A mission planning method for InSAR satellite based on multi-objective NSGA-III algorithm

Abstract

The invention discloses an InSAR satellite mission planning method based on a multi-objective NSGA-III algorithm, specifically: 1: Generate a satellite payload visible land rasterization matrix W; 2: Generate an initial population according to W; 3: Perform conflict task elimination on the initial population to obtain an updated population and satellite observation task instructions; 4: Calculate the objective function value of each individual in the population; 5: Generate a new population through a tournament selection method; 6: Perform crossover mutation on the new population, expand satellite observation task instructions and eliminate conflict tasks to obtain an updated population; 7: Merge the updated population with the previous generation population, and then select individuals to generate a selected population; 8: If the current number of iterations is greater than the maximum number of iterations, stop calculating and output the final satellite observation task instructions; otherwise, enter the next iteration. The invention provides help for InSAR satellite earth observation mission planning.

Description

Technical Field

[0001] The invention belongs to the field of satellite mission planning. Background Art

[0002] Interferometric synthetic aperture radar (InSAR) is an important remote sensing method for obtaining high-precision ground elevation models. With the rapid development of spaceborne InSAR technology, the number of imaging tasks has continued to increase and become more complex and diverse. However, problems such as the high cost of InSAR satellites and limited imaging resources still exist. Therefore, how to optimize imaging efficiency, efficiently utilize satellite resources, and improve the effectiveness of InSAR satellite use has become an urgent research direction. When formulating and generating satellite imaging plans, it is necessary to consider the diversity of mission requirements and various energy constraints of satellites during imaging, so as to ensure the safe operation of satellites and the efficient implementation of imaging tasks.

[0003] The existing research on InSAR satellite mission planning considers ideal constraints, but does not consider the constraints in actual engineering, so it cannot be applied to actual engineering. Most multi-objective evolutionary algorithms in the research on InSAR satellite intelligent mission planning are more effective when solving problems with low target dimensions, but when faced with high-dimensional target optimization problems, many methods become unsatisfactory. Summary of the invention

[0004] Purpose of the invention: In order to solve the problems existing in the above-mentioned prior art, the present invention provides an InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm.

[0005] Technical solution: The present invention provides an InSAR satellite mission planning method based on a multi-objective NSGA-III algorithm, which specifically includes the following steps:

[0006] Step 1: Generate the satellite payload visible land rasterization matrix based on satellite orbit information, satellite payload information and global land prior information where w i It represents the set of wave position numbers that can be turned on at the i-th moment. One moment corresponds to one observation task, i = 1, 2, ..., N task ; where N task Indicates the total number of observation missions in one orbit and preprocesses the satellite payload visible land rasterization matrix;

[0007] Step 2: The satellite is regarded as a chromosome, and the observation task is regarded as a gene position on the chromosome. The gene position is assigned a value according to the wave position sequence number that can be turned on in the satellite payload visible land rasterization matrix, thereby randomly generating m strings of chromosomes to form an initial population;

[0008] Step 3: resolve conflicting tasks on the initial population, obtain an updated population and satellite observation task instructions, and use the updated population as the previous generation population; the specific method of obtaining the satellite observation task instructions is: if the wave position sequence numbers of the previous and next gene positions of a certain gene position in the updated population are both 0, then the gene position is used as a task segment, and the time corresponding to the gene position is used as the start and end time of the task segment;

[0009] The continuous gene positions in a chromosome whose wave position number is not 0 form a task segment, and the time corresponding to the first gene position in the task segment is used as the starting time of the task segment, and the time corresponding to the last gene position is used as the ending time of the task segment;

[0010] Arrange each task segment from the smallest to the smallest according to the start time to obtain a task segment sequence; combine the start time, end time, number of gene bits, satellite corresponding to the gene bit, and the wave bit sequence value on the first gene bit in each task segment in the task segment sequence as the satellite observation task instruction of the task segment to obtain the satellite observation task instruction set order;

