correction data, and the various modules work together to form a complete precise positioning service capability. Attached Figure Description

[0024] Figure 1 This is a schematic flowchart of a real-time precision positioning method using multi-track fusion, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a real-time precision positioning system architecture with multi-track fusion provided in an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Please see Figure 1 This embodiment provides a real-time precise positioning method using multi-track fusion, including the following steps: Step S1: Obtain historical samples of orbit accuracy assessment from several analysis centers, and obtain the orbit fusion weight set corresponding to each analysis center based on the historical samples of orbit accuracy assessment. Step S2: Acquire raw ground observation data, broadcast ephemeris, and real-time orbital corrections from several analysis centers; Step S3: Based on the track fusion weight set corresponding to each analysis center, the real-time track corrections of each analysis center are weighted and fused to obtain the fused real-time track corrections; Step S4: Based on the fused real-time orbit corrections, the broadcast ephemeris is corrected and superimposed to obtain the real-time precise orbit; Step S5: Establish observation equations based on the real-time precise orbit and ground raw observation data, and calculate clock bias based on the observation equations to obtain precise satellite clock bias; Step S6: Based on the precise satellite clock error, the fused real-time orbit correction data, and the broadcast ephemeris, obtain and broadcast the real-time precise positioning correction data stream so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream and obtain the target positioning result.

[0027] In this embodiment, historical samples of orbit accuracy assessment from several analysis centers are acquired, and an orbit fusion weight set is obtained. Based on the orbit fusion weight set, the real-time orbit corrections of each analysis center are weighted and fused to obtain a fused real-time orbit correction, thereby suppressing the impact of anomalies or accuracy fluctuations in a single analysis center. Based on the fused real-time orbit correction, the broadcast ephemeris is corrected and superimposed to obtain a real-time precise orbit. Then, based on the real-time precise orbit and the original ground observation data, an observation equation is established to calculate the clock bias. The precise satellite clock bias is obtained by recalculating. Since the clock bias calculation directly uses the fused real-time precise orbit as the geometric reference for re-estimation, rather than directly combining multi-source clock bias products, the inconsistency of multi-source clock bias references and the error of reference jump are fundamentally avoided, ensuring high self-consistency between orbit and clock bias. Finally, based on the precise satellite clock bias, the fused real-time orbit correction, and the broadcast ephemeris, a real-time precise positioning correction data stream is obtained and broadcast. The user terminal performs positioning correction based on the real-time precise positioning correction data stream, thereby obtaining a target positioning result with higher accuracy and stronger continuity.

[0028] In one embodiment, a real-time precise positioning method based on multi-track fusion is provided, comprising the following steps: Accuracy analysis is performed on the real-time orbit products from various historical analysis centers to obtain historical samples of orbit accuracy assessment. Based on these historical samples, the root mean square error (RMS) of the historical orbits of each analysis center is statistically calculated. An inverse distance weighting method is then used to determine the weight of each analysis center in the current orbit fusion in real time, forming an orbit fusion weight set. Raw ground observation data, broadcast ephemeris, and real-time orbit corrections from several analysis centers are acquired. Based on the orbit fusion weight set, the real-time orbit corrections from each analysis center are weighted and fused to obtain the fused real-time orbit correction. The broadcast ephemeris is then corrected and superimposed based on the fused real-time orbit correction to obtain the real-time precise orbit. Using a ground reference station network and the real-time precise orbit as the geometric reference, observation equations are established and clock errors are recalculated to obtain the precise satellite clock errors.

[0029] It should be noted that, in this embodiment, after obtaining the fused real-time precise orbit, the multi-source clock bias products are not directly combined. Instead, they are used as a fixed reference to recalculate the clock bias, fundamentally avoiding inconsistencies and reference jumps in multi-source clock bias references, and ensuring high self-consistency between the orbit and clock bias. The broadcast ephemeris orbit and clock bias are calculated based on the broadcast ephemeris. The precise satellite clock bias, the fused real-time orbit correction, and the broadcast ephemeris orbit and clock bias are combined to generate a real-time precise positioning correction data stream. This stream is then broadcast via the Networked Transport of RTCM via Internet Protocol (NTRIP) so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result.

[0030] Furthermore, the step of obtaining the orbit fusion weight set corresponding to each analysis center based on historical samples of orbit accuracy assessment includes: For each analysis center, historical real-time orbit data and corresponding post-event precise orbit data are obtained based on the historical samples of orbit accuracy assessment corresponding to that analysis center. A historical orbit deviation dataset is obtained based on the aforementioned historical real-time orbit data and post-event precise orbit data. Based on the historical orbital deviation dataset, orbital error direction decomposition is performed to obtain historical radial error data, historical tangential error data, and historical normal error data; The weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data, resulting in a track fusion weight set.

