Method for correcting sky jump phenomenon of high-precision navigation data of artificial satellite
By combining the coupling of track and clock difference in navigation data processing, and using high-order Lagrangian algorithm and smooth window correction method, the problem of navigation data sky jump phenomenon is solved, efficient and accurate navigation data processing is achieved, and user burden is reduced.
Patent Information
- Application Number
- CN202411672884.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-21
- Publication Date
- 2025-03-04
AI Technical Summary
The existing navigation data generation method based on the server-side level cannot completely eliminate the sky jump phenomenon of navigation data, and the processing time is too long and the timeliness is low; the user-side level solution increases the burden on users to use services and requires too high users.
Simple mathematical methods are used to correct the precision track, and the coupling between the track and clock difference is synchronously corrected; the 11th order Lagrangian algorithm is used to interpolate and fit the track data, and the pseudo-random pulse parameter absorption data is added; the equal weight correction and time weight correction methods are used based on smooth windows to correct the track data of adjacent days, and the corrected data is sparse.
The sky jump phenomenon of navigation data is completely eliminated, which significantly reduces the computational complexity and processing time, reduces the user's usage cost, and ensures the consistency between track data and clock difference data.
Smart Images
Figure CN119556312A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of high-precision positioning and navigation, and in particular to a method for correcting sky jump phenomenon of high-precision navigation data of an artificial satellite. Background Art
[0002] The mainstream method for determining the navigation data of navigation satellites is to use simplified dynamic algorithm processing based on data from a large number of ground observation stations that are evenly distributed around the world. Satellite navigation data provided by domestic and international organizations and institutions are broadcast in a standard format. When users use these navigation data for subsequent precise positioning or precise position determination of low-orbit satellites, they need to interpolate according to their needs. The mainstream method for determining the precise position of low-orbit satellites is to use a simplified dynamic algorithm processing to receive signals from navigation satellites through the onboard receiver payload. This method requires clear and accurate information about the navigation satellite in advance, and because the observation data of the onboard receiver is usually 30s or 1s, the navigation data of the navigation satellite also needs to be interpolated.
[0003] On the one hand, whether it is a navigation satellite or a low-orbit satellite, it is usually divided into final products, rapid products, and ultra-rapid products according to different products. The arc length varies according to the organization of each organization, and is roughly divided into 1 day, 3 days, 5 days, and 7 days. The long arc segment navigation data after solution needs to be extracted again according to the standard file format requirements to only extract 24 hours of data as the final broadcast data file. The processing time limit of the adjustment method leads to the discontinuity of satellite navigation data, which will inevitably lead to the phenomenon of jumps between days when users use the above-mentioned precise navigation data for subsequent scientific research. On the other hand, when determining precise navigation data, navigation satellites and low-orbit satellites use simplified dynamic methods for processing. This method is based on satellite dynamic analysis and perturbation dynamic model construction. However, due to the complex force conditions of the satellite, it is impossible to comprehensively construct all force condition analyses; and there is currently no reasonable means to accurately model some mechanical effects. Therefore, in the process of determining precise navigation data, the influence of these error terms cannot be ignored, which will inevitably affect the continuity of satellite navigation data, especially between days.
[0004] At the server level, the current mainstream method for generating navigation data is to use a multi-day continuous solution, that is, to process arc lengths set to 3, 5, and 7 days at the same time, and then only take the middle day as the final product. This method is widely used by various analysis centers, such as the German Geoscience Center (GFZ), the European Orbit Determination Center (CODE), and the IGS Center of Wuhan University. This method can improve the accuracy of satellite navigation data to a certain extent, and can also reduce the jump phenomenon. At the user-side level, Dr. Song Chuanfeng of Wuhan University improved the orbital position interpolation algorithm by estimating the zero-order to second-order deviations between the orbits of two adjacent days, thereby reducing the jump phenomenon of satellite orbital position parameters. In order to eliminate the influence of satellite orbit day jumps, a smooth satellite orbit position can also be obtained by orbit fitting of satellite orbit file data for multiple days. Researcher Zhang Pengfei of the National Time Service Center of the Chinese Academy of Sciences adopted the edge effect strategy of the sliding window technology, and Professor Lou Yidong of Wuhan University reduced the jump of satellite clock error data through clock error modeling. Liu Zhanke and others proposed a weight function model based on the estimation of satellite orbit errors to optimize the real-time satellite clock error estimation strategy, thereby improving the accuracy of satellite clock errors. Many scholars at home and abroad, such as Zhang Xiaohong and Li Xin, have improved the accuracy and stability of ground user positioning by using ambiguity fixation technology for ground positioning.
