CN114091892A - Multi-satellite on-orbit collaborative earth observation task planning method and system - Google Patents
Multi-satellite on-orbit collaborative earth observation task planning method and system Download PDFInfo
- Publication number
- CN114091892A CN114091892A CN202111370416.9A CN202111370416A CN114091892A CN 114091892 A CN114091892 A CN 114091892A CN 202111370416 A CN202111370416 A CN 202111370416A CN 114091892 A CN114091892 A CN 114091892A
- Authority
- CN
- China
- Prior art keywords
- population
- satellite
- task
- observation
- target point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 59
- 230000035772 mutation Effects 0.000 claims abstract description 25
- 239000006185 dispersion Substances 0.000 claims abstract description 10
- 230000003044 adaptive effect Effects 0.000 claims abstract description 6
- 210000000349 chromosome Anatomy 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000010187 selection method Methods 0.000 claims description 10
- 230000001360 synchronised effect Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 4
- 230000006641 stabilisation Effects 0.000 claims description 4
- 238000011105 stabilization Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 108090000623 proteins and genes Proteins 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 abstract description 27
- 230000008901 benefit Effects 0.000 description 11
- 238000005457 optimization Methods 0.000 description 4
- 206010033307 Overweight Diseases 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 108010063499 Sigma Factor Proteins 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000005056 compaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036449 good health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Development Economics (AREA)
- Molecular Biology (AREA)
- Educational Administration (AREA)
- Genetics & Genomics (AREA)
- Game Theory and Decision Science (AREA)
- Physiology (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Radio Relay Systems (AREA)
Abstract
The invention discloses a multi-satellite on-orbit collaborative earth observation task planning method and a system, which improve a genetic algorithm for planning a multi-satellite collaborative earth observation task from various angles such as an initial solution, a selection operator, a crossover operator, a mutation operator and a stopping condition based on a population convergence coefficient by designing a heuristic initial solution generation strategy, a roulette selection algorithm based on Boltzmann selection probability, an adaptive crossover operator and a mutation operator based on a population dispersion degree and the stopping condition based on a population convergence coefficient.
Description
Technical Field
The invention belongs to the technical field of satellite observation, and particularly relates to a planning method and a system for a multi-satellite on-orbit collaborative earth observation task.
Background
The satellite flies on the orbit, and a two-dimensional scanning band with a certain width and taking the point below the satellite as a central line can be observed by utilizing the attitude maneuver. However, due to the limitation of the on-orbit motion of the satellite, the field angle of the satellite-borne remote sensing equipment, the side swing range of the satellite-borne remote sensing equipment and other factors, the satellite can observe the ground target point only in a limited time window, and sometimes the ground target point is not visible even in the planning time. In a multi-satellite scenario, a target point may have multiple time windows with a satellite or may not be visible with multiple satellite nodes. The multi-satellite cooperative earth observation task means that a user specifies a batch of target points to be observed, a time interval for executing observation and priority weights of each target point, cooperates with a plurality of satellite nodes through a scheduling strategy, the target points are observed as many as possible under the condition of meeting satellite imaging constraint conditions, and the income of the observation task is as high as possible.
The multi-satellite cooperative earth observation is proved to be an NP-hard problem, the index explosion characteristic is very obvious, the optimal solution is difficult to obtain through direct calculation, and the current solution method is to convert the problem into a combined optimization problem to obtain an approximate solution. A plurality of intelligent optimization algorithms including a genetic algorithm, an ant colony algorithm and a particle swarm algorithm are main means for solving the problem of combined optimization at present, but because a special scene of task scheduling is observed in a multi-satellite cooperation mode, the traditional intelligent optimization algorithm cannot be directly used for solving the problem, and the algorithm needs to be improved from many aspects of coding and decoding of a scheduling scheme, formulation and improvement of operator operation, setting of iteration stopping conditions and the like aiming at a constellation special environment, so that the execution effect of the multi-satellite earth observation task is effectively improved.
In addition, the traditional multi-satellite scheduling algorithm is a simulation in a modeling system, and an applicable scheduling algorithm needs to be further applied to a distributed cluster system, so that in-orbit cooperation with multiple satellites can be performed for observation, and actual engineering and economic benefits are generated.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a method and a system for planning a multi-satellite on-orbit cooperative earth observation task aiming at the defects in the prior art, the method and the system are oriented to the multi-satellite cooperative earth observation task, the genetic algorithm is improved, the total income of the observation task is improved, the number of target points to be observed successfully is increased, and the on-orbit planning of the multi-satellite cooperative earth observation task is realized.