[0031] In this embodiment, firstly, for each analysis center, historical real-time orbit data and corresponding post-hoc precise orbit data are obtained based on historical samples of orbit accuracy assessment. By introducing the post-hoc precise orbit data as a truth reference, an objective benchmark is provided for orbit accuracy assessment. Then, based on the historical real-time orbit data and the post-hoc precise orbit data, a historical orbit deviation dataset is obtained, and the orbit error is extracted in a quantified form, providing a data foundation for subsequent error analysis. Next, orbit error direction decomposition is performed based on the historical orbit deviation dataset to obtain historical radial error data, historical tangential error data, and historical normal error data. The three-dimensional orbit error is decomposed into three orthogonal directions, allowing for more refined weight determination by processing different error directions separately. Finally, weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data to obtain an orbit fusion weight set. Utilizing the inverse distance weighting principle, analysis centers with better historical accuracy performance receive higher weights, ensuring that the fusion result tends towards optimal accuracy in all directions.

[0032] Furthermore, the weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data to obtain a track fusion weight set. The preset inverse distance weighting algorithm is specifically as follows: Based on the historical radial error data, historical tangential error data, and historical normal error data, the corresponding root mean square values ​​of radial error, tangential error, and normal error are obtained respectively. The corresponding radial weight, tangential weight, and normal weight are determined based on the root mean square value of the radial error, the root mean square value of the tangential error, and the root mean square value of the normal error, respectively. The orbit fusion weight set is obtained based on the radial weight, tangential weight, and normal weight.

[0033] In this embodiment, firstly, based on the historical radial error data, historical tangential error data, and historical normal error data, the corresponding root mean square (RMS) values ​​of radial error, tangential error, and normal error are obtained respectively. The historical error sequences in each direction are then synthesized into statistics by calculating the RMS values, providing input for weight calculation. Next, based on the RMS values ​​of radial error, tangential error, and normal error, the corresponding radial weight, tangential weight, and normal weight are determined respectively, achieving weighting by direction. This ensures that the accuracy fluctuation of a certain analysis center in one direction does not affect the weights in other directions, guaranteeing independent optimization of the fusion results in each direction. Finally, based on the radial weight, tangential weight, and normal weight, the orbital fusion weight set is obtained, integrating the weights of the three directions into a complete weight set. This allows subsequent weighted fusion to apply the corresponding weights to each component of the orbital correction, achieving refined multi-directional weighted fusion, thereby ensuring that the fused real-time orbital correction achieves optimal accuracy in each direction.

[0034] In one embodiment, a precision analysis is performed on real-time orbits stored for one month to obtain orbit fusion weight sets corresponding to each analysis center. Specifically, firstly, historical real-time orbit data and corresponding post-hoc precise orbit data from each analysis center are acquired, and the stored real-time satellite orbits are... With post-event precision track By performing the subtraction, we obtain the orbital deviation vector: in, This indicates the satellite's position coordinates in three-dimensional space.

[0035] It should be noted that this embodiment, by introducing a post-hoc precise orbit as a true reference, can provide an objective and stable benchmark for orbit accuracy assessment. Then, based on the satellite orbit's position vector r and velocity vector v, the orbital deviation vector is transformed into error components in three orthogonal directions: radial, tangential, and normal. The specific transformation method is as follows: in: Therefore, the three-dimensional orbital error is decomposed into radial, tangential, and normal directions, forming a historical orbital deviation dataset. This allows subsequent weight determination to address different error directions separately, achieving a more refined evaluation. Next, for each analysis center, its radial deviation is statistically analyzed. Tangential and legal direction The root mean square (RMS) error. For the analysis center ac, the direction... Its root mean square error It can be calculated using the following formula: In the formula, for Orbital errors of analysis centers ac in different directions The number of historical samples is used. By calculating the root mean square (RMS) value, the historical error sequences in each direction are synthesized into a single statistical index, providing a basis for subsequent weight calculation. Then, a preset inverse distance weighting algorithm is used to determine the weight of each analysis center in the current orbit fusion in real time based on the RMS value of each direction. In this embodiment, the reciprocal of the RMS value in each direction is directly taken as the weight of the corresponding direction, i.e.: The physical meaning is that the analysis center with better historical accuracy and smaller RMS is assigned a higher fusion weight, thereby ensuring that the fusion result tends to the optimal accuracy in each direction. Thus, a set of orbital fusion weights including radial, tangential, and normal weights is obtained.

[0036] It should be noted that in this embodiment, each direction is weighted independently. The accuracy fluctuation of a certain analysis center in a single direction will not affect its own weight or the weight of other analysis centers in other directions. At the same time, the subsequent recalculation of clock error can absorb the small inconsistency error that may be introduced due to the different weights of the orbital directions. There is no need to perform multi-directional weighting of clock error, which not only ensures the high self-consistency between the orbit and clock error, but also reduces the computational complexity.