[0011] Step 4: Establish the objective function and calculate the objective function value of each chromosome in the updated population;

[0012] Step 5: Generate a new population from the previous generation population through tournament selection;

[0013] Step 6: Perform crossover mutation on the new population in step 5, expand each satellite observation task instruction and resolve conflicting tasks to obtain updated populations and satellite observation task instructions;

[0014] Step 7: Merge the population updated in step 6 with the previous generation population, and use the multi-objective NSGA-III algorithm to select individuals from the merged population based on the objective function to generate the selected population;

[0015] Step 8: Determine whether the current number of iterations is greater than the maximum number of iterations. If so, stop the calculation and output the final satellite observation task instruction; otherwise, go to step 9;

[0016] Step 9: Use the selected population generated in step 7 as the previous generation population, increase the number of iterations by 1 and go to step 5.

[0017] Furthermore, the step 1 is specifically as follows:

[0018] Step a: Set the six elements of the InSAR satellite main satellite orbit and its payload wave position in the STK simulation environment to establish the InSAR satellite STK simulation environment;

[0019] Step b: interconnect through matlab and STK, and record the latitude and longitude of the visible area of ​​the satellite payload wave position in units of days;

[0020] Step c: traverse the longitude and latitude of the visible area of ​​the payload wave position, and generate the south latitude ascending orbit, south latitude descending orbit, north latitude ascending orbit and north latitude descending orbit time periods according to the time when the latitude of the satellite visible area crosses the equator and the farthest point of the North and South Poles; traverse the orbit, recursively deduce the north latitude ascending orbit segment and the south latitude descending orbit segment in each orbital cycle through MATLAB and STK interconnection, and record the power on and off time of the satellite payload, determine whether there is an observation task in the imaging grid according to the SRTM global land prior data, and statistically record the areas that need to be observed, and finally output the grid matrix.

[0021] Furthermore, in step 2, when assigning a value to the i-th gene position in a chromosome, in w i A wave position number is randomly selected and assigned to the i-th gene position of a chromosome.

[0022] Furthermore, in step 6, the crossover is performed by single-point crossover, and the mutation is performed by single-point mutation.

[0023] Furthermore, the single-point crossover is specifically as follows: randomly select two parent individuals, select the gene position with wave position number 0 in the two parent chromosomes, obtain sets T1 and T2, randomly select a gene position from the intersection of T1 and T2 as the crossover point, and if the intersection of T1 and T2 is empty, randomly select one from the two parent chromosomes as the offspring individual.

[0024] Further, the single point mutation is specifically as follows: select the gene position with the middle wave position number 0 of the offspring individual to obtain the task set T3, select the tasks that are not an empty set in W to obtain the task set T4, and randomly select a gene position from the intersection of T3 and T4 as the mutation point turn up The corresponding chromosome is determined The next gene locus and the previous locus Are all the wave position numbers 0? If so, select The corresponding task w mpoint , in w mpoint Randomly select a wave number and assign it to if and There is a wave position number that is not 0 in the set w. mpoint In the case of if and The wave position numbers are not 0 and belong to the set wmpoint In and Randomly select a wave number and assign it to

[0025] Furthermore, the satellite observation task instruction in step 6 is expanded as follows: for the ath task segment on the satth chromosome, determine whether the time length of the task segment meets the single power-on time constraint of the satth satellite, sat = 1, 2, ..., m. If it meets the constraint, find the second before the start time of the task segment in the matrix W. strat -1 corresponds to the set of wave positions that can be turned on Find the second t after the end time of the task segment in matrix W end +1 corresponding wave position set that can be powered on if is not an empty set, and the gene position corresponding to the start time of the a-th task segment The wave position number on Then The wave position number on is assigned to Will The corresponding time is taken as the starting time of the ath task segment. is on the sat chromosome The previous gene position of is not an empty set, and the gene position corresponding to the end time of the a-th task segment The wave position number on Then The wave position number on is assigned to Will The corresponding time is taken as the end time of the ath task segment. is on the sat chromosome The latter gene locus.

