Multi-satellite observation planning method and device based on immune multi-objective evolutionary algorithm

Abstract

The present invention provides a multi-satellite observation planning method and device based on an immune multi-objective evolutionary algorithm, including obtaining a set of perception coverage opportunities of each satellite sensor corresponding to a grid in a disaster area, setting each chromosome to represent a combination of multi-satellite perception coverage opportunities, each gene position on the chromosome represents a perception coverage opportunity selected by a specific satellite sensor, and the number of grids in the disaster area covered by the perception coverage opportunity is used as the value of the corresponding gene position, and the fitness value of each chromosome in the entire population to multiple targets is obtained according to the value of each gene position, the entire population is sorted, the chromosomes are immune operated, and after updating, they are sorted again based on the fitness value, and the chromosomes with a preset fitness value are output from the sorted population as the optimal alternative. The present invention can plan the best solution for multiple targets such as coverage degree, cost, time, etc. of the disaster area from the numerous perception coverage opportunities of multiple satellite sensors in an emergency environment.

Description

Technical Field

[0001] The present invention relates to the field of geographic information science and technology, and in particular to a multi-satellite observation planning method and device based on an immune multi-objective evolutionary algorithm. Background Art

[0002] The occurrence of natural disasters will cause a large number of casualties and property losses. Traditional disaster investigation and disaster information acquisition methods have the problems of slow response and low efficiency. Remote sensing satellites have the advantages of wide coverage and rapid revisit. Satellite observation payloads can easily collect information on the earth's surface, which is of great significance for disaster emergency response. However, problems such as low efficiency of single satellite imaging, single sensor type and coverage that does not meet the needs are gradually exposed. How to coordinate the use of perception resources of multiple remote sensing satellites, coordinate multiple satellites to image the earth, and reasonably use different sensors carried by multiple satellites to quickly obtain disaster information during disaster emergency response. Efficient and coordinated ground observation to meet the needs of disaster emergency are the technical difficulties that the present invention intends to solve. The multi-satellite observation planning problem can be converted into a coverage optimization problem. Each satellite has many coverage opportunities. These coverage opportunities are optimized and a suitable set of satellite perception coverage opportunities is selected.

[0003] Satellite sensing coverage opportunity set optimization methods mainly include exhaustive method, linear programming method, intelligent optimization method and machine learning method. The exhaustive algorithm directly searches the coverage opportunity solution space, compares the advantages and disadvantages of all coverage opportunity combination solutions, and finally obtains the best solution. The linear programming method adjusts the service quality constraint parameters through linear weighting to solve the optimization solution. Intelligent optimization algorithm is a general term for a series of algorithms such as genetic algorithm, ant colony algorithm, particle swarm algorithm, etc. that solve the optimal combination by simulating the development laws of nature or the physical world. Machine learning methods are mainly based on deep reinforcement learning, which predicts service quality evaluation by training a recursive neural network model. Intelligent optimization algorithms are widely used because of their simple implementation, small amount of calculation, and the ability to find feasible solutions that meet service quality constraints within a limited time. However, the above methods all solve clear goals. However, in actual emergency environments, satellite emergency imaging missions of the Earth require consideration from many aspects. If multiple mission objectives are to be met at the same time, there may be conflicts between the multiple mission objectives. Specifically, the larger the coverage area, the longer the satellite imaging time will be, and the lower the satellite imaging cost will result in fewer satellites being used and a smaller coverage area. Therefore, the above method cannot achieve the best results for multiple objectives in an emergency environment. Summary of the invention

[0004] In view of the deficiencies of the prior art, the purpose of the present invention is to provide a multi-satellite observation planning method and device based on an immune multi-objective evolutionary algorithm. The present invention can enable satellite sensors to cover multiple tasks in the disaster-stricken area under emergency conditions, prevent conflicts between multiple task objectives, and achieve the best planning effect.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] In a first aspect, the present invention provides a multi-satellite observation planning method based on an immune multi-objective evolutionary algorithm, comprising:

[0007] Establish a disaster-affected area grid and obtain the sensing coverage opportunity set of each satellite sensor corresponding to the disaster-affected area;

[0008] Perform chromosome encoding of the multi-objective evolutionary algorithm, set each chromosome to represent a combination of multi-satellite sensing coverage opportunities, and each gene position on the chromosome represents a sensing coverage opportunity selected by a specific satellite sensor;

[0009] Obtain the number of grid cells in the disaster-stricken area covered by the sensing coverage opportunity of each specific satellite sensor as the value of the corresponding gene bit;

[0010] Obtain the fitness value of each chromosome in the entire population for multiple targets through the value of each gene position, and perform a preliminary sorting of all chromosomes based on the fitness value;

[0011] After performing immune operations on the chromosomes in the population, new fitness values ​​are obtained again, and all chromosomes are re-sorted based on the new fitness values;

[0012] Output the chromosome whose fitness value in the population after secondary sorting is the preset situation as the optimal alternative.