[0037] During real-time positioning, raw ground observation data, broadcast ephemeris, and real-time orbit corrections (i.e., State Space Representations, SSRs) broadcast from several analysis centers are acquired. Based on the aforementioned orbit fusion weight set, the radial, tangential, and normal components of the real-time orbit corrections from each analysis center are weighted and fused to obtain the fused real-time orbit correction. This fusion method only weights and establishes orbit parameters with clear physical laws, effectively suppressing the influence of anomalies or accuracy fluctuations from a single analysis center and improving orbit observation accuracy. Furthermore, the broadcast ephemeris is corrected and superimposed based on the fused real-time orbit correction to obtain a real-time precise orbit. Using this real-time precise orbit as a geometric reference, the observation equations are established through a ground reference station network to recalculate the precise satellite clock bias. Compared to the traditional method of directly merging multi-source real-time clock bias products, this embodiment uses the weighted high-precision orbit to recalculate the satellite clock bias, fundamentally avoiding the reference jump and inconsistency errors caused by real-time clock bias merging, thereby achieving a more accurate and continuous real-time precise positioning service.

[0038] Further, the weighted fusion of the real-time track corrections of each analysis center based on the track fusion weight set corresponding to each analysis center to obtain the fused real-time track corrections includes: Based on the real-time orbit corrections of each analysis center, the radial orbit correction component, tangential orbit correction component, and normal orbit correction component corresponding to each analysis center are obtained respectively. Based on the orbit fusion weight set corresponding to each analysis center, the radial orbit correction component, tangential orbit correction component, and normal orbit correction component of each analysis center are weighted and summed to obtain the corresponding fused radial orbit correction component, fused tangential orbit correction component, and fused normal orbit correction component. The fused real-time orbit correction number is obtained based on the fused radial orbit correction component, the fused tangential orbit correction component, and the fused normal orbit correction component.

[0039] Furthermore, the step of correcting and superimposing the broadcast ephemeris based on the fused real-time orbit corrections to obtain the real-time precise orbit includes: Based on the preset reference time and the fused real-time track correction number, the reference time track correction component and the reference time track correction velocity component are obtained; The current observation epoch, the satellite position vector at the current epoch, and the satellite velocity vector at the current epoch are obtained based on the broadcast ephemeris. The current orbit correction component is obtained based on the current observation epoch, the preset reference time, the orbit correction component at the reference time, and the orbit correction velocity component at the reference time. Based on the current epoch satellite position vector and current epoch satellite velocity vector, perform geocentric-ground-fixed coordinate transformation on the current orbit correction component to obtain the ground-fixed orbit correction component; The broadcast ephemeris is corrected based on the Earth-fixed orbit correction component to obtain the real-time precise orbit.

[0040] In one embodiment, after obtaining the fused real-time orbit correction, the weighted correction is further merged with the broadcast ephemeris to generate a real-time precise ephemeris and orbit. It should be noted that the fused real-time orbit correction is the state domain correction (SSR), which includes the reference time. Radial correction component below Tangential correction component and normal correction components And the corresponding velocity correction components in each direction. , and Specifically, the orbital correction vector in the satellite-fixed coordinate system at the current observation epoch t is first calculated. Using the correction value at the reference time as the initial value, the correction value for the current epoch is obtained through linear extrapolation, and the calculation formula is as follows: It should be noted that within the short time interval from the reference time to the current epoch, the change in the orbital correction can be approximated as linear. The above-mentioned linear extrapolation method not only ensures computational efficiency but also meets the requirements of real-time precision orbits for the timeliness of corrections.

[0041] Then, the satellite's orbital corrections are transformed from the star-fixed coordinate system to the Earth-Centered Earth-Fixed (ECEF) coordinate system to ensure that the coordinate reference upon which the corrections are based is consistent with the broadcast ephemeris, thus ensuring the superposition of subsequent corrections. Using the satellite position vector *r* and velocity vector *v* calculated from the broadcast ephemeris for the current epoch, unit vectors in three orthogonal directions of the star-fixed coordinate system are constructed. A radial unit vector is defined. ， Normal unit vector Tangential unit vector The three unit vectors mentioned above form a transformation matrix in columns. Therefore, the orbital correction vector in the geocentric-fixed coordinate system Calculate using the following formula: It should be noted that the radial direction corresponds to the direction of the satellite position vector, the tangential direction is located in the orbital plane and is orthogonal to the radial direction, and the normal direction is perpendicular to the orbital plane. The three form a right-handed coordinate system, which can accurately project the orbital correction components to the Earth-fixed coordinate system.

