Acceleration method for satellite observation orbit calculation

Abstract

The invention belongs to the technical field of satellite observation orbit calculation, and particularly relates to an acceleration method for satellite observation orbit calculation. Comprising the steps that S1, the number of two rows of orbits of all to-be-observed satellites is acquired, and all the to-be-observed satellites are stored in a telescope control system; s2, querying the core number of a CPU (Central Processing Unit) of a computer used by the telescope control system, and distributing n calculation threads according to the core number of the CPU; s3, arranging all satellites to be observed into a calculation queue in sequence; s4, all the calculation threads are started, the calculation threads in the idle state obtain the satellites to be observed of the calculation queue in sequence, and the observation orbit of the satellites to be observed is calculated according to a variable step length method till no satellites to be observed exist in the calculation queue. According to the method, the time consumption of satellite observation orbit calculation can be greatly reduced, and the working efficiency of astronomical observation research tests and the user experience of operators can be improved.

Description

Technical Field

[0001] The invention belongs to the technical field of satellite observation orbit calculation, and in particular relates to an acceleration method for satellite observation orbit calculation. Background Art

[0002] The horizontal telescope has the function of tracking and observing artificial satellites, which plays an important role in the attitude monitoring, astronomical positioning and orbit measurement of artificial satellites. Satellite tracking usually uses two-line orbit elements (TLE) to calculate the observable orbit of artificial satellites and guide the telescope to track and observe. The two-line orbit elements are a kind of orbit encoding method. At a given epoch, using a suitable prediction model, the position and velocity of the target at any point on the orbit can be calculated with a certain accuracy. The calculation model corresponding to the two-line orbit elements is the simplified perturbations model (SPG). The SGP4 model and SDP4 model are usually used to calculate the motion trajectory of artificial satellites during the observable time period. The satellite calculation parameters mentioned later refer to the two-line orbit elements.

[0003] The commonly used satellite orbit calculation method is: given a set of satellite calculation parameters and a time period for calculating the orbit. For each satellite, calculate the observation direction (azimuth and pitch) every second in the time period. If the pitch angle is greater than 0°, the satellite is observable; if the pitch angle is less than 0°, the satellite is unobservable. Arrange the continuous directions with pitch angles greater than 0° in chronological order, which is called an observable arc. For a certain station, a satellite may have 0, 1 or more observable arcs in a time period. It takes a long time to calculate the observable arcs of satellites according to the existing method, especially when the observable arcs are calculated based on multiple stations and multiple satellites. This problem is particularly prominent. Summary of the invention

[0004] In view of this, the present invention aims to provide an accelerated method for calculating satellite observation orbits to solve the problem that the existing methods take a long time to calculate the observable arc segments of satellites. When the telescope control system needs to calculate the observation orbits of a large number of satellites, the present invention can greatly reduce the time consumption of orbit calculations, which is beneficial to improving the work efficiency of astronomical observation research experiments, as well as the user experience of operators.

[0005] To achieve the above object, the technical solution created by the present invention is implemented as follows: A method for accelerating satellite observation orbit calculation, comprising the following steps: S1: Obtain the two-row orbit elements of all satellites to be observed, and store all satellites to be observed in the telescope control system; S2: Query the number of CPU cores of the computer used by the telescope control system, and allocate n computing threads according to the number of CPU cores; S3: Put all satellites to be observed into the calculation queue in order; S4: Start all computing threads. The computing threads in the idle state sequentially obtain the satellites to be observed in the computing queue and calculate the observation orbits of the satellites to be observed according to the variable step size method until there are no satellites to be observed in the computing queue.

[0006] Furthermore, in the calculation queue, medium and high orbit satellites are sorted first, and low orbit satellites are sorted later.

[0007] Further, in step S2, the number of computing threads is less than the number of CPU cores.

[0008] Furthermore, step S4 specifically includes the following steps: S41: Start all computing threads; S42: the computing thread in the idle state obtains the first-ordered satellite to be observed from the computing queue, and calculates the observation orbit of the satellite to be observed according to the variable step size method; S43: Check whether there are satellites to be observed in the calculation queue. If so, execute step S42; otherwise, end the calculation task.

[0009] Furthermore, in step S42, the specific steps of calculating the observation orbit of the satellite to be observed by the single computing thread in the idle state according to the variable step length method include: S421: Set the starting time for calculating the satellite to be observed to , the end time is ; S422: Calculate the azimuth and elevation of the satellite to be observed at the current time t, and set the initial value of t to , set the azimuth of the satellite to be observed to , the elevation direction of the satellite to be observed is ; S423: Update the current time t according to the variable step size method; S424: If the updated , then execute step S422 based on the updated t, otherwise terminate the observation orbit calculation of the satellite to be observed.

