A key target in-orbit observation planning method and system based on spatial region division

Abstract

The application provides a key target neighborhood on-orbit observation planning method and system based on spatial region division, and the planning method comprises the following steps: step 1, determining an observation region; step 2, single observation coverability judgment; step 3, region segmentation based on a minimum observation range; and step 4, observation task planning for the segmented sub-regions based on a greedy strategy. Through the minimum observation range, the maximum value of the side length of the segmented region satisfying the observation conditions of all observation resources is obtained by calculating the minimum imaging range, and the number of segmented regions of the to-be-observed region is calculated, so that the wide observation region is segmented, the wide observation problem is converted into a multiple region round observation problem, and the key target neighborhood observation problem is efficiently solved under the condition of low calculation power based on the greedy idea through the rule setting.

Description

Technical Field

[0001] This invention pertains to satellite observation mission planning, specifically involving a method and system for planning on-orbit observation of key targets in their vicinity based on spatial region division. Background Technology

[0002] With the advancement of science and technology, countries around the world have placed higher demands on acquiring information about the status and movement of space targets, leading to the vigorous development of satellite platforms capable of close-range observation. Countries, primarily the United States, have conducted numerous technical verifications of rapid-approach reconnaissance technologies. The Geosynchronous Space Situational Awareness Program (GSSAP), a highly maneuverable reconnaissance system, utilizes the unique orbital characteristics of satellites to maneuver close to and continuously monitor space targets of interest. Leveraging its high-precision reconnaissance capabilities, it obtains characteristic information about target objects, enabling further judgment on the target's behavioral intentions and activity patterns.

[0003] With the development of my country's aerospace industry, the need for close-range observation of key targets such as the Tiangong space station has become increasingly prominent. Therefore, it is particularly necessary to study the observation and sensing technology of the neighborhood of key targets and achieve observation coverage of the neighborhood of key targets.

[0004] Meanwhile, employing space-based wide-area surveillance satellites, through a network of multiple satellites, can effectively overcome many shortcomings of ground-based observations, such as susceptibility to atmospheric conditions and the inability to conduct all-weather monitoring. This provides the advantage of enabling all-day, wide-area surveillance missions. Furthermore, the use of onboard autonomous planning methods allows for autonomous observation even when communication with the ground is impossible.

[0005] Currently, there is limited research on the problem of observing the neighborhood of key targets in space, and the relevant models and methods are still immature. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention proposes a method and system for on-orbit observation planning of key target neighborhoods based on spatial region division.

[0007] This invention first proposes a spatial region-based on-orbit observation planning method for key targets' neighborhoods. This method addresses the planning problem of observation missions in the neighborhoods of key space targets, considering the limitations of onboard computing resources. To ensure efficient utilization of satellite observation resources and rapid observation coverage of the observed area, a spatial region division method is adopted, dividing the observed area based on the minimum observation range. After region division, a mathematical model for the planning problem of observation missions in the divided sub-regions is constructed. Indirect observations in the divided sub-regions are evaluated using a dynamic priority calculation method. Through a greedy strategy-based planning method for observation missions in the divided sub-regions, observation coverage of the entire key target's observed area is achieved. The method specifically includes the following steps:

[0008] Step 1, Determine the observation area: Take the target's centroid position as the center of the area, and take the maximum geometric size of the target as the reference length a of the cubic area. Determine different expansion ratios k of the cubic area according to different types of targets. The observation area is a cubic area with a side length of (1+k)a. This area is oriented relative to the ground and moves and rotates with the movement of the key target.

[0009] Step 2, Single observation coverage judgment: Calculate the time window of the observation area for m satellites in the constellation, and perform a single observation coverage judgment. If the single observation coverage condition is met, output the window with the longest observation duration. If the single observation coverage condition is not met, proceed to step 3.

[0010] Step 3, perform region segmentation based on minimum observation range: with the target centroid as the center, segment the observation area along the coordinate axis of the target's own system, and calculate the time window set ListWin for m satellites in the constellation to the segmented sub-regions;

[0011] Step 4, planning observation tasks for the segmented sub-regions based on a greedy strategy: planning observation tasks for the segmented sub-regions transforms the regional observation problem into a round-robin observation problem for the segmented sub-regions.

[0012] Furthermore, the single observation coverage condition described in step 2 is as follows:

[0013] l=2dtanθ

[0014]

[0015] In the formula, d is the actual distance between the observation platform and the target, θ is the sensor half-angle, and α is the correction coefficient.