[0042] Finally, the Earth-fixed orbit correction vector is directly applied to the satellite orbit calculated from the broadcast ephemeris to obtain a real-time precise orbit. Let the satellite position vector calculated from the broadcast ephemeris be... Real-time precision track for: Therefore, by correcting the fused high-precision orbit corrections to the broadcast ephemeris orbit, a real-time precise orbit expressed in geocentric and geofixed coordinates is obtained, providing a high-precision and self-consistent geometric reference for subsequent clock error estimation and positioning calculation.

[0043] Furthermore, the step of establishing observation equations based on the real-time precise orbit and raw ground observation data, and calculating clock bias based on the observation equations to obtain precise satellite clock bias, includes: The current epoch observation equation set is constructed based on the preset set of parameters to be estimated, the real-time precise orbit, and the raw ground observation data. Obtain the optimal state estimate of the historical epoch, and make state prediction based on the optimal state estimate of the historical epoch and the preset random walk model to obtain the current epoch state prediction value and the prediction covariance matrix. The precise satellite clock bias is obtained by estimating parameters based on the current epoch state prediction, the forecast covariance matrix, and the current epoch observation equations.

[0044] Further, the parameter estimation based on the current epoch state prediction value, the forecast covariance matrix, and the current epoch observation equations to obtain the precise satellite clock bias includes: Construct the normal equation matrix and the right normal equation matrix based on the current epoch observation equation set; Based on the current epoch state prediction value, the forecast covariance matrix, the normal equation matrix, and the right normal equation matrix, a joint parameter estimation equation is constructed. The precise satellite clock bias is obtained based on the joint parameter estimation equation.

[0045] Furthermore, the step of obtaining the real-time precise positioning correction data stream based on the precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris includes: The receiver clock bias is obtained based on the joint parameter estimation equation. The pseudorange post-test residual is obtained based on the precision satellite clock error, receiver clock error, and raw ground observation data. The spread code bias is obtained by fusing the post-pseudorange residual and the pre-acquired true inter-code bias of the satellite. The real-time precise positioning correction data stream is obtained based on the extended code deviation, precise satellite clock error, fused real-time orbit corrections, and broadcast ephemeris.

[0046] In this embodiment, the receiver clock bias is first obtained based on the joint parameter estimation equation, ensuring that the receiver clock bias and the precise satellite clock bias are obtained within the same estimation framework, guaranteeing their consistency. Then, the pseudorange post-hoc residual is obtained based on the precise satellite clock bias, receiver clock bias, and raw ground observation data. The optimal clock bias estimate is substituted back into the observation equation to calculate the residual, providing a basis for subsequent bias extraction. Next, bias fusion is performed based on the pseudorange post-hoc residual and the pre-acquired true inter-code bias of the satellites to obtain the spreading code bias. This spreading code bias simultaneously includes hardware delay and server-side systematic bias, eliminating the need for the user to process multiple bias sources separately. Finally, a real-time precise positioning correction data stream is obtained based on the spreading code bias, precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris. The orbit correction, clock bias correction, and code bias correction are integrated into a unified data stream, allowing the user to obtain complete precise positioning correction information through a single data source, simplifying the processing flow and ensuring consistency among the correction items.

[0047] In one embodiment, the satellite clock bias is recalculated using a ground reference station, and the steps are as follows: First, based on the preset set of parameters to be estimated, real-time precise orbit, and raw ground observation data, the current epoch is constructed. The ionosphere-free combined observation equations form the current epoch observation equation set. The pseudorange observation equation and the carrier phase observation equation are expressed as follows: In the formula, and These are pseudorange and carrier phase observations of Global Navigation Satellite System (GNSS) satellites from a regional ground monitoring receiver, respectively. c is the geometric distance between the satellite and the receiver; c is the speed of light in a vacuum. and These are the receiver clock bias and the satellite clock bias to be estimated, respectively. This is the tropospheric delayed projection function; To calculate the tropospheric delay in the zenith direction using an empirical model; The residual zenith-direction tropospheric delay is estimated along with the filter; For satellite wavelength; For phase ambiguity; and These are the pseudorange and phase measurement errors, respectively.

[0048] Then, the optimal state estimates for historical epochs are obtained, and a Kalman filter state prediction is performed using a random walk model to obtain the predicted state value and prediction covariance matrix for the current epoch. The state prediction equation is: In the formula, This is the predicted state for the current epoch, which includes satellite clock bias. ; This is the optimal estimate from the previous epoch; To predict the covariance matrix; The covariance matrix is ​​estimated for the previous epoch; This is the process noise matrix. Fixed-incremental covariance dilation is applied to the satellite clock error parameters: In the formula, The current epoch satellite clock error prediction variance, The variance of satellite clock error in the previous epoch. The noise intensity coefficient for the clock bias random walk process is recommended to be [value missing]. .

