Space-based Optical Observation Initial Orbit Association Method and System for Space Debris

Abstract

The present invention provides a space-based optical observation initial orbit correlation method and system for space debris, including: Step 1: Obtain imaging data through a space-based optical observation platform and process it to obtain short-arc observation data of an unknown target; Step 2: Manage the obtained short-arc observation data by numbering it according to time; Step 3: Determine the initial orbit of the short arc for the obtained short-arc observation data respectively, and save the original short-arc observation data, the number, and the processed initial orbit; Step 4: Use the processed initial orbit for rough correlation, and if the correlation is successful, execute Step 5; Step 5: Use the processed initial orbit and the original short-arc observation data for fine matching, and if the matching is successful, save the orbit determination information after correlation. The present invention can accurately correlate the initial orbits of un-cataloged objects and can be generally applicable to the initial orbit correlation of space-based optical observations for space debris.

Description

Technical Field

[0001] The present invention relates to the field of orbit correlation technology, and in particular to a method and system for initial orbit correlation of space-based optical observation of space debris. Background Art

[0002] With the rapid development of human spaceflight, the number of man-made objects in space has also increased significantly, most of which are space debris. The sources of space debris include failed spacecraft, launch vehicle bodies, objects abandoned or peeled off by spacecraft, rocket ejecta, and debris from spacecraft explosions or destruction. Space debris will seriously threaten the safety of normal spacecraft, and the world is currently paying great attention to space debris. It is said that the retired Fengyun-1C satellite produced about 3,000 fragments with a diameter of about 15 cm. It once approached the International Space Station and needed to change its orbit to avoid it.

[0003] To catalog and manage space debris using space-based optical observations, it is first necessary to determine the initial orbit of the angle measurement information of the observation arc. The higher the accuracy of the initial orbit, the easier it is to associate with cataloged targets and uncataloged initial orbits (UCTs). At present, the algorithm for associating initial orbits with cataloged targets is relatively mature. The association between a large number of unassociated initial orbits is the key to the initial cataloging, new target discovery, and abnormal movement discovery.

[0004] Reference 1 (JM Maruskin, Correlation of optical observations of objects in earth orbit, Journal of Guidance, Control, and Dynamics, 32 (1)) proposed a correlation method based on covariance propagation. In practical applications, the error matrix of the initial orbit is generally difficult to obtain accurately. Reference 2 (X. Lei, A Geometrical Approach to Association of Space-based Very Short-Arc LEO Tracks, Advances in Space Research, 63 (3)) proposed a geometric correlation algorithm, which uses the semi-major axis error of the initial orbit to iteratively adjust the semi-major axes of the two initial orbits. When the error between the two initial orbits does not exceed 100 km, the correlation accuracy rate reaches 93%. Reference 3 (Du Jianli, Research on Space-based Monitoring System for Space Debris Cataloging, Doctoral Dissertation of Wuhan University, 2018) applied the geometric correlation method to try to correlate 496 initial orbit results, with a correlation accuracy rate of 89.6%.

[0005] Patent document CN110002014A (application number: CN201910218740.5) discloses a method for associating space debris, including the following steps: Step 1, according to two sets of orbital parameters of space debris, select a common time; Step 2, propagate the two sets of orbital parameters in Step 1 to the common time respectively; calculate the difference in semi-major axis and the angle between orbital planes of the two sets of orbital parameters at the common time; Step 3, when both the difference in semi-major axis and the angle between orbital planes in Step 2 are less than the corresponding thresholds, proceed to Step 4; otherwise, determine that the two sets of orbital parameters are different space debris; Step 4, adjust the semi-major axis of the two sets of orbital parameters at the common time, and when the radial deviation, along-track deviation, and orbital plane normal direction deviation of the two sets of orbital parameters at the common time are all less than the corresponding thresholds, determine that the two sets of orbital parameters are the same space debris; otherwise, determine that the two sets of orbital parameters are different space debris.

[0006] To solve the problem of high-precision association between un-cataloged initial orbits, the present invention proposes an initial orbit association method. First, perform rough association according to the orbital plane information of the initial orbits, and then perform fine association based on the original angle measurement data, effectively improving the accuracy of initial orbit association. Summary of the Invention

[0007] Aiming at the defects in the prior art, the purpose of the present invention is to provide a space-based optical observation initial orbit association method and system for space debris.

