Spacecraft precise orbit determination parallel processing method

Abstract

The application relates to a spacecraft precise orbit determination parallel processing method. The method comprises the following steps: opening a corresponding number of integral thread handles according to the number of spacecrafts; performing orbit integral calculation on the spacecrafts to obtain ephemeris data of the spacecrafts and save the ephemeris data; opening observation data processing thread handles with a preset number, dividing observation data of the spacecrafts into the preset number of thread blocks, and allocating an upper triangular matrix and a vector to each thread block; obtaining partial derivatives of to-be-solved parameters, differences between theoretical values and measured values of observation data according to the ephemeris data and the observation data in each thread block, and substituting the partial derivatives and the differences into Givens transformation to update the upper triangular matrix and the vector; performing Givens transformation on all the updated upper triangular matrices and vectors to obtain orbit confirmation results. The application can overcome the deficiency of the existing precise orbit determination system in the aspect of calculation efficiency when processing massive observation data.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of spacecraft precise orbit determination, and in particular to a spacecraft precise orbit determination parallel processing method. BACKGROUND

[0002] At present, space powers at home and abroad are implementing space launch tasks on a large scale. New large constellations, multi-satellite formation and other applications are numerous. These application scenarios not only use traditional ground-based ranging, angle measurement, speed measurement and other observation quantities to implement orbit calculation, but also need to use inter-satellite ranging and other new observation quantities. These new requirements have greatly increased the number of satellite objects participating in orbit determination and the amount of observation data in precise orbit determination calculation, greatly reducing the time efficiency of orbit calculation. At present, multi-core and many-core architecture computer systems have become relatively popular.

[0003] Therefore, it is necessary to improve one or more problems existing in the above-mentioned related technical solutions.

[0004] It should be noted that this section aims to provide background or context to the technical solutions of the present disclosure stated in the claims. The description herein is not admitted to be prior art merely because it is included in this section. SUMMARY

[0005] The purpose of the embodiments of the present disclosure is to provide a spacecraft precise orbit determination parallel processing method, thereby at least overcoming one or more problems caused by the limitations and defects of the related art.

[0006] According to an embodiment of the present disclosure, a spacecraft precise orbit determination parallel processing method is provided, comprising:

[0007] According to the number of spacecraft, a corresponding number of integral thread handles are opened;

[0008] Orbit integral calculation is performed on the spacecraft to obtain ephemeris data of the spacecraft and save it;

[0009] A preset number of observation data processing thread handles are opened, and the observation data of the spacecraft is divided into the preset number of thread blocks, and each thread block is allocated an upper triangular matrix and a vector;

[0010] According to the ephemeris data and the observation data in each thread block, the partial derivative of the to-be-solved parameter, the difference between the theoretical value and the measured value of the observation data are obtained, the partial derivative and the difference between the theoretical value and the measured value of the observation data are substituted into the Givens transformation to update the upper triangular matrix and the vector;

[0011] All updated upper triangular matrices and vectors are decomposed by rows and subjected to Givens transformation, and an orbit confirmation result is obtained.

[0012] In an embodiment of the present disclosure, after the corresponding number of integral thread handles are opened, the method further comprises:

[0013] The initial orbit parameters of the spacecraft constellation are preset, and the initial orbit parameters comprise dynamic parameters, physical property parameters, initial orbit six elements at an orbit epoch, integral start and end times, and an orbit time interval.

[0014] In an embodiment of the present disclosure, the process of obtaining the partial derivative of the to-be-solved parameter and the difference between the theoretical value and the measured value of the observation data is as follows:

[0015] According to the ephemeris data, an observation model is obtained, and then the partial derivative of the to-be-solved parameter and the theoretical value of the observation data are obtained according to the observation model, and the difference between the theoretical value and the measured value of the observation data is calculated.

[0016] In an embodiment of the present disclosure, the measured value is obtained according to the observation data.

[0017] In an embodiment of the present disclosure, the upper triangular matrix and the vector are allocated to each thread block, and the following conditions are met:

[0018] R i ·x=d i ,i=1,2,...,M (1)

[0019] wherein R is an upper triangular matrix, d is a vector, x is a to-be-solved parameter, and M is the number of thread blocks.

