Dynamic scheduling method for multi-satellite cooperative observation task

Abstract

This invention discloses a dynamic scheduling method for multi-satellite collaborative observation tasks, mainly including the following steps: discretizing the scheduling cycle into time slices based on the minimum overlap observation time, and uniformly managing resources and time windows; using a multi-objective optimization algorithm for initial scheduling of fixed tasks, taking into account both task completion rate and minimizing the number of relays; detecting satellite idle time slices based on the initial scheduling results and constructing an index; prioritizing the insertion of temporary tasks into the schedule, if a single satellite cannot complete the task, multi-satellite relay is activated, and if all fail, collaborative rescheduling is triggered, merging fixed and temporary tasks for re-optimization, while minimizing disturbance to fixed tasks. This method, through a "prioritizing insertion and rescheduling as a fallback" strategy, improves resource utilization and dynamic response capabilities, efficiently handles temporary tasks while ensuring the execution of fixed tasks, and reduces interference with the original fixed task plan, making it suitable for complex and variable observation scenarios.

Description

Technical Field

[0001] This invention relates to aerospace remote sensing and satellite scheduling technology, and in particular to a dynamic scheduling method for multi-satellite collaborative observation missions. Background Technology

[0002] The scheduling problem of multi-satellite collaborative Earth observation missions is a typical NP-hard problem. Existing research mainly revolves around four types of methods: exact algorithms, heuristic methods, machine learning methods, and feedback-based iterative optimization methods. Exact algorithms achieve optimal mission planning by establishing mathematical models and constraints and employing rigorous mathematical solution methods. Heuristic methods complete mission allocation and scheduling by setting heuristic functions and planning rules, utilizing human experience and strategies. Machine learning methods achieve intelligent mission prediction and decision-making by training models and learning from historical data. Feedback-based methods decompose the complex multi-satellite mission planning problem into multiple sub-stages, gradually completing mission scheduling through inter-stage information feedback and iterative optimization, forming a closed-loop optimization mechanism.

[0003] The working principles of existing technologies are mainly reflected in two aspects: problem decomposition and solution strategies. Regarding problem decomposition, most methods divide multi-satellite collaborative observation mission scheduling into two sub-problems: task allocation and task planning. First, the correspondence between tasks and satellites is determined, and then specific time window scheduling is performed. In terms of solution strategies, precise algorithms use mathematical methods such as mixed-integer linear programming and branch-and-bound to construct constraint satisfaction models; heuristic methods design specific search strategies such as large neighborhood search and ant colony optimization; machine learning methods learn historical scheduling experience through deep learning networks to achieve online decision-making; and feedback-based methods establish an iterative loop mechanism of allocation-scheduling-feedback, gradually improving the scheduling scheme through multiple rounds of optimization.

[0004] Existing technologies suffer from two main drawbacks: First, they fall short in handling large-scale tasks, dynamic task insertion, and multi-objective optimization, making it difficult to meet the complex and ever-changing task requirements in practical applications. Precise algorithms have high computational complexity and cannot adapt to large-scale scenarios; heuristic methods lack versatility and adaptability; machine learning methods struggle to simultaneously handle global optimization of multiple tasks; and feedback-based methods require improvement in overall solution efficiency. Second, current multi-remote-sensing satellite scheduling methods are mostly static or single-response modes, making it difficult to simultaneously address the real-time response needs of both long-term fixed tasks and sudden temporary tasks. Existing methods lack differentiated processing mechanisms for fixed and temporary tasks, failing to flexibly insert temporary tasks while ensuring the execution of fixed tasks, and also lack effective dynamic scheduling strategies to balance resource allocation for different types of tasks.

