Region-oriented data transmission and imaging combined scheduling multi-satellite task planning method

Abstract

The application provides a region-target-oriented data transmission and imaging combined scheduling multi-satellite task planning method, and belongs to the technical field of data transmission and imaging combined scheduling multi-satellite task planning. The method uses a structure coding method to realize strip selection coding of a region target covered by multiple imaging satellites; the structure coding is used to realize observation strip selection of the maximum coverage rate of the region target by the multiple imaging satellites; a multi-layer coding method is used to realize scheduling of an optimal scheme of satellite observation data downlink; through constraint checking and chromosome repair, the imaging and data transmission schemes meet the energy and storage space constraints on the satellite, and full-coverage observation task planning of a region target by an imaging satellite cluster is realized. The application solves the problem of imaging and data transmission combined scheduling of the satellite cluster, can be suitable for satellite cluster observation task planning facing a region target, and has application prospects in autonomous and intelligent operation and maintenance of the future satellite cluster.

Description

Technical Field

[0001] This invention relates to the field of multi-satellite mission planning technology for joint scheduling of data transmission and imaging, specifically to a method for joint scheduling of data transmission and imaging for regional targets. Background Technology

[0002] Satellite swarms are a type of satellite swarm operation mode that mimics biological swarms such as bee colonies, flocks of birds, and schools of fish in nature. By sharing information and coordinating actions, satellites complement each other's capabilities, and complex group behaviors are achieved through the simple behaviors of individuals within the swarm.

[0003] Swarming behavior is a common phenomenon in nature, with typical examples including flocks of migrating birds, schools of fish swimming in formation, ant colonies working collaboratively, and bacterial communities clustering together. These phenomena share the common characteristic of a certain number of autonomous individuals cooperating and self-organizing to exhibit orderly, coordinated movement and behavior at a collective level. With the rapid development of satellite technology, the mature application of multi-satellite launch technology, and the rise of commercial space companies, the costs of satellite development, launch, and deployment have further decreased. The integrated application of multiple satellites and multiple sensors has gradually become the mainstream trend in satellite applications, and satellite constellation self-synchronization operation and maintenance technology has become one of the research hotspots in the aerospace field.

[0004] Efficient inter-satellite mission allocation and the design of efficient multi-satellite, multi-Earth observation mission execution schemes are key technologies for the self-synchronization and efficient coordination of satellite constellations. With the increasing number of Earth observation satellites and the exponential growth of Earth observation missions, traditional methods of manually scheduling Earth observation missions for satellite constellations are no longer adequate for the high-efficiency coordination requirements of modern satellite constellations. Existing multi-satellite mission planning methods for imaging satellites do not comprehensively consider data transmission scheduling and onboard energy constraints, making it difficult to meet the needs of imaging satellite constellations for planning observation missions over large areas. Summary of the Invention

[0005] The technical problem solved by this invention is that existing multi-satellite mission planning methods for imaging satellites do not comprehensively consider data transmission scheduling and on-board energy constraints, making it difficult to meet the mission planning needs of imaging satellite clusters for observation of targets over a large area.

[0006] Therefore, this invention proposes a multi-satellite mission planning method for joint scheduling of data transmission and imaging for regional targets. This invention employs a structured coding method to generate a stripe selection scheme for multi-satellite imaging of regional targets, and a multi-layer coding method to generate a satellite cluster data transmission scheme, thus realizing joint scheduling of imaging and data transmission for imaging satellite clusters. This method has potential engineering application value in the future autonomous and intelligent operation and maintenance of satellite clusters. The technical solution of this invention is as follows:

[0007] A multi-satellite mission planning method for joint scheduling of data transmission and imaging for regional targets includes the following steps:

[0008] S1, Visibility window forecasting for satellite constellations and multiple targets;

[0009] S2. Dynamic decomposition of optional observation strips for regional targets by satellite constellation;

[0010] S3. Establish a constraint satisfaction model for heterogeneous multi-star collaborative mission planning;