[0016] Furthermore, the single-observation coverage determination method described in step 2 is an algorithm based on a greedy strategy, as follows:

[0017] Step 2.1: Calculate the time windows for observing key targets from m satellites within the constellation, obtaining the observation task set TSK = (tsk1, tsk2, ..., tsk...). i ), tsk i = <sat i ,tar i ST i ,ET i >, tsk i For the time window, where sat i For satellite numbering, tar i For the target number, ST i ET is the start time of the time window. i The end time of the time window, time window tsk i The duration is ET i -ST i ;

[0018] Step 2.2, if the time window tsk i The duration is ET i -ST i Greater than or equal to the minimum duration requirement t for a single observation min_observe , t min_observe If the task requirements are set, proceed to step 2.3; if the time window is tsk i The duration is ET i -ST i Less than the minimum duration requirement for a single observation t min_observe If so, the single-coverage condition is not met;

[0019] Step 2.3: Within the time window, calculate the shortest distance d from a single satellite to the center point of the observation area. min ;

[0020] Step 2.4: Calculate the required distance d between the satellite and the center point of the observation area so that the sensor can cover the entire observation area in a single operation. request ,

[0021] Step 2.5, determine d min Is it greater than or equal to d? request If the value is greater than or equal to the value, the observation time window satisfies the single-time coverage condition; otherwise, proceed to step 2.6.

[0022] Step 2.6, determine if the spacing d is greater than d request Duration t observe Is it greater than or equal to the minimum duration requirement t for a single observation? min_observeIf the condition is met, then a segment of the time window satisfies the single-coverage condition; otherwise, the single-observation coverage condition is not met. If the single-coverage condition is met, then the window with the longest observation duration is output. If the single-coverage condition is not met, then the observation area is divided into regions, and step 3 is executed.

[0023] Furthermore, the specific steps for region segmentation described in step 3 are as follows:

[0024] Step 3.1, calculate the minimum imaging range l min :

[0025] l min =2d min tanθ;

[0026] Step 3.2, calculate the maximum side length b of the segmented sub-region. max :

[0027] Let the side length of the partitioned sub-region be b, and calculate the condition that satisfies... The maximum value of b, that is:

[0028]

[0029] Step 3.3, calculate the number of sub-regions:

[0030] Calculate the minimum value n that satisfies nb≥(1+k)a, and the number of sub-regions after segmentation is n*n*n. If the expansion in the normal direction is not considered, the number of sub-regions after segmentation is n*n.

[0031] Step 3.4: Calculate the location trajectory of the segmented sub-regions;

[0032] Step 3.5: Calculate the observation time window for the segmented sub-regions:

[0033] Based on the trajectory data of the segmented sub-regions obtained in step 3.4, the time window of the sensor for the segmented sub-regions is calculated, and the set of time windows ListWin for the segmented sub-regions of m satellites in the constellation is obtained.

[0034] Furthermore, the method for calculating the location trajectory of the segmented sub-regions in step 3.4 is as follows:

[0035] The observation area is always oriented relative to the Earth. The coordinates of the regional observation center in the J2000 coordinate system are (X... b ,Y b Z b For a target area, a point within that area with coordinates (x, y, z) in the body coordinate system has coordinates (X, Y, Z) in the J2000 system as follows:

[0036]

[0037] In the coordinate system of the regional target, the x-direction is the velocity direction, and the z-direction is always parallel to the line connecting the observation center of the regional target and the Earth's center, and points towards the zenith.

[0038] Let ψ be the elevation angle of the regional target observation center in the J2000 coordinate system, ψ be the azimuth angle, and C be the elevation angle. bo This is the rotation matrix from the regional target body coordinate system to the J2000 coordinate system;

[0039] Based on the rotation matrix at each time step, the coordinates of the center point of the observed region, and the coordinates of the center points of the segmented sub-regions in the body coordinate system, calculate the coordinates of the center points of each segmented sub-region in the J2000 system.

[0040] Furthermore, step 4 specifically involves the following steps:

[0041] Step 4.1: Perform indirect observation judgment on the segmented sub-regions. Iterate through the time window set ListWin of m satellites in the constellation for the segmented sub-regions and calculate the target number and quantity that can be indirectly observed in each time window.