[0005] It should be noted that the information disclosed in the background section above is only for understanding the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The main objective of this invention is to overcome the deficiencies in the aforementioned background technology and provide a dynamic scheduling method for multi-satellite collaborative observation tasks. In large-scale multi-satellite collaborative observation scenarios, this method enables efficient and stable collaborative scheduling of fixed tasks and sudden temporary tasks, while ensuring resource utilization and the robustness of the scheduling scheme.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A dynamic scheduling method for multi-satellite collaborative observation missions includes the following steps:

[0009] S1: Task time slice discretization: The scheduling cycle is divided into discrete observation time slices according to the minimum overlapping observation time, so as to manage satellite resources and time windows in a unified manner;

[0010] S2: Initial scheduling of fixed tasks: A multi-objective optimization algorithm is used to initially schedule fixed tasks, generating a task allocation scheme that takes into account both task completion rate and minimizing the number of relays.

[0011] S3: Idle Resource Detection: Based on the results of fixed task scheduling, calculate the set of consecutive idle time slices for each satellite and build a structured index;

[0012] S4: Dynamic scheduling of temporary tasks:

[0013] Prioritize the use of interleaved scheduling: search for continuous time slots in satellite idle time slots that meet the observation requirements of temporary tasks, and execute them independently on a single satellite;

[0014] If a single satellite cannot complete the task, a multi-satellite relay strategy is employed: tasks are allocated during overlapping idle periods of multiple satellites to ensure that the minimum overlap time slice requirement is met.

[0015] When the insertion strategy fails, a cooperative rescheduling is triggered: the temporary task is merged with the fixed task, and the scheduling scheme is re-optimized through a multi-objective optimization algorithm, while minimizing the disturbance to the fixed task.

[0016] Furthermore, in step S1:

[0017] The observation task is divided into consecutive sub-time periods, the length of which is equal to the minimum overlapping observation time.

[0018] When multiple satellites work in relay to perform a mission, the observation periods between satellites must overlap by at least one time slice to ensure data continuity and consistency.

[0019] Furthermore, in step S2:

[0020] The multi-objective optimization algorithm is a multi-objective genetic algorithm that uses chromosome encoding. Each chromosome contains multiple gene segments, and each gene segment corresponds to a task allocation scheme. The encoding parameters include task identifier, satellite identifier, start time slice, end time slice, relay flag, and associated relay satellite identifier.

[0021] Furthermore, in step S2:

[0022] The optimization objectives of multi-objective genetic algorithms include:

[0023] Maximize the completion rate of fixed tasks and impose a fitness penalty on unscheduled tasks;

[0024] Minimize the number of task relays and reduce data fusion costs;

[0025] Genetic operations include selection based on non-dominated sorting and crowding distance, task-level gene segment crossover, and parameter perturbation variation.

[0026] Furthermore, in step S3:

[0027] The structured index stores consecutive idle time periods in triplets of satellite identifier, start time slice, and end time slice, and sorts them by start time to form a timeline tree structure for fast retrieval of available gaps.

[0028] Furthermore, in step S4, the air conditioning settings are inserted:

[0029] The multi-star relay strategy is implemented as follows:

[0030] Two satellites with overlapping idle time slots are selected, and a portion of the observation time slots are allocated to each. The total number of time slots in the overlapping area, after deducting the overlapping portion, must meet the total observation requirements of the temporary mission.

[0031] Furthermore, in the coordinated rescheduling of step S4:

[0032] The new optimization objective is to minimize the scheduling disturbance, which is achieved in the following way:

[0033] The number of satellite changes and start time changes for fixed tasks are statistically analyzed, and the weighted sum is used as a disturbance penalty to maintain the stability of the scheduling scheme.

[0034] Furthermore, the overall process includes:

[0035] 1) Discretized scheduling cycle;

[0036] 2) Initial scheduling of fixed tasks;

[0037] 3) Calculate the satellite's idle time period;

[0038] 4) Determine if the temporary task can be completed in a gap: if so, execute it individually or in relay; otherwise, trigger rescheduling.

[0039] 5) Output the dynamic scheduling results.

[0040] Furthermore, in step 4):

[0041] The insertion strategy prioritizes single-satellite independent detection; if it fails, multi-satellite relay detection is performed; only when both fail is rescheduling triggered.