[0011] S4. Set the optimization algorithm parameters;

[0012] S5, Heterogeneous Multi-Star Multi-Task Multi-Layer Coding Population Initialization

[0013] Initialize a heterogeneous multi-star, multi-task, multi-layer coded population; use structure coding to encode the observation mission scheme, with each target region corresponding to multiple observations, and each observation mission consisting of a triplet. The formula for expressing the observation mission is:

[0014] ot i = i ,w i ,θ i >.,

[0015] In the above formula, s i Indicates the satellite performing the current observation mission, w i θ represents the satellite's observation window for that region. i This indicates the lateral tilt angle of the satellite's observation of targets in the area within the visible window;

[0016] A two-layer coding scheme is used to encode the data transmission task. The first layer of coding is the task sequence, and the second layer of coding is the optional satellite resource sequence. When the number of targets to be observed is n and the number of targets is t... i The required number of revisits within a planning period is m. j hour, sequence of integers;

[0017] S6. Decode and calculate the task completion rate to obtain the feasible task plan with the maximum task completion rate;

[0018] S7. Perform genetic operations on heterogeneous multi-star multi-task multi-layer coding populations;

[0019] S8. Optimize termination judgment;

[0020] S9, Optimal solution output.

[0021] ​As another aspect of the present invention, step S1 includes the following: predicting the visible windows between the onboard sensors of each satellite in the forecast satellite cluster and the targets in each region, storing the visible windows in a structured manner, and using the stored visible windows for decoding the mission sequence and determining conflicts.

[0022] As another aspect of the present invention, step S2 includes the following: using a Gaussian projection-based regional target dynamic decomposition method, the regional target is decomposed into multiple imaging satellite selectable strips according to the nadir point trajectory when the satellite passes over the regional target and the width of the satellite imaging strip.

[0023] As another aspect of the present invention, step S3 includes the following:

[0024] A constraint satisfaction model for heterogeneous multi-star cooperative mission planning, namely the CSP model, is established. The formula for the CSP model is as follows:

[0025] E =<T,S,W,Con,Obj> .,

[0026] In the above formula, T represents the set of collaborative tasks, S represents the set of resource satellites participating in mission planning and execution of observation tasks, Con represents the constraints that the scheduling method needs to satisfy, and Obj represents the set of various optimization objective functions.

[0027] As another aspect of the present invention, the optimization objective function includes a data transmission task optimization objective function and an imaging satellite optimization objective function;

[0028] The objective function for optimizing the data transmission task is:

[0029]

[0030] In the above formula, fitness(i) represents the objective function value of the data transmission task optimization, CompleteTask represents the number of subtasks successfully arranged by the data transmission task, and TotalTask ​​represents the number of all data transmission subtasks to be arranged.

[0031] The objective function for imaging satellite optimization is:

[0032]

[0033] In the above formula, This indicates the probability of finding a target in the grid. This indicates the grid coverage statistics. This represents the grid outside the target area, where α represents the horizontal index of the grid and β represents the vertical index of the grid.

[0034] As another aspect of the present invention, in step S4, the optimization algorithm parameters include population size, maximum number of generations, generation gap, crossover probability, and mutation probability.

[0035] As another aspect of the present invention, step S6 includes the following: decoding the encoded feasible solution to form a scheduling plan, resolving conflicting task windows according to heuristic rules, and obtaining the feasible task solution with the maximum task completion rate.

[0036] As another aspect of the present invention, the task completion rate is calculated as follows: the coverage rate f1 of all regional targets and the ratio f2 of the completed data transmission tasks to all pending data transmission tasks are calculated, and the two are weighted averaged as f = α1f1 + α2f2 (α1 + α2 = 1). f is used as the task completion rate, where α1 is the first proportional parameter and α2 is the second proportional parameter.