[0049] It should be noted that this embodiment uses the prediction mechanism of Kalman filtering to pass the historical best estimate to the current epoch, so that the clock difference parameter remains continuous in time and avoids the estimation instability caused by abrupt changes between epochs.

[0050] Next, based on the current epoch observation equations, a normal equation matrix and a right-hand matrix are constructed. Then, based on the current epoch state predictions, the forecast covariance matrix, the normal equation matrix, and the right-hand matrix, a joint parameter estimation equation is constructed. Parameter estimation is performed by solving this joint parameter estimation equation to obtain the optimal solution for state parameters, including satellite clock bias. The joint parameter estimation equation is expressed as: In the formula, The normal equation matrix, The right matrix of the normal equation, To design the coefficient matrix, The observation weight matrix, A constant vector, To predict the covariance matrix, These are the predicted state values. Obtained through solution. Satellite clock bias including the current epoch and receiver clock difference This allows us to obtain precise satellite clock bias and receiver clock bias.

[0051] It should be noted that this embodiment incorporates the predicted information into the normal equation in the form of a covariance matrix, forming a joint estimation framework that integrates prediction and observation. This ensures that the estimation results are simultaneously constrained by historical continuity and current observation information. Compared to existing methods that combine multi-source clock biases at the user end, this embodiment focuses on recalculating clock biases on the server side based on a large number of observations. This avoids errors caused by benchmark fluctuations or anomalies in a single analysis center, and the calculated clock biases have higher accuracy.

[0052] After obtaining the precise satellite clock bias and receiver clock bias, the pseudorange post-hoc residual is calculated based on both and the original ground observation data. The calculation method is as follows: In the formula, These are the original pseudorange observations without an ionosphere. Geometric distance For tropospheric projection functions, For the zenith tropospheric delay of the model, For residual tropospheric delay, and These are the estimated receiver clock bias and satellite clock bias, respectively.

[0053] The clock error constant deviation is calculated using the pseudorange post-verification residuals from multiple stations and multiple epochs. The calculation method is as follows: In the formula, The total number of ground monitoring stations participating in satellite observation. This represents the number of samples of the effective pseudorange residual for the satellite over consecutive epochs at each station. This clock error constant bias is then incorporated into the satellite code bias parameter to form the extended code bias parameter, expressed as: In the formula, The differential code bias (DCB) is the satellite's true inter-code bias due to hardware delay. The clock constant deviation calculated for the server.

[0054] It should be noted that this embodiment integrates the clock constant deviation and the actual inter-code deviation of the satellite into the spreading code deviation, so that the spreading code deviation includes both hardware delay and server-side systematic deviation. The user end does not need to process multiple deviation sources separately, which simplifies the user end processing flow.

[0055] Finally, the spreading code bias, precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris orbits and clock biases are combined to generate a real-time precise positioning correction data stream, which is then broadcast via the Networked Transport of RTCM via Internet Protocol (NTRIP). When user terminals perform positioning corrections based on this real-time precise positioning correction data stream, precise satellite clock bias corrections are applied to the pseudorange observations. The correction method is as follows: In the formula, For the corrected pseudorange, These are the original pseudorange observations. For the precise satellite clock bias broadcast by the server, For group delay parameters, This refers to the spreading code bias parameter. The user terminal only applies orbit and clock bias corrections to the phase observations, not spreading code bias corrections. The correction method is as follows: In the formula, For the corrected phase, These are the original phase observations. This is the track correction amount.

[0056] It should be noted that this embodiment integrates track correction, clock error correction, and code deviation correction into a unified data stream broadcast, enabling the user terminal to obtain complete precise positioning correction information through a single data source, simplifying the processing flow and ensuring consistency between each correction item, thereby obtaining target positioning results with higher accuracy and stronger continuity.

[0057] Furthermore, the step of acquiring and broadcasting a real-time precise positioning correction data stream based on the precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result, includes: Based on the broadcast ephemeris, obtain the broadcast ephemeris orbital parameters, broadcast ephemeris clock error parameters, and group delay parameters; Based on the fused real-time orbit corrections, the broadcast ephemeris orbit parameters are corrected to obtain real-time precise orbit correction parameters; Real-time precision clock error correction parameters are determined based on the precision satellite clock error and broadcast ephemeris clock error parameters; The real-time code deviation correction parameters are determined based on the aforementioned extended code deviation. Based on the real-time precise track correction parameters, real-time precise clock error correction parameters and real-time code deviation correction parameters, binary data stream encoding is performed to obtain the real-time precise positioning correction data stream. The real-time precise positioning correction data stream is continuously broadcast externally through a preset differential data network transmission protocol so that the user terminal can receive the real-time precise positioning correction data stream; Based on the real-time precise orbit correction parameters and real-time precise clock error correction parameters in the real-time precise positioning correction data stream, the original phase observation value of the user terminal is corrected to obtain the corrected phase observation value. Based on the real-time precise orbit correction parameters, real-time precise clock error correction parameters, real-time code deviation correction parameters and the group delay parameters in the real-time precise positioning correction data stream, the original pseudorange observation value of the user terminal is corrected to obtain the corrected pseudorange observation value. The target positioning result is obtained by performing positioning calculations based on the corrected phase observations and corrected pseudorange observations.

