Space debris survey observation method based on greedy algorithm and pixel ball algorithm

By adopting greedy algorithms and pixel sphere algorithms in space debris sky survey observation, combined with HEALPix pixelation and addition filtering processing, the observation strategy of large field of view small aperture telescope arrays is optimized, and the problem of inefficiency in traditional methods is solved, and the observation efficiency in traditional methods is achieved. More efficient space debris detection and cataloging support.

CN118642194BActive Publication Date: 2024-12-03SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411116548.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-03
Estimated Expiration
2044-08-15

AI Technical Summary

Technical Problem

Traditional space debris sky survey observation methods cannot effectively meet the observation needs of multiple space debris, and it is difficult to optimize the sky survey observation strategy, resulting in inefficient observation.

Method used

The space fragment sky survey observation method based on greedy algorithm and pixel sphere algorithm is adopted, and observation data is obtained using a large field of view small aperture telescope array. Through HEALPix pixelation processing and addition filtering processing, combined with greedy algorithms, the observation strategy is optimized to improve the sky survey observation efficiency.

Benefits of technology

It significantly improves the efficiency and automation of space debris observation, can detect more space debris more accurately, and provides strong data support for cataloging and collision warning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118642194B_ABST
    Figure CN118642194B_ABST
Patent Text Reader

Abstract

The invention relates to the field of satellite communication technology, in particular to a space debris survey observation method, which improves the survey observation efficiency and can effectively simplify the space debris observation plan formulation process. The method is characterized in that observation data is obtained by using a large-field-of-view small-aperture telescope array, wherein the large-field-of-view small-aperture telescope array includes four large-field-of-view optical telescopes with an aperture of 15CM, each telescope has a field of view of 10°×10°, and the total field of view of the small photoelectric array composed of the four telescopes reaches 400 square degrees. After the observation data is obtained, orbit prediction information of the space debris is obtained, the orbit prediction data is preliminarily optimized, the data is pixelated, and then summing and filtering are performed. According to a data list obtained by summing and filtering, a filter set maxΣρ with the largest number of debris in all time steps, that is, the telescope field of view direction, is obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of satellite communication technology, and in particular to a space debris survey observation method based on a greedy algorithm and a pixel ball algorithm, which improves the survey observation efficiency and can effectively simplify the space debris observation plan formulation process. Background Art

[0002] As we all know, space debris can pose a threat to the operation of satellite communication networks. In order to ensure the operational safety of satellite communication systems, it is necessary to conduct real-time observations of space debris in the orbit of satellite communication systems and their surrounding space environment, so as to provide a reliable basis for the subsequent handling of space debris. Traditional space target surveys are mainly conducted on astronomical targets. Compared with astronomical targets, space debris is more numerous, smaller in size, and in real-time change. Traditional surveys of astronomical targets cannot meet the needs of space debris observations. When there are multiple debris to observe, how to observe more space debris requires rational survey optimization of multiple space debris, and existing technologies cannot solve this problem. Summary of the invention

[0003] In view of the gaps in the prior art, the present invention proposes a space debris survey observation method based on a greedy algorithm and a pixel ball algorithm, which has high survey observation efficiency and can effectively simplify the space debris survey observation process.

[0004] The present invention is achieved by the following measures:

[0005] A space debris survey observation method based on a greedy algorithm and a pixel ball algorithm is characterized in that observation data is obtained by using a large-field-of-view small-aperture telescope array, wherein the large-field-of-view small-aperture telescope array includes four large-field-of-view optical telescopes with an aperture of 15CM, each telescope has a field of view of 10°×10°, and the total field of view of the small photoelectric array composed of the four telescopes is 400 square degrees. After obtaining the observation data, the following steps are performed:

[0006] Step 1: Obtain space debris orbit prediction information, including right ascension, declination, altitude, azimuth and debris ID data of different debris at each moment;

[0007] Step 2: Preliminary optimization of orbital prediction data. To ensure that all fragments within the set time step are valid and to prevent the situation where a fragment just enters the field of view and then runs out of the field of view, the frequency of occurrence of the same fragment within the time step is processed. When the number of occurrences is greater than or equal to the set value, it meets the processing standard. The same fragment is only counted once within the specified time step, thereby ensuring the maximum number of single non-repeating fragments.