[0005] However, the above-mentioned solution based on the server level cannot completely eliminate the jumping phenomenon of navigation data, and the processing time is too long and the timeliness is low. The solution based on the user level increases the burden of users using the service and places too high requirements on users. Summary of the invention
[0006] The purpose of the present invention is to provide a method for correcting the sky jump phenomenon of high-precision navigation data of artificial satellites, which is used to solve the problems that the existing solutions based on the server level cannot completely eliminate the sky jump phenomenon of navigation data, and the processing time is too long and the timeliness is low. The solutions based on the user level increase the burden on users to use the service and have too high requirements on users.
[0007] To achieve the above object, the present invention provides a method for correcting the sky jump phenomenon of high-precision navigation data of an artificial satellite, comprising:
[0008] S1. Use simple mathematical methods to correct the precise orbit. According to the coupling between the precise orbit and the precise clock error, the precise orbit correction value is synchronously corrected to the clock error to maintain the consistency of the orbit and the clock error.
[0009] S2. The orbital data needs to be interpolated. The interpolation density is based on the corresponding clock error interval. The interpolation method uses the 11th-order Lagrangian algorithm. Then the encrypted orbital data is fitted and predicted. On the basis of the traditional dynamical orbit fitting prediction, pseudo-random pulse parameters are added to absorb the differences in orbital data of different types and sources. The fitting time is 24 hours, and the prediction time is 1.5 hours forward and backward respectively.
[0010] S3. Correct the overlapping parts of orbital data after expansion on adjacent days, propose equal-weight correction based on smoothing window and time-weighted correction methods, and finally thin out the corrected orbital data, and intercept the 24-hour data starting from 0:00 every day as the final orbital data according to the international standard protocol;
[0011] S4. Compare the original orbit data with the corrected orbit data to obtain the tangential, normal and radial orbit correction values in the orbital coordinate system. Based on the clock error data interval, the method of recalculating the clock error data using global ground stations instead of fixed orbit data is used. On the basis of correcting only the radial orbit correction value to the clock error data method, the tangential and normal orbit correction values are corrected synchronously.
[0012] 2. A method for correcting sky jumps in high-precision navigation data of an artificial satellite according to claim 1, characterized in that the 11-order Lagrange interpolation polynomial in S2 can be expanded as:
[0013]
[0014] Where: x i Represents the horizontal coordinate of the known data point; x j Divide x i The horizontal coordinates of other data points other than y i For x i The corresponding vertical coordinate value.
[0015] 3. The method for correcting the sky jump phenomenon of high-precision navigation data of an artificial satellite according to claim 1, characterized in that the dynamic model in S2. can be written as:
[0016]
[0017] Where: Δv k is at time t k Applied velocity change (pulse); δ(tt k ) indicates that at t k A transient pulse that occurs at a certain point in time.
[0018] 4. The method for correcting sky jump phenomenon of high-precision navigation data of an artificial satellite according to claim 1, characterized in that S3. mid-orbit correction can be expressed as:
[0019] orb_smooth_beg(t)=W_beg(t)*orb_prev(t)+(1-W_beg(t))*orb_curr(t)……(a)
[0020] orb_smooth_end(t)=W_end(t)*orb_curr(t)+(1-W_end(t))*orb_next(t)……(b)
[0021] Among them: the suffix _beg indicates the beginning of the day; the suffix _end indicates the end of the day; orb_prev, orb_curr, orb_next indicate the orbital data of the previous day, the current day, and the next day respectively; W indicates the weight of the orbital data. One is to make it equal to 0.5, so that there will be no difference between the sub-formulas (a) and (b) in formula 1; the other is to establish a smoothing window based on time, as shown below:
[0022] W_beg(t)=(1+cos(ω(t-tcentroid_start+smooth_window))) / 2.0......(a)
[0023] W_end(t)=(1+cos(ω(t-tcentroid_end+smooth_window))) / 2.0......(b)
[0024] Where: tcentroid_start is 23:00 of the previous day; centroid_end is 01:00 of the next day; smooth_window is equal to 3600s; ω = 0.5π / smooth_window.
[0025] 5. The method for correcting the day jump phenomenon of high-precision artificial satellite navigation data according to claim 1, characterized in that, in S4, the original clock error data can be written as follows:
[0026]
[0027] Where: clk_raw represents the original navigation clock error data; It means projecting the corrections of the corrected orbit and the original orbit data in the tangential and normal directions to the radial direction; It represents the correction amount in the radial direction between the corrected orbit and the original orbit data; c represents the speed of light.