The invention adopts the following technical scheme:
a multi-satellite on-orbit collaborative earth observation task planning method comprises the following steps:
s1, acquiring the geographic position, task priority and task completion time interval information of each target point to be observed, accessing the etcd through the main control node to acquire real-time synchronous satellite node state data, and calculating a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites;
s2, generating a coding result based on the satellite node state data obtained in the step S1 and the geographic position, the task priority and the task completion time interval information of each target point to be observed;
s3, based on the coding result of the step S2 and the visible time window of the step S1, adopting an initial solution generation strategy with priority of task execution profit, planning a feasible observation window on the premise that each target point meets the constraint by taking the task priority as a sequence, and adopting an observation task execution time dynamic adjustment strategy to adjust the task execution time to obtain an initial population;
s4, calculating the fitness value of the initial population obtained in the step S3, selecting individuals from the initial population for heredity by using a roulette selection method based on Boltzmann selection probability, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on orbit circles on the heredity result;
s5, iterating the new generation of population obtained in the step S4 until the end, and outputting a scheduling scheme of the mission planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
Specifically, step S2 specifically includes:
firstly, carrying out chromosome coding, wherein each chromosome represents a feasible scheduling scheme, each gene in the chromosome represents scheduling of an observation target point, N +1 task scheduling queues are created for each chromosome aiming at N satellites flying in orbit, each scheduling queue corresponds to one satellite, each target point to be observed in the scheduling queues is represented by a structural body, and the N +1 scheduling queue is used for storing the target point which is not successfully scheduled in the task scheduling process to obtain a coding result.
Specifically, in step S3, the initial solution generation strategy adopting the task execution benefit priority specifically includes:
for an observation task required by a user, firstly, a plurality of target points in the observation task are sorted according to the priority specified by the user, and then an initial solution of an algorithm is generated according to the sorting sequence; for each target point, acquiring a visible time window between each satellite and the corresponding target point, and randomly selecting one time window to perform observation; when a time window is selected, whether the task meets constraint conditions after being scheduled to a satellite corresponding to the time window is judged, and if any constraint condition is not met, the next time window is selected for scheduling; and if all the selectable time windows do not meet the constraint condition, the corresponding target point fails to be scheduled.
Further, the constraint conditions comprise the rolling time of the satellite-borne remote sensing equipment required by the satellite during observation, the stabilization time of the remote sensing equipment and whether the on-satellite storage is sufficient or not.
Specifically, in step S3, the dynamic adjustment strategy for the observation task execution time specifically includes:
for the observation tasks of the front target point and the rear target point which are scheduled to the same satellite node, setting the observation execution time of the front target point as the starting time of a selected time window, and when the time window of the second target point arrives, if the sidewise swinging of the satellite-borne remote sensing equipment is not finished or a camera is not stable, waiting for the conversion time to be finished and then executing the observation; and if the conversion activity is not completed when the second target point selection time window is finished, the target point observation fails.
Specifically, in step S4, individuals are selected from the initial population for inheritance using a roulette selection method using boltzmann selection probability p (i) as follows:
wherein f isiB is the fitness value of the individual i in the population and is used to control the selection strength.
Specifically, in step S4, the crossing and variation method based on the track turns is specifically:
evaluating the dispersion degree of the population through the standard deviation of the fitness values of all individuals in the population by measuring the dispersion degree of the contemporary population and dynamically changing the cross rate and the variation rate; and when the standard deviation sigma of the parent population is greater than the preset threshold value, the probability of intersection and mutation is reduced.
Further, the standard deviation σ of the parent population is as follows:
wherein N is the total number of individuals in the population, fiIs the fitness value of the individual i in the population,
the average fitness value of all individuals in a generation population;
crossing rate PcAnd the mutation probability PmThe adaptive update of (1) is as follows:
wherein, PcAnd PmRepresents the crossing rate and the variation rate, P, of the previous generation populationcAnd PmAnd expressing the population crossing rate and the variation rate after updating, wherein alpha is an updating speed control factor, sigma is a population fitness standard deviation, and d is a preset threshold value.
Specifically, in step S5, a stopping condition is determined according to the convergence coefficient of the population, iteration is stopped when the convergence coefficient of the population of consecutive 5 generations is less than or equal to a preset threshold, and the population convergence coefficient belongs to the following according to the average fitness evaluation of the population:
wherein,
represents the average fitness of the parent population,
representing the average fitness of the progeny population.
Another technical solution of the present invention is a multi-satellite on-orbit collaborative earth observation task planning system, including:
the data module is used for acquiring the geographic position, the task priority and the task completion time interval information of each target point to be observed, acquiring real-time synchronous satellite node state data by accessing the etcd through the main control node, and calculating a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites;
the encoding module is used for generating an encoding result based on the satellite node state data obtained by the data module and the geographic position, the task priority and the task completion time interval information of each target point to be observed;
the group module adopts an initial solution generation strategy of task execution profit priority based on the coding result and the visible time window, plans a feasible observation window for each target point in sequence by taking the task priority as a sequence on the premise of meeting the constraint, and adopts an observation task execution time dynamic adjustment strategy to adjust the task execution time to obtain an initial group;
the calculation module is used for calculating the fitness value of the initial population obtained by the population module, selecting individuals from the initial population for heredity by using a roulette selection method based on Boltzmann selection probability, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on track circles on the heredity result;
the planning module is used for iterating the new generation of population obtained by the calculation module until the end of the iteration, and outputting a scheduling scheme of the task planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
Compared with the prior art, the invention has at least the following beneficial effects:
the invention relates to a multi-satellite on-orbit collaborative earth observation task planning method, which aims at multi-satellite collaborative earth observation tasks, improves a genetic algorithm and enables high-weight tasks to be scheduled preferentially by a heuristic initial solution generation strategy of task profit priority; the dominant individual has greater probability inheritance due to the benefit of boltzmann selection probability; the self-adaptive cross rate and the self-adaptive variation rate enable the convergence to be faster when the population diversity is higher, and avoid falling into the local optimum when the population diversity is lower; the convergence stopping condition can make the individuals in the population tend to be consistent, so that the average income is improved, and the algorithm has higher stability; finally, the task planning algorithm is used for improving the scheduling strategy of the traditional cloud computing system, the on-orbit scheduling of the multi-satellite cooperative earth observation task is realized, and the key scheduling support is provided for constructing the constellation cloud computing system.