[0008] The space-based optical observation initial orbit association method for space debris provided by the present invention includes:

[0009] Step 1: Obtain imaging data through a space-based optical observation platform and perform processing to obtain short-arc observation data of an unknown target, including the imaging time of the moving target, the position of the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epoch, and the unit vector of the target relative to the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epoch.

[0010] Step 2: Manage the obtained short-arc observation data by numbering according to time.

[0011] Step 3: Determine the short-arc initial orbits for the obtained short-arc observation data respectively, and save the original short-arc observation data and its number, as well as the initial orbits obtained by processing.

[0012] Step 4: Use the initial orbits obtained by processing to perform rough association. If the association is successful, execute Step 5.

[0013] Step 5: Use the initial orbits obtained by processing and the original short-arc observation data to perform fine matching. If the matching is successful, save the orbit determination information after association.

[0014] Preferably, the initial orbit of the first unassociated arc segment is: time t1, semi-major axis a1, eccentricity e1, inclination i1, right ascension of the ascending node Ω1, argument of perigee ω1, mean anomaly M1; the initial orbit of the second unassociated arc segment is: time t2, semi-major axis a2, eccentricity e2, inclination i2, right ascension of the ascending node Ω2, argument of perigee ω2, mean anomaly M2;

[0015] The rough association is carried out by using the orbital plane threshold method;

[0016] Calculate the coefficient p in the non-singular orbital elements. For the first arc segment, it is p1, and for the second arc segment, it is p2. The expression is:

[0017]

[0018] If |p1 - p2| < Th, it is considered that the rough association is successful, where Th is the rough association threshold.

[0019] Preferably, the imaging times in the original observation data of the first arc segment are successively t 11 , t 12 , …, t 1k , and the position of the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epoch is The pointing of the target relative to the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epoch is i = 0, 1, …, k, where k is the number of samples of the first arc segment;

[0020] The imaging times in the original observation data of the second arc segment are successively t 21 , t 22 , …, t 2j , and the position of the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epoch is The pointing of the target relative to the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epoch is i = 0, 1, …, j, where j is the number of samples of the second arc segment;

[0021] The fine matching in step 5 includes the following steps:

[0022] Step 5.1: Convert the initial orbit of the first arc segment from Keplerian elements to Cartesian elements to obtain the position [rx1, ry1, rz1] T , and the velocity [vx1, vy1, vz1] T , and denote it as the initial value The orbital elements to be determined are denoted as Before the iteration starts

[0023] Step 5.2: Calculate the current orbital elements The corresponding partial derivative matrix H. If from the orbital elements The predicted angle measurement value corresponding to a certain imaging moment obtained by derivation is Then the partial derivative matrix of the predicted value relative to is as follows:

[0024]

[0025] The partial derivative matrix H satisfies:

[0026]

[0027] Step 5.3: First, based on the orbital elements derive the target positions corresponding to t 11 , t 12 , …, t 1k , t 21 , t 22 , …, t 2j moments respectively. Then, calculate the predicted pointing value relative to the space-based optical observation platform from the target positions and normalize it to obtain the current predicted value y g . Then, calculate the residual matrix y between the current predicted value y g and the measured value y c as y = y g - y c ;

[0028] where the expression of the measured value y c is:

[0029]

[0030] Step 5.4: Calculate the root mean square error u, and the expression is:

[0031]

[0032] Step 5.5: If the root mean square error u is less than the preset convergence threshold or reaches the preset upper limit of the number of iteration times, stop the iteration; otherwise, calculate the orbital improvement amount and update the orbital elements and repeat Steps 5.2 to 5.5;

[0033] Step 5.6: Compare the root mean square error at the end of the iteration with the fine correlation threshold. If it is less than the fine correlation threshold, it is determined that the fine correlation is successful.

[0034] Preferably, the coarse correlation threshold Th is set to 0.01;

[0035] The fine correlation threshold is set to 1×10 -4 .

[0036] Preferably, after the fine association is successful, the orbit determination information is managed as follows:

[0037] Calculate the duration t of the associated arc segment, where t = t 2j -t 11 , if the arc segment duration t is greater than the threshold for entering the catalog library, then the updated orbital elements will be converted to the catalog format and managed in the catalog library; otherwise, the short-arc observation data of the two associated arc segments will be merged into one arc segment for management, and the merged short-arc observation data, as well as the number and initial orbit, will be saved.