[0037] As another aspect of the present invention, step S7 includes the following:

[0038] Based on the linear fitness method, a roulette wheel selection strategy is used to select individual chromosomes in a heterogeneous multi-star multi-task multi-layer coding population, eliminating individual chromosomes whose fitness ranks in the bottom 10% of the heterogeneous multi-star multi-task multi-layer coding population.

[0039] Crossover operations are performed between individual chromosomes in a heterogeneous multi-star multi-task multi-layer coding population. Partial gene sequences are exchanged between two individual chromosomes, and invalid solutions generated during the crossover process are repaired.

[0040] In heterogeneous multi-star multi-task multi-layer coding populations, gene mutations occur in the chromosomes of individuals, altering the order of tasks or the resources used to perform a task, and repairing redundant or missing chromosomes generated during the mutation process.

[0041] As another aspect of the present invention, step S8 includes the following: determining whether the optimization should be terminated based on the convergence status of the current population; if the chromosome similarity of individuals in the population is greater than 0.99, it indicates that the optimization algorithm has converged early, so the optimization is terminated and the next step is entered; if the current optimization process has reached the maximum number of generations, the optimization is terminated and the next step is entered; otherwise, proceed to step S6.

[0042] As another aspect of the present invention, step S9 includes the following: decoding the feasible task scheme with the highest task completion rate and outputting the optimal task arrangement scheme, namely the imaging scheme and data transmission scheme of all satellites.

[0043] The beneficial effects of this invention are:

[0044] (1) This invention establishes a CSP (constraint satisfaction) model that considers illumination, energy and on-board storage capacity, and the constraint description is simple and clear; it uses structure coding and multi-layer coding methods to realize the observation mission planning of imaging satellites for regional targets, and solves the problem of difficult joint scheduling of satellite cluster imaging and data transmission;

[0045] (2) The imaging scheme of the present invention adopts the structure coding method. The structure coding method realizes the scheduling scheme coding design that maximizes the cumulative coverage of multiple observations of regional targets by the imaging satellite cluster; the data transmission scheme of the present invention adopts the multi-layer coding method; the multi-layer coding realizes the joint scheduling of data transmission tasks and observation tasks of multiple satellites to multiple ground stations. Attached Figure Description

[0046] Figure 1 This is a flowchart of the multi-satellite mission planning method for joint scheduling of data transmission and imaging for regional targets according to the present invention.

[0047] Figure 2 This is the overall algorithm flowchart of the present invention;

[0048] Figure 3 This is a flowchart illustrating the heuristic rules of this invention;

[0049] Figure 4 This is a schematic diagram of the remote sensing scheduling plan structure encoding and the multi-layer encoding of data transmission scheduling calculation in this invention;

[0050] Figure 5 This is a graph showing the on-board data storage and energy reserve of each satellite in this invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0052] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0053] It should be understood that although the terms first, second, third, etc., may be used to describe... in the embodiments of the present invention, these... should not be limited to these terms. These terms are only used to distinguish... For example, first... may also be referred to as second... without departing from the scope of the embodiments of the present invention, and similarly, second... may also be referred to as first...

[0054] The present invention will now be described in detail based on specific embodiments. This embodiment describes a multi-satellite mission planning method for joint scheduling of data transmission and imaging for regional targets, such as... Figure 1 , Figure 2 As shown, it includes the following steps:

[0055] S1, Visibility Window Forecasting of Satellite Constellations and Multiple Targets

[0056] The visible windows between the onboard sensors of each satellite in the forecast satellite constellation and targets in each region are predicted and stored in a structured manner. The stored visible windows are then used for decoding mission sequences and determining conflicts.

[0057] In this embodiment, a co-simulation method using STK and MATLAB software is employed to calculate structured multi-satellite, multi-target visible window data; the STK Matlab Connector is used to establish a communication interface between the software programs. The specific process is as follows:

[0058] Create a new mission scenario and set the start time, end time, and epoch time according to the mission planning cycle; create all resource satellite objects based on the satellite orbit elements and sensor parameters, and create sensor sub-objects for all satellites; create all target objects based on the target location information; traverse all satellites and targets, calculate the visible window between each target and each satellite in each region, and store all window data in a structured manner.