[0020] In an embodiment of the present disclosure, before the upper triangular matrix and the vector calculated for all the thread blocks are subjected to Givens transformation, the following process is further included:

[0021] The upper triangular matrix and the vector are decomposed by row, that is:

[0022]

[0023] wherein K is the number of parameters of an orbit calculation program.

[0024] In an embodiment of the present disclosure, after the upper triangular matrix and the vector calculated for all the thread blocks are decomposed by row, the upper triangular matrix and the vector are subjected to Givens transformation, and finally all the upper triangular matrices and the vectors are integrated to obtain:

[0025] R·x=d (3)

[0026] Solving the equation can obtain an improved spacecraft orbit state.

[0027] In an embodiment of the present disclosure, the measurement data is on-board observation data and ground station observation data.

[0028] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0029] In the embodiments of this disclosure, the above-described parallel processing method for precise orbit determination of spacecraft is used to process the orbit integration of each spacecraft in parallel and to divide the massive observation data into multiple thread blocks for parallel processing, which greatly improves the computational efficiency and overcomes the shortcomings of existing precise orbit determination systems in terms of computational efficiency when processing massive observation data of large constellations. Attached Figure Description

[0030] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0031] Figure 1 This diagram illustrates the steps of the parallel processing method for precise orbit determination of spacecraft disclosed herein.

[0032] Figure 2 The flowchart of the parallel processing method for precise orbit determination of spacecraft disclosed herein is shown. Detailed Implementation

[0033] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0034] Furthermore, the accompanying drawings are merely illustrative diagrams of embodiments of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities.

[0035] This example implementation first provides a parallel processing method for precise orbit determination of spacecraft. (Reference) Figure 1 As shown, the parallel processing method for precise orbit determination of a spacecraft may include steps S101 to S105.

[0036] Step S101: Create a corresponding number of integration thread handles based on the number of spacecraft;

[0037] Step S102: Orbit integration calculation is performed on each of the spacecrafts to obtain ephemeris data of each of the spacecrafts and save the ephemeris data;

[0038] Step S103: A preset number of observation data processing thread handles are opened, and observation data of the spacecrafts are divided into the preset number of thread blocks, and an upper triangular matrix and a vector are allocated to each of the thread blocks.

[0039] Step S104: Partial derivatives of to-be-solved parameters, differences between theoretical values and measured values are obtained according to the ephemeris data and observation data in each of the thread blocks, the partial derivatives of the pairs of unconnected parameters, the differences between the theoretical values and the measured values are substituted into Givens transformation to update the upper triangular matrix and the vector.

[0040] Step S105: The upper triangular matrix and the vector obtained after updating are decomposed by row and then subjected to Givens transformation to obtain orbit confirmation results.

[0041] Through the above-mentioned spacecraft precise orbit determination parallel processing method, orbit integration of each spacecraft is processed in parallel, and massive observation data is divided into multiple thread blocks and processed in parallel, which greatly improves the calculation efficiency and overcomes the deficiencies of the existing precise orbit determination system in terms of calculation efficiency when processing massive observation data of a large constellation.

[0042] In the following, the above-mentioned steps of the spacecraft precise orbit determination parallel processing method in the present example embodiment will be described in more detail. Figure 1 The above-mentioned steps of the spacecraft precise orbit determination parallel processing method in the present example embodiment will be described in more detail.

[0043] Step S101: A corresponding number of integration thread handles are opened according to the number of spacecrafts (satellites).

[0044] Specifically, integration thread handles are opened according to the number of spacecrafts participating in orbit calculation in the constellation, for example, N spacecrafts, and N integration thread handles are opened.

[0045] In an embodiment, before the corresponding number of integration thread handles are opened, it further includes: presetting initial orbit parameters of a spacecraft constellation, the initial orbit parameters including dynamic parameters, physical property parameters, orbit epoch time initial orbit six elements, integration start and end time, and orbit time interval. Specifically, the dynamic parameters and physical property parameters of all satellites participating in orbit determination are initialized first.