[0013] According to the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm provided by the present invention, the continuous roll angle is abstracted into a discrete set of perception coverage opportunities of satellite sensors, the width of the perception coverage opportunity is obtained, and the cell whose width of the perception coverage opportunity is smaller than the corresponding cell of the maximum width of the satellite is taken as the candidate left cell u. After all the left cells u are determined, all grids are traversed, and a perpendicular line Lt perpendicular to the sub-satellite point trajectory line is made through the cell t, and a perpendicular line L″u perpendicular to the sub-satellite point trajectory line is made through the left cell, and the cell t whose perpendicular line L″u is below the perpendicular line Lt is retained as the required upper cell t′. After all the upper cells t′ are determined, the Lt line is taken as the starting point, and the power-on and power-off time with a certain time step length is selected, and the length of the perception coverage opportunity of a single satellite transit is calculated by the satellite orbit speed, and the width and length of the perception coverage opportunity are multiplied to obtain a perception coverage opportunity, and the discrete perception coverage opportunity set of a single satellite sensor is obtained cyclically, and the discrete perception coverage opportunity set of all satellites is used as the solution space of multi-objective planning.

[0014] According to the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm provided by the present invention, the width of the sensing coverage opportunity is represented by D, and the calculation formula is:

[0015] D=D1+D2

[0016]

[0017]

[0018] Among them, P is the deflection angle of the satellite, H is the orbital height, and fov is the field of view angle;

[0019] The calculation formula for the field of view angle fov is:

[0020]

[0021] Where d is the distance between the subsatellite point trajectory line and the straight line Lu passing through the left cell u and parallel to the subsatellite point trajectory line.

[0022] According to the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm provided by the present invention, the multi-objectives are set as the number of coverage grids, the price index, and the imaging time, all chromosomes in the entire population are calculated, and the non-dominated ranking value of each chromosome to the population is used as the fitness value of the chromosome to the multi-objectives.

[0023] According to the multi-satellite observation planning method based on immune multi-objective evolutionary algorithm, the number of coverage grids f is obtained. cov :

[0024]

[0025]

[0026] Among them, a i Indicates the normalized value of the callable capacity of each satellite; grid i is a decision variable, which indicates the number of ground grids covered by the sensing coverage opportunities of each satellite sensor in the chromosome; i is each satellite sensor in the chromosome; S is the set of candidate satellite sensors involved in the calculation; G is the number of grids in the disaster area;

[0027] Get imaging time f tim :

[0028]

[0029] Among them, t i , i∈[0,q) represents the latest completion time of the sensing coverage opportunity of each satellite sensor in the chromosome; T s is the time when the disaster monitoring mission starts;

[0030] Get the price index f cos :

[0031]

[0032] Among them, b i represents the cost of each satellite sensor to the cell area, It represents the number of grids covered by the coverage opportunity of each satellite sensor, m is the number of coverage opportunities of a specific satellite sensor, J is the maximum value of the coverage opportunity of each satellite sensor, and j is the number of perception coverage opportunities corresponding to each satellite.

[0033] According to the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm provided by the present invention, the immune mutation operation updating method is: immune cloning and crossover mutation are performed on each chromosome of the population in turn, and then the non-dominated order and crowding degree of all chromosomes are calculated, and the chromosome with the highest non-dominated order and the lowest crowding degree is retained as the optimal chromosome to the updated population, and the updated population is combined with a random new population, and an elite retention strategy is used to obtain a new generation of sub-populations, and each gene of the chromosome is subjected to mutation judgment, and the genes that need to be mutated are randomly selected within a certain range of the original genes. This range gradually decreases as the number of iterations increases. The formula for random selection of the range is as follows:

[0034] scope=g i +[-1,1]×(m / gen)

[0035] Among them, g i is the original chromosome gene, m is the number of coverage opportunities of a specific satellite sensor, and gen is the current iteration number.

[0036] In a second aspect, the present invention provides a multi-satellite observation planning device based on an immune multi-objective evolutionary algorithm, comprising:

[0037] The first acquisition module is used to establish a grid of disaster-stricken areas and obtain a set of sensing coverage opportunities of each satellite sensor corresponding to the disaster-stricken area;

[0038] The chromosome encoding module is used to perform chromosome encoding of the multi-objective evolutionary algorithm. Each chromosome is set to represent a combination of multi-satellite sensing coverage opportunities, and each gene position on the chromosome represents a sensing coverage opportunity selected by a specific satellite sensor.