[0038] According to the space-based optical observation initial orbit association system for space debris provided by the present invention, it includes:

[0039] Module M1: Obtain imaging data through a space-based optical observation platform and process it to obtain short-arc observation data of an unknown target, including the imaging time of the moving target, the position of the space-based optical observation platform in the epoch geocentric celestial coordinate system, and the unit vector of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system;

[0040] Module M2: Manage the obtained short-arc observation data by numbering according to time;

[0041] Module M3: Determine the initial orbit of the short arc for the obtained short-arc observation data respectively, and save the original short-arc observation data, the number, and the processed initial orbit;

[0042] Module M4: Use the processed initial orbit for rough association, and if the association is successful, call Module M5;

[0043] Module M5: Use the processed initial orbit and the original short-arc observation data for fine matching, and if the matching is successful, save the associated orbit determination information.

[0044] Preferably, the initial orbit of the first unassociated arc segment is: time t1, semi-major axis a1, eccentricity e1, inclination i1, right ascension of the ascending node Ω1, argument of perigee ω1, mean anomaly M1; the initial orbit of the second unassociated arc segment is: time t2, semi-major axis a2, eccentricity e2, inclination i2, right ascension of the ascending node Ω2, argument of perigee ω2, mean anomaly M2;

[0045] The rough association is carried out using the orbital plane threshold method;

[0046] Calculate the coefficient p in the non-singular orbital elements, which is p1 for the first arc segment and p2 for the second arc segment. The expression is:

[0047]

[0048] If |p1 - p2| < Th, it is considered that the rough association is successful, where Th is the rough association threshold.

[0049] Preferably, the imaging times in the original observation data of the first arc segment are successively \(t\) 11 , \(t\) 12 , …, \(t\) 1k , and the positions of the space-based optical observation platform in the epoch geocentric celestial coordinate system corresponding thereto are The directions of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system are where \(i = 0, 1, …, k\), and \(k\) is the number of samples in the first arc segment;

[0050] The imaging times in the original observation data of the second arc segment are successively \(t\) 21 , \(t\) 22 , …, \(t\) 2j , and the positions of the space-based optical observation platform in the epoch geocentric celestial coordinate system corresponding thereto are The directions of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system are where \(i = 0, 1, …, j\), and \(j\) is the number of samples in the second arc segment;

[0051] The fine matching in the module M5 includes the following modules:

[0052] Module M5.1: Convert the initial orbit of the first arc segment from Keplerian elements to Cartesian elements to obtain the position \([r_{x1}, r_{y1}, r_{z1}]\) T , and the velocity \([v_{x1}, v_{y1}, v_{z1}]\) T , and denote them as the initial values The orbit elements to be determined are denoted as Before the iteration starts

[0053] Module M5.2: Calculate the current orbit elements The corresponding partial derivative matrix \(H\). If the predicted angle value corresponding to a certain imaging time derived from the orbit elements is then the partial derivative matrix of the predicted value with respect to is as follows:

[0054]

[0055] The partial derivative matrix \(H\) satisfies:

[0056]

[0057] Module M5.3: First, according to the orbit elements , derive the values corresponding to \(t\) 11 , \(t\) 12 , …, \(t\) 1k , \(t\) 21 , \(t\)22 , …, t 2j The target position at the moment, then calculate the pointing prediction value relative to the space-based optical observation platform from the target position and normalize it to obtain the current prediction value y g , and then calculate the current prediction value y g and the measured value y c The residual matrix y = y g - y c ;

[0058] Among them, the expression of the measured value y c is:

[0059]

[0060] Module M5.4: Calculate the root mean square error u, and the expression is:

[0061]

[0062] Module M5.5: If the root mean square error u is less than the preset convergence threshold or reaches the preset upper limit of the number of iteration times, stop the iteration; otherwise, calculate the orbit improvement amount and update the orbital elements Repeat Modules M5.2 to 5.5;

[0063] Module M5.6: Compare the root mean square error at the end of the iteration with the fine correlation threshold. If it is less than the fine correlation threshold, it is determined that the fine correlation is successful.