[0046] Step S102: Orbit integration calculation is performed on each of the spacecrafts to obtain ephemeris data of each of the spacecrafts and save the ephemeris data.

[0047] Specifically, the dynamics, physical properties, orbit ephemeris initial orbit six parameters, integral start and end time, orbit time interval and other parameters of each satellite are input into the integral function, and the thread calculation module of all satellites is started. After waiting for all thread calculations to be completed, the ephemeris of each satellite calculated is stored in the global storage area.

[0048] Step S103: opening a handle with a preset number of observation data processing threads, and dividing the observation data of the spacecraft into the preset number of thread blocks, and assigning an upper triangular matrix and a vector to each thread block.

[0049] Specifically, the number of observation data thread blocks is set, and a corresponding number of observation data processing thread handles are opened. Assuming that the number of observation data thread blocks is M, M observation data processing thread handles are opened, and the observation data of the spacecraft is divided into M thread blocks, and an upper triangular matrix and a vector are assigned to each observation data thread block, satisfying:

[0050] R i ·x=d i ,i=1,2,...,M (1)

[0051] Where R is an upper triangular matrix, d is a vector, x is a parameter to be solved, and M is the number of thread blocks.

[0052] Step S104: obtaining the partial derivative of the parameter to be solved, the difference between the theoretical value and the measured value according to the ephemeris data and the observation data in each thread block, and substituting the partial derivative of the unconnected parameter pair and the difference between the theoretical value and the measured value into the Givens transformation to update the upper triangular matrix and the vector;

[0053] Specifically, the data processing thread is started, the observation model is obtained according to the ephemeris data, the partial derivative of the parameter to be solved and the theoretical value are obtained, and the difference between the theoretical value and the measured value is calculated. Substitute these two calculation parameters into the Givens transformation to update the matrix and vector of each thread block to obtain M sets of linear equations.

[0054] Step S105: performing Givens transformation on the upper triangular matrix and the vector obtained after updating all, and obtaining the orbit confirmation result.

[0055] Specifically, after the matrix and vector of each thread block are updated, the observation information of each thread block is contained in its respective normal equation, and for the final orbit determination calculation, they need to be integrated. R i and d i are decomposed by row, and if the number of parameters to be solved by the orbit calculation program is K, then:

[0056]

[0057] r ij and δ ij , i = 1, 2,..., M, j = 1, 2,..., K are input into the Givens transformation, and finally M groups of R i and d i are integrated into one group of R and d to form the normal equation

[0058] R·x=d (3)

[0059] Solving the equation can obtain the improved spacecraft orbit state.

[0060] Through the above-mentioned spacecraft precise orbit determination parallel processing method, the orbit integration of each spacecraft is processed in parallel, and the massive observation data is divided into multiple thread blocks for parallel processing, which greatly improves the calculation efficiency and overcomes the deficiencies of the existing precise orbit determination system in terms of calculation efficiency when processing massive observation data of a large constellation.

[0061] The application will be further described below in combination with specific examples.

[0062] Taking the orbit determination of a constellation composed of 10 simulated GPS navigation satellites as an example, the embodiments of the application are as follows:

[0063] 1. Assume that the initial orbit parameters of the satellite constellation are set as follows:

[0064] Table 1. Initial orbit parameters of the satellite constellation

[0065]

[0066] The satellite dynamics parameters are set as follows: 50x50 order non-spherical perturbation, three-body gravitational perturbation considering the influence of the sun and the moon, and 9-parameter ECOM model is used for the light pressure model. The number of integration threads is set to 10, and the number of data processing threads is set to 32. The types of measurement data used are inter-satellite link ranging data and ground station ranging data.

[0067] 2. Constellation orbit integration

[0068] Ten orbit integration calculation threads are started, and the initial orbit parameters of the above 10 satellites are input into the integrator for integration calculation within the observation data time period. The orbit integration results of the 10 satellites are stored in the storage area at a step of 10s.