[0039] The second acquisition module is used to obtain the number of grids in the disaster-stricken area covered by the sensing coverage opportunity of each specific satellite sensor as the value of the corresponding gene bit;

[0040] The third acquisition module is used to obtain the fitness value of each chromosome in the entire population to multiple targets through the value of each gene position, and perform a preliminary sorting of all chromosomes based on the fitness value;

[0041] An updating module is used to perform immune operations on the chromosomes in the population, reacquire new fitness values, and perform secondary sorting on all chromosomes based on the new fitness values;

[0042] The output module is used to output the chromosome with a preset fitness value in the population after the secondary sorting as the optimal alternative.

[0043] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm is implemented.

[0044] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm.

[0045] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm.

[0046] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0047] The present invention integrates multiple objectives, calculates non-dominated ranking values, obtains the dominance of each satellite sensor coverage opportunity set, and in order to avoid convergence to a local optimum, adds immune operations in a continuous iterative process, so that the method can jump out of the local optimum while efficiently generating solutions, obtains the best multi-satellite coverage opportunity set for multiple objectives, obtains multi-satellite coverage opportunity set solutions for different mission objectives, can be applied to multi-satellite planning problems in disaster emergency environments, and greatly improves the capabilities of satellite resources in disaster emergency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0049] Figure 1 is the application environment of the present invention;

[0050] Figure 2It is a flow chart of the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm of the present invention;

[0051] Figure 3 A schematic diagram of a population of a combination of multi-satellite sensing coverage opportunities of the present invention;

[0052] Figure 4 It is a schematic diagram of immune operation of a population of a multi-satellite sensing coverage opportunity combination of the present invention;

[0053] Figure 5 A schematic diagram of chromosome immune variation operation of a population of a multi-satellite sensing coverage opportunity combination of the present invention;

[0054] Figure 6 A schematic diagram of chromosome immune crossover operation of a population of a multi-satellite sensing coverage opportunity combination of the present invention;

[0055] Figure 7 This is a schematic diagram of the relationship between satellite sensing coverage opportunities and ground grids;

[0056] Figure 8 Schematic diagram of a multi-satellite observation planning device based on the immune multi-objective evolutionary algorithm. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0058] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0059] It should be understood that the present invention can be implemented in different forms and should not be interpreted as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the disclosure thorough and complete and to fully convey the scope of the present invention to those skilled in the art.

[0060] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0061] 1. Multi-satellite perception planning: also known as multi-satellite imaging planning, refers to the coordinated imaging of multiple satellites within a certain time to plan the optimal combination of multi-satellite perception coverage opportunities.

[0062] 2. Perception coverage opportunity set of satellite sensors: A satellite sensor's earth imaging will present a rectangular coverage opportunity. Depending on the different deflection angles, field of view angles, and power-on and power-off times of the satellite sensor, many perception coverage opportunities will be presented. These perception coverage opportunities are counted together to form the perception coverage opportunity set of the satellite sensor.

[0063] 3. Multi-satellite perception coverage opportunity combination: refers to a chromosome in the population, in which each gene corresponds to a perception coverage opportunity in the perception coverage opportunity set of each satellite sensor. In this way, a multi-satellite perception coverage opportunity combination contains a perception coverage opportunity of all satellite sensors participating in the calculation.

[0064] 4. Immune operation: including immune cloning, immune mutation, and immune selection. Immune cloning is to replicate antibodies (chromosomes), immune mutation is to perform crossover mutation operations on the replicated antibodies (chromosomes), and immune selection is to select the clones after mutation and select the best individuals to be retained for the next generation.

[0065] First embodiment

[0066] Traditional disaster investigation and disaster information acquisition methods have the problems of slow response and low efficiency. Satellite observation payloads can easily collect information on the earth's surface and obtain disaster intelligence. Optical images can reflect rich ground information, and radar, near-infrared and other sensors can penetrate clouds and fog to provide support for disaster information acquisition. However, single satellites have problems such as low imaging efficiency, single sensor type and coverage that does not meet needs. In related technologies, the allocation of satellite sensors is to solve clear goals, while in actual emergency environments, satellite emergency imaging tasks for the ground need to be considered from many aspects and need to meet multiple mission goals at the same time. Therefore, there may be conflicts between goals, which is specifically manifested in that the larger the coverage area, the longer the satellite imaging time. Reducing the cost of satellite imaging will result in fewer satellites being used and a smaller coverage area. Therefore, the above method cannot achieve the best results for multiple goals in an emergency environment.