[0008] Step 3: Data pixelation: The spherical data is discretized using HEALPix pixelation. When mapping data points to the HEALPix grid, the spherical coordinates of the HEALPix grid, including right ascension and declination, are used. Each data point is assigned to the nearest pixel, and the coordinates are converted into pixel indices.

[0009] Step 4: Perform summing and filtering on the data processed in step 3;

[0010] Step 5: Based on the data list obtained by summing and filtering, take the filter set maxΣρ with the largest number of fragments in each time step i , delete the observed fragment id from the total fragment library, update the step information, and then calculate the next time step until the time step ends, and obtain the filter set maxΣρ with the largest number of fragments in all time steps, that is, the telescope field of view.

[0011] The summing and filtering process in step 4 of the present invention specifically includes:

[0012] Step 4-1: Take the i-th time step as an example, 1, 2, 3, 4, ... j-1, j represents different pixels, and assume that different pixels N pix The internal debris density is ρ i,j , the fragment density set of all pixels in the i-th time step is {ρ i,j}:

[0013] {ρ i,j}={ρ i,1, ρ i,2, ρ i,3,…… ρ i,j-1, ρ i,j}(5);

[0014] Step 4-2: Consider the effective field of view as a filter. The center of the filter is the center coordinate of the pixel. If the spherical angle distance of the surrounding pixels from the center coordinate is less than half of the field of view, the pixel is considered to be within the filter.

[0015] Step 4-3: Get a data list based on the filter centered on each pixel (k), traverse each non-empty pixel, and get the distribution of fragments in different filters, including the filter's right ascension and declination coordinates, the number of fragments, and id), forming a set {Σρ i,j,k}, as shown in formula (6): {Σρ i,j,k}={Σρ i,j,1, Σρ i,j,2,……, Σρ i,j,k}(6), using the sum filtering method, we can obtain the fragment distribution of the effective field of view in each time step for a total of i time steps throughout the night.

[0016] In order to optimize the scheduling, in step 5 of the present invention, a greedy algorithm is used to arrange the observation every night, that is, the field of view points to the sky area where the maximum filter is located in the current time step, and the telescope points at the next moment to: the sky area closest to the current pointing and where the maximum filter is located in the time step. Before each observation, the sky area pointing with the highest reward metric is selected to obtain a local optimal solution, which specifically includes:

[0017] Step 5-1: Initialization: Calculate the field of view pointing set where the filter is the largest at all time steps to obtain the initial solution and solution space;

[0018] Step 5-2: Select the optimal solution to check the constraints: Select the current optimal solution from the solution space, that is, different solutions with the same density exist at the same time, that is, different sky area directions. When this happens, take the solution closest to the current time (that is, the telescope moves the shortest distance) as the optimal solution;

[0019] Step 5-3: Update the problem state: Update the state of the solution space according to the selected solution, and so on to obtain the optimal solution for each time step;

[0020] Step 5-4: Determine the termination condition: Determine whether the termination condition is met. If all observation times of the current observation night have been traversed, the algorithm ends; otherwise, return to step 5-2.

[0021] The present invention is based on a large-field-of-view small-aperture telescope array, comprehensively considers the characteristics of the telescope and the observation conditions, and aims to observe more space debris. A basic sky survey observation strategy is formulated, which constructs an objective function and formulates reasonable evaluation indicators based on factors such as the characteristics of the telescope, the observation conditions and the orbital prediction information of space debris. The sky area is finely divided by using the pixelated sphere method. The greedy algorithm is used in the form of sum filtering to maximize the benefit of the objective function, thereby generating the optimal sky survey observation strategy. At the same time, it can improve the degree of automation in the formulation of space debris observation plans, significantly improve the execution efficiency of telescope debris observations, and provide strong data support for the cataloging of space debris and collision warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the pixelated ball method of the present invention.