[0026] Furthermore, the objective functions in step 4 and step 7 are:

[0027]

[0028] Among them, f1 represents the objective function of the duration of the task in the conventional area, f2 represents the objective function of the duration of the task in the key area, and f3 is the objective function of the task optimization level. i1. time represents the observation time of the i1th satellite observation mission instruction, o i1. keyarea indicates the observation time of the key area of ​​the i1th satellite observation mission instruction, o i1. priority represents the optimization level of the i1th satellite observation task instruction.

[0029] Furthermore, the conflicting task resolution in step 3 and step 6 is specifically as follows: determine whether there is a data transmission window in the current orbit, and whether the storage capacity of all satellites exceeds the set threshold value; if there is a data transmission window in the current orbit and there is a satellite whose storage capacity exceeds the set threshold value, the data transmission task is executed, and the wave position sequence number of the gene bit corresponding to the observation task that conflicts with the data transmission window is set to 0; then traverse each satellite, form a task set with the gene bits in the satellite whose wave position sequence number is not 0, determine whether two adjacent observation tasks meet the satellite payload constraint, and if not, randomly delete one of the two observation tasks until all tasks in the task set meet the satellite payload constraint; then traverse each observation task in the task set, and delete the repeated observation tasks; finally, delete the tasks that do not meet the satellite resource constraint;

[0030] The satellite payload constraints are:

[0031]

[0032] In the formula, wave on Indicates the number of power-on wave positions. and They represent the start-up time and shutdown time of the satellite payload corresponding to the i'th satellite observation mission instruction, respectively. Indicates the start-up time of the satellite payload corresponding to the i'+1th satellite observation mission instruction; and They represent the start-up time and shutdown time of the satellite wave position for the j-th satellite observation task instruction respectively; Indicates the start time of the satellite wave position for the j+1th satellite observation task instruction; time on Indicates the maximum working time of a single satellite startup, time restart Indicates the satellite restart time, time change Indicates the wave position switching time.

[0033] The data transmission communication constraints are:

[0034]

[0035] In the formula, represents the data transmission task of the kth track, represents the empty set;

[0036] The satellite resource constraints are:

[0037]

[0038] In the formula, in the formula, times on and times changeRespectively represent the number of satellite power-on times and the number of wave position switching times. and They represent the maximum number of satellite startup times and the maximum number of wave position switching times. Represents the storage capacity of the satellite in the kth orbit, memory max Indicates the maximum storage capacity of the satellite.

[0039] Furthermore, the preprocessing of the satellite payload visible land rasterization matrix in step 1 is specifically as follows: deleting the set of wave position numbers that can be turned on and are not in the configuration imaging area, and deleting the wave positions that cannot be turned on in the latitude band at the i-th moment.

[0040] Beneficial effects: The present invention adopts a multi-chromosome encoding method to make the multi-satellite mission planning problem clearer and more straightforward; the NSGA-III algorithm is used to solve high-dimensional multi-objective problems, with better optimization effects, and full consideration of actual engineering constraints, which provides assistance for InSAR satellite earth observation mission planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of an embodiment of the present invention;

[0042] Figure 2 Schematic diagram of a multiple chromosome encoding embodiment of the present invention. DETAILED DESCRIPTION

[0043] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0044] like Figure 1 As shown, the technical solution provided by the present invention is an InSAR satellite mission planning method based on a multi-objective NSGA-III algorithm, and the specific steps are as follows:

[0045] Step 1: Generate a satellite payload visible land rasterization matrix based on satellite orbit information, satellite payload information and global land prior information.