[0010] Further, step S423 specifically includes the following steps: Set the benchmark calculation step size to , the unit of the benchmark calculation step value is seconds; When the satellite to be observed is a medium-orbit satellite, if the pitch direction , then calculate the step length ,in , if the pitch is pointing , then the calculation step size is set in minutes; When the satellite to be observed is a low-orbit satellite, if the pitch direction ,but , if the pitch is pointing , then set the calculation step size in seconds. If the pitch direction , then set the calculation step size in minutes.

[0011] Furthermore, when a single computing thread is in the process of acquiring a satellite to be observed, if the computing queue is empty, the current computing thread is destroyed.

[0012] Compared with the prior art, the invention can achieve the following beneficial effects: The present invention creates the acceleration method for satellite observation orbit calculation, uses multi-threaded calculation to improve CPU utilization, and changes the original single-core serial calculation into multi-core parallel calculation; adopts a variable step-size method to greatly shorten the orbit calculation time of a single satellite; when using the variable step-size method, due to the particularity of the observation orbits of medium and high-orbit satellites, the difference in calculation time is large, while the calculation time of the observation orbit of low-orbit satellites is relatively stable. The present invention adopts a method of giving priority to processing medium and high-orbit satellites, which is conducive to more balanced calculation loads of various calculation threads. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings constituting part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments and descriptions of the present invention are used to explain the present invention and do not constitute an improper limitation on the present invention. In the drawings: Figure 1 A schematic flow chart of a method for accelerating satellite observation orbit calculation according to an embodiment of the present invention; Figure 2 A schematic diagram of a process for accelerating the calculation of multiple satellite orbits according to an embodiment of the present invention; Figure 3 A schematic diagram of a flow chart of single satellite orbit calculation according to an embodiment of the present invention; Figure 4 A schematic diagram of the structure of the computing queue arrangement method described in the embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the invention more clear, the invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the invention and do not constitute a limitation of the invention.

[0015] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0016] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0017] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the invention can be understood according to specific circumstances.

[0018] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0019] like Figure 1-Figure 2 As shown, the acceleration method for calculating satellite observation orbits proposed by the present invention specifically includes the following steps: S1: obtaining the two-row orbit elements of all satellites to be observed, and storing all satellites to be observed in the telescope control system; S2: querying the number of CPU cores of the computer used by the telescope control system, and allocating n computing threads according to the number of CPU cores; S3: arranging all satellites to be observed into a computing queue in sequence; S4: starting all computing threads, and the computing threads in the idle state obtain the satellites to be observed in the computing queue in sequence, and calculate the observation orbits of the satellites to be observed according to the variable step size method, until there are no satellites to be observed in the computing queue.

[0020] The acceleration method for calculating satellite observation orbits of the present invention is applicable to horizontal telescopes, and solves the problem that batch calculation of satellite observation orbits takes a long time in telescope control systems. The present invention accelerates the calculation of a single satellite orbit through a variable step size method, uses multi-threading technology to achieve parallel calculation of multiple satellite orbits, and sorts the calculation order of the satellites to be observed in advance, and dynamically allocates the satellites to be observed to each calculation thread during calculation, so that the load of multiple calculation threads is balanced, and the total time consumption of satellite observation orbit calculation is greatly shortened.

[0021] The main principles of the present invention are summarized as follows: first, the number of cores of the CPU (Central Processing Unit) of the computer used by the telescope control system is queried, and computing threads are allocated according to the number of cores; then, all satellites to be observed are divided into two categories, one is medium- and high-orbit satellites, and the other is low-orbit satellites, and they are arranged in a computing queue in order; finally, each computing thread is enabled to take the computing parameters of a satellite to be measured from the head of the computing queue when idle, and calculate the observation orbit of the satellite to be measured according to the variable step size method until the computing queue is empty.

[0022] It should be noted that the two rows of orbital elements used in the present invention are taken from a public database on the Internet; the simplified perturbation model (SPG) for calculating satellite observation orbits is a public technology and has open source code; the observation station address (longitude, latitude and elevation) of the telescope can be obtained by measuring instruments. In addition, the present invention only needs to discuss the orbit generation acceleration method for a single station address. For multiple stations, only the orbit generation steps for a single station address need to be repeated. In the present invention, because the orbit generation time for a single station address is greatly shortened, the orbit generation efficiency for multiple stations is also significantly improved.