[0042] Step 4.2: Calculate the dynamic priority of the time window;

[0043] Step 4.3: Select the time window with the highest window priority from ListWin. select And remove win from ListWin select and win select Conflicting time windows exist;

[0044] Step 4.4: Update the number of observations for each sub-region. If all segmented sub-regions have been observed, then the process ends; otherwise, proceed to step 4.5.

[0045] Step 4.5: Determine the number of windows in the time window set ListWin of the center points of the sub-regions after the constellation pair is segmented. If the number of windows in ListWin is greater than 0, then execute step 4.2 to update the dynamic priority of each window. If the number of windows is equal to 0, then end.

[0046] Further, in step 4.1, the indirect observation judgment method specifically involves: determining whether the segmented sub-region b can be indirectly observed when the satellite directly observes the segmented sub-region a, based on the coordinates (x, y) of the segmented sub-region a in the J2000 system. a ,y a ,z a The coordinates (x, y) of the segmented sub-region b in the J2000 system. b ,y b ,z b) and the satellite's coordinates in the J2000 system (x s ,y s ,z s ), calculate the vector from the satellite to the directly observed segmented sub-region a. sa ,

[0047] Vector sa =[x a -x s ,y a -y s ,z a -z s ] T

[0048] Vector from satellite to sub-region b after indirect observation segmentation. sb ,

[0049] Vector sb =[x b -x s ,y b -y s ,z b -z s ] T

[0050] According to the vector sa and Vector sb Calculate the angle θ between the two vectors. v ,

[0051]

[0052] If the included angle θ v Must meet When directly observing the segmented sub-region a, indirect observation of the segmented sub-region b can be achieved, where θ is the half-angle of the sensor's field of view.

[0053] Furthermore, in step 4.2, the function for calculating the dynamic window priority is as follows:

[0054] priority=α1·priority1+α2·priority2

[0055] priority1 is the observation capability benefit, α1 is the observation capability benefit weight, priority2 is the window substitutability benefit, and α2 is the window substitutability benefit weight.

[0056] Furthermore, when considering the expansion in three directions—tangential, radial, and normal to the orbital plane—and the number of sub-regions after segmentation is n*n*n, the formula for calculating the observation capability gain priority1 is:

[0057]

[0058] In the formula, j is the number of times the target is observed. When the target is observed multiple times, the observation weight decreases. brightness is the normalized value of the observed brightness when the i-th target is indirectly observed within the time window.

[0059] The formula for calculating the window substitution benefit priority2 is as follows:

[0060]

[0061] In the formula, x represents the number of windows that can observe the target within the unplanned window. The more windows that can observe the target, the greater the substitutability of the windows and the smaller the benefit. ObseverTimes(i) = 0 indicates that the target is not observed.

[0062] When expansion along the normal direction of the orbital plane is not considered, and the final number of sub-regions is n*n, the calculation formula for the observation capability benefit priority1 is as follows:

[0063]

[0064] The formula for calculating the window substitution benefit priority2 is as follows:

[0065]

[0066] The present invention also discloses a system for implementing the above-mentioned on-orbit observation planning method for key target neighborhoods based on spatial region division, the system comprising a computer module containing the on-orbit observation planning method for key target neighborhoods based on spatial region division.

[0067] Beneficial effects:

[0068] 1. This invention provides a method for determining the observation area based on the scale of key targets, whereby the area to be observed is always oriented towards the Earth.

[0069] 2. This invention uses a minimum observation range approach to calculate the minimum imaging range, obtain the maximum side length of the segmented region that satisfies the observation conditions of all observation resources, and use this to calculate the number of segments of the region to be observed, thereby enabling the segmentation of a wide observation area and transforming the wide-area observation problem into a multi-regional rotation observation problem.

[0070] 3. Considering the limitations of onboard computing power, intelligent optimization algorithms such as genetic algorithms require high computing power and take a long time to solve, making it difficult to complete the calculation on the onboard computer. Based on the greedy idea, this invention achieves efficient solution of the problem of observing the neighborhood of key targets by setting rules, under the condition of using less computing power. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the observation area;

[0072] Figure 2 A schematic diagram for determining whether a single observation can cover the area;

[0073] Figure 3 Flowchart for single-observation coverage determination based on greedy strategy;

[0074] Figure 4 This is a schematic diagram of the region segmentation method;

[0075] Figure 5 This is a schematic diagram showing the relative relationship between the J2000 coordinate system and the coordinate system of the target body in the region;

[0076] Figure 6 A schematic diagram for determining the location trajectory of segmented cells;

[0077] Figure 7 This is a schematic diagram for indirect observation and judgment.