[0046] According to the strict return orbit of the main satellite of the InSAR satellite system and the SAR payload wave position information, the STK simulation environment of the formation main satellite with SAR payload is established. The orbit of the main satellite is recursively calculated in seconds, with the north latitude ascending orbit and the south latitude descending orbit, and the global attribute grid matrix of the payload is generated by combining the SRTM global land prior data. Among them, w i N represents the set of wave position numbers that can be turned on at the i-th moment. One moment corresponds to one observation task. taskis the total number of missions on an orbit. The matrix mainly includes the satellite payload wave position visible grid attributes, revisit period orbit number, sub-satellite point longitude and latitude and other data information at each moment.

[0047] Step a: Set the six elements of the InSAR satellite main satellite orbit and its payload wave position in the STK simulation environment to establish the InSAR satellite STK simulation environment;

[0048] Step b: Interconnect with STK through MATLAB, call STK to generate the longitude and latitude of the satellite's sub-satellite point, and record the longitude and latitude of the visible area of ​​the satellite payload wave position in days;

[0049] Step c: traverse the longitude and latitude of the visible area of ​​the payload wave position, and generate the south latitude ascending orbit, south latitude descending orbit, north latitude ascending orbit and north latitude descending orbit time periods according to the time when the latitude of the satellite visible area crosses the equator and the farthest point of the North and South Poles; traverse the orbit, recursively deduce the north latitude ascending orbit segment and the south latitude descending orbit segment in each orbital cycle through MATLAB and STK interconnection, and record the power on and off time of the satellite payload, determine whether there is an observation task in the imaging grid according to the SRTM global land prior data, and statistically record the areas that need to be observed, and finally output the grid matrix.

[0050] The satellite payload visible land rasterization matrix is ​​preprocessed; the set of power-on wave positions that are not in the configuration imaging area is deleted considering the configuration constraints; the set of power-on wave positions at each moment in the rasterization matrix is ​​further adjusted considering the latitude band constraints (different latitude bands correspond to different power-on wave positions), and the wave positions that cannot be powered on in the imaging latitude band at that moment are deleted, and finally the global rasterization matrix of payload wave positions that meets the configuration constraints is obtained. The imaging latitude band constraints are that the higher the latitude, the smaller the diameter of the earth, and the fewer areas need to be imaged. If all wave positions are used for imaging, it will cause repeated imaging of the area. Therefore, the higher the latitude, the fewer wave positions need to be powered on.

[0051] Step 2: Under the constraints of the InSAR satellite configuration, according to the information of the power-on wave position at each moment in the payload wave position rasterization matrix, a multi-chromosome encoding method is adopted, and the imaging wave position sequence number is used as the encoding information to randomly generate m strings of multi-chromosomes to form an initial population such as Figure 2 shown.

[0052] The initial population is obtained by using the encoding method of multiple chromosomes x sat represents the sat chromosome, represents the Nth on the sat chromosome sat gene loci, the length of the chromosome is N task , whose gene positions correspond one to one with the number of track tasks; for the i-th gene position in a chromosome s When assigning a value,i A random wave position number b is selected and assigned to the i-th gene position of a chromosome. like but Represents the empty set.

[0053] Step 3: Considering constraints such as time window constraints, resource constraints, and data transmission constraints, a conflict resolution operator is introduced to transform infeasible solutions in the population into feasible solutions.

[0054] First, determine whether there is a data transmission window in the orbit and whether the storage capacity of the satellite exceeds the set threshold value. If there is a data transmission window and there is a satellite whose storage capacity exceeds the set threshold value, the data transmission task is executed, and the wave position number of the gene bit corresponding to the observation task that conflicts with the data transmission window is set to 0; then traverse each satellite, and form a satellite execution task set with the task of wave position number 0 in each satellite, and determine whether the adjacent tasks meet the satellite payload constraint. If not, the tasks are eliminated until the satellite execution task set meets the satellite payload constraint; then traverse each observation task, delete the tasks that are observed repeatedly to ensure that the observation task is only observed once; finally, determine the satellite resource constraint, delete the tasks that do not meet the constraint, and generate a new offspring individual offspring (that is, a new population) and a satellite observation task instruction set order.