[0064] Preferably, the coarse correlation threshold Th is set to 0.01;

[0065] The fine correlation threshold is set to 1×10 -4 .

[0066] Preferably, after the fine correlation is successful, the orbit determination information is managed as follows:

[0067] Calculate the duration t of the arc segment after correlation t = t 2j - t 11 , if the arc segment duration t is greater than the threshold for entering the catalog library, convert the updated orbital elements to the catalog format for catalog library management; otherwise, merge the short arc observation data of the two associated arc segments into one arc segment for management, and save the merged short arc observation data as well as the number and initial orbit.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] The present invention can provide a space-based optical observation initial orbit correlation method for space debris. By first performing rough correlation of the initial orbit and then performing fine correlation based on the original angle measurement data, the accuracy of the initial orbit correlation can be generally improved. The method of the present invention is reasonable, simple in calculation, and easy to implement, and can be generally applied to the initial orbit correlation of space-based optical observation platforms and the cataloging of space debris. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] By reading the following detailed description of the non-limiting embodiments with reference to the accompanying drawings, other features, objects, and advantages of the present invention will become more apparent:

[0071] Figure 1 is a flowchart of the present invention;

[0072] Figure 2 is a schematic diagram of the number of groups for rough correlation and screening of 500 arc segments. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] The present invention will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that those of ordinary skill in the art can make several changes and improvements without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0074] Embodiment:

[0075] In order to catalog space debris, it is necessary to determine its orbit information. For a space-based optical observation platform, only the angle measurement information (lacking distance information) of the space debris relative to the space-based optical observation platform can be obtained in one imaging. Therefore, theoretically, three imaging operations are required to determine the orbit of the space debris in the absence of measurement errors. When there are errors in both the angle measurement information of the space-based optical observation platform and its own orbit determination information, longer observation time data is required to ensure the orbit accuracy of the calculated space debris.

[0076] As Figure 1 , since the continuous imaging time of a space-based optical observation platform for the same target at one time may be only a few seconds or a few minutes, the imaging data obtained by the space-based optical observation platform is processed to obtain short arc observation data of unknown targets, including the imaging time of the moving target, the position of the space-based optical observation platform in the epoch geocentric celestial coordinate system, and the pointing information (unit vector) of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system. The initial orbits of multiple observations need to be correlated before further application. First, the short arc observation data obtained by the space-based optical observation platform is numbered and managed according to time.

[0077] The short-arc initial orbit determination is performed separately on the short-arc observation data obtained by the space-based optical observation platform. The original short-arc observation data, as well as the numbers, and the processed initial orbits are saved.

[0078] Due to the ill-posed nature of the initial orbit determination problem, there are multiple solutions and inaccuracies in the semi-major axis and eccentricity of the orbit. Coarse correlation is carried out using the processed initial orbits. If the initial orbit of an uncorrelated arc segment is: time t1, semi-major axis a1, eccentricity e1, inclination i1, right ascension of the ascending node Ω1, argument of perigee ω1, mean anomaly M1, and the initial orbit of another uncorrelated arc segment is: time t2, semi-major axis a2, eccentricity e2, inclination i2, right ascension of the ascending node Ω2, argument of perigee ω2, mean anomaly M2, the general deviation of the orbital plane obtained from the initial orbit determination is relatively small. Therefore, the coarse correlation method is the orbital plane threshold method.

[0079] Calculate the coefficient p in the non-singular orbital elements (also known as regular orbital elements or equinoctial elements), that is, p1 for the first arc segment and p2 for the second arc segment. Then there is:

[0080]

[0081] If |p1 - p2| < Th, it is considered that the coarse correlation is successful. Here, Th is the coarse correlation threshold. The coarse correlation threshold Th is set to 0.01.

[0082] Coarse correlation only screens out the short-arc segments with large differences in the orbital plane. For the arc segments with close orbital planes, fine matching is carried out using the processed initial orbits and the original short-arc observation data.

[0083] If the imaging times in the original observation data of the first arc segment are successively t 11 , t 12 , …, t 1k , and the positions of the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epochs are (i = 0, 1, …, k), and the directions of the target relative to the space-based optical observation platform in the geocentric celestial sphere coordinate system are (i = 0, 1, …, k). k is the number of samples in the first arc segment. The imaging times in the original observation data of the second arc segment are successively t 21 , t 22 , …, t 2j , and the positions of the space-based optical observation platform in the geocentric celestial sphere coordinate system at the epochs are (i = 0, 1, …, j), and the directions of the target relative to the space-based optical observation platform in the geocentric celestial sphere coordinate system are (i = 0, 1, …, j). j is the number of samples in the second arc segment.