[0067] The multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm provided in this embodiment generates a chromosome representing a combination of multi-satellite perception coverage opportunities based on a genetic algorithm, sorts the chromosomes according to the fitness values ​​of multiple objectives, and then optimizes them through an immune algorithm to select the best alternative solution, which can be applied to Figure 1In the application environment shown. The terminal 100 communicates with the server 200 through the network. The server 200 can collect the grid of the disaster area and send a multi-satellite observation planning instruction to the terminal 100. The terminal 100 can obtain the multi-satellite observation planning instruction sent by the server 200 and obtain the grid of the disaster area from the server 200, so that the multi-satellite observation planning can be performed based on the grid of the disaster area. The terminal 100 can be, but is not limited to, various personal computers, laptops, smart phones and tablet computers, and the server 200 can be implemented by an independent server or a server cluster composed of multiple servers.

[0068] In one embodiment, see Figure 2 , is a flowchart of a multi-satellite observation planning method based on an immune multi-objective evolutionary algorithm in a first embodiment, and provides a multi-satellite observation planning method based on an immune multi-objective evolutionary algorithm, and the method is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0069] Step S1: Establish a disaster-affected area grid and obtain the sensing coverage opportunity set of each satellite sensor corresponding to the disaster-affected area.

[0070] In this embodiment, a discretized grid of the disaster-affected area is established, and a set of grids of the disaster-affected area G={1,2,...,k,...,n} is defined, where k is the grid number of the disaster-affected area, and n is the total number of grids of the disaster-affected area;

[0071] Assume that the satellite sensor set corresponding to the grid in the disaster-stricken area is I = {1, 2, ..., i, ..., u}, i is the i-th satellite sensor, u is the total number of satellite sensors, and the sensing coverage opportunity set of the i-th satellite sensor is m is the total number of sensing coverage opportunities of the i-th satellite sensor.

[0072] The width of the perceived coverage opportunity is represented by D, and the calculation formula is:

[0073] D=D1+D2

[0074]

[0075]

[0076] Among them, P is the deflection angle of the satellite, H is the orbital altitude, and fov is the field of view angle.

[0077] The calculation formula for the field of view angle fov is:

[0078]

[0079] Among them, d is the distance between the sub-satellite point trajectory line and the straight line Lu passing through the left cell u and parallel to the sub-satellite point trajectory line.

[0080] The continuous roll angle is abstracted into a discrete set of perception coverage opportunities of satellite sensors. The cells whose width of perception coverage opportunities is less than the corresponding maximum width of the satellite are taken as candidate left cells u. After all left cells u are determined, all grids are traversed, and Lt is made perpendicular to the sub-satellite point trajectory line through cell t, and L″u is made perpendicular to the sub-satellite point trajectory line through the left cell. The upper cell t′ is filtered by judging whether the vertical line L″u is below the vertical line Lt, and the cell t whose vertical line L″u is below the vertical line Lt is retained as the required upper cell t′. After all upper cells t′ are determined, the Lt line is taken as the starting point, and the power-on and power-off time with a certain time step length is selected. The length of the perception coverage opportunity of a single satellite pass is calculated by the satellite orbit speed, and the discrete perception coverage opportunity set of a single satellite sensor is obtained cyclically. The discrete perception coverage opportunity set of all satellites is used as the solution space of multi-objective planning.

[0081] Step S2: Perform chromosome encoding of the multi-objective evolutionary algorithm, set each chromosome to represent a combination of multi-satellite sensing coverage opportunities, and each gene position on the chromosome represents a sensing coverage opportunity selected by a specific satellite sensor.

[0082] In this embodiment, an immune multi-objective evolutionary algorithm framework is established to perform chromosome encoding of the multi-objective evolutionary algorithm, specifically including: defining a chromosome group Among them, t is the number of chromosome group iterations; q is the chromosome number, corresponding to the specific satellite sensor; u is the chromosome population size, that is, the total number of chromosomes; It is the result of the tth chromosome group iteration of the qth chromosome.

[0083] In some specific implementations of this embodiment, the mutation probability p_mutation=0.7 and the crossover probability p_crossover=0.7 are set, and all chromosomes in the chromosome group are initialized randomly. Each gene of the chromosome is a perception coverage opportunity of a satellite sensor, corresponding to a coverage opportunity of the satellite sensor, and the cloning parameter of the immune clone is set to 1 / 10 of the population number.