[0055] The specific method of obtaining the satellite observation task instruction is: if the wave position sequence numbers of the previous and next gene positions of a certain gene position in the updated population are all 0, then the gene position is used as a task segment, and the time corresponding to the gene position is used as the start and end time of the task segment.

[0056] The continuous gene positions in a chromosome whose wave position number is not 0 are grouped into a task segment. The time corresponding to the first gene position in the task segment is taken as the starting time of the task segment, and the time corresponding to the last gene position is taken as the ending time of the task segment.

[0057] Arrange each task segment from the smallest to the smallest according to the start time to obtain a task segment sequence; in the task segment sequence, combine the start time, end time, number of gene bits, orbit where the gene bit is located, corresponding satellite, and the wave bit sequence value on the first gene bit in each task segment as the satellite observation task instruction of the task segment to obtain the satellite observation task instruction set order, order = {o1, o2, ..., o n},o n Indicates the nth satellite observation mission instruction.

[0058] The specific steps of the conflict resolution operator are as follows:

[0059]

[0060]

[0061]

[0062] The satellite payload constraint is that the single startup time of the satellite payload shall not exceed the maximum working time and the time interval between two startups must meet the payload restart time. The payload can only use one wave position for imaging each time it is started, and the wave position switching must meet the payload switching time. This constraint is described by the following formula:

[0063]

[0064] In the formula, wave on Indicates the number of power-on wave positions. and They represent the start-up time and shutdown time of the satellite payload corresponding to the i'th satellite observation mission instruction, respectively. Indicates the start-up time of the satellite payload corresponding to the i'+1th satellite observation mission instruction; and They represent the start-up time and shutdown time of the satellite wave position for the j-th satellite observation task instruction respectively; Indicates the start time of the satellite wave position for the j+1th satellite observation task instruction; time on Indicates the maximum working time of a single satellite startup, time restart Indicates the satellite restart time, time change Indicates the wave position switching time.

[0065] Due to digital communication constraints, the satellite cannot perform imaging when performing digital transmission with the ground station.

[0066]

[0067] In the formula, represents the data transmission task of the kth orbit; this formula indicates that there is no conflict between the data transmission task and the satellite imaging task instructions.

[0068] Satellite resource constraints: The number of satellite startups, the number of wave position switching, and the storage capacity in each orbit are all limited by satellite resources. This constraint is described by the following formula:

[0069]

[0070] times on and times change Respectively represent the number of satellite power-on times and the number of wave position switching times. and They represent the maximum number of satellite startup times and the maximum number of wave position switching times. Represents the storage capacity of the satellite in the kth orbit, memory max Indicates the maximum storage capacity of the satellite.

[0071] Step 4: Construct a multi-objective function, namely the three optimization objectives of observation time in conventional areas, observation time in key areas, and task priority.

[0072] Traverse the satellite observation task instructions generated in step 4, calculate the total observation task time and total task priority of each instruction execution task, determine whether the execution task includes the key area observation task and calculate the key area observation time. The multi-objective function expression is as follows:

[0073]

[0074] Among them, f1 represents the objective function of the duration of the task in the conventional area, f2 represents the objective function of the duration of the task in the key area, and f3 is the objective function of the task optimization level. i1. time represents the observation time of the i1th satellite observation mission instruction, o i1. keyarea indicates the observation time of the key area of ​​the i1th satellite observation mission instruction, o i1. priority represents the optimization level of the i1th satellite observation task instruction.

[0075] Step 5: Select individuals from the population through the tournament selection method, and use crossover, mutation operators and expansion operations to generate a new population.

[0076] Firstly, the non-dominated relationship is calculated according to the objective function value of each individual, and the first m individuals are selected to enter the next generation population according to the non-dominated sorting level and crowding distance of the individuals in the population, and then the crossover mutation operator is used for operation.