[0069] 3. Parallel processing of observation data

[0070] In this embodiment, the observation data before and after the coverage range involves 3 days, a total of 100081. In order to implement parallel computing, the block is divided according to the set number of threads. The 1-31 thread block is allocated 3127 observation data, and the 32 thread block is allocated 3144 observation data. Start the data processing flow, and each thread block processes its own observation data at the same time, and performs Givens transformation respectively. Thus each thread block gets its own normal equation:

[0071]

[0072] 4, the normal equation is synthesized

[0073] After completing step 3, the observation information of each thread block is contained in its respective normal equation, in order to perform the final orbit determination calculation, it needs to be merged. R i and d i are decomposed by row, here the number of parameters is 6 satellite numbers, so

[0074]

[0075] For each observation thread block, respectively r ij and δ ij , i=1,2,...,32, j=1,2,...,6 are input to Givens transformation, and finally 32 groups of R i and d i are synthesized into one group of R and d to form the normal equation

[0076] R·x=d

[0077] Solving the linear equation set can obtain the orbit improvement.

[0078] Table 2. Comparison table of single-thread computing and parallel computing

[0079]

[0080]

[0081] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In the specification, the illustrative description of the above terms is not necessarily for the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in the specification.

[0082] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the features disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Claims

1. A method for precise orbit determination of a spacecraft in parallel processing, characterized in that, The method comprises the following steps: Opening a corresponding number of integral thread handles according to the number of spacecrafts; Carrying out orbit integral calculation on the spacecrafts to obtain ephemeris data of the spacecrafts and save them; Opening a preset number of observation data processing thread handles, dividing observation data of the spacecrafts into the preset number of thread blocks, and allocating an upper triangular matrix and a vector to each thread block; Obtaining partial derivatives of to-be-solved parameters, differences between theoretical values and measured values of observation data according to the ephemeris data and observation data in each thread block, and substituting the partial derivatives and the differences between theoretical values and measured values of observation data into Givens transformation to update the upper triangular matrix and the vector; Carrying out Givens transformation on all updated upper triangular matrices and vectors after row decomposition, and obtaining orbit confirmation results; The step of carrying out orbit integral calculation on the spacecrafts to obtain ephemeris data of the spacecrafts and save them comprises the following steps: Inputting dynamic parameters, physical attribute parameters, orbit epoch initial orbit six elements, integral start and end time, and orbit time interval of each satellite into an integral function, simultaneously starting a thread calculation module of all satellites, and waiting for completion of calculation of all threads, and then storing ephemeris of each satellite obtained by calculation in a global storage area. The step of allocating an upper triangular matrix and a vector to each thread block comprises the following steps: （1） Wherein, R is an upper triangular matrix, d is a vector, x is a to-be-solved parameter, and M is a thread block number.

2. The method of claim 1, wherein the method further comprises: After opening the corresponding number of integral thread handles, the following steps are further included: First, preset initial orbit parameters of a spacecraft constellation, the initial orbit parameters including dynamic parameters, physical attribute parameters, orbit epoch initial orbit six elements, integral start and end time, and orbit time interval.

3. The method of claim 1, wherein the method further comprises: The process of obtaining partial derivatives of to-be-solved parameters and differences between theoretical values and measured values of observation data comprises the following steps: Obtaining an observation model according to the ephemeris data, obtaining partial derivatives of to-be-solved parameters and theoretical values of observation data according to the observation model, and calculating differences between theoretical values and measured values of observation data.

4. The method of claim 1, wherein the method further comprises: The measured values are obtained according to the observation data.

5. The method of claim 1, wherein the method further comprises: Before implementing Givens transformation on all upper triangular matrices and vectors calculated by all thread blocks, the following steps are further included: Row decomposition of the upper triangular matrix and the vector, that is: （2） Wherein, K is a number of orbit calculation program solution parameters.

6. The method of claim 1, wherein the method further comprises: After row decomposition of all upper triangular matrices and vectors calculated by all thread blocks, Givens transformation is implemented to finally integrate all upper triangular matrices and vectors, and obtain: （3） Solving the equation can obtain an improved orbit state of the spacecraft.

7. The method of claim 1, wherein the method is implemented in a parallel processing system. The observation data are on-board observation data and ground station observation data.

Images