[0084] In one embodiment of the present invention, Figure 3As shown in the figure, each row represents a combination of multi-satellite sensing coverage opportunities, where the number of each gene corresponds to the sensing coverage opportunities of different satellite sensors. For example, the 163rd sensing coverage opportunity of the Resource 1F satellite IRS sensor, the 672nd sensing coverage opportunity of the Environment 2B satellite CCD sensor, the 1152nd sensing coverage opportunity of the High-Resolution 3C satellite sensor, the 1621st sensing coverage opportunity of the Sentinel 2B satellite sensor... The dotted box in the figure represents the different sensing coverage opportunities of a specific satellite sensor, and the range of its numbers is determined by the number of sensing coverage opportunities of each satellite sensor. A chromosome represents a combination of multi-satellite sensing coverage opportunities.

[0085] by Figure 3 Chromosome 1 in the example shows a complete description of a specific combination of multi-satellite sensing coverage opportunities: {163rd coverage opportunity of ZY1F satellite IRS sensor, 672nd coverage opportunity of HJ2B satellite CCD sensor, 1152nd coverage opportunity of GF3C satellite SAR sensor, 1621st coverage opportunity of SENTINEL2B satellite MSI sensor, …, 2760th coverage opportunity of GF7 satellite DLC sensor}. The order of specific satellite sensors in the list remains unchanged, and is encoded using an integer sequence as {163, 672, 1152, 1621, …, 2760}, which is the chromosome encoding of the combination of multi-satellite sensing coverage opportunities.

[0086] Step S3: Obtain the number of grid cells in the disaster-stricken area covered by the sensing coverage opportunity of each specific satellite sensor as the value of the corresponding gene position.

[0087] In this embodiment, the coverage of each satellite sensor on the grid of the disaster-stricken area is calculated according to its coverage opportunity, and the chromosome phenotype is converted into genotype. Figure 3 The chromosome of the multi-satellite sensor coverage opportunity combination of chromosome 1 is converted into {0, 1, 24, 18, …, 0}, that is, the number of coverage of the grid of the disaster-stricken area by the satellite sensor coverage opportunity, which is convenient for the subsequent objective function calculation.

[0088] Step S4: Obtain the fitness value of each chromosome in the entire population for multiple targets through the value of each gene position, and perform initial sorting on all chromosomes based on the fitness value.

[0089] The multi-objectives are set as the number of coverage grids, price index, and imaging time, and all chromosomes in the entire population are calculated to obtain the non-dominated sorting value of the population.

[0090] Individuals are sorted according to their fitness values, so that all individuals are divided into multiple dominance levels. Individuals with high dominance levels are selected first, and crossover mutation is performed on the selected individuals to generate a new generation of population, which is then iterated.

[0091] The goal of the coverage grid number is to evaluate the coverage rate of the regional target, using all cells x g The cumulative sum of the callable capabilities is expressed as, and the number of covered grids f is obtained cov The calculation formula is:

[0092]

[0093]

[0094] Among them, a i Indicates the normalized value of the callable capacity of each satellite; grid i is a decision variable, which indicates the coverage quantity of ground grid by each satellite sensor in chromosome. i If it is not covered by a strip, its value is 0; if it is covered by more satellite sensor strips, its value is smaller; when the accumulated value of all grids is larger, it means that the distribution between strips is more uniform and the overlapping grids are fewer, so f cov The available capacity values ​​of all grids in the study area are accumulated to determine the combination of satellite sensors that have the most grid coverage opportunities; i is the satellite sensor in the chromosome; S is the set of candidate satellite sensors involved in the calculation; and G is the number of grids in the affected area.

[0095] The imaging time objective is to evaluate the completion time of satellite imaging in the satellite imaging combination, measured by the latest completion time of the satellite sensing coverage opportunity in each chromosome, and obtain the imaging time f tim The calculation formula is:

[0096]

[0097] Among them, t i ,i∈[0,S) represents the latest completion time of each satellite sensing coverage opportunity in the chromosome; T s is the time when the disaster monitoring mission starts. The goal is to find the minimum completion time of satellite imaging, that is, the shortest time used by the satellite in the combination of satellite scheduling.

[0098] The price index objective is to evaluate the cost of satellite sensing coverage in the satellite imaging combination and obtain the price index f cos The calculation formula is:

[0099]

[0100] Among them, b i represents the cost of each satellite sensor to the cell area, Represents the number of grids covered by the coverage opportunity of each satellite sensor, and finds the minimum cost of the satellite combination, that is, the minimum satellite sensor combination cost index. m is the number of coverage opportunities of a specific satellite sensor, J is the maximum value of the coverage opportunities of each satellite sensor, and j is the number of perception coverage opportunities corresponding to each satellite.