[0028] Compared with the prior art, the advantages of the present invention are:
[0029] 1. The solution of the present invention directly uses the navigation data of the server for correction and secondary processing, and does not need to use large-scale ground stations to solve the server-side products. This solution greatly reduces the computational complexity, shortens the processing time, and can completely eliminate the sky jump phenomenon of navigation data;
[0030] 2. The solution of the present invention can directly provide navigation orbit data and clock error data with sufficient accuracy and no sky jump phenomenon. This solution does not require the user end to perform redundant processing, which greatly reduces the user's usage cost.
[0031] 3. The solution of the present invention takes into account the coupling of orbit data and clock error data, and adopts a simple and scientific method to correct the clock error data to ensure the consistency of orbit data and clock error data.
[0032] 4. The solution of the present invention corrects the clock error through the error absorption characteristics between orbit data and clock error data, and does not need to re-use large-scale ground stations to solve the server-side product. This solution greatly reduces the computational complexity and processing time. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 The present invention is a structural schematic diagram of a method for correcting the sky jump phenomenon of high-precision navigation data of artificial satellites.
[0034] Figure 2 Fitting and prediction accuracy of navigation orbit data from different sources.
[0035] Figure 3 Simplified dynamic orbit accuracy for LEO satellites with different sources of navigation data as input.
[0036] Figure 4 Kinematic orbit accuracy of low-orbit satellites using different source navigation data as input.
[0037] Figure 5 Simplified dynamical orbit sky-jump variables for low-orbit satellites.
[0038] Figure 6 is the kinematic orbit sky jump variable of the low-orbit satellite.
[0039] Figure 7 It is the time series of dynamic precise point positioning results of a single GPS system.
[0040] Figure 8 It is the time series of dynamic precise point positioning results of a single BDS-2 system.
[0041] Fig. 9It is the time series of dynamic precise point positioning results of a single BDS-3 system. DETAILED DESCRIPTION
[0042] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.
[0043] like Figure 1 As shown, in an embodiment of the present invention, a method for correcting a sky jump phenomenon of high-precision navigation data of an artificial satellite includes:
[0044] S1. Use simple mathematical methods to correct the precise orbit. According to the coupling between the precise orbit and the precise clock error, the precise orbit correction value is synchronously corrected to the clock error to maintain the consistency of the orbit and the clock error.
[0045] S2. The orbital data needs to be interpolated. The interpolation density is based on the corresponding clock error interval. The interpolation method uses the 11th-order Lagrangian algorithm. Then the encrypted orbital data is fitted and predicted. On the basis of the traditional dynamical orbit fitting prediction, pseudo-random pulse parameters are added to absorb the differences in orbital data of different types and sources. The fitting time is 24 hours, and the prediction time is 1.5 hours forward and backward respectively.
[0046] Furthermore, on the basis of traditional dynamic orbit fitting and prediction, pseudo-random pulse parameters are added to absorb the errors caused by differences in dynamic parameters, error models, etc. in the server-side solution process of different navigation data. The fitting and prediction errors of navigation orbit data from different sources are as follows: Figure 2 As shown, the accuracy can reach within 2mm and 4mm, which is negligible compared with the centimeter-level orbit jump variable of orbit data. This greatly reduces the deviation of the orbit data day jump variable quantified by the error introduced by the orbit data fitting prediction method.
[0047] S3. Correct the overlapping parts of orbital data after expansion on adjacent days, propose equal-weight correction based on smoothing window and time-weighted correction methods, and finally thin out the corrected orbital data, and intercept the 24-hour data starting from 0:00 every day as the final orbital data according to the international standard protocol;
[0048] S4. Compare the original orbit data with the corrected orbit data to obtain the orbit correction values in the tangential, normal and radial directions in the orbit coordinate system. Based on the clock error data interval, the method of recalculating the clock error data using global ground stations without using fixed orbit data, and on the basis of correcting only the radial direction orbit correction value to the clock error data method, the tangential and normal orbit correction values are corrected synchronously;
[0049] For the corrected navigation data, use the low-orbit satellite precise orbit determination results to verify whether its accuracy meets the standards. Figure 3 , Figure 4 These are the simplified dynamic and kinematic orbit accuracies of low-orbit satellites using different source navigation data as input. The difference in the accuracy of the simplified dynamic orbit of the onboard GPS is about 1mm, and the difference in the accuracy of the onboard GPS kinematic orbit is about 2mm; the difference in the accuracy of the simplified dynamic orbit of the onboard BDS is within 10mm. It can be seen that the corrected precise orbit and clock error products can still be well applied to subsequent scientific research and applications. INPUTⅠ, Ⅱ, and Ⅲ represent the use of original orbit data, orbit data after equal weight correction, and orbit data after time weight correction as input, respectively.