[0023] Figure 2 It is a heat map of the density distribution of non-empty pixel space fragments within a given time step in the present invention.

[0024] Figure 3 It is the spatial distribution diagram of the debris density of the first six time steps in the present invention.

[0025] Figure 4 It is a schematic diagram of summing and filtering in the present invention.

[0026] Figure 5 It is the effective observation field of random observation in the present invention.

[0027] Figure 6 It is a schematic diagram of the distribution of the telescope field of view on the celestial sphere during random observation in the present invention.

[0028] Figure 7 It is a schematic diagram of the effective observation field of the present invention.

[0029] Figure 8 It is a schematic diagram of the distribution of the telescope field of view on the celestial sphere during observation by the present invention. DETAILED DESCRIPTION

[0030] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0031] When using ground-based optical telescopes to survey space debris, different observation strategies need to be formulated according to different mission requirements, debris information, equipment conditions and other factors. Appropriate observation strategies can not only improve data quality, but also improve the execution efficiency of the telescope to maximize benefits. At the same time, when there are multiple fragments to observe, how to solve multi-objective optimization problems, the formulation of observation strategies becomes a complicated matter. Formulating reasonable evaluation indicators and constructing objective functions, selecting and sorting candidate fragments, a series of problems will become more difficult as the number of fragments increases. In view of the above difficulties, the present invention proposes a method for surveying space debris based on a greedy algorithm and a pixelated ball algorithm.

[0032] In order to accurately and efficiently pixelate the celestial sphere, the present invention uses the HEALPix (HierarchicalEqual Area isoLatitude Pixelation) algorithm to discretize the spherical data. HEALPix is ​​an advanced method for discretization and rapid analysis of high-resolution spherical data. Its main features include equal-area pixelation and hierarchical structure, which makes the data more evenly distributed on the sphere and effectively avoids the distortion problems common in other pixelation methods.

[0033] Specifically, the pixelation scheme of the present invention is to divide the sphere into 12 basic pixels through HEALPix, and these pixels can be recursively subdivided in multiples of 4 in subsequent levels to form a hierarchical equal-area pixel structure. This ensures that the data on the sphere is evenly distributed, avoids the deformation problem of other pixelation methods, and facilitates unbiased statistical analysis of the data.

[0034] Resolution parameter: The resolution of HEALPix is ​​defined by the `Nside` parameter. The higher the `Nside` value, the more pixels there are and the higher the resolution. In this paper, `Nside = 32` is selected, which means that the sphere is divided into 12,288 pixels of equal area. The choice of `Nside` balances computational efficiency and spatial resolution. This choice ensures data accuracy while maintaining computational manageability. pix =12×N side 2 , (1).

[0035] The present invention first needs to perform data mapping and processing: data mapping refers to the use of spherical coordinates (right ascension and declination) when mapping data points to the HEALPix grid. Each data point is assigned to the nearest pixel, and the coordinates are converted to pixel indices using the `ang2pix` function in the HEALPix library; after the data is pixelated, statistical calculations such as averages and sums can be performed directly on the pixelated data. This method simplifies data processing and increases the speed of calculation; the implementation tool uses the HEALPix software package (C++ and Python versions are provided) to implement pixelation and data analysis. Specifically, the Python library `healpy` is used because it is easy to integrate with other data processing tools and is easy to use.

[0036] The latest two orbital elements of the day are obtained from Space-track. Through orbital calculations, including the time of the designated observation site at night, the time when the debris is visible, the time when the debris is visible to the observation station at night, and the time when the debris is illuminated by the sun during the visible time, the right ascension, declination, altitude angle, azimuth angle and debris number id of different debris at each moment are obtained.