[0023] In some embodiments, in the calculation queue, medium and high orbit satellites are sorted first, and low orbit satellites are sorted later.

[0024] In some embodiments, the number of computing threads is less than the number of cores of the CPU.

[0025] In some embodiments, step S4 specifically includes the following steps: S41: start all computing threads; S42: the computing thread in idle state obtains the first-ordered satellite to be observed from the computing queue, and calculates the observation orbit of the satellite to be observed according to the variable step size method; S43: check whether there is a satellite to be observed in the computing queue, if so, execute step S42, otherwise end the computing task.

[0026] It should be noted that in step S42, if the number of computing threads in the idle state is greater than or equal to 2, taking two as an example, the two computing threads take the two satellites to be observed that are first sorted in the computing queue for calculation. Furthermore, each computing thread randomly assigns one of the two satellites to be observed for calculation, and the satellites to be observed assigned to the two computing threads are different, and there is no absolute order of priority in the acquisition of the computing threads.

[0027] In some embodiments, when a single computing thread is in the process of acquiring a satellite to be observed, if the computing queue is empty, the current computing thread is destroyed to release resources.

[0028] The traditional satellite trajectory calculation method is to calculate the satellite's pointing direction to the station every whole second within a specified time period, and then connect all the pointing directions with a pitch angle greater than 0° to form several observable arcs. However, calculating all the whole seconds will obviously result in a serious waste of time.

[0029] In the present invention, taking a certain high-orbit satellite as an example, the altitude of the high-orbit satellite is about 36000km, and the azimuth and elevation pointing changes are less than If the high-orbit satellite is within the observable range (pitch angle greater than 0°) at a certain point in time, the calculation step size of its observation trajectory can be appropriately increased, and the observation direction of other non-calculation time points can be calculated using the trajectory fitting method. If the satellite is within the unobservable range (pitch angle less than or equal to 0°) at a certain point in time, it is highly likely that it will not enter the observable range for a long time, and the calculation step size can be set to a larger value.

[0030] Take a low-orbit satellite as an example. The satellite orbit height of a low-orbit satellite is about 500km, and its azimuth and elevation usually change rapidly. If the low-orbit satellite is within the observable range at a certain point in time (the elevation angle is greater than 0°), the orientation of the low-orbit satellite for all full seconds is still calculated. If the low-orbit satellite is within the unobservable range at a certain point in time (the elevation angle is less than 0°), the calculation step size can be dynamically adjusted according to the elevation angle pointed at that point in time to ensure that the next observable start time is found in a shorter time.

[0031] Based on the above reasons, the present invention adopts different variable step length calculation methods for medium-high orbit and low-orbit satellites.

[0032] In some embodiments, Figure 3 As shown, in step S42, the specific steps of calculating the observation orbit of the satellite to be observed according to the variable step length method by a single computing thread in an idle state include: S421: setting the starting time of calculating the satellite to be observed to , the end time is ; S422: Calculate the azimuth and elevation of the satellite to be observed at the current time t, and set the initial value of t to , set the azimuth of the satellite to be observed to , the elevation direction of the satellite to be observed is ; S423: Update the current time t according to the variable step length method; S424: If the updated time t in step S423 , then execute step S422 based on the updated t, otherwise terminate the observation orbit calculation of the satellite to be observed.

[0033] Set the benchmark calculation step size to , the unit of the benchmark calculation step value is seconds; When the satellite to be observed is a medium-orbit satellite, if the pitch direction , then calculate the step length ,in , if the pitch is pointing , then the calculation step size is set in minutes; It should be noted that the value of n is related to the satellite orbit height (i.e., pointing speed). The greater the satellite orbit height, the smaller the pointing speed, and the larger the value of n. The smaller the satellite orbit height, the greater the pointing speed, and the smaller the value of n. To ensure tracking accuracy, the value range of n is ,Through experimental statistics, it is found that for all the medium and high orbit satellites used in the ,experiment, setting n=10 can not only meet the tracking accuracy but ,can significantly reduce the calculation amount and the program writing complexity.

[0034] When the satellite to be observed is a low-orbit satellite, if the pitch direction ,but , if the pitch is pointing , then set the calculation step size in seconds. If the pitch direction , then set the calculation step size in minutes.