[0059] STK, developed by Analytical Graphics, is a leading commercial analytics software in the aerospace field. STK supports the entire aerospace mission lifecycle, including design, testing, launch, operation, and mission application. STK stands for Satellite Tool Kit. It provides an analytics engine for data computation and can display various forms of 2D maps, showing satellites and other objects such as launch vehicles, missiles, aircraft, ground vehicles, and targets. STK's core capabilities are generating position and attitude data, acquiring time data, and performing remote sensor coverage analysis. STK Professional extends STK's basic analytical capabilities, including additional orbit prediction algorithms, attitude definition, coordinate types and systems, remote sensor types, advanced constraint definitions, and databases of satellites, cities, ground stations, and stars. For specific analytical tasks, STK provides additional analysis modules to address communication analysis, radar analysis, coverage analysis, orbital maneuvers, precise orbit determination, and real-time operation. Furthermore, STK has a 3D visualization module, providing a leading 3D display environment for STK and other add-on modules.

[0060] MATLAB is a commercial mathematical software produced by MathWorks, Inc. It is used in fields such as data analysis, wireless communication, deep learning, image processing and computer vision, signal processing, quantitative finance and risk management, robotics, and control systems. MATLAB is a combination of "matrix" and "laboratory," meaning matrix factory (matrix laboratory). The software primarily targets a high-tech computing environment for scientific computing, visualization, and interactive programming. It integrates powerful functions such as numerical analysis, matrix computation, scientific data visualization, and modeling and simulation of nonlinear dynamic systems into an easy-to-use graphical environment. It provides a comprehensive solution for scientific research, engineering design, and many scientific fields that require effective numerical computation, largely eliminating the need for the editing modes of traditional non-interactive programming languages ​​(such as C and Fortran).

[0061] S2. Dynamic decomposition of selectable observation strips for regional targets by satellite constellation.

[0062] A dynamic decomposition method for regional targets based on Gaussian projection is adopted. According to the nadir point trajectory when the satellite passes over the regional target and the width of the satellite imaging strip, the regional target is decomposed into multiple imaging satellite selectable strips.

[0063] Understandably, the regional target in step S2 refers to a closed convex polygonal region formed by connecting multiple latitude and longitude coordinate points on the Earth's surface in a clockwise direction; the observation strip refers to the strip of ground accumulated by the imaging satellite's sensors when it flies in a fixed attitude, that is, the area that the imaging satellite can observe.

[0064] Understandably, the method for dynamic decomposition of regional targets based on Gaussian projection is as follows:

[0065] A potential region for the target in a geodetic rectangular coordinate system is generated. The visible time window of the available resource set for the target in this region is calculated, and the nadir trajectory of the resources within the visible time window is calculated. The nadir trajectory is then projected onto a rectangular coordinate system. Due to the small observation area, the projection result is an irregular curve with very little curvature. To facilitate subsequent calculations, the least squares method is used to fit this curve to a straight line equation, in the form of:

[0066] x = a ij y+b ij ,

[0067] The potential region is segmented into strips; the maximum side angle of the imaging satellite constitutes the boundary of its maximum possible observation area, and the intersection of this possible observation area and the target potential region is the range that the observation strip needs to cover; within this region, strips are arranged from left to right, with an offset distance of Δλ each time, and the strip length is determined by the intersection of the right boundary of the strip and the observation area, until the entire area that needs to be covered is completely covered.

[0068] The specific execution time for segmentation is determined by the time window between the specific segment and the visible area within the mission execution period. Since the imaging satellite's orbit is relatively low and the ground observation area is small, it can be processed according to the planar assumption. If the instantaneous orbital altitude of the imaging satellite is H, and the previously given nadir trajectory is x = a... ij y+b ij Then the side tilt angle of the imaging satellite is:

[0069]

[0070] S3. Establish a heterogeneous multi-star collaborative mission planning constraint satisfaction model.