[0084] Convert the initial orbit of the first arc segment (time t1, semi-major axis a1, eccentricity e1, inclination i1, right ascension of the ascending node Ω1, argument of perigee ω1, mean anomaly M1) from Keplerian elements to Cartesian elements (time t1, position [rx1, ry1, rz1] T , velocity [vx1, vy1, vz1] T ), and denote it as the initial value The orbit elements to be determined are denoted as Before the iteration starts

[0085] Calculate the current orbit elements The corresponding partial derivative matrix H. If the predicted angle value corresponding to a certain imaging time derived from the orbit elements is Then the partial derivative matrix of the predicted value with respect to is (3 rows, 6 columns) is The partial derivative matrix H satisfies:

[0086]

[0087] Calculate the current predicted value y g and the residual matrix y between the measured value y c is y = y g - y c .

[0088] y g The calculation method is as follows: First, according to the orbit elements derive the target positions corresponding to t 11 , t 12 , …, t 1k , t 21 , t 22 , …, t 2j moments respectively, and then calculate the pointing predicted value relative to the space-based optical observation platform from the target positions and normalize it.

[0089] y c Satisfies:

[0090]

[0091] Calculate the root mean square error u:

[0092]

[0093] If the root mean square error u is less than the convergence threshold or the iteration count reaches the upper limit, stop the iteration; otherwise, calculate the orbit improvement amount and update the orbit elements Repeat steps 5.2 - 5.5.

[0094] The root mean square error at the iteration stop is compared with the threshold Tr, and if it is less than the threshold, the precise correlation is considered successful. The threshold Tr is set to 1×10 -4 .

[0095] If the correlation is successful, the orbit determination information after correlation is managed in the following manner: Calculate the duration t of the arc segment after correlation, where t = t 2j -t 11 . If the arc segment duration t is greater than the threshold for entering the catalog library, the updated orbital elements are converted to the catalog format and managed in the catalog library; otherwise, the short arc observation data of the two associated arc segments are merged into one arc segment for management, and the merged short arc observation data, as well as the number (uniformly the number of the first arc segment) and the initial orbit (converted from the orbital elements to Keplerian elements) are saved.

[0096] The effectiveness of the method of the present invention is verified by combining with the simulation of STK. Two observation arc segment data of 500 groups of space debris are generated by simulation. The simulation data contains measurement noise, where the standard deviation of the angle measurement data error is 6 arcseconds, and the standard deviation of the position error of the space-based optical observation platform is 10 meters. The first arc segment and the second arc segment in the two arc segments are sequentially traversed and tried to be correlated, that is, 250,000 pairs of data, and the correlation success rate is 100%, without false matches and without missed matches. Figure 2 Shown is the number of groups screened out by rough correlation when sequentially traversing the first arc segment to try to correlate with 500 groups of the second arc segment. Taking the processing of a certain arc segment and 500 groups of arc segments to be correlated as an example, 492 groups are screened out by rough matching. The root mean square error of successful precise correlation is 1.94×10 -5 , and the root mean square errors screened out by precise correlation are 0.30, 0.47, 0.33, 0.55, 0.57, 0.41, 0.43.

[0097] Those skilled in the art know that in addition to implementing the system, device and its various modules provided by the present invention in the form of pure computer-readable program code, the method steps can be logically programmed to enable the system, device and its various modules provided by the present invention to be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers, etc. to achieve the same program. Therefore, the system, device and its various modules provided by the present invention can be regarded as a kind of hardware component, and the modules included therein for implementing various programs can also be regarded as the structure within the hardware component; the modules for implementing various functions can also be regarded as both software programs for implementing the method and the structure within the hardware component.