[0058] In one embodiment, a real-time precise positioning correction data stream is obtained and broadcast based on the precise satellite clock bias, fused real-time orbit correction data, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result, as detailed below.

[0059] Broadcast ephemeris orbit parameters, broadcast ephemeris clock bias parameters, and group delay parameters are obtained based on broadcast ephemeris data. The broadcast ephemeris orbit parameters are corrected based on fused real-time orbit corrections to obtain real-time precise orbit correction parameters; real-time precise clock bias correction parameters are determined based on precise satellite clock bias and broadcast ephemeris clock bias parameters; and real-time code bias correction parameters are determined based on spreading code bias. These real-time precise orbit correction parameters, real-time precise clock bias correction parameters, and real-time code bias correction parameters are encoded using RTCM 3.x binary format to form a real-time precise positioning correction data stream.

[0060] It should be noted that this embodiment effectively improves data transmission efficiency by controlling the magnitude of each correction parameter within a moderate range and encoding them in a compact binary format. Typical correction update rates are 5 seconds for both orbital corrections and clock error corrections.

[0061] The real-time precise positioning correction data stream is continuously broadcast externally via a preset differential data network transmission protocol. This preset differential data network transmission protocol is the Networked Transport of RTCM via Internet Protocol (NTRIP). When a user terminal accesses the network, it obtains the source information table through the NTRIP client, selects the mount point, completes basic authentication using the Hypertext Transfer Protocol, establishes a connection with the broadcaster, and continuously receives the real-time precise positioning correction data stream in binary data stream format.

[0062] After receiving the real-time precise positioning correction data stream, the user terminal corrects its original phase observations based on the real-time precise orbit correction parameters and real-time precise clock bias correction parameters to obtain corrected phase observations. Similarly, the user terminal corrects its original pseudorange observations based on the real-time precise orbit correction parameters, real-time precise clock bias correction parameters, real-time code bias correction parameters, and the group delay parameter to obtain corrected pseudorange observations. It should be noted that this embodiment implements differentiated corrections for the phase observations and pseudorange observations, ensuring that the phase observations maintain high accuracy and are unaffected by code bias, while the pseudorange observations receive complete system error correction. Finally, positioning calculations are performed based on the corrected phase observations and corrected pseudorange observations to obtain high-precision, highly continuous target positioning results.

[0063] Please see Figure 2 This embodiment also provides a real-time precision positioning system with multi-track fusion, including: The weight determination module is used to obtain historical samples of orbit accuracy assessment from several analysis centers, and to obtain the orbit fusion weight set corresponding to each analysis center based on the historical samples of orbit accuracy assessment. The data acquisition module is used to acquire raw ground observation data, broadcast ephemeris data, and real-time orbit corrections from several analysis centers; The track weighted fusion module is used to perform weighted fusion of the real-time track corrections of each analysis center based on the track fusion weight set corresponding to each analysis center, so as to obtain the fused real-time track corrections. The orbit correction module is used to correct and superimpose the broadcast ephemeris based on the fused real-time orbit correction data to obtain a real-time precise orbit. The clock error calculation module is used to establish observation equations based on the real-time precise orbit and raw ground observation data, and to calculate the clock error based on the observation equations to obtain the precise satellite clock error. The data broadcasting module is used to acquire and broadcast real-time precise positioning correction data streams based on the precise satellite clock error, fused real-time orbit correction data, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data streams and obtain the target positioning result.