Further, the chromosome coding is a problem to be solved primarily by using a genetic algorithm, and based on real-time state data of the satellite, the chromosome can establish scheduling queues for satellite nodes with normal flight and good health check, and each scheduling queue stores a target point task scheduled to the satellite node to execute observation. Each task stores detailed scheduling information through a structure body so as to facilitate detailed output of a final scheduling scheme.
Furthermore, a heuristic initial solution generation strategy with task benefit priority is used for sequencing target points to be scheduled according to the task weight, and tasks with high weight are preferentially selected for observation, so that the tasks with high weight can be completed as much as possible, and the overall benefit of the earth observation task planning is ensured to a certain extent.
Further, whether each target point observation task can be successfully scheduled to a certain satellite depends on the following constraints: the target point observation task must be performed within a visible time window; the capacity required by satellite imaging cannot exceed the maximum storage capacity supported by the satellite nodes; two target point observation tasks are executed before and after the same satellite, and the time interval between the tasks is longer than the attitude stabilization time of the remote sensing equipment between the two tasks.
Furthermore, because a certain yaw conversion time is required for executing two tasks on the same satellite in sequence, the compact strategy means that for the observation task of each selected time window, the initial time of the time window should be used as the starting execution time of observation, and the influence of the task observed first on the same satellite on the task observed later is avoided as much as possible.
Furthermore, due to the heuristic solution generation strategy and the fitness calculation method, the fitness values of the individuals in the initial population are not very different, and the traditional roulette algorithm tends to be a general random selection algorithm under the condition, so that the dominant individuals are difficult to highlight; to alleviate this problem, elite individuals are better inherited using boltzmann selection probabilities instead of the original selection probabilities in roulette methods.
Furthermore, the crossover and mutation probabilities of the traditional genetic algorithm are fixed values and set by experience, and the too large and too small crossover and mutation probabilities are not beneficial to population evolution. The self-adaptive crossover operator and the genetic operator can dynamically adjust the crossover and mutation probabilities in the iterative process of the algorithm, so that the genetic algorithm can better balance between global search and local search, the search space is prevented from being reduced, and the convergence rate of the algorithm is accelerated.
Furthermore, the population standard deviation reflects the deviation degree of the fitness value of all individuals in the population and the population fitness mean value, reflects the dispersion degree of the population, and is an important parameter for determining the dynamic adjustment of crossover and variation probability; and when the crossing and mutation probabilities are updated, alpha is used for indicating the updating speed, L indicates the updating size of each time, the crossing and mutation probabilities are reduced when the population standard deviation is larger than the threshold, and the crossing and mutation probabilities are increased when the population standard deviation is smaller than the threshold.
Furthermore, the population convergence coefficient represents the average fitness of a generation of population, and the fact that the average fitness of continuous iteration for several times is within a certain threshold means that the algorithm tends to converge, so that the stability of the algorithm is guaranteed.
In summary, the invention provides an important support for the constellation cloud computing system to automatically cooperate with multiple nodes to execute observation task scheduling on the orbit.
The technical solution of the present invention is further described in detail by the following examples.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be understood that the terms "comprises" and/or "comprising" indicate the presence of the described features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
The invention provides a multi-satellite on-orbit collaborative earth observation task planning method, which improves a genetic algorithm for planning a multi-satellite collaborative earth observation task from various angles such as an initial solution, a selection operator, a crossover operator, a mutation operator and a stopping condition based on a population convergence coefficient by designing a heuristic initial solution generation strategy, a roulette selection algorithm based on Boltzmann selection probability, an adaptive crossover operator and a mutation operator based on a population dispersion degree and a stopping condition based on a population convergence coefficient.
The invention discloses a multi-satellite on-orbit collaborative earth observation task planning method, which comprises the following steps of:
s1, the constellation cloud computing system acquires the geographic position, task priority and task completion time interval information of each target point to be observed, real-time synchronous satellite node state data are acquired by accessing the etcd through the main control node, and a visible time window between each satellite node and the target point to be observed is calculated according to the real-time orbit information of the N satellites;
the constellation cloud computing system is based on a constellation system, a lightweight virtualization technology is utilized, physical separation between single-satellite nodes is broken, all satellite nodes with limited resources are organically integrated in a space resource pooling mode, and through establishment of a task scheduling and resource perception platform facing the constellation system, the automatic processing of the whole process of receiving an observation instruction, performing on-orbit cooperation multi-satellite earth observation, performing on-orbit cooperation multi-satellite target detection and returning a detection result to a ground control station is realized, so that the satellite-earth link transmission bandwidth is saved, the capability of completing tasks in cooperation of a constellation group is enhanced, the perception capability and decision timeliness of the constellation system are improved, the constellation cloud computing system is the brain of on-orbit unified planning and cooperative control of the constellation system, and a solid foundation is laid for space information guarantee in China.