[0071] A constraint satisfaction model for heterogeneous multi-star cooperative mission planning, namely the CSP model, is established. The formula for the CSP model is as follows:

[0072] E =<T,S,W,Con,Obj> .,

[0073] In the above formula, T represents the set of collaborative tasks, S represents the set of resource satellites participating in mission planning and execution of observation tasks, Con represents the constraints that the scheduling method needs to satisfy, and Obj represents the set of various optimization objective functions.

[0074] In this embodiment, T includes the IDs of all targets to be observed, the regional target boundaries, and the observation time period; S includes the IDs of each satellite, orbital elements, orbital epoch time, sensor type, and sensor field of view parameter configuration; Con represents the constraints that the mission plan needs to satisfy, which will be elaborated in detail below; Obj represents the set of various optimization objective functions, which will be elaborated in detail below.

[0075] First, the CSP model symbol system is defined, as shown in Table 1.

[0076] Table 1. CSP Model Symbol System

[0077]

[0078]

[0079] For the constraint set Con, the constraint modeling is as follows. The formula for the optional resources for performing a certain task is:

[0080]

[0081] The formula for expressing that a target cannot occupy multiple resources simultaneously is:

[0082]

[0083] The available resources for subtask execution are limited and fixed. The formula for expressing that any subtask can only choose one resource from its candidate resources for execution is:

[0084]

[0085] The formula for stating that the interval between adjacent satellite sub-tasks must be greater than the time spent by the satellite performing attitude maneuvers is as follows:

[0086]

[0087] Subtasks are atomic; any task must be scheduled or abandoned on a selected resource. Once execution begins, it cannot be exited midway or switched to other resources. The formula for expressing this is:

[0088]

[0089] The formula for expressing that the imaging observation time for each subtask must be greater than the required observation duration is:

[0090]

[0091] The formula for expressing the timeliness requirement that the visible window of the resource targets that can be scheduled for observation must meet the following is:

[0092]

[0093] The optimization objective function Obj is modeled as follows. Data transmission task optimization objective: For data transmission task planning, the task planning objective is to maximize the data transmission task scheduling rate. The data transmission task optimization objective function is:

[0094]

[0095] In the above formula, CompleteTask represents the number of subtasks successfully scheduled for the data transmission mission, and TotalTask ​​represents the total number of data transmission subtasks to be scheduled. Imaging satellite optimization objective: For observation missions, the mission planning objective is to maximize the coverage of the area to be observed.

[0096] The objective function for imaging satellite optimization is as follows:

[0097]

[0098] In the above formula, This indicates the probability of finding a target in the grid. This indicates the grid coverage statistics. This represents the grid outside the target area, where α represents the horizontal index of the grid and β represents the vertical index of the grid.

[0099] S4. Set the optimization algorithm parameters, including population size, maximum number of generations, generation gap, crossover probability, and mutation probability.

[0100] In this embodiment, a genetic algorithm is selected as the optimization algorithm, and the parameters of the optimization algorithm are set as follows:

[0101] Population size: NINE = 500; Maximum number of generations: MAXGEN = 1000; Generation gap: GGAP = 0.9; Crossover probability: XOVR = 0.8; Mutation probability: MUTR = 5e-3.

[0102] S5, Heterogeneous Multi-Star Multi-Task Multi-Layer Coding Population Initialization

[0103] Initialize a heterogeneous multi-star, multi-task, multi-layer coded population; use structure coding to encode the observation mission scheme, with each target region corresponding to multiple observations, and each observation mission consisting of a triplet. The formula for expressing the observation mission is:

[0104] ot i = i ,w i ,θ i >

[0105] In the above formula, s i ​Indicates the satellite performing the current observation mission, w i θ represents the satellite's observation window for that region. i This indicates the lateral tilt angle of the satellite's observation of targets in the area within the visible window;

[0106] A two-layer coding scheme is used to encode the data transmission task. The first layer of coding is the task sequence, and the second layer of coding is the optional satellite resource sequence. When the number of targets to be observed is n and the number of targets is t... i The required number of revisits within a planning period is m. j hour, sequence of integers.