[0037] Parameters are set according to the observatory site, telescope conditions, and observation requirements. To ensure that all fragments within the set time step are valid and to prevent the fragments from running out of the field of view just after entering it, the frequency of occurrence of the same fragment within the time step is processed. When the number of occurrences is greater than or equal to a certain value, it meets the processing standard (in preparation for subsequent mobile debris detection); the same fragment is counted only once within the specified time step, thereby ensuring the maximum number of single non-repeating fragments.

[0038] In terms of the division of the entire celestial sphere, the pixelated sphere method is used. According to the definition of HEALPix resolution, N is taken. side The parameter is 32, which divides the celestial sphere into 12,288 pixels of equal area. The $ang2pix$ function in the HEALPix library is used to convert the celestial right ascension and declination coordinates to pixel indices. The ang2pix function of HEALPix is ​​used to convert the pixel index to the spherical coordinates of the pixel center. The function calculates the spherical coordinates of the four boundary points of each pixel to form a pixel area. Within the specified observation time step, if the right ascension and declination of the space debris are within the pixel range, the count is increased by one, and the debris density ρ of the specified pixel area within the current time step is i,j , obtained by the following formula 2, as Figure 2 Shown is the heat map distribution of spatial debris density within non-empty pixels at a certain time step.

[0039] (2)

[0040] Where N i,j represents the number of fragments of the jth pixel in the i-th time step, and S represents the pixel area

[0041] The size of the filter is set according to the size of the telescope's field of view. To prevent the fragments from being missed, the fields of view in the declination and right ascension directions are ensured to overlap, and the theoretical field of view is converted into an effective field of view. The effective field of view is regarded as a filter, and the size of the filter is the size of the effective observation field of view. The center of the filter is the center coordinate of the pixel. If the spherical angle distance of the surrounding pixels from the center coordinate is less than half of the field of view, the pixel is considered to be in the filter. A data list is obtained by traversing each non-empty pixel to obtain the distribution of fragments in different filters (including the right ascension and declination coordinates of the filter, the number of fragments, and the ID). The sum filtering method is used to obtain the distribution of fragments in the field of view of the whole sky. The specific process is shown in Table 1.

[0042] Table 1. Distribution of debris in the whole sky field of view obtained by summing and filtering

[0043]

[0044] Different observation tasks and different research objectives have different evaluation indicators for space debris. In the present invention, no matter what the observation conditions are, the goal is to observe the most debris in the shortest time, so that the remaining time can be used for other observation tasks to improve the utilization rate of the telescope. According to the data list obtained by summing and filtering, the filter set maxΣρ with the largest number of debris in each time step is taken. i , delete the observed fragment ID from the total fragment library and update the step information; then calculate the next time step until the time step ends; get the filter set maxΣρ with the largest number of fragments in all time steps, that is, the telescope field of view direction, the specific process is shown in Algorithm 2.

[0045]

[0046] Table 2 The telescope field of view pointing is obtained based on the data list obtained by summing and filtering

[0047]

[0048] In order to optimize the scheduling, a greedy algorithm is further used to arrange the observations every night. That is, the field of view points to the area of ​​the sky where the largest filter is located in the current time step, and the telescope points to the area of ​​the sky closest to the current point and where the largest filter is located in the time step. This optimization algorithm has the advantages of simplicity, efficiency, and easy implementation. The basic idea is to take the best choice in the current state at each step of the selection so that the final result is the best. The optimization of the scheduling can be decomposed into sub-problems, specifically, the optimization problem of each night or each individual observation. Before each observation, the sky area with the highest reward metric is selected to obtain a local optimal solution. Once the optimal observation value is collected, the entire schedule can be optimized.

[0049] The greedy algorithm of the present invention comprises the following steps:

[0050] Step (a): Calculate the field of view point set maxΣρ where the filter is at its maximum at all time steps to obtain the initial solution and solution space;

[0051] Step (b): Select the optimal solution to check the constraints: Select the current optimal solution from the solution space, that is, maxΣρ i , there are different solutions with the same density at the same time, that is, different sky area directions. When this happens, the solution closest to the current moment (that is, the telescope moves the shortest distance) is taken as the optimal solution.