S2, generating a coding result based on the satellite node state data obtained in the step S1 and the geographic position, the task priority and the task completion time interval information of each target point to be observed;
each chromosome in the genetic algorithm represents a feasible solution, each chromosome in the multi-star earth observation represents a feasible scheduling scheme, and each gene in the chromosome represents the observation of a certain ground target point.
If N satellites are operated in the current satellite cloud computing system, N +1 task scheduling queues are created for each chromosome, each scheduling queue corresponds to one satellite, and observation targets arranged to be executed by the satellite are arranged according to the sequence of task execution time. Each observation target in the scheduling queue is represented by a structural body, and the structural body comprises information such as a target point position, a target point priority, specific execution time of the target point, and which satellite the target point is scheduled to observe. And the (N + 1) th scheduling queue is used for storing target points which are not successfully scheduled in the task planning process.
S3, based on the coding result of the step S2 and the visible time window of the step S1, adopting an initial solution generation strategy with priority of task execution profit, planning a feasible observation window on the premise that each target point meets the constraint by taking the task priority as a sequence, and adopting an observation task execution time dynamic adjustment strategy to adjust the task execution time to obtain an initial population;
in order to improve the execution yield of the observation task, an initial solution generation strategy taking the execution yield of the task as guidance is designed, and a time-compaction strategy is adopted for observing more target points.
For an observation task required by a user, firstly, a plurality of target points in the observation task are sorted according to priorities designated by the user, and then an initial solution of a genetic algorithm is generated according to the sorting sequence. And calculating the visible time between each satellite in the constellation cloud computing system and the target point for each target point, and randomly selecting a time window from the visible times to perform observation. It should be noted that when a time window is selected, it is necessary to determine whether a certain satellite scheduled this time meets constraint conditions, including the satellite-borne remote sensing device side-sway time required by the satellite observing the target point during observation, the stabilization time of the remote sensing device, whether the on-satellite storage is sufficient, and the like. If either constraint is not met, an alternative suitable time window may be required for scheduling.
Furthermore, in order to accomplish the observation of the target points as much as possible, a time-compact strategy is employed. Specifically, for the observation of two target points in front and behind in a certain scheduling queue in a certain chromosome, the observation time of the target point is set as the starting time of the selected time window as much as possible. And if the remote sensing equipment side sway is not finished or stabilized when the later-observed target point time window arrives, the observation is executed after the conversion time is finished. And if the conversion time is not completed when the target point time window of the later observation is finished, the target point observation fails.
S4, calculating the fitness value of the initial population obtained in the step S3, selecting individuals from the initial population for heredity by using a roulette selection method based on Boltzmann selection probability, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on orbit circles on the heredity result;
the selection operator commonly used in genetic algorithms is the roulette algorithm, whose main idea is that the probability that a chromosome is inherited to the next generation is proportional to the fitness of that chromosome. If the fitness value of a chromosome is recorded as fiThen the probability that the chromosome is inherited to the next generation is:
where N represents the total number of chromosomes in a generation population.
However, due to the heuristic initial solution generation strategy oriented to the gains in the constellation scene and the calculation method of the fitness, the fitness value of the chromosomes in the initial population is not very different, and in this case, the traditional roulette algorithm tends to be a general random selection algorithm, and the high-quality chromosomes are difficult to highlight.
To alleviate this problem, consider incorporating the boltzmann selection algorithm into the roulette selection algorithm, i.e., replacing the selection probability of roulette with the boltzmann selection probability, which is defined as follows:
where b is used to control the selection strength, the individual selection probability of high fitness is increased when b >0, thereby solving the above-described problem.
Self-adaptive cross and variation probability based on population discrete degree
In the genetic algorithm, a crossover operator determines the global searching capability of the genetic algorithm, and a mutation operator determines the local searching capability of the genetic algorithm. Therefore, crossover operators and mutation operators profoundly influence the trend of genetic algorithms. Too large a cross probability causes the genetic algorithm mechanism to be destroyed, and too small a cross probability reduces the search speed of the algorithm. The genetic algorithm tends to search randomly due to the excessively large mutation probability, and the excessively small mutation probability is easy to fall into local optimum.