[0078] Figure 8 The flowchart for planning the observation task of the segmented sub-region based on the greedy strategy is shown below.

[0079] Figure 9 This is a Gantt chart showing the task planning results of Example 1;

[0080] Figure 10 This is a statistical chart showing the number of observations of sub-regions after region segmentation in Example 1;

[0081] Figure 11 This is a statistical chart showing the number of sub-regions that can be covered by each window observation in Example 1;

[0082] Figure 12 This is a Gantt chart showing the planning results of Example 2;

[0083] Figure 13 This is a statistical chart showing the number of observations of sub-regions after region segmentation in Example 2;

[0084] Figure 14 This is a statistical chart showing the number of sub-regions that can be covered by each window observation in Example 2. Detailed Implementation

[0085] The following discloses a specific implementation method for on-orbit observation planning of key target neighborhoods based on spatial region division, including the following steps:

[0086] Step 1, determine the observation area: such as Figure 1 As shown, with the target's centroid position O as the center of the region, and the maximum value of the target's geometric dimensions as the reference length a of the cubic region, different expansion ratios k of the cubic region are determined according to different types of targets. The observation region is a cubic region with a side length of (1+k)a. Considering that the application of regional observation is mainly the observation of potential targets around key targets such as spacecraft, the observation region should maintain the relative position with the center point of the region to be observed. Therefore, the region to be observed is always oriented towards the ground and moves and rotates with the movement of the target.

[0087] For a key target with a reference length of *a*, the observation area is a cube with a side length of (1+k)a. When *k* = 0, the observation area is the key target itself. The size of the observation area increases with the value of *k*. Considering practical engineering applications, the value of *k* should be between 10 and 10. 3 ~10 4 Magnitude;

[0088] Step 2, Single observation coverage judgment: Calculate the time window of the observation area for m satellites in the constellation, and perform a single observation coverage judgment. If the single observation coverage condition is met, output the window with the longest observation duration. If the single observation coverage condition is not met, proceed to step 3.

[0089] If, within the visible window of the target center from the satellite, there exists a period during which the satellite sensor's field of view can cover the entire observation area, then a single observation can cover it.

[0090] like Figure 2 As shown, when the sensor's optical axis points to the target center, the actual imaging range of the sensor is l×l, where l = 2dtanθ, and d is the actual distance between the observation platform and the target, and θ is the sensor's half-angle. To ensure coverage of the observation cube region regardless of the observation angle, the determination of whether a single observation can achieve coverage is achieved by judging the relationship between l and the volume diagonal of the region cube. The volume diagonal of the region cube is... To ensure coverage in a single observation, and considering the impact of calculation errors, a coefficient α needs to be added. The final geometric condition for determining coverage is as follows: Where d is the actual distance between the observation platform and the target, θ is the sensor half-angle, and α is the correction coefficient.

[0091] The single-observation coverage determination method described in step 2 is an algorithm based on a greedy strategy, as detailed below:

[0092] Step 2.1: Calculate the time windows for observing key targets from m satellites within the constellation, obtaining the observation task set TSK = (tsk1, tsk2, ..., tsk...). i ), tsk i = <sat i ,tar i ST i ,ET i >, tsk i For the time window, where sat i For satellite numbering, tar i For the target number, ST i ET is the start time of the time window. i The end time of the time window, time window tsk i The duration is ET i -ST i ;

[0093] Step 2.2, if the time window tsk i The duration is ET i -ST i Greater than or equal to the minimum duration requirement t for a single observation min_observe , t min_observe If the task requirements are set, proceed to step 2.3; if the time window is tsk i The duration is ET i -ST i Less than the minimum duration requirement for a single observation t min_observe If so, the single-coverage condition is not met;

[0094] Step 2.3: Within the time window, calculate the shortest distance d from a single satellite to the center point of the observation area. min ;

[0095] Step 2.4: Calculate the required distance d between the satellite and the center point of the observation area so that the sensor can cover the entire observation area in a single operation. request ,

[0096] Step 2.5, determine d min Is it greater than or equal to d? request If the value is greater than or equal to the value, the observation time window satisfies the single-time coverage condition; otherwise, proceed to step 2.6.