[0107] Understandably, in step S5, the regional target refers to a closed convex polygonal region formed by connecting multiple latitude and longitude coordinate points on the Earth's surface in a clockwise direction. The location, range, and number of regional targets are the inputs to the task planning algorithm.

[0108] In this embodiment, each individual chromosome corresponds one-to-one with a feasible solution to the optimization problem. For the observation task planning and coding, the structure coding is as follows: Figure 4 The diagram illustrates a structured encoding method that incorporates the observation count of each target into the encoded variable. The dimension of the observation activity sequence for each regional target is equal to the observation count. Each dimension includes the satellite ID, the visible window IDs of the satellite and the regional target, and the side-swing angle ID within that visible window. The upper and lower limits of the observation count PNumber for each target are determined using the following method:

[0109]

[0110] P * ≤PNumber≤ηP *

[0111] Among them, Width i The maximum width of the observable area of ​​the target perpendicular to the nadir point. ViewWidth is the satellite's field of view width. η is a parameter greater than 1. For data transmission task planning and coding, when the total number of satellites to be transmitted is n, and the target is n... i The number of revisits within a planning period is m j At that time, the length of the chromosome integer string is For example, data transmission scheduling genes:

[0112] [2 4 3 1 1 2 3 4||2 1 3 3 2 2 1 3]

[0113] This individual represents the observation sequence of four satellites, each transmitting data twice, observed by three ground stations. The first eight digits indicate the satellite data transmission mission order: Satellite 2 → Satellite 4 → Satellite 3 → Satellite 1 → Satellite 1 → Satellite 2 → Satellite 3 → Satellite 4. Digits 9 to 16 indicate the order of the ground stations performing the data transmission mission: Ground Station 2 → Ground Station 1 → Ground Station 3 → Ground Station 3 → Ground Station 2 → Ground Station 2 → Ground Station 1 → Ground Station 3.

[0114] S6. Decoding and task completion rate calculation to obtain feasible task solutions with the maximum task completion rate.

[0115] The feasible solutions generated by encoding are decoded to form a scheduling plan. Conflicting task windows are resolved according to heuristic rules to obtain the feasible task solution with the maximum task completion rate.

[0116] In this embodiment, the decoding method transforms the chromosome into a feasible scheduling scheme. For observation mission planning, a sequential constraint check decoding method is used to calculate the satellite's remaining energy and onboard storage capacity in real time during the mission planning period, delete missions that do not meet the mission constraints, and generate an observation mission scheme. For data transmission mission schemes, a heuristic rule is used for decoding. A feasible window is selected according to the sequence of missions on each satellite, and the duration already occupied by the satellite is recorded. The decoding process is as follows: Figure 2 As shown.

[0117] Understandably, the task completion rate is calculated as follows: calculate the coverage rate f1 of all target areas, and the ratio f2 of completed data transmission tasks to all pending data transmission tasks. Take the weighted average of the two: f = α1f1 + α2f2 (α1 + α2 = 1). Use f as the task completion rate, where α1 is the first proportional parameter and α2 is the second proportional parameter.

[0118] In this embodiment, the heuristic rule implementation process is as follows: Figure 3 As shown.

[0119] S7. Perform genetic operations on heterogeneous multi-star, multi-task, multi-layer coding populations.

[0120] Based on the linear fitness method, a roulette wheel selection strategy is used to select individual chromosomes in a heterogeneous multi-star multi-task multi-layer coding population, eliminating individual chromosomes whose fitness ranks in the bottom 10% of the heterogeneous multi-star multi-task multi-layer coding population.

[0121] Crossover operations are performed between individual chromosomes in a heterogeneous multi-star multi-task multi-layer coding population. Partial gene sequences are exchanged between two individual chromosomes, and invalid solutions generated during the crossover process are repaired.

[0122] In heterogeneous multi-star multi-task multi-layer coding populations, gene mutations occur in the chromosomes of individuals, altering the order of tasks or the resources used to perform a task. Redundant or missing chromosomes generated during the mutation process are repaired, specifically by deleting redundant chromosomes and supplementing missing chromosomes.

[0123] In this embodiment, the selection operator uses the roulette wheel selection method. The probability of selecting a particular offspring individual in the roulette wheel selection is directly proportional to the individual's fitness; chromosomes with higher fitness have a higher probability of being selected, while chromosomes with lower fitness have a lower probability of being selected. The roulette wheel selection method ensures that the population evolves towards higher fitness while preserving population diversity and preventing premature convergence of the genetic algorithm. The crossover operator uses random single-point crossover. After exchanging a segment between two chromosomes, there may be missing or redundant genes in the satellite sequences. Therefore, it is necessary to check the validity of the chromosomes and repair any invalid chromosomes. The mutation operator uses random locus mutation, i.e., randomly exchanging the order of the two satellite numbers. After mutation, there may be cases where antenna genes are invalid. Therefore, it is also necessary to check the validity of the chromosomes and repair any invalid chromosomes.

[0124] S8, Optimize Termination Judgment

[0125] Based on the current convergence status of the population, determine whether the optimization should terminate; if the chromosome similarity of individuals in the population is greater than 0.99, it indicates that the optimization algorithm has converged early, so terminate the optimization and proceed to the next step; if the current optimization process has reached the maximum number of generations, terminate the optimization and proceed to the next step; otherwise, proceed to step S6.

[0126] Optionally, this embodiment uses Euclidean distance to calculate the similarity of individual chromosomes in the population. All chromosomes are treated as N-dimensional real vectors, and the mean cosine of all vector directions is >0.99, indicating that the chromosome similarity of individuals in the current population is too high.

[0127]

[0128] In the above formula, For the i-th individual chromosome, Let N be the chromosome of the j-th individual, and N be the number of real vectors in the individual chromosome.

[0129] S9, Optimal Solution Output

[0130] The feasible mission scheme with the highest mission completion rate is decoded, and the optimal mission orchestration scheme is output, namely the imaging scheme and data transmission scheme of all satellites.

[0131] In this embodiment, the imaging scheme includes the start and end times of each payload power-on and power-off cycle and the side swing angle; the data transmission scheme includes the start and end times of each data transmission mission and the corresponding ground station.

[0132] In this embodiment, the final observation mission scheme shows that multiple regional targets with different topological configurations are fully covered. Onboard real-time data storage and remaining energy curves are shown below. Figure 5 As shown, the onboard storage space did not overflow during the mission planning period, and the remaining available onboard power-on time did not exceed the critical time.

Claims

1. A multi-satellite mission planning method for joint scheduling of data transmission and imaging for regional targets, characterized in that, Includes the following steps: S1, Visibility window forecasting for satellite constellations and multiple targets; S2. Dynamic decomposition of optional observation strips for regional targets by satellite constellation, including the following: Using a dynamic decomposition method for regional targets based on Gaussian projection, the regional targets are decomposed into multiple optional strips for imaging satellites according to the nadir point trajectory when the satellite passes over the regional targets and the width of the satellite imaging strip; S3. Establish a constraint satisfaction model for heterogeneous multi-star collaborative mission planning, including the following: A constraint satisfaction model for heterogeneous multi-star cooperative mission planning, namely the CSP model, is established. The formula for the CSP model is as follows: .， In the above formula, T Represents a set of collaborative tasks. S This refers to the set of resource satellites that participate in mission planning and the execution of observation missions. Con This represents the constraints that the scheduling method must satisfy. Obj This represents the set of various optimization objective functions; these include optimization objective functions for data transmission tasks and optimization objective functions for imaging satellites; the optimization objective function for data transmission tasks is: In the above formula, fitness ( i () represents the objective function value for optimizing the data transmission task. This represents the number of subtasks successfully scheduled for the data transmission task. This represents the number of all data transfer subtasks to be scheduled. The objective function for imaging satellite optimization is: .， In the above formula, This indicates the probability of finding a target in the grid. This indicates the grid coverage statistics. Represents the grid outside the target area. α Indicates the horizontal index of the grid. β Indicates the vertical index of the grid; S4. Set the optimization algorithm parameters; S5, Heterogeneous Multi-Star Multi-Task Multi-Layer Coding Population Initialization Initialize a heterogeneous multi-star, multi-task, multi-layer coded population; use structure coding to encode the observation mission scheme, with each target region corresponding to multiple observations, and each observation mission consisting of a triplet. The formula for expressing the observation mission is: .， In the above formula, Indicates the satellite performing the current observation mission. This indicates the satellite's observation window for that region. This indicates the lateral tilt angle of the satellite's observation of targets in the area within the visible window; A two-layer coding scheme is used to encode the data transmission task. The first layer of coding is the task sequence, and the second layer of coding is the optional satellite resource sequence. When the number of targets to be observed is... n And the target t i The required number of revisits within a planning period is: m j At that time, the individual chromosome is encoded as sequence of integers; S6. Decode and calculate the task completion rate to obtain the feasible task plan with the maximum task completion rate; S7. Perform genetic operations on heterogeneous multi-star multi-task multi-layer coding populations; S8. Optimize termination judgment; S9, Optimal solution output.

2. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 1, characterized in that, Step S1 includes the following: forecasting the visible windows between the onboard sensors of each satellite in the satellite cluster and the targets in each region, storing the visible windows in a structured manner, and using the stored visible windows for decoding the mission sequence and determining conflicts.

3. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 1, characterized in that, In step S4, the optimized algorithm parameters include population size, maximum number of generations, generation gap, crossover probability, and mutation probability.

4. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 1, characterized in that, Step S6 includes the following: decoding the encoded feasible solution to form a scheduling plan, resolving conflicting task windows according to heuristic rules, and obtaining the feasible task solution with the maximum task completion rate.

5. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 4, characterized in that, The task completion rate is calculated as follows: calculate the coverage rate of all target areas. And the ratio of completed data transmission tasks to all pending data transmission tasks. The weighted average of the two is taken. ,Will f As a measure of task completion rate, among which α 1 is the first proportional parameter. α 2 is the second proportional parameter.

6. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 1, characterized in that, Step S7 includes the following: Based on the linear fitness method, a roulette wheel selection strategy is used to select individual chromosomes in a heterogeneous multi-star multi-task multi-layer coding population, eliminating individual chromosomes whose fitness ranks in the bottom 10% of the heterogeneous multi-star multi-task multi-layer coding population. Crossover operations are performed between individual chromosomes in a heterogeneous multi-star multi-task multi-layer coding population. Partial gene sequences are exchanged between two individual chromosomes, and invalid solutions generated during the crossover process are repaired. In heterogeneous multi-star multi-task multi-layer coding populations, gene mutations occur in the chromosomes of individuals, altering the order of tasks or the resources used to perform a task, and repairing redundant or missing chromosomes generated during the mutation process.

7. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 1, characterized in that, Step S8 includes the following: The optimization process is determined based on the current convergence status of the population. If the chromosome similarity of individuals in the population is greater than 0.99, it indicates that the optimization algorithm has converged prematurely, so the optimization is terminated and the next step is initiated. If the current optimization process has reached the maximum number of generations, the optimization is terminated and the next step is initiated. Otherwise, proceed to step S6.

8. The multi-satellite mission planning method for joint scheduling of data transmission and imaging as described in claim 1, characterized in that, Step S9 includes the following: The feasible mission scheme with the highest mission completion rate is decoded, and the optimal mission orchestration scheme is output, namely the imaging scheme and data transmission scheme of all satellites.

Images