[0098] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for initial orbit correlation of space - based optical observation for space debris, characterized in that, Including: Step 1: Obtain imaging data through a space-based optical observation platform and process it to obtain short-arc observation data of unknown targets, including the imaging time of moving targets, the position of the space-based optical observation platform in the epoch geocentric celestial coordinate system, and the unit vector of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system; Step 2: Manage the obtained short-arc observation data by numbering them according to time; Step 3: Determine the initial short-arc orbits for the obtained short-arc observation data respectively, and save the original short-arc observation data, their numbers, and the processed initial orbits; Step 4: Conduct rough correlation using the processed initial orbits. If the correlation is successful, execute Step 5; Step 5: Conduct fine matching using the processed initial orbits and the original short-arc observation data. If the matching is successful, save the orbit determination information after correlation; The initial orbit of the first uncorrelated arc segment is: time t1, semi-major axis a1, eccentricity e1, inclination i1, right ascension of the ascending node Ω1, argument of perigee ω1, mean anomaly M1; The initial orbit of the second uncorrelated arc segment is: time t2, semi-major axis a2, eccentricity e2, inclination i2, right ascension of the ascending node Ω2, argument of perigee ω2, mean anomaly M2; Adopt the orbital plane threshold method for rough correlation; Calculate the coefficient p in the non-singular orbital elements. For the first arc segment, it is p1, and for the second arc segment, it is p2. The expression is: If |p1 - p2| < Th, it is considered that the rough correlation is successful, where Th is the rough correlation threshold; The imaging times in the original observation data of the first arc segment are successively \(t\) 11 , \(t\) 12 , …, \(t\) 1k . The position of the space-based optical observation platform in the epoch geocentric celestial coordinate system corresponding to these times is The pointing of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system is where \(i = 0, 1, \ldots, k\), and \(k\) is the number of samples in the first arc segment; The imaging times in the original observation data of the second arc segment are successively \(t\) 21 , \(t\) 22 , …, \(t\) 2j , and the position of the space-based optical observation platform in the epoch geocentric celestial coordinate system is The pointing of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system is where \(i = 0, 1, \ldots, j\), and \(j\) is the number of samples in the second arc segment; The fine matching in Step 5 includes the following steps: Step 5.1: Convert the initial orbit of the first arc segment from Keplerian elements to Cartesian elements to obtain the position [rx1, ry1, rz1] T , the velocity [vx1, vy1, vz1] T , and denote it as the initial value The orbital elements to be determined are denoted as Before the iteration starts Step 5.2: Calculate the current orbital elements The corresponding partial derivative matrix H. If the angle measurement prediction value corresponding to a certain imaging time derived from the orbital elements is then the partial derivative matrix of the prediction value with respect to is as follows: The partial derivative matrix H satisfies: Step 5.3: First, based on the orbital elements derive the target positions corresponding to t 11 , t 12 , …, t 1k , t 21 , t 22 , …, t 2j moments respectively. Then, calculate the pointing prediction values relative to the space-based optical observation platform from the target positions and normalize them to obtain the current prediction value y g . Then, calculate the residual matrix between the current prediction value y g and the measured value y c : y = y g - y c ; where the measured value y c is expressed as: Step 5.4: Calculate the root mean square error u. The expression is: Step 5.5: If the root mean square error u is less than the preset convergence threshold or the preset upper limit of the number of iterations is reached, stop the iteration; otherwise, calculate the orbit improvement amount and update the orbital elements Repeat steps 5.2 to 5.5; Step 5.6: Compare the root mean square error at the end of iteration with the fine correlation threshold. If it is less than the fine correlation threshold, it is determined that the fine correlation is successful.

2. The method for initial orbit correlation of space-based optical observation for space debris according to claim 1, wherein The rough correlation threshold Th is set to 0.01; The fine correlation threshold is set to 1×10 -4 .

3. The space-based optical observation initial orbit correlation method for space debris according to claim 1, characterized in that After the fine correlation is successful, manage the orbit determination information in the following manner: The duration t of the associated arc segment after calculation is t 2j - t 11 . If the duration t of the arc segment is greater than the threshold for entering the catalog library, the updated orbital elements will be converted to the catalog format for management in the catalog library; otherwise, the short-arc observation data of the two associated arc segments will be merged into one arc segment for management, and the merged short-arc observation data, as well as the number and initial orbit, will be saved.