[0101] Step S5: After performing immune operations on the chromosomes in the population, new fitness values ​​are obtained again, and all chromosomes are re-sorted based on the new fitness values.

[0102] In this embodiment, the method for updating the immune mutation operation is: perform immune cloning and crossover mutation on each chromosome of the population in turn, then calculate the non-dominated order and crowding of all chromosomes after immune cloning and crossover mutation, retain the chromosome with the highest non-dominated order and the lowest crowding as the optimal chromosome to the updated population, combine the updated population with the random new population, and use the elite retention strategy to obtain a new generation of sub-population.

[0103] Specifically, Figure 3 As shown, in the parent population, the chromosomes are immune cloned, that is, a single chromosome is taken out from the parent population and copied a number, where a value is the cloning parameter. For example: in the present invention, a=10, and 9 of the 10 chromosomes after replication are mutated, such as cloning chromosome m-2 to chromosome m-10 in the figure, and the initial chromosomes are retained, such as chromosome m-1 in the figure, to ensure the paternal characteristics of the chromosomes. Subsequently, the non-dominated sorting value and crowding degree of the 10 chromosome populations after mutation are calculated, and the optimal chromosome is selected and passed into the child population. This immune operation ends after operating on all chromosomes in the parent population.

[0104] Specifically, the mutation operation is shown as follows: Figure 4 Taking chromosome 1-1 as an example, each gene of the chromosome is judged for mutation. The genes that need to be mutated are randomly selected within a certain range of the original genes. The range gradually decreases as the number of iterations increases, so as to prevent falling into the local optimal result in advance. The formula for random range selection is as follows:

[0105] scope=g i +[-1,1]×(m / gen)

[0106] Among them, g i is the original chromosome gene, m is the number of coverage opportunities of a specific satellite sensor, and gen is the current iteration. After this operation, the gene becomes another satellite sensor coverage opportunity, thereby increasing population diversity. For example: Figure 4 , g1 mutates, changing from around 223 to 873 according to the formula, g2 gene does not mutate, g iThe gene mutates from 89 to 67 according to the random selection formula, and so on, traversing all the genes of the chromosome and transforming into chromosome m-1″.

[0107] Crossover operation such as Figure 5 , randomly select two crossover points of the chromosome, exchange the chromosome fragments between the crossover points, and realize the chromosome crossover operation. For example: Figure 5 The phenotype of chromosome m-1′ is {223,718,1357,24,…,10}, of which the crossover fragment is {1357,24,…}, and the phenotype of chromosome m-2′ is {100,523,1273,89,…,2945}, of which the crossover fragment is {1273,89,…}. The two fragments are exchanged to form new chromosome m-1″{223,718,1273,89,…,10} and chromosome m-2″{100,523,1357,24,…,2945}.

[0108] Specifically, the non-dominated sorting of chromosomes in the population is to re-sort the population by calculating the dominance relationship between chromosomes in the population. For example, two chromosomes A and B are selected in the population, and the fitness values ​​of the three objective functions of chromosomes A and B are calculated respectively. There are four preset cases for the fitness value: ① If f cov (A) is better than f cov (B) f tim (A) is better than f tim (B) f cos (A) is better than f cos (B), indicating that A dominates B; ② If f cov (A) Not better than f cov (B) f tim (A) is better than f tim (B) f cos (A) is better than f cos (B), indicating that A does not dominate B; ③ If f cov (A) Not better than f cov (B) f tim (A) Not better than f tim (B) f cos (A) is better than f cos (B), indicating that A does not dominate B; ④ If f cov (A) Not better than f cov (B) f tim (A) Not better than f tim (B) f cos (A) Not better than f cos (B), indicating that A does not dominate B. In the population sorting, the priority is ①②③④. In this way, the chromosomes of the entire population are sorted, completing the non-dominated sorting of the population.

[0109] The updated population generated in this way is combined with a random new population, and an elite retention strategy is used to obtain a new generation of sub-populations. After continuous iterations, the combination corresponding to each chromosome in the population becomes better and better.

[0110] In summary, the multi-objective optimization method for multi-satellite planning proposed in the present invention can obtain the optimal solution for the combination of multi-satellite coverage opportunities in disaster emergency response, and improve the planning capability of multi-satellite perception tasks that take into account coverage, cost and time in emergency environments.

[0111] Step S6: Output the chromosomes with preset fitness values ​​in the population after secondary sorting as the optimal alternative.

[0112] Specifically, if the chromosome meets the optimization termination condition, if it does not meet the optimization termination condition, the chromosome is subjected to an immune operation to obtain a new chromosome until the optimization termination condition is met.