[0052] Step (c): Update the problem state: Based on the selected solution, update the state of the solution space, and so on, to obtain the optimal solution for each time step.

[0053] Step (d): Determine the termination condition: Determine whether the termination condition is met. If all observation times of the current observation night have been traversed, the algorithm ends; otherwise, return to step (b). Example

[0054] This example is based on a large field of view small aperture telescope array, four 15CM large field of view optical telescopes, each telescope field of view is about 10°×10°, the total field of view of the small photoelectric array composed of four telescopes can reach 400 square degrees. Usually, the main observation mode at night when there is no observation task is sky survey observation, and the main observation object is the tracking observation of moving targets such as space debris. The four telescopes of the small photoelectric array are installed on a base in pairs.

[0055] This example uses Space-track to obtain the TLE data of all space debris on May 19, June 6, and June 8, 2024. Based on the physical conditions of the observation station, the altitude angle is set to be greater than 20°, the data point sampling interval is 2 seconds, and orbit prediction calculations are performed to obtain the visible time, right ascension, declination, altitude angle, azimuth, etc. of the debris. The observation on May 19 is taken as an example.

[0056] According to the observation of the small photoelectric array, the observation adopts the constant motion mode. The telescope observes each field of view for two minutes, that is, the field of view is changed every 2 minutes, the exposure time is 2 seconds and 0.1 seconds, the data readout time is 1 second, and the interval exposure method is adopted. In order to ensure that all fragments within two minutes are valid and prevent the situation where they just enter the field of view and run out of the field of view, the frequency of the same fragment within two minutes is processed. When the number of occurrences is greater than or equal to 5, it meets the processing standard (to prepare for the subsequent mobile fragment detection). The same fragment is counted only once within two minutes to ensure the maximum number of single non-repeating fragments.

[0057] Using the pixelated sphere method, the celestial sphere is divided into 12,288 pixels of equal area. The spatial debris density distribution map for all time steps on May 19 is obtained by formula 1, as shown in Figure 3 as shown (only the first six steps are listed here).

[0058] The small photoelectric array used in this example includes two groups of telescopes (two in each group), which are installed on two turntables. The field of view of one telescope is 10°x10°, and the field of view of a group of telescopes in the form of field splicing is 10°x20°. This invention takes one group as an example. To prevent the fragments from being missed, the fields of view of 1° in the declination direction and 2° in the right ascension direction are ensured to overlap, and the effective field of view becomes 9°x18°. Next, the summing and filtering process is performed, as shown in the schematic diagram. Figure 4 As shown:

[0059] Step 4-1: Take the i-th time step as an example, 1, 2, 3, 4, ... j-1, j represents different pixels, and assume that different pixels N pix The internal debris density is ρ i,j , the fragment density set of all pixels in the i-th time step is {ρ i,j}:

[0060] {ρ i,j}={ρ i,1, ρ i,2, ρ i,3,…… ρ i,j-1, ρ i,j}(5);

[0061] Step 4-2: Consider the effective field of view as a filter. The center of the filter is the center coordinate of the pixel. If the spherical angle distance of the surrounding pixels from the center coordinate is less than half of the field of view, the pixel is considered to be within the filter.

[0062] Step 4-3: Get a data list based on the filter centered on each pixel (k), traverse each non-empty pixel, and get the distribution of fragments in different filters, including the filter's right ascension and declination coordinates, the number of fragments, and id), forming a set {Σρ i,j,k}, as shown in formula (6): {Σρ i,j,k}={Σρ i,j,1, Σρ i,j,2,……, Σρ i,j,k}(6), using the sum filtering method, we can obtain the fragment distribution of the effective field of view in each time step for a total of i time steps throughout the night.