[0077] The crossover operator uses a single-point crossover operation. In order to avoid the selected crossover point in the continuously executed tasks, resulting in the generated offspring individual offspring not satisfying the constraint conditions, the following operation is taken to select the crossover point Generate two sets of unallocated satellite resource tasks (gene positions with the start-up wave position number of 0) T1 and T2 for the parent individuals, and select a gene from the intersection of T1 and T2 as the intersection point to avoid it appearing in the continuously executed task segments. If the intersection of T1 and T2 is empty, randomly select a child individual from the parent generation.

[0078] The specific steps of the crossover operator are as follows:

[0079]

[0080]

[0081] The mutation operator uses a single-point mutation operation and an expansion operation based on the task sequence. The single-point mutation operation is as follows: First, the task set T3 of the unallocated satellite resources of the offspring individuals and the task set T4 that is not an empty set in the payload wave position matrix W are generated, and a gene position is randomly selected from the intersection of T3 and T4 as the mutation point. turn up The corresponding chromosome is determined The next gene locus and the previous locus Are all the wave position numbers 0? If so, select The corresponding task w mpoint , in w mpoint Randomly select a wave number and assign it to if and There is a wave position number that is not 0 in the set w. mpoint , then assign the wave position number that is not 0 to if and The wave position numbers are not 0 and belong to the set w mpoint , then in and Randomly select a wave number and assign it to

[0082] The expansion operation based on the task sequence is as follows: for the ath task segment on the satth chromosome, determine whether the duration of the task segment meets the single power-on duration constraint of the satth satellite, sat = 1, 2, ..., m. If it does, find the second before the start time of the task segment in the matrix W. strat -1 corresponds to the set of wave positions that can be turned on Find the second t after the end time of the task segment in matrix W end +1 corresponding wave position set that can be powered on if is not an empty set, and the gene position corresponding to the start time of the a-th task segment The wave position number on Then The wave position number on is assigned to Will The corresponding time is taken as the start time of the ath task segment. is on the sat chromosome The previous gene position of is not an empty set, and the gene position corresponding to the end time of the a-th task segment The wave position number on Then The wave number on the Will The corresponding time is taken as the end time of the ath task segment. is on the sat chromosome The latter gene locus.

[0083] The specific steps of the mutation operator are as follows:

[0084]

[0085]

[0086] represents the data transmission time window of the kth track; Indicates the total storage capacity of the k-th orbital satellite; mem sill Indicates the threshold of storage capacity. Once the threshold is exceeded, data transmission operation is required.

[0087] In step 6, the new population generated after step 5 is merged with the previous generation population, and a selected population is generated based on the objective function value and the reference point selection mechanism in the multi-objective NSGA-III algorithm.

[0088] The size of the merged population is 2m. First, sort the merged population by non-dominated level and associate the reference point; then, put the population of that level into the new population X according to the non-dominated ranking from small to large. i+1 , until a certain level F is placed j X appears when the population i+1 The size of exceeds the population size m; finally, according to F j The number of individual associated reference points in X continues to increase from small to large. i+1 This process ends when the population size reaches m.

[0089] Step 7, determine whether the current population meets the requirements for exiting iteration. If so, output all observation task instructions, otherwise return to step 2.

[0090] Determine whether the current number of iterations g is not less than the maximum number of iterations g max If the termination condition is not met, the selected population obtained this time is used as the next generation population, and the optimization calculation is continued in step 5. Otherwise, all the final observation task instructions are output.