[0097] Step 2.6, determine if the spacing d is greater than d request Duration t observe Is it greater than or equal to the minimum duration requirement t for a single observation? min_observeIf the conditions are met, then a portion of the time window satisfies the single-observation coverage condition; otherwise, the single-observation coverage condition is not met. If the single-observation coverage condition is met, the window with the longest observation duration is output. If the single-observation coverage condition is not met, the observation area is divided into regions, and step 3 is executed. See the flowchart for the single-observation coverage determination algorithm. Figure 3 .

[0098] Step 3, perform region segmentation based on minimum observation range: Using the target's centroid as the center, segment the observation area along the coordinate axes of the target's own system (or orbital system if attitude is involved). A segmentation diagram is shown below. Figure 4 As shown, the time window set ListWin for the segmented sub-regions of m satellites within the constellation is calculated.

[0099] The specific steps for region segmentation described in step 3 are as follows:

[0100] Step 3.1, calculate the minimum imaging range l min :

[0101] l min =2d min tanθ;

[0102] Step 3.2, calculate the maximum side length b of the segmented sub-region. max :

[0103] Let the side length of the partitioned sub-region be b, and calculate the condition that satisfies... The maximum value of b, that is:

[0104]

[0105] Step 3.3, calculate the number of sub-regions:

[0106] Calculate the minimum value n that satisfies nb≥(1+k)a, and the number of sub-regions after segmentation is n*n*n. If the expansion in the normal direction is not considered, the number of sub-regions after segmentation is n*n.

[0107] Step 3.4: Calculate the location trajectory of the segmented sub-regions. The specific method is as follows:

[0108] like Figure 5 , 6 As shown, the observation area is always oriented relative to the Earth. The coordinates of the regional observation center in the J2000 coordinate system are (X... b ,Y b Z b For a target area, a point within that area with coordinates (x, y, z) in the body coordinate system has coordinates (X, Y, Z) in the J2000 system as follows:

[0109]

[0110] In the coordinate system of the regional target, the x-direction is the velocity direction, and the z-direction is always parallel to the line connecting the observation center of the regional target and the Earth's center, and points towards the zenith.

[0111] Let ψ be the elevation angle of the regional target observation center in the J2000 coordinate system, ψ be the azimuth angle, and C be the elevation angle. bo This is the rotation matrix from the regional target body coordinate system to the J2000 coordinate system;

[0112] Based on the rotation matrix at each moment, the coordinates of the center point of the observation area, and the coordinates of the center point of the segmented sub-region in the body coordinate system, calculate the coordinates of the center point of each segmented sub-region in the J2000 system.

[0113] Step 3.5: Calculate the observation time window for the segmented sub-regions:

[0114] Based on the trajectory data of the segmented sub-regions obtained in step 3.4, the time window of the sensor for the segmented sub-regions is calculated, and the set of time windows ListWin for the segmented sub-regions of m satellites in the constellation is obtained.

[0115] Step 4, planning observation tasks for the segmented sub-regions based on a greedy strategy: planning observation tasks for the segmented sub-regions transforms the regional observation problem into a round-robin observation problem for the segmented sub-regions.

[0116] Step 4, the task planning process, is shown below. Figure 8 The specific steps are as follows:

[0117] Step 4.1: Perform indirect observation judgment on the segmented sub-regions. Iterate through the time window set ListWin of m satellites in the constellation for the segmented sub-regions and calculate the target number and quantity that can be indirectly observed in each time window.

[0118] Considering the uncertainty of the expansion of the segmented region, the number of targets in the segmented unit may be large. If we adopt the method of observing each segmented sub-region completely, it is difficult to guarantee that all sub-regions can be quickly circulated in a short time. Therefore, it is necessary to fully consider the observation overlap of the sub-regions and adopt a combination of direct observation and indirect observation.

[0119] Direct observation is defined as the observation method in which the central axis of the sensor's field of view points to the center point of the region, while indirect observation is the observation method in which the sensor can cover the entire region after the segmentation when the center of the sensor's field of view points to the center point of the other sub-regions.