[0113] In this embodiment, it is checked whether the number of chromosome group iterations is less than the maximum number of iterations T. If t≤T, the chromosome group is updated and the chromosome group iteration continues; otherwise, the chromosome group iteration ends.

[0114] This embodiment is based on a multi-objective optimization algorithm. On the basis of considering multi-objective optimization at the same time, the population is continuously updated through immune operations, and finally the chromosome that meets the multi-objective optimality in the population is output as the optimal alternative solution, and a multi-satellite perception coverage opportunity combination solution for different mission objectives is obtained. It can be applied to multi-satellite planning problems in disaster emergency environments, greatly improving the ability of satellite resources in disaster emergency response and preventing conflicts between objectives.

[0115] In summary, the multi-satellite planning method provided by the present invention comprehensively considers the coverage target, cost target, and time target, calculates the non-dominated ranking value, obtains the dominance of each satellite sensor perception coverage opportunity set, and in order to avoid convergence to the local optimum, an immune operation is added in the continuous iteration process, so that the method can jump out of the local optimum while efficiently generating a solution, and obtain the multi-satellite perception coverage opportunity set that is the best for all three targets.

[0116] The multi-satellite planning method provided by the present invention is based on a multi-objective optimization algorithm. On the basis of considering multi-objective optimization at the same time, the population is continuously updated through operations such as immune cloning, immune selection, and immune suppression, and finally a scheme of chromosomes in the population that meet the three optimal objectives is output to obtain a multi-satellite sensing coverage opportunity set scheme for different mission objectives. It can be applied to multi-satellite planning problems in disaster emergency environments, greatly improving the ability of satellite resources in disaster emergency response.

[0117] Second embodiment

[0118] The second embodiment provides a multi-satellite observation planning device based on an immune multi-objective evolutionary algorithm, such as Figure 4 As shown, including:

[0119] The first acquisition module is used to establish a grid of disaster-stricken areas and obtain a set of sensing coverage opportunities of each satellite sensor corresponding to the disaster-stricken area;

[0120] The chromosome encoding module is used to perform chromosome encoding of the multi-objective evolutionary algorithm. Each chromosome is set to represent a combination of multi-satellite sensing coverage opportunities, and each gene position on the chromosome represents a sensing coverage opportunity selected by a specific satellite sensor.

[0121] The second acquisition module is used to obtain the number of grids in the disaster-stricken area covered by the sensing coverage opportunity of each specific satellite sensor as the value of the corresponding gene bit;

[0122] The third acquisition module is used to obtain the fitness value of each chromosome in the entire population to multiple targets through the value of each gene position, and perform a preliminary sorting of all chromosomes based on the fitness value;

[0123] The updating module is used to perform immune operations on the chromosomes in the population, re-obtain the fitness value of each chromosome, and perform secondary sorting on all chromosomes based on the fitness value;

[0124] The output module is used to output the chromosome with a preset fitness value in the population after the secondary sorting as the optimal alternative.

[0125] Third embodiment

[0126] The third embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm when executing the program.

[0127] Fourth embodiment

[0128] A fourth embodiment provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm is implemented.

[0129] Fifth embodiment

[0130] The fifth embodiment provides a computer program product, including a computer program, which implements the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm when executed by a processor.

[0131] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0132] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A multi-satellite observation planning method based on an immune multi-objective evolutionary algorithm, characterized in that: include: Establish a disaster-affected area grid and obtain the sensing coverage opportunity set of each satellite sensor corresponding to the disaster-affected area; Perform chromosome encoding of the multi-objective evolutionary algorithm, set each chromosome to represent a combination of multi-satellite sensing coverage opportunities, and each gene position on the chromosome represents a sensing coverage opportunity selected by a specific satellite sensor; Obtain the number of grid cells in the disaster-stricken area covered by the sensing coverage opportunity of each specific satellite sensor as the value of the corresponding gene bit; Obtain the fitness value of each chromosome in the entire population for multiple targets through the value of each gene position, and perform a preliminary sorting of all chromosomes based on the fitness value; The multi-objectives are set as the number of covered grids, price index, and imaging time, and all chromosomes in the entire population are calculated. The non-dominated ranking value of each chromosome to the population is taken as the fitness value of the chromosome to the multi-objectives. Get the number of covered grids f cov : Among them, a i Indicates the normalized value of the callable capacity of each satellite; grid i is a decision variable, which indicates the number of ground grids covered by the sensing coverage opportunities of each satellite sensor in the chromosome; i is each satellite sensor in the chromosome; S is the set of candidate satellite sensors involved in the calculation; G is the number of grids in the disaster area; Get imaging time f tim : Among them, t i , i∈[0,S) represents the latest completion time of the sensing coverage opportunity of each satellite sensor in the chromosome; T s is the time when the disaster monitoring mission starts; Get the price index f cos : Among them, b i represents the cost of each satellite sensor to the cell area, represents the number of grids covered by the coverage opportunity of each satellite sensor, m is the number of coverage opportunities of a specific satellite sensor, J is the maximum value of the coverage opportunities of each satellite sensor, and j is the number of perception coverage opportunities corresponding to each satellite sensor; After performing immune operations on the chromosomes in the population, new fitness values ​​are obtained again, and all chromosomes are re-sorted based on the new fitness values; Output the chromosome whose fitness value in the population after secondary sorting is the preset situation as the optimal alternative.