[0050] In the stability verification of low-orbit satellites using corrected navigation data:
[0051] InputⅠ, Ⅱ, and Ⅲ represent the original orbit data, orbit data after equal weight correction, and orbit data after time weight correction as input. The sky jump phenomenon of the simplified dynamics and kinematic orbits of low-orbit satellites was compared using different input schemes. The results are shown in Figure 2. Figure 5 , Figure 6 As shown in the figure. Compared with InputⅡ and InputⅢ, InputⅠ has smaller sky jump variable values; except for FY 3D_BDS 2 mode, the sky jump variable of most simplified dynamic low-orbit satellite orbits is about 20mm, and the sky jump variable of kinematic low-orbit satellite orbits is between 30mm and 40mm. This is because the simplified dynamic orbit itself has continuous characteristics, rather than the randomness of kinematic scatter points, and it is also related to the higher orbit accuracy of the former; the sky jump variable of the simplified dynamic orbit in FY 3D_BDS2 mode reaches more than 300mm, because the precise BDS2 orbit and clock error used in this mode have quality defects, and the error of BDS2 satellite greatly affects the orbit accuracy of low-orbit satellites.
[0052] In the stability verification of user positioning using corrected navigation data:
[0053] InputⅠ, Ⅱ, and Ⅲ represent the use of original orbit data, orbit data after equal weight correction, and orbit data after time weight correction as input. The time series of single epoch dynamic precise point positioning are evaluated using different input schemes. Figure 7 , Figure 8 ,as well as Fig. 9These are the precise single-point positioning results using data from a single GPS system, a single BDS-2 system, and a single BDS-3 system. The two correction methods for the discontinuity of the sky boundary of GNSS products proposed above are also effective in the dynamic PPP experiment. After correction, the GNSS products have a significant improvement effect mainly in the U direction of dynamic PPP. For the GPS and BDS-3 systems, the positioning results near the sky boundary that are much larger than the non-boundary period can be pulled back to normal accuracy; for the BDS-2 system, the above two improvement methods can effectively improve the positioning accuracy near the boundary, but due to the small number of available satellites, its positioning accuracy is still poor.
[0054] In the embodiment of the present invention, the 11th-order Lagrange interpolation polynomial in S2 can be expanded as follows:
[0055]
[0056] Where: x i Represents the horizontal coordinate of the known data point; x j Divide x i The horizontal coordinates of other data points other than y i For x i The corresponding vertical coordinate value.
[0057] In the embodiment of the present invention, the kinetic model in S2. can be written as:
[0058]
[0059] Where: Δv k is at time t k Applied velocity change (pulse); δ(tt k ) indicates that at t k A transient pulse that occurs at a certain point in time.
[0060] In the embodiment of the present invention, the orbit correction in S3. can be expressed as:
[0061] orb_smooth_beg(t)=W_beg(t)*orb_prev(t)+(1-W_beg(t))*orb_curr(t)……(a)
[0062] orb_smooth_end(t)=W_end(t)*orb_curr(t)+(1-W_end(t))*orb_next(t)……(b)
[0063] Among them: the suffix _beg indicates the beginning of the day; the suffix _end indicates the end of the day; orb_prev, orb_curr, orb_next indicate the orbital data of the previous day, the current day, and the next day respectively; W indicates the weight of the orbital data. One is to make it equal to 0.5, so that there will be no difference between the sub-formulas (a) and (b) in formula 1; the other is to establish a smoothing window based on time, as shown below:
[0064] W_beg(t)=(1+cos(ω(t-tcentroid_start+smooth_window))) / 2.0......(a)
[0065] W_end(t)=(1+cos(ω(t-tcentroid_end+smooth_window))) / 2.0......(b)
[0066] Where: tcentroid_start is 23:00 of the previous day; centroid_end is 01:00 of the next day; smooth_window is equal to 3600s; ω = 0.5π / smooth_window.
[0067] In the embodiment of the present invention, in S4, the original clock error data for settlement can be written as follows:
[0068]
[0069] Where: clk_raw represents the original navigation clock error data; It means projecting the corrections of the corrected orbit and the original orbit data in the tangential and normal directions to the radial direction; It represents the correction amount in the radial direction between the corrected orbit and the original orbit data; c represents the speed of light.