[0035] It should be noted that the pitch pointing , indicating that the satellite moves below the horizon and is unobservable. When the satellite to be observed is a medium-orbit satellite, since the medium-orbit satellite moves slowly, the satellite will be unobservable for a long time in the future, so the calculation step is set in minutes to reduce the number of calculations. When the satellite to be observed is a low-orbit satellite, the pitch direction is from the time the satellite moves below the horizon to the next time it moves above the horizon. It may go through one or more processes of decreasing first and then increasing. During this process, the pitch pointing satisfies . And in this process, The calculation step size is gradually increased as the By gradually reducing the calculation step size as the value increases, the purpose of variable step size can be achieved.

[0036] In some embodiments, when the satellite to be observed is a medium-orbit satellite, the following method is used to update the calculation step size: if the pitch direction is ,but ; If the pitch is pointing ,but .

[0037] When the satellite to be observed is a low-orbit satellite, the calculation step size is updated using the following method: If the pitch direction ,but ; If the pitch pointing satisfies ,but ; If the pitch pointing satisfies ,but ; If the pitch pointing satisfies ,but ; If the pitch pointing satisfies ,but ; If the pitch direction ,but .

[0038] If the above strategy is used, the orbit calculation time of a single satellite will be greatly shortened. Taking the calculation of the satellite orbit of a certain station within 24 hours as an example, for different satellites, the optimization effect of the number of calculations is as follows: 1) If the calculation is for medium and high orbit satellites, and all time periods are within the observable range, the number of calculations before optimization is 86400, and the number of calculations after optimization is 8640, a decrease of 90%.

[0039] 2) If the calculation is for medium and high orbit satellites, and all time periods are not within the observable range, the number of calculations before optimization is 86,400, and the number of calculations after optimization is 144, a decrease of 99.8%.

[0040] 3) If low-orbit satellites are calculated, taking 100 randomly selected low-orbit satellites as an example, the number of calculations is counted. Before optimization, the number of calculations for each satellite is 86,400. After optimization, the average number of calculations for each satellite is 4,383, a decrease of 94.9%.

[0041] Medium and high orbit satellites are not completely stationary, especially medium orbit satellites move faster than high orbit satellites. Medium and high orbit satellites are sometimes observable during the calculation period and sometimes unobservable. Here we only discuss the extreme case, which has little impact on the overall performance optimization.

[0042] The multi-thread load balancing in the present invention mainly refers to: the total execution time of each computing thread should be as close as possible. Since the present invention uses a variable step size method to calculate the observation orbit of a single satellite, the number of observations required to calculate a certain satellite orbit cannot be predicted before calculating its observation orbit. If the principle of evenly distributing satellite calculation parameters to all computing threads is followed, and the computing threads happen to All calculations in the calculation are high-orbit satellites within the observable range, and the calculation threads If all calculations are for high-orbit satellites that are not observable, the calculation thread A high load will result in an increase in the overall trajectory calculation time.

[0043] Taking into account the fact that the calculation times of a single medium-high orbit satellite fluctuate greatly, while the calculation times of a single low-orbit satellite fluctuate less, the present invention puts all satellite calculation parameters into the calculation queue in the order of medium-high orbit first and low-orbit later, and the arrangement is as follows: Figure 4 Each computing thread is required to obtain the computing parameters of a satellite from the queue during the idle computing time, thereby ensuring the load balance of all computing threads.

[0044] Overall, the time for generating satellite observation orbits has been greatly shortened. Taking 10 different stations as an example, all observable arcs of 200 satellites within 24 hours are calculated.

[0045] Under the premise of using the existing method, if a computer with a CPU model of Intel(R) Core(TM) i9-11900K @3.50GHz is used for calculation, the average pointing calculation time for a single satellite at a single site per full second is 0.0012ms, and the pointing time for 200 satellites at 10 sites per full second within 24 hours is a total of 207 seconds.

[0046] If a computer with an Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz is used for calculation, the average pointing calculation time for a single satellite at a single site is 0.0026ms per full second. The pointing calculation time for 200 satellites at 10 sites per full second in 24 hours is 449 seconds in total.

[0047] Under the premise of using the present invention, if the computer CPU model is Intel(R) Core(TM) i9-11900K @3.50GHz and 8 threads are used for calculation, the total calculation time is 2.8 seconds, which is a reduction of 98.6%.

[0048] If the computer CPU model is Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz, using 4 threads for calculation, the total calculation time is 18.3 seconds, a reduction of 95.9%.