Based on a chromosome coding mode, the invention adopts a crossing and variation mode based on orbit cycles. For two chromosomes in a certain generation of population, target points in a certain circle are randomly selected in a scheduling queue corresponding to the same satellite to be crossed. For a single chromosome in a certain generation of population, a target point in a certain circle is randomly selected in a scheduling queue corresponding to the same satellite for mutation. However, the cross probability and the mutation probability in the operator are fixed values and are set by depending on experience, so that the evolution of the population is not facilitated. Therefore, an adaptive crossover and mutation operator based on the population dispersion degree is designed, and the crossover rate and the mutation rate are dynamically changed by measuring the dispersion degree of the contemporary population, so that the genetic algorithm can better balance between global search and local search. The degree of population dispersion is assessed by the standard deviation of the fitness values of all chromosomes in the population, as follows:
and when the standard deviation of the parent population is smaller than a preset threshold value, increasing the probability of crossing and mutation, and when the standard deviation of the parent population is larger than the preset threshold value, reducing the probability of crossing and mutation.
Crossing rate PcAnd the mutation probability PmThe updated calculation formula of (c) is as follows:
Pc`＝c+L
Pm`＝m+L
wherein, PcAnd PmRepresents the crossing rate and the variation rate, P, of the previous generation populationc' and Pm"denotes the population crossing rate and variation rate after updating, α is the update speed control factor, σ is the population fitness standard deviation, and d is the preset threshold value.
In addition, the invention limits the change range of the cross rate to [0.8,0.95], and limits the change range of the change rate to [0.05,0.15 ].
S5, iterating the new generation of population obtained in the step S4 until the end, and outputting a scheduling scheme of the mission planning; and the main control node of the constellation cloud computing system schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to perform earth observation.
The invention judges the stopping time of the iteration of the algorithm through the convergence coefficient of the population, namely, if the convergence coefficient of the population of 5 successive generations is less than or equal to the preset threshold value, the execution of the algorithm is stopped.
The convergence factor of the population is defined as:
wherein,
represents the average fitness of the parent population,
representing the average fitness of the progeny population.
In another embodiment of the present invention, a multi-satellite on-orbit collaborative earth observation task planning system is provided, which can be used for implementing the above multi-satellite on-orbit collaborative earth observation task planning method, and specifically, the multi-satellite on-orbit collaborative earth observation task planning system includes a data module, a coding module, a population module, a calculation module, and a planning module.
The data module acquires the geographic position, task priority and task completion time interval information of each target point to be observed, acquires real-time synchronous satellite node state data by accessing the etcd through the main control node, and calculates a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites;
the encoding module is used for generating an encoding result based on the satellite node state data obtained by the data module and the geographic position, the task priority and the task completion time interval information of each target point to be observed;
the group module adopts an initial solution generation strategy of task execution profit priority based on the coding result and the visible time window, plans a feasible observation window for each target point in sequence by taking the task priority as a sequence on the premise of meeting the constraint, and adopts an observation task execution time dynamic adjustment strategy to adjust the task execution time to obtain an initial group;
the calculation module is used for calculating the fitness value of the initial population obtained by the population module, selecting individuals from the initial population for heredity by using a roulette selection method based on Boltzmann selection probability, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on track circles on the heredity result;
the planning module is used for iterating the new generation of population obtained by the calculation module until the end of the iteration, and outputting a scheduling scheme of the task planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
In yet another embodiment of the present invention, a terminal device is provided that includes a processor and a memory for storing a computer program comprising program instructions, the processor being configured to execute the program instructions stored by the computer storage medium. The Processor may be a Central Processing Unit (CPU), or may be other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable gate array (FPGA) or other Programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., which is a computing core and a control core of the terminal, and is adapted to implement one or more instructions, and is specifically adapted to load and execute one or more instructions to implement a corresponding method flow or a corresponding function; the processor provided by the embodiment of the invention can be used for the operation of the multi-satellite on-orbit cooperative earth observation task planning method, and comprises the following steps:
acquiring the geographic position, task priority and task completion time interval information of each target point to be observed, accessing the etcd through the master control node to acquire real-time synchronous satellite node state data, and calculating a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites; generating a coding result based on the obtained satellite node state data and the geographic position, task priority and task completion time interval information of each target point to be observed; based on the coding result and the visible time window, an initial solution generation strategy of task execution profit priority is adopted, a feasible observation window is planned for each target point in sequence by taking the task priority as a sequence on the premise of meeting the constraint, and the task execution time is adjusted by adopting an observation task execution time dynamic adjustment strategy to obtain an initial population; calculating the fitness value of the initial population, selecting individuals from the initial population by using a roulette selection method based on Boltzmann selection probability to inherit, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on track circles on the genetic result; iterating the new generation of population until the end, and outputting a scheduling scheme of the task planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
In still another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a Memory device in the terminal device and is used for storing programs and data. It is understood that the computer readable storage medium herein may include a built-in storage medium in the terminal device, and may also include an extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also, one or more instructions, which may be one or more computer programs (including program code), are stored in the memory space and are adapted to be loaded and executed by the processor. It should be noted that the computer readable storage medium may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory.