[0070] Furthermore,
[0071] Compared with the prior art, the advantages of the present invention are:
[0072] 1. The solution of the present invention directly uses the navigation data of the server for correction and secondary processing, and does not need to use large-scale ground stations to solve the server-side products. This solution greatly reduces the computational complexity, shortens the processing time, and can completely eliminate the sky jump phenomenon of navigation data;
[0073] 2. The solution of the present invention can directly provide navigation orbit data and clock error data with sufficient accuracy and no sky jump phenomenon. This solution does not require the user end to perform redundant processing, which greatly reduces the user's usage cost.
[0074] 3. The solution of the present invention takes into account the coupling of orbit data and clock error data, and adopts a simple and scientific method to correct the clock error data to ensure the consistency of orbit data and clock error data.
[0075] 4. The solution of the present invention corrects the clock error through the error absorption characteristics between orbit data and clock error data, and does not need to re-use large-scale ground stations to solve the server-side product. This solution greatly reduces the computational complexity and processing time.
[0076] The preferred embodiments of the present invention are described in detail above, but the present invention is not limited to the above embodiments. Various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.
Claims
1. A method for correcting sky jump phenomenon of high-precision navigation data of artificial satellites, characterized in that: include: S1. Use simple mathematical methods to correct the precise orbit. According to the coupling between the precise orbit and the precise clock error, the precise orbit correction value is synchronously corrected to the clock error to maintain the consistency of the orbit and the clock error. S2. The orbital data needs to be interpolated. The interpolation density is based on the corresponding clock error interval. The interpolation method uses the 11th-order Lagrangian algorithm. Then the encrypted orbital data is fitted and predicted. On the basis of the traditional dynamical orbit fitting prediction, pseudo-random pulse parameters are added to absorb the differences in orbital data of different types and sources. The fitting time is 24 hours, and the prediction time is 1.5 hours forward and backward respectively. S3. Correct the overlapping parts of orbital data after expansion on adjacent days, propose equal-weight correction based on smoothing window and time-weighted correction methods, and finally thin out the corrected orbital data, and intercept the 24-hour data starting from 0:00 every day as the final orbital data according to the international standard protocol; S4. Compare the original orbit data with the corrected orbit data to obtain the tangential, normal and radial orbit correction values in the orbital coordinate system. Based on the clock error data interval, the method of recalculating the clock error data using global ground stations instead of fixed orbit data is used. On the basis of correcting only the radial orbit correction value to the clock error data method, the tangential and normal orbit correction values are corrected synchronously.
2. The method for correcting sky jump phenomenon of high-precision artificial satellite navigation data according to claim 1, characterized in that: The 11th-order Lagrange interpolation polynomial described in S2 can be expanded as: Where: x i Represents the horizontal coordinate of the known data point; x j Divide x i The horizontal coordinates of other data points other than y i For x i The corresponding vertical coordinate value.
3. The method for correcting sky jump phenomenon of high-precision artificial satellite navigation data according to claim 1, characterized in that: The kinetic model in S2. can be written as: Where: Δv k is at time t k Applied velocity change (pulse); δ(tt k ) indicates that at t k A transient pulse that occurs at a certain point in time.
4. The method for correcting sky jump phenomenon of high-precision artificial satellite navigation data according to claim 1, characterized in that: S3. The mid-track correction can be expressed as: orb_smooth_beg(t)=W_beg(t)*orb_prev(t)+(1-W_beg(t))*orb_curr(t)……(a) orb_smooth_end(t)=W_end(t)*orb_curr(t)+(1-W_end(t))*orb_next(t)……(b) Among them: the suffix _beg indicates the beginning of the day; the suffix _end indicates the end of the day; orb_prev, orb_curr, orb_next indicate the orbital data of the previous day, the current day, and the next day respectively; W indicates the weight of the orbital data. One is to make it equal to 0.5, so that there will be no difference between the sub-formulas (a) and (b) in formula 1; the other is to establish a smoothing window based on time, as shown below: W_beg(t)=(1+cos(ω(t-tcentroid_start+smooth_window))) / 2.0......(a) W_end(t)=(1+cos(ω(t-tcentroid_end+smooth_window))) / 2.0......(b) Where: tcentroid_start is 23:00 of the previous day; centroid_end is 01:00 of the next day; smooth_window is equal to 3600s; ω = 0.5π / smooth_window.
5. The method for correcting sky jump phenomenon of high-precision artificial satellite navigation data according to claim 1, characterized in that: In S4, the original settlement clock error data can be written as follows: Where: clk_raw represents the original navigation clock error data; It means projecting the corrections of the corrected orbit and the original orbit data in the tangential and normal directions to the radial direction; It represents the correction amount in the radial direction between the corrected orbit and the original orbit data; c represents the speed of light.