[0049] This shows that the use of the present invention to accelerate the generation of satellite observation orbits can not only accelerate the calculation of multiple satellite orbits using multi-core processors, but also significantly reduce the orbit calculation time of a single satellite in each calculation thread. Therefore, when generating observation orbits for hundreds of satellites at multiple sites, the user experience and work efficiency of operators will be greatly improved.

[0050] The existing method of accelerating the calculation of satellite orbits by combining CPU and GPU usually uses NVIDIA graphics cards for accelerated calculations in terms of equipment selection. However, in the deployment plan of optoelectronic telescopes, the computers used to calculate satellite orbits are rarely equipped with high-end graphics cards. In most cases, they do not have the conditions for GPU accelerated calculations, and the GPU acceleration effect of using low-end graphics cards cannot meet expectations, which greatly increases the cost of equipment. At present, multi-core CPUs have become standard for computers. The CPU-based optimization solution proposed by the present invention based on actual conditions can greatly save equipment costs and improve the portability of the software. Secondly, from the perspective of the professional skills of software developers, developers in the fields of image processing or artificial intelligence are usually proficient in using programming methods based on CUDA or OpenCL parallel computing, while satellite observation orbit calculations usually belong to the work content of control software developers, and require a certain understanding of the principles of two-row orbital elements and orbit solution. It is relatively high for developers to have the above capabilities. However, developers can achieve satellite observation orbit calculations with lower difficulty and shorter development cycles by applying the method proposed by the present invention. Furthermore, the prior art does not propose a specific purpose for solving satellite orbits, and all solve satellite orbits with a step size of 1s. The present invention solves satellite orbits for ground-based photoelectric telescopes, and only focuses on arcs with pitch angles greater than 0°. When the pitch angle is less than 0°, in order to quickly find the next observable arc segment, not all whole second points are calculated, which greatly reduces the time cost of calculation.

[0051] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the disclosure of the present invention can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0052] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for accelerating satellite observation orbit calculation, characterized in that: The specific steps include: S1: Obtain the two-row orbit elements of all satellites to be observed, and store all satellites to be observed in the telescope control system; S2: Query the number of CPU cores of the computer used by the telescope control system, and allocate n computing threads according to the number of CPU cores; S3: Put all satellites to be observed into the calculation queue in order; S4: Start all computing threads. The computing threads in the idle state sequentially obtain the satellites to be observed in the computing queue and calculate the observation orbits of the satellites to be observed according to the variable step size method until there are no satellites to be observed in the computing queue.

2. The method for accelerating satellite observation orbit calculation according to claim 1, characterized in that: In the calculation queue, medium and high orbit satellites are sorted first, and low orbit satellites are sorted later.

3. The method for accelerating satellite observation orbit calculation according to claim 1, characterized in that: In step S2, the number of computing threads is less than the number of CPU cores.

4. The method for accelerating satellite observation orbit calculation according to claim 1, characterized in that: Step S4 specifically includes the following steps: S41: Start all computing threads; S42: the computing thread in the idle state obtains the first-ordered satellite to be observed from the computing queue, and calculates the observation orbit of the satellite to be observed according to the variable step size method; S43: Check whether there are satellites to be observed in the calculation queue. If so, execute step S42; otherwise, end the calculation task.

5. The method for accelerating satellite observation orbit calculation according to claim 1, characterized in that: In step S42, the specific steps of calculating the observation orbit of the satellite to be observed by the single computing thread in the idle state according to the variable step length method include: S421: Set the starting time for calculating the satellite to be observed to , the end time is ; S422: Calculate the azimuth and elevation of the satellite to be observed at the current time t, and set the initial value of t to , set the azimuth of the satellite to be observed to , the elevation direction of the satellite to be observed is ; S423: Update the current time t according to the variable step size method; S424: If the updated , then execute step S422 based on the updated t, otherwise terminate the observation orbit calculation of the satellite to be observed.

6. The method for accelerating satellite observation orbit calculation according to claim 1, characterized in that: Step S423 specifically includes the following steps: Set the benchmark calculation step size to , the unit of the benchmark calculation step value is seconds; When the satellite to be observed is a medium-orbit satellite, if the pitch direction , then calculate the step length ,in , if the pitch is pointing , then set the calculation step size in minutes; When the satellite to be observed is a low-orbit satellite, if the pitch direction ,but , if the pitch is pointing , then set the calculation step size in seconds. If the pitch direction , then set the calculation step size in minutes.

7. The method for accelerating satellite observation orbit calculation according to claim 1, characterized in that: When a single computing thread is in the process of acquiring satellites to be observed, if the computing queue is empty, the current computing thread will be destroyed.

Images