[0064] In this embodiment, the weight determination module obtains an orbit fusion weight set based on historical samples of each orbit's accuracy assessment. It achieves differentiated weighting across multiple analysis centers through accuracy analysis of historical data, providing a weighting basis for subsequent fusion. The data acquisition module acquires raw ground observation data, broadcast ephemeris data, and real-time orbit corrections from each analysis center, responsible for multi-source data collection and aggregation, providing raw input to the system. The orbit weighted fusion module performs weighted fusion of the real-time orbit corrections from each analysis center based on the orbit fusion weight set, obtaining a fused real-time orbit correction, achieving robust fusion of multi-source orbit products and suppressing the impact of anomalies from a single analysis center. The orbit correction module corrects and superimposes the broadcast ephemeris based on the fused real-time orbit correction, obtaining a real-time precise orbit. The fused high-precision correction is applied to the broadcast ephemeris to improve orbit accuracy. The clock error calculation module establishes observation equations based on the real-time precise orbit and raw ground observation data and performs clock error calculation to obtain precise satellite clock errors. The clock errors are recalculated using the fused orbit as the geometric reference, avoiding inconsistencies in multi-source clock error references. The data broadcasting module acquires and broadcasts real-time precise positioning correction data streams based on precise satellite clock bias, real-time orbit correction data, and broadcast ephemeris data. This enables users to perform positioning corrections to obtain target positioning results. The server can uniformly generate and broadcast high-precision correction data, and the various modules work together to form a complete precise positioning service capability.

[0065] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A real-time precise positioning method using multi-track fusion, characterized in that, Includes the following steps: Obtain historical samples of orbit accuracy assessment from several analysis centers, and obtain the orbit fusion weight set corresponding to each analysis center based on the historical samples of orbit accuracy assessment. Acquire raw ground observation data, broadcast ephemeris, and real-time orbital corrections from several analysis centers; Based on the track fusion weight set corresponding to each analysis center, the real-time track corrections of each analysis center are weighted and fused to obtain the fused real-time track corrections. Based on the fused real-time orbit corrections, the broadcast ephemeris is corrected and superimposed to obtain a real-time precise orbit. Based on the real-time precise orbit and ground raw observation data, an observation equation is established, and based on the observation equation, the clock error is calculated to obtain the precise satellite clock error; Based on the precise satellite clock bias, the fused real-time orbit correction data, and the broadcast ephemeris, a real-time precise positioning correction data stream is obtained and broadcast, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result.

2. The real-time precise positioning method using multi-track fusion according to claim 1, characterized in that, The method of obtaining the orbit fusion weight set corresponding to each analysis center based on historical samples of orbit accuracy assessment includes: For each analysis center, historical real-time orbit data and corresponding post-event precise orbit data are obtained based on the historical samples of orbit accuracy assessment corresponding to that analysis center. A historical orbit deviation dataset is obtained based on the aforementioned historical real-time orbit data and post-event precise orbit data. Based on the historical orbital deviation dataset, orbital error direction decomposition is performed to obtain historical radial error data, historical tangential error data, and historical normal error data; The weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data, resulting in a track fusion weight set.

3. The real-time precise positioning method using multi-track fusion according to claim 2, characterized in that, The weights are determined based on a preset inverse distance weighting algorithm, historical radial error data, historical tangential error data, and historical normal error data to obtain a track fusion weight set. The preset inverse distance weighting algorithm is specifically as follows: Based on the historical radial error data, historical tangential error data, and historical normal error data, the corresponding root mean square values ​​of radial error, tangential error, and normal error are obtained respectively. The corresponding radial weight, tangential weight, and normal weight are determined based on the root mean square value of the radial error, the root mean square value of the tangential error, and the root mean square value of the normal error, respectively. The orbit fusion weight set is obtained based on the radial weight, tangential weight, and normal weight.

4. The real-time precise positioning method using multi-track fusion according to claim 1, characterized in that, The weighted fusion of real-time track corrections from each analysis center based on the track fusion weight set corresponding to each analysis center, to obtain the fused real-time track correction, includes: Based on the real-time orbit corrections of each analysis center, the radial orbit correction component, tangential orbit correction component, and normal orbit correction component corresponding to each analysis center are obtained respectively. Based on the orbit fusion weight set corresponding to each analysis center, the radial orbit correction component, tangential orbit correction component, and normal orbit correction component of each analysis center are weighted and summed to obtain the corresponding fused radial orbit correction component, fused tangential orbit correction component, and fused normal orbit correction component. The fused real-time orbit correction number is obtained based on the fused radial orbit correction component, the fused tangential orbit correction component, and the fused normal orbit correction component.

5. The real-time precise positioning method using multi-track fusion according to claim 1, characterized in that, The process of correcting and superimposing the broadcast ephemeris based on the fused real-time orbit corrections to obtain a real-time precise orbit includes: Based on the preset reference time and the fused real-time track correction number, the reference time track correction component and the reference time track correction velocity component are obtained; The current observation epoch, the satellite position vector at the current epoch, and the satellite velocity vector at the current epoch are obtained based on the broadcast ephemeris. The current orbit correction component is obtained based on the current observation epoch, the preset reference time, the orbit correction component at the reference time, and the orbit correction velocity component at the reference time. Based on the current epoch satellite position vector and current epoch satellite velocity vector, perform geocentric-ground-fixed coordinate transformation on the current orbit correction component to obtain the ground-fixed orbit correction component; The broadcast ephemeris is corrected based on the Earth-fixed orbit correction component to obtain the real-time precise orbit.