[0120] like Figure 7As shown, the indirect observation judgment method is as follows: Determine whether the segmented sub-region b can be indirectly observed when the satellite directly observes the segmented sub-region a. This is based on the coordinates (x, y) of the segmented sub-region a in the J2000 system. a ,y a ,z a The coordinates (x, y) of the segmented sub-region b in the J2000 system. b ,y b ,z b ) and the satellite's coordinates in the J2000 system (x s ,y s ,z s ), calculate the vector from the satellite to the directly observed segmented sub-region a. sa ,

[0121] Vector sa =[x a -x s ,y a -y s ,z a -z s ] T

[0122] Vector from satellite to sub-region b after indirect observation segmentation. sb ,

[0123] Vector sb =[x b -x s ,y b -y s ,z b -z s ] T

[0124] According to the vector sa and Vector sb Calculate the angle θ between the two vectors. v ,

[0125]

[0126] If the included angle θ v Must meet When directly observing the segmented sub-region a, indirect observation of the segmented sub-region b can be achieved, where θ is the half-angle of the sensor's field of view.

[0127] Step 4.2, calculate the dynamic priority of the time window:

[0128] The function for calculating the priority of a dynamic window is as follows:

[0129] priority=α1·priority1+α2·priority2

[0130] priority1 is the observation capability benefit, α1 is the observation capability benefit weight, priority2 is the window substitutability benefit, and α2 is the window substitutability benefit weight.

[0131] When considering expansion in three directions—tangential, radial, and normal to the orbital plane—and the number of sub-regions after segmentation is n*n*n, the formula for calculating the observation capability gain priority1 is:

[0132]

[0133] In the formula, j is the number of times the target is observed. When the target is observed multiple times, the observation weight decreases. brightness is the normalized value of the observed brightness when the i-th target is indirectly observed within the time window.

[0134] The formula for calculating the window substitution benefit priority2 is as follows:

[0135]

[0136] In the formula, x represents the number of windows that can observe the target within the unplanned window. The more windows that can observe the target, the greater the substitutability of the windows and the smaller the benefit. ObseverTimes(i) = 0 indicates that the target is not observed.

[0137] When expansion along the normal direction of the orbital plane is not considered, and the final number of sub-regions is n*n, the calculation formula for the observation capability benefit priority1 is as follows:

[0138]

[0139] The formula for calculating the window substitution benefit priority2 is as follows:

[0140]

[0141] Step 4.3: Select the time window with the highest window priority from ListWin. select And remove win from ListWin select and win select Conflicting time windows exist;

[0142] Step 4.4: Update the number of observations for each sub-region. If all segmented sub-regions have been observed, then the process ends; otherwise, proceed to step 4.5.

[0143] Step 4.5: Determine the number of windows in the time window set ListWin of the center points of the sub-regions after the constellation pair is segmented. If the number of windows in ListWin is greater than 0, then execute step 4.2 to update the dynamic priority of each window. If the number of windows is equal to 0, then end.

[0144] Example 1: This example is a simulation result of spatial 3D extended simulation.

[0145] A key target with an orbital altitude of 380km and an orbital inclination of 51° is selected. The target's baseline length is a = 110m, the cube region expansion ratio is k = 2000, and the correction factor is α = 0.05. The search area is a 220×220×220km cube region. The planning start time is from 02:00:00.000 to 03:00:00.000 on July 4, 2023 (UTCG time), with a planning duration of 1 hour.

[0146] The observation windows of the six satellites in the constellation for observing key targets are calculated, totaling seven. None of the seven time windows meet the single observation conditions in step 2.

[0147] The search region is segmented based on the minimum observation range, and the maximum side length b of the segmented region is calculated. max = 84.932km, and the number of sub-regions after division is 3×3×3.

[0148] For a task scenario involving searching and observing the vicinity of a key target within a 1-hour planning period, observation task planning was completed for 27 segmented sub-regions after region segmentation. The resulting Gantt chart, a statistical chart of the number of observations of each segmented sub-region, and a statistical chart of the number of segmented sub-regions that can be covered by observations in each window are shown below. Figures 9-11 As shown.

[0149] Depend on Figure 9 , 10 It can be seen that the search and observation of all sub-regions after segmentation of the search area were completed within 1600s, and 33.3% of the segmented sub-regions were observed twice, which can effectively achieve a rapid response to the needs of regional observation tasks; Figure 11 It can be seen that in most cases, each observation window in the planning results can cover at least 3 or more sub-regions at the same time, and up to 5 sub-regions at the same time, which can achieve efficient solution of the observation problem of the neighborhood of key targets under the condition of low computing power.

[0150] Example 2: This example is a simulation result of spatial 2D extended simulation.