The processor can load and execute one or more instructions stored in the computer readable storage medium to realize the corresponding steps of the multi-satellite on-orbit collaborative earth observation task planning method in the embodiment; one or more instructions in the computer-readable storage medium are loaded by the processor and perform the steps of:
acquiring the geographic position, task priority and task completion time interval information of each target point to be observed, accessing the etcd through the master control node to acquire real-time synchronous satellite node state data, and calculating a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites; generating a coding result based on the obtained satellite node state data and the geographic position, task priority and task completion time interval information of each target point to be observed; based on the coding result and the visible time window, an initial solution generation strategy of task execution profit priority is adopted, a feasible observation window is planned for each target point in sequence by taking the task priority as a sequence on the premise of meeting the constraint, and the task execution time is adjusted by adopting an observation task execution time dynamic adjustment strategy to obtain an initial population; calculating the fitness value of the initial population, selecting individuals from the initial population by using a roulette selection method based on Boltzmann selection probability to inherit, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on track circles on the genetic result; iterating the new generation of population until the end, and outputting a scheduling scheme of the task planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
The invention is based on a constellation cloud computing system to develop simulation tests. The low-orbit observation satellite in the satellite cloud computing system takes remote sensing No. 6, remote sensing No. 11, remote sensing No. 13 and remote sensing No. 16-A emitted by China as examples, the initial orbit parameters of each satellite are shown in table 1, and the performance setting of the satellite-borne observation equipment on each satellite is shown in table 2.
TABLE 1 orbit six numbers of each low orbit observation node in the Star cloud computing System
TABLE 2 Observation Performance of each Low-Earth Observation satellite
To enhance the realism of the experimental example, the example sets the observation period of the low orbit satellite: 2021/3/184:00: 00-2021/3/194: 00:00, and selecting the city of the first 100 ranking in the 2020 Kornia city index report for observation, wherein the priority weight of each target point is selected within the range of [1,50], and higher priority indicates higher observation yield of the target point. In addition, the software STK widely used in the aerospace field is adopted to obtain the visibility data between the low-orbit observation satellite and the ground observation target point so as to avoid complex coordinate transformation and dynamics calculation.
The performance of the algorithm is tested by setting 50, 75 and 100 observation tasks of 3 target points with different scales to form 4 groups of arithmetic examples, each arithmetic example is operated 20 times, the optimal benefit of the operation result, the number of targets for completing observation, the average benefit of the tasks and the execution time of the algorithm are recorded, and the average value of each record is taken as the final experiment result.
Based on the test scene and the experimental parameter setting, the improved genetic algorithm IGA provided by the invention is compared with the genetic algorithm GA which is crossed and mutated based on the circle, and the effectiveness of the method is verified. The IGA algorithm parameter settings are shown in Table 3, the GA algorithm parameter settings are shown in Table 4, and the quality comparison of the two methods is shown in Table 5.
TABLE 3 IGA Algorithm parameter set
TABLE 4 GA Algorithm parameter set
| Parameter(s) | Value taking |
| Number of population | 50 |
| Probability of crossing | 0.9 |
| Probability of variation | 0.1 |
| Maximum number of iterations | 400 |
TABLE 5 comparison of the results of the genetic Algorithm with the improved genetic Algorithm
As can be seen from the table, the improved genetic algorithm of the invention achieves more excellent results in a plurality of test examples than the genetic algorithm based on circle-based cross variation, and compared with the latter, the overall average benefit of the observation task is improved by 2.58%, the overall best benefit is improved by 1.92%, and the number of successfully observed target points is improved by 4.84%. The method benefits from Boltzmann selection probability, and the dominant individual has higher probability to inherit, thereby improving observation income. The convergence stopping condition can make the individuals in the population tend to be consistent, so that the average income is improved, and the algorithm has higher stability. And the cross rate and the variation rate are dynamically adjusted, so that the convergence is faster when the population diversity is higher, and the local optimum is avoided when the population diversity is lower. Meanwhile, the operation time of the algorithm is increased correspondingly by the strategies, and the operation time is still within an acceptable range
In summary, the multi-satellite on-orbit collaborative earth observation task planning method and system provided by the invention are improved aiming at heredity based on a multi-satellite collaborative observation scene, experiments prove that the method improves the overall income of observation tasks and improves the completion number of tasks to be observed, and the method is taken as a core scheduling strategy to improve the traditional distributed cloud computing system, so that the on-orbit scheduling of the multi-satellite on-earth observation tasks is realized, important support is provided for the construction of a constellation cloud computing system, and the method and system have positive engineering benefits and strategic significance.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.