[0091] The observation task instruction set obtained in this embodiment is shown in Table 1 below:

[0092] Table 1

[0093]

[0094]

[0095] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

Claims

1. An InSAR satellite mission planning method based on a multi-objective NSGA-III algorithm, characterized by: The specific steps include: Step 1: Generate the satellite payload visible land rasterization matrix based on satellite orbit information, satellite payload information and global land prior information where w i It represents the set of wave position numbers that can be turned on at the i-th moment. One moment corresponds to one observation task, i = 1, 2, ..., N task ; where N task Indicates the total number of observation missions in one orbit and preprocesses the satellite payload visible land rasterization matrix; Step 2: The satellite is regarded as a chromosome, and the observation task is regarded as a gene position on the chromosome. The gene position is assigned a value according to the wave position sequence number that can be turned on in the satellite payload visible land rasterization matrix, thereby randomly generating m strings of chromosomes to form an initial population; Step 3: resolve conflicting tasks on the initial population, obtain an updated population and satellite observation task instructions, and use the updated population as the previous generation population; the specific method of obtaining the satellite observation task instructions is: if the wave position sequence numbers of the previous and next gene positions of a certain gene position in the updated population are both 0, then the gene position is used as a task segment, and the time corresponding to the gene position is used as the start and end time of the task segment; The continuous gene positions in a chromosome whose wave position number is not 0 form a task segment, and the time corresponding to the first gene position in the task segment is used as the starting time of the task segment, and the time corresponding to the last gene position is used as the ending time of the task segment; Arrange each task segment from small to large according to the start time to obtain a task segment sequence; in the task segment sequence, combine the start time, end time, number of gene bits, satellite corresponding to the gene bit, and the wave bit sequence value on the first gene bit in each task segment as the satellite observation task instruction of the task segment to obtain the satellite observation task instruction set order; Step 4: Establish the objective function and calculate the objective function value of each chromosome in the updated population; Step 5: Generate a new population from the previous generation population through tournament selection; Step 6: Perform crossover mutation, satellite observation task instruction expansion and conflict task elimination on the new population in step 5 to obtain the updated population and satellite observation task instructions; Step 7: Merge the updated population in step 6 with the previous generation population, select individuals from the merged population based on the objective function value and the multi-objective NSGA-III algorithm, and generate the selected population; Step 8: Determine whether the current number of iterations is greater than the maximum number of iterations. If so, stop the calculation and output the final satellite observation task instruction; Otherwise, go to step 9; Step 9: Use the selected population generated in step 7 as the previous generation population, increase the number of iterations by 1 and go to step 5; The satellite observation task instruction in step 6 is specifically expanded as follows: for the ath task segment on the satth chromosome, determine whether the time length of the task segment meets the single power-on time constraint of the satth satellite, sat = 1, 2, ..., m. If it does, find the second before the start time of the task segment in the matrix W. strat -1 corresponds to the set of wave positions that can be turned on Find the second t after the end time of the task segment in matrix W end +1 corresponding wave position set that can be powered on if is not an empty set, and the gene position corresponding to the start time of the a-th task segment The wave position number on Then The wave position number on is assigned to Will The corresponding time is taken as the starting time of the ath task segment. is on the sat chromosome The previous gene position of is not an empty set, and the gene position corresponding to the end time of the a-th task segment The wave position number on Then The wave position number on is assigned to Will The corresponding time is taken as the end time of the ath task segment. is on the sat chromosome The latter gene locus; The objective function in step 4 and step 7 is: Among them, f1 represents the objective function of the duration of the task in the conventional area, f2 represents the objective function of the duration of the task in the key area, and f3 is the objective function of the task optimization level. i1 .time represents the observation time of the i1th satellite observation mission instruction, o i1 .keyarea indicates the observation time of the key area of ​​the i1 satellite observation mission instruction. i1 .priority indicates the optimization level of the i1th satellite observation task instruction; The conflict task resolution in step 3 and step 6 is specifically as follows: determine whether there is a data transmission window in the current orbit, and whether the storage capacity of all satellites exceeds the set threshold value; if there is a data transmission window in the current orbit and there is a satellite whose storage capacity exceeds the set threshold value, execute the data transmission task, and set the wave position number of the gene bit corresponding to the observation task that conflicts with the data transmission window to 0; then traverse each satellite, form a task set with the gene bits in the satellite whose wave position number is not 0, determine whether two adjacent observation tasks meet the satellite payload constraint, and if not, randomly delete one of the two observation tasks until all tasks in the task set meet the satellite payload constraint; then traverse each observation task in the task set, delete the repeated observation tasks, and ensure that the observation task is only observed once; finally, delete the tasks that do not meet the satellite resource constraint; The satellite payload constraints are: In the formula, wave on Indicates the number of power-on wave positions. and They represent the start-up time and shutdown time of the satellite payload corresponding to the i'th satellite observation mission instruction, respectively. Indicates the start-up time of the satellite payload corresponding to the i'+1th satellite observation mission instruction; and They represent the start-up time and shutdown time of the satellite wave position for the j-th satellite observation task instruction respectively; Indicates the start time of the satellite wave position for the j+1th satellite observation task instruction; time on Indicates the maximum working time of a single satellite startup, time restart Indicates the satellite restart time, time change Indicates the wave position switching time; The data transmission communication constraints are: In the formula, represents the data transmission task of the kth track, represents the empty set; The satellite resource constraints are: In the formula, times on and times change Respectively represent the number of satellite power-on times and the number of wave position switching times. and They represent the maximum number of satellite startup times and the maximum number of wave position switching times. Represents the storage capacity of the satellite in the kth orbit, memory max Indicates the maximum storage capacity of the satellite.