[0151] A key target with an orbital altitude of 380km and an orbital inclination of 51° is selected. The target's baseline length is a = 110m, the cube region expansion ratio is k = 2000, and the correction factor is α = 0.05. The search area is a 220×220km planar region. The planning start time is from 02:00:00.000 to 03:00:00.000 on July 4, 2023 (UTCG time), with a planning duration of 1 hour.

[0152] The observation windows of the six satellites in the constellation for observing key targets are calculated, totaling seven. None of the seven time windows meet the single observation conditions in step 2.

[0153] The search region is segmented based on the minimum observation range, and the maximum side length b of the segmented region is calculated. max = 84.932km, and the number of sub-regions after division is 3×3.

[0154] For a task scenario involving searching and observing the vicinity of a key target within a 1-hour planning period, observation task planning was completed for nine sub-regions after region segmentation. The resulting Gantt chart, a statistical chart of the number of observations of each sub-region, and a statistical chart of the number of sub-regions covered by observations in each window are shown below. Figures 12-14 As shown.

[0155] Depend on Figure 12 , 13 It can be seen that the search and observation of all sub-regions after segmentation of the area to be observed was completed within 220 seconds, which can effectively achieve a rapid response to the needs of regional observation tasks; by Figure 14 It can be seen that each observation window in the planning results can cover at least 2 segmented sub-regions at the same time, and up to 5 segmented sub-regions at the same time, which can achieve efficient solution of the observation problem of the neighborhood of key targets under the condition of low computing power.

Claims

1. A method for planning on-orbit observation of key targets in their vicinity based on spatial region division, comprising the following steps: Step 1, Determine the observation area: Using the target's centroid as the center of the area, and the maximum geometric size of the target as the base length *a* of the cubic region, determine different expansion ratios *k* for the cubic region based on different target categories. The observation area has a side length of... A cubic region that is oriented relative to the ground and moves and rotates in response to the movement of the target. Step 2, Single observation coverage judgment: Calculate the time window of the observation area for m satellites in the constellation, and perform a single observation coverage judgment. If the single observation coverage condition is met, output the window with the longest observation duration. If the single observation coverage condition is not met, proceed to step 3. Step 3, perform region segmentation based on minimum observation range: Using the target's centroid as the center, segment the observation region along the target's intrinsic coordinate axes, and calculate the time window set for each of the m satellites within the constellation for the segmented sub-region. ; Step 4, planning observation tasks for the segmented sub-regions based on a greedy strategy: planning observation tasks for the segmented sub-regions, transforming the regional observation problem into a round-robin observation problem for the segmented sub-regions; The single observation coverage condition mentioned in step 2 is: In the formula, d is the actual distance between the observation platform and the target. The sensor's half-angle. This is a correction factor; The single-observation coverage determination method described in step 2 is an algorithm based on a greedy strategy, as detailed below: Step 2.1: Calculate the time window for the observation target from the m satellites within the constellation to obtain the set of observation tasks. , , , For the time window, where Number the satellite. Number the target. The start time of the time window. The end time of the time window, the time window The duration is ; Step 2.2, if the time window The duration is Greater than or equal to the minimum duration requirement for a single observation , If the task requirements are set, proceed to step 2.3; if the time window... The duration is Less than the minimum duration requirement for a single observation If so, the single-coverage condition is not met; Step 2.3: Within the time window, calculate the shortest distance from a single satellite to the center point of the observation area. ; Step 2.4: Calculate the required distance between the satellite and the center point of the observation area so that the sensor can cover the entire observation area in a single operation. , ; Step 2.5, Determine Is it greater than or equal to? If the value is greater than or equal to the value, the observation time window satisfies the single-time coverage condition; otherwise, proceed to step 2.