Claims (10)
1. A multi-satellite on-orbit collaborative earth observation task planning method is characterized by comprising the following steps:
s1, acquiring the geographic position, task priority and task completion time interval information of each target point to be observed, accessing the etcd through the main control node to acquire real-time synchronous satellite node state data, and calculating a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites;
s2, generating a coding result based on the satellite node state data obtained in the step S1 and the geographic position, the task priority and the task completion time interval information of each target point to be observed;
s3, based on the coding result of the step S2 and the visible time window of the step S1, adopting an initial solution generation strategy with priority of task execution profit, planning a feasible observation window on the premise that each target point meets the constraint by taking the task priority as a sequence, and adopting an observation task execution time dynamic adjustment strategy to adjust the task execution time to obtain an initial population;
s4, calculating the fitness value of the initial population obtained in the step S3, selecting individuals from the initial population for heredity by using a roulette selection method based on Boltzmann selection probability, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on orbit circles on the heredity result;
s5, iterating the new generation of population obtained in the step S4 until the end, and outputting a scheduling scheme of the mission planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
2. The method for planning the multi-satellite on-orbit collaborative earth observation task according to claim 1, wherein the step S2 specifically comprises:
firstly, carrying out chromosome coding, wherein each chromosome represents a feasible scheduling scheme, each gene in the chromosome represents scheduling of an observation target point, N +1 task scheduling queues are created for each chromosome aiming at N satellites flying in orbit, each scheduling queue corresponds to one satellite, each target point to be observed in the scheduling queues is represented by a structural body, and the N +1 scheduling queue is used for storing the target point which is not successfully scheduled in the task scheduling process to obtain a coding result.
3. The method for planning a multi-satellite on-orbit collaborative earth observation task according to claim 1, wherein in step S3, the initial solution generation strategy adopting task execution yield priority specifically comprises:
for an observation task required by a primary user, firstly, a plurality of target points in the observation task are sorted according to the priority specified by the user, and then an initial solution of an algorithm is generated according to the sorting sequence; for each target point, acquiring a visible time window between each satellite and the corresponding target point, and randomly selecting one time window to perform observation; when a time window is selected, whether the task meets constraint conditions after being scheduled to a satellite corresponding to the time window is judged, and if any constraint condition is not met, the next time window is selected for scheduling; and if all the selectable time windows do not meet the constraint condition, the corresponding target point fails to be scheduled.
4. The method for planning the multi-satellite in-orbit collaborative earth observation task according to claim 3, wherein the constraint conditions comprise the required yaw time of the satellite-borne remote sensing equipment during observation, the stabilization time of the remote sensing equipment and whether the on-satellite storage is sufficient.
5. The method for planning the multi-satellite on-orbit cooperative earth observation task according to claim 1, wherein in step S3, the dynamic adjustment strategy for the execution time of the observation task specifically comprises:
for the observation tasks of the front target point and the rear target point which are scheduled to the same satellite node, setting the observation execution time of the front target point as the starting time of the selected time window, and when the time window of the second target point arrives, if the sidewise swinging of the satellite-borne remote sensing equipment is not finished or the camera is not stable, waiting for the conversion time to be finished and then executing the observation; and if the conversion activity is not completed when the second target point selection time window is finished, the target point observation fails.
6. The method for planning a multi-satellite in-orbit collaborative earth observation mission according to claim 1, wherein in step S4, individuals are selected from the initial population for inheritance using a roulette selection method using boltzmann selection probability p (i) as follows:
wherein f isiIs the fitness value of individual i in the population, b is used to control the selection strength, and N represents the total number of chromosomes in the generation population.
7. The method for planning a multi-satellite on-orbit collaborative earth observation task according to claim 1, wherein in step S4, the adaptive intersection and variation mode based on orbit turns is specifically:
evaluating the dispersion degree of the population through the standard deviation of the fitness values of all individuals in the population by measuring the dispersion degree of the contemporary population and dynamically changing the cross rate and the variation rate; and when the standard deviation sigma of the parent population is greater than the preset threshold value, the probability of intersection and mutation is reduced.
8. The multi-satellite on-orbit collaborative earth observation task planning method according to claim 7, wherein a parent population standard deviation σ is as follows:
wherein N is the total number of individuals in the population, fiIs the fitness value of the individual i in the population,
the average fitness value of all individuals in a generation population;
crossing rate PcAnd probability of mutation PmThe adaptive update of (1) is as follows:
wherein, PcAnd PmRepresents the crossing rate and the variation rate, P, of the previous generation populationc"and Pm"represents the population crossing rate and the variation rate after updating, alpha is the control factor of the updating speed, sigma is the standard deviation of the population fitness, and d is a preset threshold value.
9. The method for planning a multi-satellite on-orbit collaborative earth observation task according to claim 1, wherein in step S5, a stopping condition is determined according to a convergence coefficient of a population, iteration is stopped when the convergence coefficient of the population of 5 consecutive generations is less than or equal to a preset threshold, and the convergence coefficient e of the population is evaluated according to an average fitness of the population as follows:
wherein,
the average fitness of the parent population is represented,
representing the average fitness of the progeny population.