[0063] According to the results of orbit calculation and pixelated ball data processing, the fragment distribution of the effective field of view in each time step of 290 time steps on May 19, 2024 is obtained. Using the random sampling method, one effective field of view is taken for observation in each time step, and finally 290 effective observation fields of view in all time steps are obtained, such as Figure 4 As shown in the figure (the first 16 steps are used as a schematic diagram), the final small photoelectric array telescope field of view on the celestial sphere is pointed as follows Figure 5 shown.

[0064] According to the results of orbit calculation and pixelated ball data processing, the fragment distribution of the effective field of view in each time step of 290 time steps on May 19, 2024 is obtained. According to the different data lists of filters in 290 time steps, the filter set maxΣρ with the largest number of fragments in each time step is taken. i ; Delete the observed fragment ID from the total fragment library and update the step information; then calculate the next time step until the time step ends; obtain the filter set maxΣρ with the largest number of fragments in all time steps, use the greedy algorithm method, take an effective field of view observation in each time step, and simulate a time step of two minutes based on the 9°x 18° field of view of a group of telescopes in the small photoelectric array to obtain the optimal telescope pointing for each time step. Finally, 290 effective observation fields of view for all time steps are obtained, that is, the telescope field of view pointing. Figure 7 As shown in the figure (the first 6 steps are used as a schematic diagram), the final small photoelectric array telescope field of view on the celestial sphere is pointed as follows Figure 8 shown.

[0065] The results are compared as follows, where the number of space debris is shown in Table 3:

[0066] Table 3 Comparison of space debris numbers

[0067]

[0068] The number of arc segments is shown in Table 4:

[0069] Table 4 Comparison of arc segment numbers

[0070]

[0071] This example simulates three days of data. Comparing the above two results, it can be seen from the simulation results that in terms of quantity statistics, the number of non-repeating space debris observed by the random observation strategy every night is in the range of 1800-2000, and the number of non-repeating space debris observed by the greedy algorithm is in the range of 4300-4600; in terms of arc segment statistics, the number of arc segments observed by the random observation is in the range of 2700-3500, and the number of arc segments observed by the greedy algorithm is in the range of 5300-5800. Through the early pixel ball processing and sum filtering processing, both random observation and greedy algorithm observation have greatly improved in terms of the number of space debris and arc segments. Compared with random observation, the number of space debris observed by the greedy algorithm is 2.4 times that of random observation, and the number of arc segments is 1.9 times that of random observation. Moreover, the greedy algorithm is more purposeful and will not miss the space debris gathering area. The observation is more scientific, which significantly improves the speed of making telescope debris observation plans and optimizes the degree of automation of debris observation plan making.

[0072] Compared with the prior art, the present invention significantly improves the efficiency of space debris survey observation, which is specifically reflected in: 1. Orbital data optimization, setting parameters according to the observation station site, telescope conditions and observation requirements, simulating the observation of a small photoelectric array, mainly considering the altitude and azimuth angle restrictions, field of view range, and the observation requirements with the largest number of space debris, and obtaining optimized data, which is more conducive to later data processing, such as mobile target detection; 2. Application of the pixelated ball healpix algorithm, using the pixelated ball method for the entire celestial sphere, according to the definition of the resolution of HEALPix, taking the `Nside` parameter as 32, the celestial area is divided into 12,288 equal-area pixel blocks, and the distribution of debris in each pixel block is obtained by calculation, which effectively avoids the deformation problem of the celestial area near the pole, and is helpful for unbiased statistical analysis of data. 3. By adding and filtering, the field of view of the telescope is equivalent to a filter of equal size, which effectively speeds up the data processing speed. 4. Using the greedy algorithm, the best choice in the current state is taken in each time step, the best observation value is collected, and the entire scheduling process is continuously optimized to make the final result the best. Compared with random observation, the number of space debris observed by the greedy algorithm is 2.4 times that of random observation, and the number of arc segments is 1.9 times that of random observation. More space debris and arc segments are conducive to the cataloging of space debris, and the catalog library is continuously enriched, providing strong data support for the cataloging of space debris and collision warning.