6. Step 2.6, determine if the spacing d is greater than Duration Is it greater than or equal to the minimum duration requirement for a single observation? If the condition is met, then a segment of the time window satisfies the single-coverage condition; otherwise, the single-observation coverage condition is not met. If the single-coverage condition is met, then the window with the longest observation duration is output. If the single-coverage condition is not met, then the observation area is divided into regions, and step 3 is executed. The specific steps for region segmentation described in step 3 are as follows: Step 3.1, calculate the minimum imaging range : ； Step 3.2, calculate the maximum side length of the segmented sub-region. : Let the side length of the partitioned sub-region be b, and calculate the condition that satisfies... The maximum value of b, that is: ； Step 3.3, calculate the number of sub-regions: Calculation satisfies The minimum value of n is used to obtain the number of sub-regions after segmentation. If we disregard expansion along the normal direction, the number of sub-regions obtained after segmentation is: ; Step 3.4: Calculate the location trajectory of the segmented sub-regions; Step 3.5: Calculate the observation time window for the segmented sub-regions: Based on the trajectory data of the segmented sub-regions obtained in step 3.4, the time window of the sensor for the segmented sub-regions is calculated, resulting in the set of time windows for the segmented sub-regions for the m satellites in the constellation. ; Step 4 consists of the following steps: Step 4.1: Perform indirect observation and judgment on the segmented sub-regions, traversing the time window set of m satellites within the constellation for the segmented sub-regions. Calculate the target number and quantity that can be indirectly observed in each time window; Step 4.2: Calculate the dynamic priority of the time window; Step 4.3, in Select the time window with the highest priority. and in Delete and with Conflicting time windows exist; Step 4.4: Update the number of observations for each sub-region. If all segmented sub-regions have been observed, the process ends; otherwise, proceed to step 4.

5. Step 4.5: Determine the time window set of the center points of the sub-regions after the constellation pair is segmented. The number of windows in the middle, if If the number of windows is greater than 0, proceed to step 4.2 to update the dynamic priority of each window; if the number of windows is equal to 0, then the process ends.

2. The method for planning on-orbit observation of key targets based on spatial region division according to claim 1, characterized in that, The method for calculating the location trajectory of the segmented sub-regions in step 3.4 is as follows: The observation area is always oriented relative to the Earth. The coordinates of the regional observation center in the J2000 coordinate system are... For regional targets, the coordinates within that region are in the body coordinate system. The coordinates of a point in the J2000 system for: In the coordinate system of the regional target, the x-direction is the velocity direction, and the z-direction is always parallel to the line connecting the observation center of the regional target and the Earth's center, and points towards the zenith. The elevation angle of the regional target observation center in the J2000 coordinate system. It is the azimuth angle. This is the rotation matrix from the regional target body coordinate system to the J2000 coordinate system; Based on the rotation matrix at each moment, the coordinates of the center point of the observation area, and the coordinates of the center point of the segmented sub-region in the body coordinate system, calculate the coordinates of the center point of each segmented sub-region in the J2000 system.

3. The method for planning on-orbit observation of key targets based on spatial region division according to claim 1, characterized in that, In step 4.1, the indirect observation judgment method is specifically as follows: determine whether the segmented sub-region b can be indirectly observed when the satellite directly observes the segmented sub-region a, based on the coordinates of the segmented sub-region a in the J2000 system. The coordinates of the segmented sub-region b in the J2000 system coordinates of the satellite in the J2000 system Calculate the vector from the satellite to the directly observed segmented sub-region a. , The vector from the satellite to the sub-region b after indirect observation segmentation. , According to the vector and Calculate the angle between two vectors. , If the included angle Must meet Therefore, when directly observing the segmented sub-region a, indirect observation of the segmented sub-region b can be achieved. This is the half-angle of the sensor's field of view.

4. The method for planning on-orbit observation of key targets based on spatial region division according to claim 1, characterized in that, In step 4.2, dynamic window priority The calculation function is as follows: For the benefit of observation capabilities, Weighting of observation capability gains, For window substitution benefit, The window substitution rate is weighted.

5. The method for planning on-orbit observation of key targets based on spatial region division according to claim 4, characterized in that, When considering expansion in three directions—tangential, radial, and normal to the orbital plane—the number of sub-regions after segmentation is: At that time, the benefits of observation capabilities The calculation formula is: In the formula, j represents the number of times the target is observed. When the target is observed multiple times, the observation weight decreases. This is the normalized value of the observed brightness when indirectly observing the i-th target within this time window; Window Substitution Benefit The calculation formula is: In the formula, In an unplanned window, the number of windows that can observe the target is greater. The more windows that can observe the target, the greater the substitutability of the windows and the smaller the benefit. This indicates that the target has not been observed; Without considering expansion along the normal direction of the orbital plane, the final number of sub-regions is: At that time, the benefits of observation capabilities The calculation formula is: Window Substitution Benefit The calculation formula is: 。 6. A system for implementing the on-orbit observation planning method for key target neighborhoods based on spatial region division as described in any one of claims 1 to 5, characterized in that, The system includes a computer module containing a method for planning on-orbit observations of key target neighborhoods based on spatial region division.

Images