2. The InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm according to claim 1, characterized in that: The step 1 is specifically as follows: Step a: Set the six elements of the InSAR satellite main satellite orbit and its payload wave position in the STK simulation environment to establish the InSAR satellite STK simulation environment; Step b: interconnect through matlab and STK, and record the latitude and longitude of the visible area of ​​the satellite payload wave position in units of days; Step c: Traverse the longitude and latitude of the visible area of ​​the payload wave position, and generate the south latitude ascending orbit, south latitude descending orbit, north latitude ascending orbit and north latitude descending orbit time periods according to the moment when the latitude of the satellite visible area crosses the equator and the farthest point of the North and South Poles; perform orbit recursion on the north latitude ascending orbit segment and the south latitude descending orbit segment in each orbit, determine whether there is an observation task in the imaging grid based on the SRTM global land prior data, and statistically record the areas that need to be observed, and finally output the grid matrix.

3. The InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm according to claim 1, characterized in that: In step 2, when assigning a value to the i-th gene position in a chromosome, in w i A wave position number is randomly selected and assigned to the i-th gene position of a chromosome.

4. The InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm according to claim 1, characterized in that: In step 6, the crossover adopts single-point crossover, and the mutation adopts single-point mutation.

5. The InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm according to claim 4, characterized in that: The single-point crossover is specifically as follows: randomly select two parent individuals, select the gene position with the wave position number 0 in the two parent chromosomes, obtain sets T1 and T2, randomly select a gene position from the intersection of T1 and T2 as the crossover point, and if the intersection of T1 and T2 is empty, randomly select one from the two parent chromosomes as the offspring individual.

6. The InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm according to claim 4, characterized in that: The single-point mutation is specifically as follows: select the gene position with the middle wave position number 0 of the offspring individual to obtain the task set T3, select the tasks that are not an empty set in W to obtain the task set T4, and randomly select a gene position from the intersection of T3 and T4 as the mutation point turn up The corresponding chromosome is determined The next gene locus and the previous locus Are all the wave position numbers 0? If so, select The corresponding task w mpoint , in w mpoint Randomly select a wave number and assign it to if and There is a wave position number that is not 0 in the set w. mpoint , then assign the wave position number that is not 0 to if and The wave position numbers are not 0 and belong to the set w mpoint , then in and Randomly select a wave number and assign it to 7. The InSAR satellite mission planning method based on the multi-objective NSGA-III algorithm according to claim 1, characterized in that: The preprocessing of the satellite payload visible land rasterization matrix in step 1 is specifically as follows: deleting the set of wave position numbers that can be turned on and are not in the InSAR configuration imaging area, and deleting the wave positions that cannot be turned on in the latitude band at the i-th moment.

Images