[0042] A computer program product includes a computer program that, when executed by a processor, implements the dynamic scheduling method for the multi-satellite collaborative observation mission.

[0043] The present invention has the following beneficial effects:

[0044] This invention addresses the dynamic scheduling of multi-satellite collaborative observations in scenarios where fixed and temporary tasks coexist. Considering task completion and scheduling efficiency as optimization objectives, it proposes a dynamic scheduling method for multi-satellite collaborative observation tasks, ensuring efficient insertion of temporary tasks and coordinated completion of the overall task while fulfilling the execution requirements of fixed tasks. This invention designs and implements a dynamic scheduling strategy to address the differentiated processing needs of fixed and temporary tasks. It enables the insertion of temporary tasks into available slots and allows for global rescheduling when necessary. Furthermore, a multi-objective genetic algorithm optimizes task allocation and time scheduling schemes, ensuring the algorithm's effectiveness in solving the dynamic scheduling problem of multi-satellite collaborative observation tasks. This invention's method, tailored to practical applications with both fixed and temporary tasks, implements an intelligent scheduling mechanism that balances task completion, resource utilization, and minimizing scheduling disturbances. Key innovations include: 1. Discretizing the task scheduling cycle based on the minimum overlapping observation time, unifying the management of satellite resources and time windows; 2. Prioritizing the use of remaining satellite observation time slots for scheduling temporary tasks, triggering coordinated rescheduling upon failure to minimize interference with the original fixed task plan while ensuring the completion of temporary tasks.

[0045] Compared to existing static scheduling methods, this invention possesses stronger dynamic response and resource optimization capabilities. The proposed unified time-slice modeling and multi-satellite relay observation mechanism effectively improves resource utilization; the "priority of insertion and fallback scheduling" strategy achieves efficient dynamic scheduling of temporary tasks; and the multi-objective genetic algorithm optimizes task allocation, balancing completion rate and scheduling stability while significantly reducing interference with the original plan, making it suitable for complex and ever-changing real-world observation scenarios.

[0046] Other beneficial effects of the embodiments of the present invention will be further described below. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the observation task relay in an embodiment of the present invention.

[0048] Figure 2 This is a schematic diagram of chromosome encoding according to an embodiment of the present invention.

[0049] Figure 3 This is a flowchart of a dynamic scheduling method based on insertion and rescheduling according to an embodiment of the present invention.

[0050] Figure 4 This is the overall flowchart of the dynamic scheduling method for multi-satellite collaborative observation missions of the present invention. Detailed Implementation

[0051] The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary and not intended to limit the scope and application of the present invention.

[0052] This invention addresses the coexistence of fixed and temporary tasks in multi-satellite collaborative observation missions. Considering task completion and scheduling efficiency as optimization objectives, it proposes a dynamic scheduling method based on insertion and rescheduling, along with a multi-objective optimization algorithm. This method ensures efficient insertion of temporary tasks and coordinated completion of the overall mission while satisfying the execution of fixed tasks. To achieve these objectives, this invention designs and implements a dynamic scheduling strategy to address the differentiated processing needs of fixed and temporary tasks. This strategy enables the insertion of temporary tasks and allows for global rescheduling when necessary. Furthermore, a multi-objective optimization algorithm optimizes task allocation and time scheduling schemes, ensuring the effectiveness of the algorithm for the dynamic scheduling problem of multi-satellite collaborative observation missions.

[0053] See Figure 1 This invention provides a dynamic scheduling method for multi-satellite collaborative observation missions, comprising the following steps:

[0054] Step S1: Task time slice discretization: Divide the scheduling cycle into discrete observation time slices according to the minimum overlapping observation time, and manage satellite resources and time windows in a unified manner.

[0055] In some embodiments, the observation task is divided into consecutive sub-time periods, the length of which is equal to the minimum overlapping observation time. When multiple satellites relay the task, the observation time periods between satellites are required to overlap by at least one time slice to ensure data continuity and consistency.