6. The real-time precise positioning method using multi-track fusion according to claim 1, characterized in that, The process of establishing observation equations based on the real-time precise orbit and raw ground observation data, and calculating clock bias based on the observation equations to obtain precise satellite clock bias includes: The current epoch observation equation set is constructed based on the preset set of parameters to be estimated, the real-time precise orbit, and the raw ground observation data. Obtain the optimal state estimate of the historical epoch, and make state prediction based on the optimal state estimate of the historical epoch and the preset random walk model to obtain the current epoch state prediction value and the prediction covariance matrix. The precise satellite clock bias is obtained by estimating parameters based on the current epoch state prediction, the forecast covariance matrix, and the current epoch observation equations.

7. The real-time precise positioning method using multi-track fusion according to claim 6, characterized in that, The precise satellite clock bias is obtained by estimating parameters based on the current epoch state prediction, the forecast covariance matrix, and the current epoch observation equations, including: Construct the normal equation matrix and the right normal equation matrix based on the current epoch observation equation set; Based on the current epoch state prediction value, the forecast covariance matrix, the normal equation matrix, and the right normal equation matrix, a joint parameter estimation equation is constructed. The precise satellite clock bias is obtained based on the joint parameter estimation equation.

8. The real-time precise positioning method using multi-track fusion according to claim 7, characterized in that, The process of obtaining real-time precise positioning correction data stream based on the precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris includes: The receiver clock bias is obtained based on the joint parameter estimation equation. The pseudorange post-test residual is obtained based on the precision satellite clock error, receiver clock error, and raw ground observation data. The spread code bias is obtained by fusing the post-pseudorange residual and the pre-acquired true inter-code bias of the satellite. The real-time precise positioning correction data stream is obtained based on the extended code deviation, precise satellite clock error, fused real-time orbit corrections, and broadcast ephemeris.

9. The real-time precise positioning method using multi-track fusion according to claim 8, characterized in that, The process of acquiring and broadcasting a real-time precise positioning correction data stream based on the precise satellite clock bias, fused real-time orbit corrections, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data stream to obtain the target positioning result, includes: Based on the broadcast ephemeris, obtain the broadcast ephemeris orbital parameters, broadcast ephemeris clock error parameters, and group delay parameters; Based on the fused real-time orbit corrections, the broadcast ephemeris orbit parameters are corrected to obtain real-time precise orbit correction parameters; Real-time precision clock error correction parameters are determined based on the precision satellite clock error and broadcast ephemeris clock error parameters; The real-time code deviation correction parameters are determined based on the aforementioned extended code deviation. Based on the real-time precise track correction parameters, real-time precise clock error correction parameters and real-time code deviation correction parameters, binary data stream encoding is performed to obtain the real-time precise positioning correction data stream. The real-time precise positioning correction data stream is continuously broadcast externally through a preset differential data network transmission protocol so that the user terminal can receive the real-time precise positioning correction data stream; Based on the real-time precise orbit correction parameters and real-time precise clock error correction parameters in the real-time precise positioning correction data stream, the original phase observation value of the user terminal is corrected to obtain the corrected phase observation value. Based on the real-time precise orbit correction parameters, real-time precise clock error correction parameters, real-time code deviation correction parameters and the group delay parameters in the real-time precise positioning correction data stream, the original pseudorange observation value of the user terminal is corrected to obtain the corrected pseudorange observation value. The target positioning result is obtained by performing positioning calculations based on the corrected phase observations and corrected pseudorange observations.

10. A real-time precision positioning system with multi-track fusion, characterized in that, include: The weight determination module is used to obtain historical samples of orbit accuracy assessment from several analysis centers, and to obtain the orbit fusion weight set corresponding to each analysis center based on the historical samples of orbit accuracy assessment. The data acquisition module is used to acquire raw ground observation data, broadcast ephemeris data, and real-time orbit corrections from several analysis centers; The track weighted fusion module is used to perform weighted fusion of the real-time track corrections of each analysis center based on the track fusion weight set corresponding to each analysis center, so as to obtain the fused real-time track corrections. The orbit correction module is used to correct and superimpose the broadcast ephemeris based on the fused real-time orbit correction data to obtain a real-time precise orbit. The clock error calculation module is used to establish observation equations based on the real-time precise orbit and raw ground observation data, and to calculate the clock error based on the observation equations to obtain the precise satellite clock error. The data broadcasting module is used to acquire and broadcast real-time precise positioning correction data streams based on the precise satellite clock error, fused real-time orbit correction data, and broadcast ephemeris, so that the user terminal can perform positioning correction based on the real-time precise positioning correction data streams and obtain the target positioning result.

Images