10. A multi-satellite on-orbit collaborative earth observation task planning system is characterized by comprising:
the data module is used for acquiring the geographic position, the task priority and the task completion time interval information of each target point to be observed, acquiring real-time synchronous satellite node state data by accessing the etcd through the main control node, and calculating a visible time window between each satellite node and the target point to be observed according to the real-time orbit information of the N satellites;
the encoding module is used for generating an encoding result based on the satellite node state data obtained by the data module and the geographic position, the task priority and the task completion time interval information of each target point to be observed;
the group module adopts an initial solution generation strategy of task execution profit priority based on the coding result and the visible time window, plans a feasible observation window for each target point in sequence by taking the task priority as a sequence on the premise of meeting the constraint, and adopts an observation task execution time dynamic adjustment strategy to adjust the task execution time to obtain an initial group;
the calculation module is used for calculating the fitness value of the initial population obtained by the population module, selecting individuals from the initial population for heredity by using a roulette selection method based on Boltzmann selection probability, and obtaining a new generation of population by adopting a self-adaptive intersection and variation mode based on track circles on the heredity result;
the planning module is used for iterating the new generation of population obtained by the calculation module until the end of the iteration, and outputting a scheduling scheme of the task planning; and the main control node schedules the observation task of the target point to the corresponding satellite node through the inter-satellite link according to the scheduling scheme to observe the earth.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111370416.9A CN114091892A (en) | 2021-11-18 | 2021-11-18 | Multi-satellite on-orbit collaborative earth observation task planning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111370416.9A CN114091892A (en) | 2021-11-18 | 2021-11-18 | Multi-satellite on-orbit collaborative earth observation task planning method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091892A true CN114091892A (en) | 2022-02-25 |
Family
ID=80301867
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111370416.9A Pending CN114091892A (en) | 2021-11-18 | 2021-11-18 | Multi-satellite on-orbit collaborative earth observation task planning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091892A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115686874A (en) * | 2023-01-03 | 2023-02-03 | 中南大学 | Dynamic inter-satellite multi-satellite cooperative computing method, system, equipment and medium |
| CN116882142A (en) * | 2023-06-27 | 2023-10-13 | 中国空间技术研究院 | Method, equipment and medium for earth observation multi-level planning strategy based on loose coupling |
| CN118153671A (en) * | 2024-05-09 | 2024-06-07 | 中国人民解放军国防科技大学 | Fixed station satellite observation method, system, equipment and medium based on genetic algorithm |
-
2021
- 2021-11-18 CN CN202111370416.9A patent/CN114091892A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115686874A (en) * | 2023-01-03 | 2023-02-03 | 中南大学 | Dynamic inter-satellite multi-satellite cooperative computing method, system, equipment and medium |
| CN116882142A (en) * | 2023-06-27 | 2023-10-13 | 中国空间技术研究院 | Method, equipment and medium for earth observation multi-level planning strategy based on loose coupling |
| CN116882142B (en) * | 2023-06-27 | 2024-03-05 | 中国空间技术研究院 | Method, equipment and medium for earth observation multi-level planning strategy based on loose coupling |
| CN118153671A (en) * | 2024-05-09 | 2024-06-07 | 中国人民解放军国防科技大学 | Fixed station satellite observation method, system, equipment and medium based on genetic algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114091892A (en) | Multi-satellite on-orbit collaborative earth observation task planning method and system | |
| CN108388958B (en) | Method and device for researching two-dimensional attitude maneuvering satellite mission planning technology | |
| CN107679748B (en) | Star-ground combined operation method for autonomous planning of constellation observation task | |
| CN112001585B (en) | Multi-agent decision method, device, electronic equipment and storage medium | |
| CN109933842A (en) | A kind of mobile target list star mission planning method based on constraint satisfaction genetic algorithm | |
| CN111664852B (en) | Unmanned aerial vehicle path planning method and device | |
| CN111913787A (en) | Imaging satellite scheduling method and system based on genetic algorithm | |
| WO2019127945A1 (en) | Structured neural network-based imaging task schedulability prediction method | |
| CN110515708A (en) | The in-orbit autonomous mission planning method and system of satellite | |
| CN105512479A (en) | Evaluation method and device of multi-satellite earth observation task planning algorithms | |
| CN112434435B (en) | Imaging satellite intensive task scheduling method based on task synthesis | |
| CN109918731B (en) | Satellite task planning simulation analysis method and system based on critical path | |
| CN111861230A (en) | Multi-satellite cooperative observation task planning method and device for disaster emergency | |
| CN115481779A (en) | Satellite resource scheduling optimization method based on federal reinforcement learning | |
| CN114625506A (en) | Edge cloud collaborative task unloading method based on adaptive covariance matrix evolution strategy | |
| CN117455193A (en) | Task allocation method and device | |
| CN111157002B (en) | Aircraft 3D path planning method based on multi-agent evolutionary algorithm | |
| CN116634498A (en) | Low orbit satellite constellation network edge calculation multistage unloading method based on reinforcement learning | |
| CN113657718B (en) | Multi-robot dynamic alliance task allocation method and related device | |
| CN114815891A (en) | PER-IDQN-based multi-unmanned aerial vehicle enclosure capture tactical method | |
| CN117008641B (en) | Distribution method and device for cooperative low-altitude burst prevention of multiple heterogeneous unmanned aerial vehicles | |
| CN114371925A (en) | Edge cloud cooperative multi-task scheduling method and system for guaranteeing edge cloud load ratio | |
| CN114371634A (en) | Unmanned aerial vehicle combat analog simulation method based on multi-stage after experience playback | |
| CN117806346A (en) | Unmanned plane path planning method based on self-adaptive dung beetle optimizer | |
| CN116578354A (en) | Method and device for unloading edge calculation tasks of electric power inspection unmanned aerial vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |