A heterogeneous space platform cooperative earth observation and data transmission scheduling method

Abstract

This invention discloses a method for collaborative Earth observation and data transmission scheduling across heterogeneous space platforms. The method employs integer encoding to represent the execution sequence of observation tasks from multiple space platforms, proposes a three-stage decoding strategy to repair infeasible chromosomes, and designs feasible observation time window selection strategies, as well as observation task insertion strategies based on time-based benefits, task urgency, and attitude transition time, to construct an initial chromosome population. Crossover operations are performed on the current population, constructing three neighborhood search operators: inter-platform exchange, intra-platform exchange, and insertion based on relaxation time, to perform local searches on the chromosomes within the current population. The algorithm terminates if the maximum number of iterations or the maximum computation time limit is reached, generating a scheme for collaborative Earth observation and data transmission scheduling across heterogeneous space platforms, providing technical support for space-based Earth observation.

Description

Technical Field

[0001] This invention relates to the field of space-based Earth observation technology, and in particular to a method for collaborative Earth observation and data transmission scheduling of heterogeneous space platforms based on a cultural gene algorithm. Background Technology

[0002] Space-based Earth observation plays a crucial role in earthquake relief and scientific research. Remote sensing data is acquired through various space platforms and transmitted back to Earth for processing, providing information to users. However, the observation efficiency of a single type of space platform is limited, making it difficult to meet the increasingly complex needs of observation missions. Therefore, integrated space-air collaborative Earth observation, which effectively coordinates multiple types of space platforms to achieve collaborative work among them, thereby improving the utilization efficiency of space platforms and the overall observation benefits, has become a development trend in aerospace technology. However, existing space platforms differ significantly in operating altitude, motion characteristics, and working methods, greatly increasing the complexity of mission allocation. Therefore, it is difficult to coordinate heterogeneous space platforms for Earth observation while simultaneously considering solution efficiency and solution quality, thus affecting the effectiveness of space-based Earth observation. Summary of the Invention

[0003] This invention provides a method for collaborative Earth observation and data transmission scheduling of heterogeneous space platforms based on the cultural gene algorithm, in order to solve the existing observation problems of collaborative heterogeneous space platforms.

[0004] This invention provides a method for scheduling collaborative Earth observation and data transmission among heterogeneous space platforms based on a cultural gene algorithm. The method includes: using integer encoding for chromosomes, where each chromosome represents the execution sequence of observation tasks for each space platform completing the current Earth observation task; judging the feasibility of the chromosomes through a preset feasibility judgment strategy; and repairing infeasible chromosomes to make them feasible. The feasibility judgment strategy includes feasibility of observation task time windows, data storage feasibility, and feasibility of executing data transmission tasks.

[0005] An observation task insertion selection strategy based on time gain, task urgency, and attitude transition time determines the initial population of chromosomes among feasible chromosomes.

[0006] The fitness of chromosomes in the initial population is evaluated, and the total benefit obtained by each space platform in completing its observation task and its corresponding data transmission task is evaluated.

[0007] A crossover operation is performed on the population, along with local search and adaptive neighborhood selection, to obtain the current solution;

[0008] The algorithm repeatedly evaluates the fitness of the current population, performs crossover operations and local searches. If the maximum number of iterations or the maximum computation time limit is reached, the algorithm terminates, outputs the best solution, and uses this solution to coordinate with heterogeneous space platforms to complete Earth observation.

[0009] Optionally, the chromosome uses integer encoding to represent the execution sequence of observation tasks across multiple space platforms. The numbers represent observation task numbers, and the observation task sequences executed by each space platform are separated by 0. Different colors correspond to observation task sequences for different types of space platforms, and each chromosome contains observation task sequences for all space platforms. For a scenario with two AEOSs, one airship, and one UAV, the chromosome 0-9-11-5-0-7-4-2-0-12-8-1-0-3-6-0 indicates that the observation task execution sequences for the first and second AEOS are 9-11-5 and 7-4-2, respectively. The observation task execution sequences for the airship and the UAV are 12-8-1 and 3-6, respectively.

[0010] Optionally, when the space platform includes an Agile Earth Observation Satellite (AEOS), an airship, and an unmanned aerial vehicle (UAV), the feasibility of the chromosome is determined based on the feasibility of the observation mission time window, including:

[0011] The intersection of the observation time requirement of the observation task and the visible time window of AEOS for that task is defined as the feasible time window of AEOS for that task. The start time of the feasible time window of AEOS for that task is taken as the start time of the first observation task in the observation task sequence. The start time of each observation task is calculated sequentially. The next observation task... start time The calculation formula is:

[0012]

[0013] in, This indicates that AEOS is for observation tasks. The start time of the feasible time window. Indicates the previous observation task The start time, Indicates observation task The duration of observation, This indicates that AEOS is in the observation mission. and observation mission The attitude transition time between observation tasks. If the start time of an observation task exceeds the end time of the feasible time window for that observation task by AEOS, the observation task is removed from the observation task sequence.

[0014] For the first observation task in the observation task sequence of airships and UAVs, the airship base station or UAV base station closest to the observation task is selected as the starting point of the airship or UAV. The cruise time from the starting point to the first observation task is calculated. If the cruise time exceeds the time requirement of the task, the task is removed from the observation task sequence. Otherwise, the cruise time from the execution position of the first observation task to the execution position of the next observation task is calculated, and it is determined whether the time requirement of the next observation task is violated. This process is repeated until all observation tasks in the observation task sequence meet the time requirements of the observation tasks.

[0015] Optionally, the feasibility of the chromosome is determined based on data storage feasibility, including: if the total data storage capacity occupied by the observation tasks in the observation task sequence exceeds the maximum data storage capacity of the space platform, then one or more observation tasks are removed until the data storage capacity is met.

[0016] Optionally, the feasibility of the chromosome is determined based on the feasibility of the data transmission task, including:

[0017] The AEOS's visible time window for the ground station is set as the feasible data transmission time window after deleting the time period occupied by the observation task. According to the priority of the corresponding observation task, the data transmission task is inserted into the feasible data transmission time window with the earliest start time. This continues until the remaining feasible data transmission time window is insufficient to execute the next data transmission task. All remaining data transmission tasks are traversed, and data transmission tasks are inserted in the remaining time of the feasible data transmission time window according to the priority of the corresponding observation task, until the remaining feasible data transmission time is insufficient.

[0018] Within the airship's coverage area of ​​the ground station, data transmission tasks are inserted into the feasible observation task sequence according to the priority of the corresponding observation tasks, and the observation task sequence is repaired to ensure feasibility. If the data storage capacity exceeds the airship's maximum data storage capacity after inserting the data transmission task, one or more observation tasks are moved after the insertion point until the data storage capacity no longer exceeds the limit. If there is insufficient remaining time to execute the data transmission task, one or more observation tasks are removed before the insertion point.

[0019] When the drone reaches its maximum data storage capacity, it returns to base to perform data transmission tasks.

[0020] Optionally, the fitness of chromosomes in the population is evaluated. The fitness function is defined as the sum of the total rewards obtained from observation tasks and their corresponding data transmission tasks completed by satellites, aerostats, and UAVs, respectively. Optionally, feasible observation time window selection strategies, time-based rewards, and observation task insertion strategies based on task urgency and attitude transition time are used to construct the initial population of chromosomes, including:

[0021] Based on a feasible observation time window selection strategy, a feasible time window is assigned to each observation task to reduce time conflicts between observation tasks. A feasible time window selection strategy is proposed, which selects a feasible time window from the set of feasible time windows for all observation tasks to reduce conflicts with other observation tasks. An observation task set is defined and initialized at the beginning of the scheduling cycle. Feasible time windows that do not conflict with other observation tasks are identified and used as the feasible time window for each observation task, and the observed task is removed. If an observation task has multiple feasible time windows that do not conflict with other observation tasks, a feasible time window is randomly selected. The earliest observation time window among all observation tasks is selected as the feasible time window for its corresponding observation task. The latest observation time window among all observation tasks is selected as the time window for its corresponding observation task. If a feasible time window for an observation task is selected, the observation task is removed. This process is repeated until all observation tasks are assigned a feasible time window.

[0022] Optionally, based on time-dependent rewards, task urgency, and attitude transition time, a feasible observation task sequence is generated for each space platform to obtain the initial population of the chromosome, including:

[0023] At the start of the scheduling cycle, the set of unexecuted observation tasks is initialized. Under the constraints of AEOS payload type and resolution, a feasible observation task is randomly selected as the first AEOS observation task. The lower bound of the selected feasible time window is used as the start time for AEOS to execute the observation task. For each feasible subsequent observation task, AEOS is calculated for two consecutive observation tasks. and Costs between: ,in Indicates linear normalization, It is the reciprocal of the time-dependent reward for the observation task. This indicates the time AEOS takes to perform the observation task. This is the latest start time for AEOS to execute an observation task, used to measure the urgency of the observation task. This indicates that AEOS is performing two consecutive observation tasks. and Given the attitude transition time between observations, select the observation task with the lowest cost, denoted as . Repeat the calculation of AEOS in two consecutive observation tasks until there are no more feasible observation tasks.

[0024] After completing the AEOS observation task allocation, any unexecuted observation tasks are assigned to the aerostat to generate a feasible sequence of aerostat observation tasks, consisting of two consecutive observation tasks executed by the aerostat. and The costs involved also consist of three parts: , It is the reciprocal of the gains from the observation missions performed by the aerostat. This is the latest start time for the aerostat to perform its observation mission, used to indicate the urgency of the mission. It is the cruise time between two consecutive observation missions performed by an airship at two different geographical locations;

[0025] After the aerostat's task assignment is completed, any unexecuted observation tasks are assigned to UAVs to generate a feasible sequence of UAV observation tasks. The cost of the UAVs... Two consecutive observation missions were carried out and The costs between them are calculated in the same way.

[0026] Optionally, a crossover operation is performed on the initial population, including: selecting two crossover points from a sequence of observation tasks executed by the same space platform; after crossover, repairing the two offspring chromosomes to ensure their feasibility. This includes: deleting duplicate observation tasks from the offspring chromosomes; inserting observation tasks into the offspring chromosomes to increase their fitness values; constructing an observation task set from the deleted observation tasks in the previous step; adding the highest-priority observation task that has not been executed by any space platform to the offspring chromosomes; inserting each observation task between any two adjacent observation tasks in the repaired offspring chromosomes and calculating their fitness values; if the chromosome is infeasible after insertion, the fitness value is negative; inserting the observation task with a 1 / 2 probability into the position with the maximum fitness value, or randomly inserting an observation task into a position with a positive fitness value; then deleting the observation task from the observation task set; repeating the insertion operation until all observation tasks in the insertion set have no feasible insertion positions.

[0027] Optionally, the neighborhood search operator includes an inter-platform exchange operator, an intra-platform exchange operator, and a relaxation-time-based insertion operator. The inter-platform exchange operator selects two observation tasks from the observation task sequences of two random space platforms, determines whether the two observation tasks can be executed by the exchanged space platform, and if not, selects a new observation task to ensure the feasibility of the chromosome after the observation task exchange. The intra-platform exchange operator randomly selects a space platform and randomly swaps the positions of two observation tasks in the observation task sequence of that platform. The relaxation-time-based insertion operator randomly selects a space platform, calculates the relaxation time of the observation task sequence for that platform (i.e., the interval between the current execution time and the latest execution time of an observation task), and randomly selects an unassigned observation task to insert into the relaxation time of the observation task sequence.

[0028] The process of improving the chromosome using a preset neighborhood search operator, and performing local search and adaptive neighborhood selection to obtain the current optimal solution includes:

[0029] The inter-platform exchange operator, the intra-platform exchange operator, and the relaxation-time-based insertion operator are executed with equal probability. If a neighborhood search operator generates a better solution, the probability of executing that neighborhood search operator is increased; otherwise, the probability is decreased, and a local search operator is used for local searching. The probability is, , Represents search operators The quality, through calculation An adaptive neighborhood selection is performed to obtain the current optimal solution, where... .

[0030] In a second aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements any of the methods described above.

[0031] The beneficial effects of this invention are as follows:

[0032] This invention employs integer encoding to represent the execution sequence of observation tasks across multiple space platforms. It proposes a three-stage decoding strategy to repair infeasible chromosomes and designs feasible observation time window selection strategies, as well as observation task insertion strategies based on time-based benefits, task urgency, and attitude transition time, constructing an initial chromosome population. Crossover operations are performed on the current population, constructing three neighborhood search operators: inter-platform exchange, intra-platform exchange, and relaxation-time-based insertion, to perform local searches on the chromosomes within the current population. The algorithm terminates if the maximum number of iterations or the maximum computation time limit is reached, generating a scheme for collaborative Earth observation and data transmission scheduling across heterogeneous space platforms, providing technical support for space-based Earth observation.

[0033] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0034] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0035] Figure 1 This is a schematic diagram of the collaborative Earth observation framework for a spatial platform provided in an embodiment of the present invention;

[0036] Figure 2 This is a schematic diagram of a method for collaborative Earth observation and data transmission scheduling of heterogeneous space platforms based on a cultural gene algorithm, provided in an embodiment of the present invention.

[0037] Figure 3 This is a schematic diagram of chromosome encoding provided in an embodiment of the present invention; Detailed Implementation

[0038] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of the invention.

[0039] Existing methods for solving AEOS scheduling problems using heuristic / metaheuristic algorithms still have limitations when addressing the problem of collaborative Earth observation and data transmission scheduling among heterogeneous space platforms. These limitations include difficulty in effectively balancing computational efficiency and solution quality, and the potential for a sharp increase in computational complexity when pursuing high-quality solutions. This invention provides a method for collaborative Earth observation and data transmission scheduling among heterogeneous space platforms based on the cultural gene algorithm. (See [link to relevant documentation]). Figure 1 and Figure 2 The method includes:

[0040] S101. Identify the possible chromosomes for completing the current Earth observation mission;

[0041] Specifically, this invention decomposes the current Earth observation mission into an execution sequence of observation tasks that each space platform can perform. This execution sequence is named a chromosome. The feasibility of each plan and its final benefits are then determined based on this chromosome, allowing for a simple and accurate evaluation of the merits of each plan. In other words, this invention uses chromosomes to represent the execution sequence of observation tasks for each space platform completing the current Earth observation mission. A preset feasibility determination strategy is then used to determine the feasibility of the chromosomes, and infeasible chromosomes are modified to be feasible. The feasibility determination strategy in this invention includes the feasibility of the observation task time window, the feasibility of data storage, and the feasibility of executing data transmission tasks.

[0042] In specific implementation, the chromosomes in this embodiment of the invention use integer encoding, with numbers representing observation mission numbers. Preset identifiers are used to separate observation mission sequences executed by different space platforms, and different colors are used to identify observation mission sequences of different types of space platforms. The aforementioned preset identifier can be a number, such as "0," or any other character. Those skilled in the art can set it according to actual needs, and this invention does not impose specific limitations on it.

[0043] See Figure 3 As an example of a chromosome in an embodiment of the present invention, through Figure 3 As can be seen, the chromosomes in this embodiment of the invention use integer encoding to represent the execution sequence of observation tasks across multiple space platforms. The numbers represent the observation task numbers, and the observation task sequences executed by each space platform are separated by 0. Different colors correspond to observation task sequences of different types of space platforms, and each chromosome contains observation task sequences from all space platforms. For example... Figure 3As shown, for a scenario with two AEOS satellites, one aerostat, and one drone, chromosome 0-9-11-5-0-7-4-2-0-12-8-1-0-3-6-0 indicates that the observation mission execution sequences for the first and second AEOS satellites are 9-11-5 and 7-4-2, respectively. The observation mission execution sequences for the aerostat and drone are 12-8-1 and 3-6, respectively.

[0044] It should be noted that the heterogeneous space platform in the embodiments of the present invention includes various space platforms such as AEOS, stratospheric airships, and UAVs. Those skilled in the art can make any configuration according to actual needs, and the present invention does not impose any specific limitations on this.

[0045] See Figure 1 The ground support resources for the heterogeneous space platform collaborative Earth observation framework in this embodiment of the invention mainly include AEOS ground stations, airship ground stations, UAV bases, operation control centers, and data processing centers, among which...

[0046] AEOS provides long-term observation capabilities, offering a visible time window for missions and enabling continuous, highly stable tracking and situational awareness of wide-area targets. Aerial vehicles provide area enhancement capabilities, with long cruise cycles, allowing them to hover at target locations for long-term fixed-point surveillance, but their cruise speed is relatively slow. Unmanned aerial vehicles (UAVs) provide emergency response capabilities, enabling rapid acquisition of specific target locations within areas of focus, but their high operating costs typically limit their use to emergency missions. The operations control center is responsible for allocating observation and data transmission tasks to the space platform and ground stations. Observational data is processed by the data processing center to meet user requirements.

[0047] When the space platform includes AEOS, airships, and UAVs, the feasibility of the chromosome is determined based on the feasibility of the observation mission time window, including:

[0048] The intersection of the observation time requirement of the observation task in the chromosome and the visible time window of AEOS for the observation task is calculated as the feasible time window of AEOS for the observation task. The start time of the feasible time window is taken as the start time of the first observation task in the observation task sequence. The start time of each observation task is calculated in turn. If the start time of an observation task exceeds the end time of the feasible time window of AEOS for the observation task, the observation task is determined to be not feasible in the observation task time window and is removed from the observation task sequence.

[0049] For the first observation task in the observation task sequence of airships and UAVs, the airship base station or UAV base station closest to the observation task is selected as the starting point of the airship or UAV. The cruise time from the starting point to the first observation task is calculated. If the cruise time exceeds the time requirement of the task, the task is removed from the observation task sequence. Otherwise, the cruise time from the execution position of the first observation task to the execution position of the next observation task is calculated, and it is determined whether the time requirement of the next observation task is violated. This process is repeated until all observation tasks in the observation task sequence meet the time requirements of the observation tasks.

[0050] In other words, the embodiments of the present invention provide a three-stage decoding strategy to repair infeasible chromosomes into feasible solutions, specifically including:

[0051] The first phase involves assessing the feasibility of the space platform within the observation mission's time window:

[0052] (1) Calculate the intersection of the observation time requirement of the observation task and the visible time window of AEOS for that task, and define it as the feasible time window of AEOS for that task. Take the start time of the feasible time window of AEOS for that task as the start time of the first observation task in the observation task sequence. Calculate the start time of each observation task in turn. The next observation task... start time for: ,in, This indicates that AEOS is for observation tasks. The start time of the feasible time window. Indicates the previous observation task The start time, Indicates observation task The duration of observation, This indicates that AEOS is in the observation mission. and observation mission The attitude transition time between observation tasks. If the start time of an observation task exceeds the end time of the feasible time window for that observation task by AEOS, the observation task is removed from the observation task sequence.

[0053] (2) For the first observation task in the observation task sequence of the airship (UAV), select the airship base station (UAV base station) closest to the task as the starting point of the airship (UAV), calculate the cruise time from the starting point to the first observation task. If the cruise time exceeds the time requirement of the task, remove the task from the observation task sequence. Otherwise, calculate the cruise time from the execution position of the first observation task to the execution position of the next observation task, and determine whether the time requirement of the next observation task is violated. This process continues until all observation tasks in the observation task sequence of the space platform meet the time requirements of the observation tasks.

[0054] The second phase involves assessing the feasibility of data storage for the AEOS or the aerostat. If the total data storage capacity occupied by the observation tasks in the observation task sequence exceeds the maximum data storage capacity of the AEOS or the aerostat, one or more observation tasks need to be removed until the data storage capacity is met.

[0055] The third stage involves assessing the feasibility of executing the data transmission task.

[0056] (1) Define the visible time window of AEOS to the ground station, excluding the time period occupied by the observation task, as the feasible data transmission time window. According to the priority of the corresponding observation task, insert the data transmission task into the feasible data transmission time window with the earliest start time, until the remaining feasible data transmission time window is insufficient to execute the next data transmission task. Iterate through all remaining data transmission tasks, and according to the priority of the corresponding observation task, insert the data transmission task within the remaining time of the feasible data transmission time window, until the remaining feasible data transmission time is insufficient.

[0057] (2) Within the airship's coverage area of ​​the ground station, insert data transmission tasks into the feasible observation task sequence according to the priority of the corresponding observation tasks, and repair the observation task sequence to ensure feasibility. If the data storage capacity exceeds the airship's maximum data storage capacity after inserting a data transmission task, one or more observation tasks need to be moved after the insertion point until the data storage capacity no longer exceeds the limit. If there is insufficient remaining time to execute the data transmission task, one or more observation tasks need to be removed before the insertion point.

[0058] (3) When the UAV reaches its maximum data storage capacity, it returns to the base to perform data transmission tasks.

[0059] S102. An observation task insertion selection strategy based on time gain, task urgency and attitude transition time determines the initial population of chromosomes among feasible chromosomes.

[0060] Specifically, the embodiments of the present invention are based on a feasible observation time window selection strategy, which selects a feasible time window from the set of feasible time windows for all observation tasks, assigns a feasible time window to each observation task, and generates a feasible observation task sequence for each space platform based on time-dependent rewards, task urgency, and attitude transition time, thereby obtaining the initial population of the chromosome.

[0061] In simple terms, the embodiments of the present invention perform screening based on several conditions, including feasible observation time windows, time-dependent rewards, task urgency, and attitude transition time, and then form the initial population with the feasible chromosomes obtained from the screening.

[0062] In specific implementation, this embodiment of the invention initializes the set of unexecuted observation tasks at the beginning of the scheduling period. Under the constraints of AEOS payload type and resolution, a feasible observation task is randomly selected as the first AEOS observation task. The lower bound of the selected feasible time window is used as the start time for AEOS to execute the observation task. For each feasible subsequent observation task, AEOS is calculated between two consecutive observation tasks. and Costs between: ,in Indicates linear normalization, It is the reciprocal of the time-dependent reward for the observation task. This indicates the time AEOS takes to perform the observation task. This is the latest start time for AEOS to execute an observation task, used to measure the urgency of the observation task. This indicates that AEOS is performing two consecutive observation tasks. and Given the attitude transition time between observations, select the observation task with the lowest cost, denoted as . Repeat the calculation of AEOS in two consecutive observation tasks until there are no more feasible observation tasks.

[0063] After completing the AEOS observation task allocation, any unexecuted observation tasks are assigned to the aerostat to generate a feasible sequence of aerostat observation tasks, consisting of two consecutive observation tasks executed by the aerostat. and The costs involved also consist of three parts: , It is the reciprocal of the gains from the observation missions performed by the aerostat. This is the latest start time for the aerostat to perform its observation mission, used to indicate the urgency of the mission. It is the cruise time between two consecutive observation missions performed by an airship at two different geographical locations;

[0064] After completing the task allocation for the aerostat, any unexecuted observation tasks are assigned to the UAVs to generate a feasible sequence of UAV observation tasks. The cost of the UAVs... Two consecutive observation missions were carried out and The costs between them are calculated in the same way.

[0065] S103. The fitness of chromosomes in the initial population is evaluated, and the total benefit obtained by each space platform in completing its observation task and its corresponding data transmission task is evaluated.

[0066] Specifically, the embodiments of the present invention evaluate the sum of the total gains obtained by chromosomes in the initial population in completing their observation tasks and their corresponding data transmission tasks.

[0067] S104. Perform a crossover operation on the current population, improve the population using a preset neighborhood search operator, and perform local search and adaptive neighborhood selection to obtain the current optimal solution.

[0068] It is understood that, in this embodiment of the invention, the population is subjected to crossover and neighborhood search, and then the current optimal solution is obtained through local search and adaptive neighborhood selection.

[0069] In this embodiment of the invention, the crossover operation on the initial population includes: selecting two crossover points from the sequence of observation tasks executed by the same space platform; after crossover, repairing the chromosomes of the two offspring to ensure its feasibility.

[0070] The neighborhood search operator in this embodiment of the invention includes an inter-platform exchange operator, an intra-platform exchange operator, and a relaxation-time-based insertion operator. The inter-platform exchange operator selects two observation tasks from the observation task sequences of two random space platforms, determines whether the two observation tasks can be executed by the exchanged space platform, and if not, selects a new observation task to ensure the feasibility of the chromosome after the exchange. The intra-platform exchange operator randomly selects a space platform and randomly swaps the positions of two observation tasks in the observation task sequence of that platform. The relaxation-time-based insertion operator randomly selects a space platform, calculates the relaxation time of the observation task sequence for that platform (i.e., the interval between the current execution time and the latest execution time of an observation task), and randomly selects an unassigned observation task to insert into the relaxation time of the observation task sequence.

[0071] The process of improving the population using a preset neighborhood search operator, and performing local search and adaptive neighborhood selection to obtain the current optimal solution includes:

[0072] The inter-platform exchange operator, the intra-platform exchange operator, and the relaxation-time-based insertion operator are executed with equal probability. If a neighborhood search operator generates a better solution, the probability of executing that neighborhood search operator is increased; otherwise, the probability is decreased, and a local search operator is used for local searching. The probability is, , Represents search operators The quality, through calculation An adaptive neighborhood selection is performed to obtain the current optimal solution, where... .

[0073] In specific implementation, this embodiment of the invention performs a crossover operation on the population, including: first, deleting duplicate observation tasks from the offspring chromosomes, and then inserting observation tasks into the offspring chromosomes to improve the fitness value of the offspring chromosomes. An observation task set is constructed from the observation tasks deleted in the previous step. The highest priority observation task that has not been executed by any space platform is added to the offspring chromosome. Each observation task is inserted between any two adjacent observation tasks in the repaired offspring chromosome, and its fitness value is calculated. If the inserted chromosome is infeasible, the fitness value is negative. An observation task is inserted with a 1 / 2 probability into the position with the maximum fitness value, or randomly inserted into a position with a positive fitness value. Then, the observation task is deleted from the observation task set. The insertion operation is repeated until all observation tasks in the insertion set have no feasible insertion positions.

[0074] S105. Repeat steps S103 and S104 until the maximum number of iterations or the maximum computation time limit is reached, at which point the algorithm terminates.

[0075] Specifically, in this embodiment of the invention, S103 and S104 are repeatedly executed, and the algorithm terminates when the maximum number of iterations or the maximum computation time limit is reached. Then, the best solution is determined from all the solutions, and finally, this solution is used to coordinate with heterogeneous space platforms to complete Earth observation.

[0076] In summary, this invention employs chromosomes with integer encoding to set the execution sequence of observation tasks on heterogeneous space platforms. It proposes a three-stage decoding strategy, designs feasible observation time window selection strategies, and insertion strategies based on time-based benefits, task urgency, and attitude transition time. An initial chromosome population is constructed. Then, by constructing a local search operator consisting of three neighborhood operators—inter-platform exchange, intra-platform exchange, and insertion based on relaxation time—and an adaptive neighborhood selection mechanism, an optimal scheme for collaborative Earth observation and data transmission scheduling on heterogeneous space platforms is generated, thus providing technical support for space-based Earth observation.

[0077] The following will combine Figure 1 , Figure 2 and Figure 3 The method described in the embodiments of the present invention will be explained and illustrated in detail through a specific example:

[0078] First, we make the following assumptions about the observation and data transmission tasks:

[0079] (1) The scheduling period is [0, T], and all observation task information is known and all are point targets.

[0080] (2) Each observation task corresponds to an observation benefit. The benefit can only be obtained after the observation task and its corresponding data transmission task are completed.

[0081] (3) Each observation task has an execution time requirement and can only be executed once.

[0082] (4) Each observation task has payload type and resolution requirements. The payload types include panchromatic light and multispectral light.

[0083] The specific assumptions regarding the space platform are as follows:

[0084] (1) Each space platform can only perform one observation or data transmission task at a time.

[0085] (2) The use of different types of space platforms has priority. The observation tasks performed by different types of space platforms have different benefits. AEOS has the highest benefits, followed by airships, and UAVs have the lowest benefits.

[0086] (3) Each space platform carries two types of payloads: panchromatic payloads and multispectral payloads.

[0087] (4) AEOS performs data transmission tasks within the visible time window of the ground station. It can establish a communication link with one ground station each time. A data transmission task cannot be down to multiple ground stations.

[0088] (5) AEOS, airships, and UAVs all have a maximum data storage capacity limit. For satellites and airships, storage space is released after the data transmission task is completed. For UAVs, storage space is released after returning to the base station to complete the data transmission task.

[0089] (6) Due to energy limitations, UAVs have a cruise cycle, meaning that when performing a mission, due to energy consumption, they need to return to the base station to replenish energy after a certain cruise time.

[0090] The present invention relates to the problem of scheduling collaborative Earth observation and data transmission between heterogeneous space platforms, which requires decision-making on the space platform to perform the observation mission, the ground station to perform the data transmission mission, determining the start time of the observation and data transmission mission, and the spatial location of the airship or UAV to perform the observation and data transmission mission.

[0091] To address the aforementioned scheduling problem, this invention proposes a collaborative Earth observation and data transmission scheduling method for heterogeneous spatial platforms based on a cultural gene algorithm. It employs integer encoding and proposes a three-stage decoding strategy. A heuristic initial solution generation strategy is designed. A local search operator composed of three neighborhood operators and an adaptive neighborhood selection mechanism are also proposed. Figure 2 The flowchart of the multi-space platform collaborative Earth observation and data transmission scheduling method based on the cultural gene algorithm proposed in this invention is shown, and specifically includes:

[0092] S1: Chromosomes use integer encoding to represent the execution sequence of observation tasks across multiple space platforms. The numbers represent the observation task numbers, and the sequence of observation tasks executed by each space platform is separated by 0. Different colors correspond to observation task sequences for different types of space platforms, and each chromosome contains observation task sequences for all space platforms. For example... Figure 3 As shown, for a scenario with two AEOS satellites, one aerostat, and one drone, chromosome 0-9-11-5-0-7-4-2-0-12-8-1-0-3-6-0 indicates that the observation mission execution sequences for the first and second AEOS satellites are 9-11-5 and 7-4-2, respectively. The observation mission execution sequences for the aerostat and drone are 12-8-1 and 3-6, respectively.

[0093] S2: A three-stage decoding strategy is proposed to repair infeasible chromosomes into feasible solutions. The specific steps are as follows:

[0094] S201: The first stage involves assessing the feasibility of the space platform within the observation mission's time window. The specific steps are as follows:

[0095] (1) Calculate the intersection of the observation time requirement of the observation task and the visible time window of the satellite for that task, and define it as the feasible time window of the satellite for that task. Take the start time of the feasible time window of the satellite for that task as the start time of the first observation task in the observation task sequence. Calculate the start time of each observation task in turn. The next observation task... start time The calculation formula is:

[0096]

[0097] in, This indicates that AEOS is for observation tasks. The start time of the feasible time window. Indicates the previous observation task The start time, Indicates observation task The duration of observation, This indicates that AEOS is in the observation mission. and observation mission The attitude transition time between observation tasks. If the start time of an observation task exceeds the end time of the feasible time window for that observation task by AEOS, the observation task is removed from the observation task sequence.

[0098] (2) For the first observation task in the observation task sequence of the airship (UAV), select the airship base station (UAV base station) closest to the task as the starting point of the airship (UAV), calculate the cruise time from the starting point to the first observation task. If the cruise time exceeds the time requirement of the task, remove the task from the observation task sequence. Otherwise, calculate the cruise time from the execution position of the first observation task to the execution position of the next observation task, and determine whether the time requirement of the next observation task is violated. This process continues until all observation tasks in the observation task sequence of the space platform meet the time requirements of the observation tasks.

[0099] S202: Second phase, assess the feasibility of AEOS (aerial Observatory) data storage. If the total data storage capacity occupied by the observation tasks in the observation task sequence exceeds the maximum data storage capacity of the AEOS (aerial Observatory), one or more observation tasks need to be removed until the data storage capacity is met.

[0100] S203: The third stage involves determining the feasibility of executing the data transmission task. The specific steps are as follows:

[0101] (1) Define the visible time window of AEOS to the ground station, excluding the time period occupied by the observation task, as the feasible data transmission time window. According to the priority of the corresponding observation task, insert the data transmission task into the feasible data transmission time window with the earliest start time, until the remaining feasible data transmission time window is insufficient to execute the next data transmission task. Iterate through all remaining data transmission tasks, and according to the priority of the corresponding observation task, insert the data transmission task within the remaining time of the feasible data transmission time window, until the remaining feasible data transmission time is insufficient.

[0102] (2) Within the airship's coverage area of ​​the ground station, insert data transmission tasks into the feasible observation task sequence according to the priority of the corresponding observation tasks, and repair the observation task sequence to ensure feasibility. If the data storage capacity exceeds the airship's maximum data storage capacity after inserting a data transmission task, one or more observation tasks need to be moved after the insertion point until the data storage capacity no longer exceeds the limit. If there is insufficient remaining time to execute the data transmission task, one or more observation tasks need to be removed before the insertion point.

[0103] (3) When the UAV reaches its maximum data storage capacity, it returns to the base to perform data transmission tasks.

[0104] S3: A feasible observation time window selection strategy, an observation task insertion strategy based on time-based benefits, task urgency and attitude transition time are proposed to construct the initial population of chromosomes.

[0105] S301: Based on a feasible observation time window selection strategy, a feasible time window is assigned to each observation task to reduce time conflicts between observation tasks. A feasible time window selection strategy is proposed, which selects a feasible time window from the set of feasible time windows for all observation tasks to reduce conflicts with other observation tasks. This strategy can be divided into two steps:

[0106] (1) An observation task set is defined and initialized at the beginning of the scheduling period. All feasible time windows that do not conflict with other observation tasks are identified and used as the feasible time windows of the observation tasks, and the observation tasks are removed. If an observation task has multiple feasible time windows that do not conflict with other observation tasks, a feasible time window is randomly selected.

[0107] (2) Select the earliest observation window among all observation tasks as the feasible time window for its corresponding observation task. Select the latest observation window among all observation tasks as the time window for its corresponding observation task. If a feasible time window for an observation task is selected, remove that observation task. Repeat this process until all observation tasks are assigned to a feasible time window.

[0108] S302: A time-dependent reward and task urgency-based observation task insertion strategy is proposed, which generates a feasible observation task sequence for each space platform without considering data storage capacity. The specific steps are as follows:

[0109] (1) At the start of the scheduling period, the set of unexecuted observation tasks is initialized. Under the constraints of AEOS payload type and resolution, a feasible observation task is randomly selected as the first observation task of AEOS, and the lower bound of the selected feasible time window is used as the start time for AEOS to execute the observation task. For each feasible subsequent observation task, AEOS is calculated between two consecutive observation tasks. and Costs between:

[0110]

[0111] in This indicates linear normalization. The first term... It is the reciprocal of the time-dependent reward for the observation task. This indicates the time AEOS spent performing the observation task. (Second item) This is the latest start time for AEOS to execute an observation task, used to measure the urgency of the task. The last item indicates that AEOS is executing two consecutive observation tasks. and The attitude transition time between [the two points]. The observation task with the lowest cost is selected and denoted as [the task name]. Repeat the above process until no more feasible observation tasks are available.

[0112] (2) After completing the AEOS observation task allocation, the unexecuted observation tasks are assigned to the aerostat to generate a feasible sequence of aerostat observation tasks. Two consecutive observation tasks are executed by the aerostat. and The costs involved also consist of three parts:

[0113]

[0114] First item It is the reciprocal of the gains from the observation missions performed by the aerostat. (Second term) This is the latest start time for the aerostat to perform its observation mission, used to indicate the urgency of the mission. (Last item) It is the cruise time between two consecutive observation missions performed by an airship at two different geographical locations.

[0115] After the aerostat's task assignment is completed, any unexecuted observation tasks are assigned to UAVs to generate a feasible sequence of UAV observation tasks. The cost of the UAVs... Two consecutive observation missions were carried out and The costs between them are calculated in the same way.

[0116] S4: Assess the fitness of chromosomes in the population. The fitness function is defined as the sum of the total rewards obtained from the observation tasks and corresponding data transmission tasks completed by AEOS, airships, and UAVs, respectively.

[0117] S5: Perform a crossover operation on the current population. Select two parent chromosomes to generate two offspring chromosomes. Select two crossover points from the observation mission sequence executed by the same space platform. After crossover, repair the two offspring chromosomes to ensure feasibility. The specific steps are as follows:

[0118] S501: Deletion of duplicate observation tasks in offspring chromosomes.

[0119] S502: Insert observation tasks into the progeny chromosome to improve its fitness value. Construct an observation task set from the observation tasks deleted in the previous step. Add the highest-priority observation task that has not been executed by any space platform to the progeny chromosome. Insert each observation task between any two adjacent observation tasks in the repaired progeny chromosome and calculate its fitness value. If the chromosome is infeasible after insertion, the fitness value is negative. Insert the observation task with a 1 / 2 probability into the position with the maximum fitness value, or randomly insert it into a position with a positive fitness value. Then delete the observation task from the observation task set. Repeat the insertion operation until there are no feasible insertion positions for any observation task in the insertion set.

[0120] S6: Construct three neighborhood search operators: inter-platform exchange, intra-platform exchange, and relaxation-time-based insertion. Perform local searches on individuals in the current population and propose an adaptive neighborhood selection mechanism to select the optimal search operator. The specific steps are as follows:

[0121] S601: Space Platform Exchange Operator. Two observation tasks are randomly selected from the observation task sequences of two space platforms. It is determined whether the two observation tasks can be executed by the exchanged space platform. If not, a new observation task is selected to ensure the feasibility of the exchanged observation tasks.

[0122] S602: Intra-platform exchange operator. Randomly select a space platform and randomly swap the positions of two observation tasks in the observation task sequence of that space platform.

[0123] S603: Relaxation-time-based insertion operator. Randomly select a space platform and calculate the relaxation time of the observation task sequence for that platform; this is the interval between the current execution time and the latest execution time of an observation task. Randomly select an unassigned observation task and insert it into the relaxation time of the observation task sequence.

[0124] S604: Adaptive Neighborhood Selection Mechanism. Three neighborhood search operators are executed with equal probability. If a neighborhood search operator generates a better solution, the probability of executing that operator is increased; otherwise, the probability is decreased. Local Search Operator The probability of can be expressed as, , Represents search operators The quality is calculated iteratively using the following formula:

[0125] ,

[0126] S7: Repeat S5 and S6. If the maximum number of iterations or the maximum computation time limit is reached, the algorithm terminates and outputs the best solution.

[0127] The collaborative Earth observation and data transmission scheduling method based on the cultural gene algorithm of this invention is compared with the results of independent task planning for three types of heterogeneous space platforms to verify the effectiveness of the method. Independent task planning refers to the independent execution of all observation tasks by various types of space platforms in a random order. When a type of space platform is fully utilized, the observation task will be assigned to the next type of space platform until all space platforms are utilized or all observation tasks have been assigned.

[0128] This invention addresses the problem of collaborative Earth observation and data transmission scheduling for heterogeneous space platforms by constructing multiple sets of computational examples of different scales. These examples are compared with independent planning for three types of space platforms to verify the effectiveness of the proposed method. The scheduling period is 24 hours (October 1, 2023, 00:00:00.000 to October 1, 2023, 24:00:00.000). The observation tasks are distributed in the Taiwan Strait and surrounding waters in southeastern China (120°E–125°E, 20°N–25°N). Twenty-four sets of computational examples of different scales are used, including six sets of observation task scales: 300, 600, 900, 1200, 1500, and 1800; and four sets of three different types of space platforms (AEOS / airships / UAVs): (4 / 5 / 5), (8 / 10 / 10), (12 / 15 / 15), and (16 / 20 / 20).

[0129] The rewards for observation tasks are generated randomly, following a uniform distribution of (0,1). The observation duration is randomly generated between 15 and 30 seconds. The start time for the observation is randomly generated within 24 hours, and the end time is determined based on the start time, with an interval of 12 hours. Both panchromatic and multispectral payloads are considered. The panchromatic resolution is set to 0.5 m, 1 m, and 2 m, while the multispectral resolution is set to 2 m, 4 m, and 8 m.

[0130] The algorithm was coded in Python 3.8.8 and ran on a computer with a Windows 10 Enterprise x64 operating system and an Intel(R) Core(TM) i7-6700 CPU @ 3.40 GHz with 24 GB of memory. The orbital elements of AEOS are shown in Table 1. The base station locations and performance parameters of the aerostat and UAV are shown in Tables 2 and 3.

[0131] Table 1 Partial AEOS Orbital Roots

[0132] Semi-major axis (m) Track inclination angle (°) Right ascension of the ascending node (°) Eccentricity Perigee argument (°) 1 7569345 100.74 83.2388 0.0025382 10.9656 2 7467869 63.40 338.3441 0.0278024 4.928 3 7580228 100.63 39.861 0.0005816 260.4264 4 7467799 63.40 116.9629 0.018683 4.7392 5 7129366 98.41 280.6872 0.0001215 60.3322 6 7011361 98.30 84.095 0.001895 218.1331 7 7620677 100.28 190.5985 0.0021603 30.9156 8 7580045 100.05 283.1983 0.0009035 207.7205 9 7077757.11 98.27 157.06 0.0003 120.38 10 7015608.55 97.82 155.47 0.0016 343.06 11 7468028.03 63.39 56.47 0.0265 8.69 12 7578252.29 100.13 246.80 0.0008 229.46 13 7467993.12 63.41 285.35 0.0048 351.09 14 6878105.59 97.61 169.13 0.0001 78.13 15 7002879.50 97.95 238.62 0.0001 81.45 16 7074354.90 98.21 107.56 0.0001 154.05 17 7018391.617 97.8913 349.901 0.0010973 122.6651 18 6879953.25 97.3644 348.8307 0.0001544 87.1164 19 6829913.842 97.1385 241.5936 0.0023805 49.3746 20 7021123.685 98.1399 143.6798 0.0025489 344.8311

[0133] Table 2 Locations of Aerial Vehicle and UAV Base Stations

[0134] Base station type latitude and longitude airship base station (118.2°E, 24.8°N), (118.4°E, 24.8°N), (118.6°E, 24.8°N), (118.8°E,24.8°N), (118.2°E, 24.6°N), (118.4°E, 24.6°N), (118.6°E, 24.6°N), (118.2°E, 24.4°N), (118.4°E, 24.4°N), (118.2°E, 24.2°N) drone base station (118.1°E, 24.9°N), (118.2°E, 24.9°N), (118.3°E, 24.9°N), (118.4°E, 24.9°N), (118.5°E, 24.9°N), (118.6°E, 24.9°N), (118.7°E, 24.9°N), (118.8°E, 24.9°N), (118.1°E, 24.7°N), (118.2°E, 24.7°N), (118.3°E,24.7°N), (118.4°E, 24.7°N), (118.5°E, 24.7°N), (118.6°E, 24.7°N), (118.1°E, 24.5°N), (118.2°E, 24.5°N), (118.3°E, 24.5°N), (118.4°E, 24.5°N), (118.1°E, 24.3°N), (118.2°E, 24.3°N)

[0135] Table 3 Performance parameters of airships and drones

[0136] parameter airship drones Maximum battery life 12h 5h Observe the maximum pitch angle 30° 30° Flight altitude 20km 2km movement speed 70km / h 100km / h Observation radius 11.5km 1.15 km Data storage 2TB 1TB

[0137] The collaborative task planning results of the heterogeneous space platform collaborative Earth observation and data transmission scheduling method based on the cultural gene algorithm were obtained by running it 10 times. The results included the maximum, average, and minimum total benefits, as well as the average running time of MA. The independent task planning results of three types of heterogeneous space platforms were obtained by modifying MA (keeping the initial solution generation strategy of MA, removing the inter-space platform exchange operator in the local search), and running it 10 times. The results included the maximum, average, and minimum total benefits, as well as the average running time. The difference between the average total benefits of the heterogeneous space platform collaborative task planning and the average total benefits of the independent space platform planning was calculated as: difference = (average total benefits of collaborative task planning - average total benefits of independent task planning) / average total benefits of collaborative task planning × 100%.

[0138] As shown in Table 4 below, the collaborative task planning method of the heterogeneous space platform collaborative Earth observation and data transmission scheduling method based on the cultural gene algorithm has improved the total benefit compared with the independent task planning of the three types of heterogeneous space platforms, with the highest improvement reaching 20.12%.

[0139] Table 4 Comparison of Collaborative Task Planning and Independent Planning Results of Heterogeneous Space Platforms

[0140]

[0141] Accordingly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described above. Related content regarding the storage medium embodiments of the present invention can be understood by referring to the method embodiments of the present invention, and will not be elaborated upon here.

[0142] Although preferred embodiments of the invention have been disclosed for illustrative purposes, those skilled in the art will recognize that various modifications, additions, and substitutions are possible, and therefore the scope of the invention should not be limited to the embodiments described above.

Claims

1. A method for collaborative Earth observation and data transmission scheduling of heterogeneous space platforms based on a cultural gene algorithm, characterized in that, The method includes: The chromosome uses integer encoding. The chromosome represents the execution sequence of observation tasks of each space platform to complete the current Earth observation task. The feasibility of the chromosome is judged by a preset feasibility judgment strategy, and infeasible chromosomes are repaired to be feasible. The feasibility judgment strategy includes the feasibility of the observation task time window, the feasibility of data storage, and the feasibility of executing data transmission tasks. An observation task insertion selection strategy based on time gain, task urgency, and attitude transition time determines the initial population of chromosomes among feasible chromosomes. The fitness of chromosomes in the initial population is evaluated, and the total benefit obtained by each space platform in completing its observation task and its corresponding data transmission task is evaluated. A crossover operation is performed on the population, along with local search and adaptive neighborhood selection, to obtain the current solution; The algorithm assesses the fitness of the current population, performs crossover operations and local searches. If the maximum number of iterations or the maximum computation time limit is reached, the algorithm terminates, outputs the best current solution, and uses this solution to coordinate with heterogeneous space platforms to complete Earth observation.

2. The method according to claim 1, characterized in that, The chromosome uses integer encoding to represent the execution sequence of observation tasks on multiple space platforms. The numbers represent the observation task numbers. The observation task sequences executed by each space platform are separated by 0. Different colors correspond to the observation task sequences of different types of space platforms. Each chromosome contains the observation task sequences of all space platforms.

3. The method according to claim 1, characterized in that, When the space platform includes AEOS, airships, and UAVs, the feasibility of the chromosome is determined based on the feasibility of the observation mission time window, including: The intersection of the observation time requirement of the observation task in the chromosome and the visible time window of AEOS for the observation task is calculated as the feasible time window of AEOS for the observation task. The start time of the feasible time window is taken as the start time of the first observation task in the observation task sequence. The start time of each observation task is calculated in turn. If the start time of an observation task exceeds the end time of the feasible time window of AEOS for the observation task, the observation task is determined to be not feasible in the observation task time window and is removed from the observation task sequence. For the first observation task in the observation task sequence of airships and UAVs, the airship base station or UAV base station closest to the observation task is selected as the starting point of the airship or UAV. The cruise time from the starting point to the first observation task is calculated. If the cruise time exceeds the time requirement of the task, the task is removed from the observation task sequence. Otherwise, the cruise time from the execution position of the first observation task to the execution position of the next observation task is calculated, and it is determined whether the time requirement of the next observation task is violated. This process is repeated until all observation tasks in the observation task sequence meet the time requirements of the observation tasks.

4. The method according to claim 3, characterized in that, The feasibility of the chromosome is assessed based on data storage feasibility, including: If the total data storage capacity occupied by the observation tasks in the observation task sequence exceeds the maximum data storage capacity of the space platform, one or more observation tasks will be removed until the data storage capacity is met.

5. The method according to claim 3, characterized in that, The feasibility of the chromosome is determined based on the feasibility of the data transmission task, including: The AEOS's visible time window for the ground station is set as the feasible data transmission time window after deleting the time period occupied by the observation task. According to the priority of the corresponding observation task, the data transmission task is inserted into the feasible data transmission time window with the earliest start time. This continues until the remaining feasible data transmission time window is insufficient to execute the next data transmission task. All remaining data transmission tasks are traversed, and data transmission tasks are inserted in the remaining time of the feasible data transmission time window according to the priority of the corresponding observation task, until the remaining feasible data transmission time is insufficient. Within the airship's coverage area of ​​the ground station, data transmission tasks are inserted into the feasible observation task sequence according to the priority of the corresponding observation tasks, and the observation task sequence is repaired to ensure feasibility. If the data storage capacity exceeds the airship's maximum data storage capacity after inserting the data transmission task, one or more observation tasks are moved after the insertion point until the data storage capacity no longer exceeds the limit. If there is insufficient remaining time to execute the data transmission task, one or more observation tasks are removed before the insertion point. When the drone reaches its maximum data storage capacity, it returns to base to perform data transmission tasks.

6. The method according to any one of claims 1-5, characterized in that, An observation task insertion selection strategy based on time gain, task urgency, and attitude transition time determines the initial population of chromosomes among feasible chromosomes, including: Based on a feasible observation time window selection strategy, a feasible time window is selected from the set of feasible time windows for all observation tasks. A feasible time window is assigned to each observation task. Based on time-dependent rewards, task urgency, and attitude transition time, a feasible observation task sequence is generated for each space platform to obtain the initial population of the chromosome.

7. The method according to claim 6, characterized in that, Based on time-dependent rewards, task urgency, and attitude transition time, a feasible observation task sequence is generated for each space platform, resulting in the initial population of the chromosome, including: At the start of the scheduling cycle, the set of unexecuted observation tasks is initialized. Under the constraints of AEOS payload type and resolution, a feasible observation task is randomly selected as the first AEOS observation task. The lower bound of the selected feasible time window is used as the start time for AEOS to execute the observation task. For each feasible subsequent observation task, AEOS is calculated for two consecutive observation tasks. and Costs between: ,in Indicates linear normalization, It is the reciprocal of the time-dependent reward for the observation task. This indicates the time AEOS takes to perform the observation task. This is the latest start time for AEOS to execute an observation task, used to measure the urgency of the observation task. This indicates that AEOS is performing two consecutive observation tasks. and Given the attitude transition time between observations, select the observation task with the lowest cost, denoted as . Repeat the calculation of AEOS in two consecutive observation tasks until there are no more feasible observation tasks. After completing the AEOS observation task allocation, any unexecuted observation tasks are assigned to the aerostat to generate a feasible sequence of aerostat observation tasks, consisting of two consecutive observation tasks executed by the aerostat. and The costs involved also consist of three parts: , It is the reciprocal of the gains from the observation missions performed by the aerostat. This is the latest start time for the aerostat to perform its observation mission, used to indicate the urgency of the mission. It is the cruise time between two consecutive observation missions performed by an airship at two different geographical locations; After completing the task allocation for the aerostat, any unexecuted observation tasks are assigned to the UAVs to generate a feasible sequence of UAV observation tasks; the cost of the UAVs... Two consecutive observation missions were carried out and The costs between them are calculated in the same way.

8. The method according to any one of claims 1-5, characterized in that, Performing a crossover operation on the initial population includes: Two intersection points are selected from the sequence of observation tasks executed by the same space platform. After intersection, the two progeny chromosomes are repaired to ensure their feasibility. This involves deleting duplicate observation tasks from the progeny chromosomes and inserting new observation tasks into them to improve their fitness values. An observation task set is constructed from the deleted observation tasks. The highest priority observation task that has not been executed by any space platform is added to the progeny chromosome. Each observation task is inserted between any two adjacent observation tasks in the repaired progeny chromosome, and its fitness value is calculated. If the chromosome is not feasible after insertion, the fitness value is negative. The observation task is then inserted with a 1 / 2 probability into the position with the maximum fitness value, or randomly inserted into a position with a positive fitness value. The observation task is then deleted from the observation task set. This insertion operation is repeated until there are no feasible insertion positions for any observation task in the insertion set.

9. The method according to any one of claims 1-5, characterized in that, The neighborhood search operator includes an inter-platform exchange operator, an intra-platform exchange operator, and a relaxation-time-based insertion operator. The inter-platform exchange operator randomly selects two observation tasks from the observation task sequences of two space platforms, determines whether the two tasks can be executed by the exchanged platform, and if not, selects a new observation task to ensure the feasibility of the chromosome after the exchange. The intra-platform exchange operator randomly selects a space platform and randomly swaps the positions of two observation tasks in its observation task sequence. The relaxation-time-based insertion operator randomly selects a space platform, calculates the relaxation time of its observation task sequence (the interval between the current execution time and the latest execution time of an observation task), and randomly selects an unassigned observation task to insert into the relaxation time of the observation task sequence. The process of expanding the seed chromosome using a preset neighborhood search operator, performing local search and adaptive neighborhood selection to obtain the current optimal solution includes: The inter-platform exchange operator, the intra-platform exchange operator, and the relaxation-time-based insertion operator are executed with equal probability. If a neighborhood search operator generates a better solution, the probability of executing that neighborhood search operator is increased; otherwise, the probability is decreased, and a local search operator is used for local searching. The probability is, , Represents search operators The quality, through calculation An adaptive neighborhood selection neighborhood search operator is used, where... .

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of any one of claims 1-9.

Images