Claims

1. A space debris survey observation method based on a greedy algorithm and a pixel ball algorithm, characterized in that: The observation data is obtained by using a large-field-of-view small-aperture telescope array, wherein the large-field-of-view small-aperture telescope array includes four large-field-of-view optical telescopes with an aperture of 15CM, each telescope has a field of view of 10°×10°, and the total field of view of the small photoelectric array composed of four telescopes is 400 square degrees. After obtaining the observation data, the following steps are performed: Step 1: Obtain space debris orbit prediction information, including right ascension, declination, altitude, azimuth and debris ID data of different debris at each moment; Step 2: Preliminary optimization of orbital prediction data. To ensure that all fragments within the set time step are valid and to prevent the situation where a fragment just enters the field of view and then runs out of the field of view, the frequency of occurrence of the same fragment within the time step is processed. When the number of occurrences is greater than or equal to the set value, it meets the processing standard. The same fragment is only counted once within the specified time step, thereby ensuring the maximum number of single non-repeating fragments. Step 3: Data pixelation: The spherical data is discretized using HEALPix pixelation. When mapping data points to the HEALPix grid, the spherical coordinates of the HEALPix grid, including right ascension and declination, are used. Each data point is assigned to the nearest pixel, and the coordinates are converted into pixel indices. Step 4: Perform summing and filtering on the data processed in step 3; Step 5: Based on the data list obtained by summing and filtering, take the filter set maxΣρ with the largest number of fragments in each time step i , delete the observed fragment id from the total fragment library, update the step information, and then calculate the next time step until the time step ends, and obtain the filter set maxΣρ with the largest number of fragments in all time steps, that is, the telescope field of view. In step 5, a greedy algorithm is used to arrange observations every night, that is, the field of view points to the sky area where the largest filter is located in the current time step. The telescope points at the next moment: the sky area closest to the current point and where the largest filter is located in the time step. Before each observation, select the sky area with the highest reward metric to obtain the local optimal solution, which includes: Step 5-1: Initialization: Calculate the field of view pointing set where the filter is the largest at all time steps to obtain the initial solution and solution space; Step 5-2: Select the optimal solution to check the constraints: Select the current optimal solution from the solution space, that is, different solutions with the same density exist at the same time, that is, different sky area directions. When this happens, take the solution with the shortest distance to the current time, that is, the smallest telescope movement distance, as the optimal solution; Step 5-3: Update the problem state: Update the state of the solution space according to the selected solution, and so on to obtain the optimal solution for each time step; Step 5-4: Determine the termination condition: Determine whether the termination condition is met. If all observation times of the current observation night have been traversed, the algorithm ends; otherwise, return to step 5-2.

2. The space debris survey observation method based on the greedy algorithm and the pixel ball algorithm according to claim 1 is characterized in that: The summing and filtering process in step 4 specifically includes: Step 4-1: Take the i-th time step as an example, 1, 2, 3, 4, ... j-1, j represents different pixels, and assume that different pixels N pix The internal debris density is ρ i,j , the fragment density set of all pixels in the i-th time step is {ρ i,j }: {p} i,j }={p i,1, r i,2, r i,3,…… r i,j-1, r i,j }(5); Step 4-2: Consider the effective field of view as a filter. The center of the filter is the center coordinate of the pixel. If the spherical angle distance of the surrounding pixels from the center coordinate is less than half of the field of view, the pixel is considered to be within the filter. Step 4-3: Get a data list based on the filter centered on each pixel (k), traverse each non-empty pixel, and get the distribution of fragments in different filters, including the filter's right ascension and declination coordinates, the number of fragments, and id), forming a set {Σρ i,j,k }, as shown in formula (6): {Σρ i,j,k }={Σρ i,j,1, Σρ i,j,2,……, Σρ i,j,k }(6), using the sum filtering method, we can obtain the fragment distribution of the effective field of view in each time step for a total of i time steps throughout the night.