2. The multi-satellite observation planning method based on immune multi-objective evolutionary algorithm according to claim 1 is characterized in that: Abstract the continuous roll angle into a discrete set of perception coverage opportunities of satellite sensors, obtain the width of the perception coverage opportunity, and take the cell whose width of the perception coverage opportunity is less than the corresponding cell of the maximum width of the satellite as the candidate left cell u. After determining all the left cells u, traverse all grids, draw a perpendicular line Lt through the cell t and the sub-satellite point trajectory line, draw a perpendicular line L″u through the left cell and the sub-satellite point trajectory line, and retain the cell below the perpendicular line Lt as the required upper cell t′. After determining all the upper cells t′, take the Lt line as the starting point, select the power on and off time with a certain time step, calculate the length of the perception coverage opportunity of a single satellite pass by the satellite orbit speed, multiply the width and length of the perception coverage opportunity to get a perception coverage opportunity, and cyclically obtain the discrete perception coverage opportunity set of a single satellite sensor. The discrete perception coverage opportunity set of all satellites is used as the solution space of multi-objective planning.

3. The multi-satellite observation planning method based on immune multi-objective evolutionary algorithm according to claim 2 is characterized in that: The width of the perceived coverage opportunity is represented by D, and the calculation formula is: D=D1+D2 Among them, P is the deflection angle of the satellite, H is the orbital height, and fov is the field of view angle; The calculation formula for the field of view angle fov is: Where d is the distance between the subsatellite point trajectory line and the straight line Lu passing through the left cell u and parallel to the subsatellite point trajectory line.

4. The multi-satellite observation planning method based on immune multi-objective evolutionary algorithm according to claim 1 is characterized in that: The method for updating the immune mutation operation is: Perform immune cloning and crossover mutation on each chromosome of the population in turn, then calculate the non-dominated order and crowding of all chromosomes, retain the chromosome with the highest non-dominated order and the lowest crowding as the optimal chromosome to the updated population, combine the updated population with the random new population, and use the elite retention strategy to obtain a new generation of sub-populations, perform mutation judgment on each gene of the chromosome, and randomly select genes that need to be mutated within a certain range of the original genes. This range gradually decreases with the increase of the number of iterations. The formula for random selection of the range is as follows: scope=g i +[-1,1]×(m / gen) Among them, g i is the original chromosome gene, m is the number of coverage opportunities of a specific satellite sensor, and gen is the current iteration number.

5. A multi-satellite observation planning device based on an immune multi-objective evolutionary algorithm, applied to a multi-satellite observation planning method based on an immune multi-objective evolutionary algorithm according to any one of claims 1 to 4, characterized in that: The first acquisition module is used to establish a grid of disaster-stricken areas and obtain a set of sensing coverage opportunities of each satellite sensor corresponding to the disaster-stricken area; The chromosome encoding module is used to perform chromosome encoding of the multi-objective evolutionary algorithm. Each chromosome is set to represent a combination of multi-satellite sensing coverage opportunities, and each gene position on the chromosome represents a sensing coverage opportunity selected by a specific satellite sensor. The second acquisition module is used to obtain the number of grids in the disaster-stricken area covered by the sensing coverage opportunity of each specific satellite sensor as the value of the corresponding gene bit; The third acquisition module is used to obtain the fitness value of each chromosome in the entire population to multiple targets through the value of each gene position, and perform a preliminary sorting of all chromosomes based on the fitness value; An updating module is used to perform immune operations on the chromosomes in the population, re-acquire new fitness values, and perform secondary sorting on all chromosomes based on the new fitness values; The output module is used to output the chromosome with a preset fitness value in the population after the secondary sorting as the optimal alternative.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm as described in any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm as described in any one of claims 1 to 4 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the multi-satellite observation planning method based on the immune multi-objective evolutionary algorithm as described in any one of claims 1 to 4 is implemented.

Images