4. A space-based optical observation initial orbit correlation system for space debris, characterized in that, Including: Module M1: Obtain imaging data through a space-based optical observation platform and process it to obtain short-arc observation data of unknown targets, including the imaging time of moving targets, the position of the space-based optical observation platform in the epoch geocentric celestial coordinate system, and the unit vector of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system; Module M2: Manage the obtained short-arc observation data by numbering them according to time; Module M3: Determine the initial short-arc orbits for the obtained short-arc observation data respectively, and save the original short-arc observation data, their numbers, and the processed initial orbits; Module M4: Conduct rough correlation using the processed initial orbits. If the correlation is successful, call Module M5; Module M5: Conduct fine matching using the processed initial orbits and the original short-arc observation data. If the matching is successful, save the orbit determination information after correlation; The initial orbit of the first uncorrelated arc segment is: time t1, semi-major axis a1, eccentricity e1, inclination i1, right ascension of the ascending node Ω1, argument of perigee ω1, mean anomaly M1; The initial orbit of the second uncorrelated arc segment is: time t2, semi-major axis a2, eccentricity e2, inclination i2, right ascension of the ascending node Ω2, argument of perigee ω2, mean anomaly M2; Adopt the orbital plane threshold method for rough correlation; Calculate the coefficient p in the root number without singularities. For the first arc segment, it is p1, and for the second arc segment, it is p2. The expression is as follows: If |p1 - p2| < Th, it is considered that the rough correlation is successful, where Th is the rough correlation threshold; The imaging times in the original observation data of the first arc segment are successively \(t\) 11 , \(t\) 12 , …, \(t\) 1k , and the positions of the space-based optical observation platform in the epoch geocentric celestial coordinate system corresponding to them are The pointing of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system is where \(i = 0, 1, \ldots, k\), and \(k\) is the number of samples in the first arc segment; The imaging times in the original observation data of the second arc segment are successively \(t\) 21 , \(t\) 22 , …, \(t\) 2j . The position of the space-based optical observation platform in the epoch geocentric celestial coordinate system corresponding to this is The pointing of the target relative to the space-based optical observation platform in the epoch geocentric celestial coordinate system is where \(i = 0, 1, \ldots, j\), and \(j\) is the number of samples in the second arc segment; The fine matching in the module M5 includes the following modules: Module M5.1: Convert the initial orbit of the first arc segment from Keplerian elements to Cartesian elements to obtain the position [rx1, ry1, rz1] T , velocity [vx1, vy1, vz1] T , and denote it as the initial value The orbital elements to be determined are denoted as Before the iteration starts Module M5.2: Calculate the current orbital elements The corresponding partial derivative matrix H. If the angle measurement prediction value corresponding to a certain imaging time derived from the orbital elements is then the partial derivative matrix of the prediction value with respect to is as follows: The partial derivative matrix H satisfies: Module M5.3: First, based on the orbital elements derive the target positions corresponding to t 11 , t 12 , …, t 1k , t 21 , t 22 , …, t 2j moments respectively. Then, calculate the pointing prediction value relative to the space-based optical observation platform from the target positions and normalize it to obtain the current prediction value y g . Then, calculate the residual matrix between the current prediction value y g and the measured value y c as y = y g - y c ; Among them, the measured value y c has the following expression: Module M5.4: Calculate the root mean square error u. The expression is as follows: Module M5.5: Stop iteration if the root mean square error u is less than the preset convergence threshold or the preset upper limit of the number of iterations is reached; otherwise, calculate the orbit improvement amount and update the orbital elements Repeat Modules M5.2 to 5.5; Module M5.6: Compare the root mean square error at the end of iteration with the fine correlation threshold. If it is less than the fine correlation threshold, it is determined that the fine correlation is successful.

5. The space-based optical observation initial orbit correlation system for space debris according to claim 4, characterized in that The rough correlation threshold Th is set to 0.01; The fine correlation threshold is set to 1×10 -4 .

6. The space-based optical observation initial orbit correlation system for space debris according to claim 4, characterized in that After the fine correlation is successful, manage the orbit determination information in the following manner: The duration t of the arc segment after calculation association is t 2j -t 11 . If the duration t of the arc segment is greater than the threshold for entering the catalog library, the updated orbital elements will be converted to the catalog format for catalog library management; otherwise, the short arc observation data of the two associated arc segments will be merged into one arc segment for management, and the merged short arc observation data, as well as the number and initial orbit, will be saved.

Images