[0056] Step S2: Initial scheduling of fixed tasks: A multi-objective optimization algorithm is used to perform initial scheduling of fixed tasks, generating a task allocation scheme that takes into account both task completion rate and minimizing the number of relays.

[0057] In some embodiments, the multi-objective optimization algorithm is a multi-objective genetic algorithm that uses chromosome encoding. Each chromosome contains multiple gene segments, and each gene segment corresponds to a task allocation scheme. The encoding parameters include task identifier, satellite identifier, start time slice, end time slice, relay flag, and associated relay satellite identifier.

[0058] In some embodiments, the optimization objectives of the multi-objective genetic algorithm include: maximizing the completion rate of fixed tasks and imposing fitness penalties on unscheduled tasks; minimizing the number of task relays and reducing data fusion costs; and genetic operations including selection based on non-dominated sorting and crowding distance, task-level gene segment crossover, and parameter perturbation mutation.

[0059] Step S3: Idle resource detection: Based on the fixed task scheduling results, calculate the set of consecutive idle time slices for each satellite and construct a structured index.

[0060] In some embodiments, the structured index stores consecutive idle time periods as a triplet of satellite identifier, start time slice, and end time slice, and sorts them by start time to form a timeline tree structure for quick retrieval of available gaps.

[0061] Step S4: Dynamic scheduling of temporary tasks: The insertion scheduling is given priority. The continuous time period that meets the observation requirements of the temporary task is searched in the satellite idle time slice, and each satellite executes independently. If a single satellite cannot complete the task, a multi-satellite relay strategy is adopted: the task is allocated in the overlapping idle time periods of multiple satellites to ensure that the minimum overlap time slice requirement is met. When the insertion strategy fails, a collaborative rescheduling is triggered: the temporary task is merged with the fixed task, and the scheduling scheme is re-optimized through a multi-objective genetic algorithm, while minimizing the disturbance to the fixed task.

[0062] In some embodiments, the multi-satellite relay strategy in the interleaving schedule is implemented as follows: two satellites with overlapping idle time slots are selected, and a portion of the observation time slots are allocated to each satellite. The total number of time slots in the overlapping area, after deducting the overlapping portion, must meet the total observation requirements of the temporary mission.

[0063] In some embodiments, the new optimization objective in the collaborative rescheduling is to minimize the scheduling disturbance, which is achieved by: statistically analyzing the number of satellite changes and start time changes for fixed tasks, and using the weighted sum as a disturbance penalty term to maintain the stability of the scheduling scheme.

[0064] In some embodiments, the overall process of the dynamic scheduling method for the multi-satellite collaborative observation mission includes:

[0065] 1) Discretized scheduling cycle;

[0066] 2) Initial scheduling of fixed tasks;

[0067] 3) Calculate the satellite's idle time period;

[0068] 4) Determine if the temporary task can be completed in a gap: if so, execute it individually or in relay; otherwise, trigger rescheduling.

[0069] 5) Output the dynamic scheduling results.

[0070] In some embodiments, in step 4):

[0071] The insertion strategy prioritizes single-satellite independent detection; if it fails, multi-satellite relay detection is performed; only when both fail is rescheduling triggered.

[0072] The dynamic scheduling method proposed in this invention addresses the core challenge of coexisting fixed and temporary tasks in multi-satellite collaborative observation. Through the synergistic effect of unified time-slice discretization modeling and a multi-level dynamic scheduling mechanism, it effectively overcomes the limitations of existing technologies in large-scale dynamic task scheduling. Specifically: 1) It pioneers a time-slice discretization method based on minimum overlap observation time, uniformly managing satellite resources and time windows, providing fundamental support for multi-satellite relay observation, and significantly improving resource utilization; 2) It adopts a hierarchical response strategy of "prioritizing insertion and rescheduling as a fallback," prioritizing the use of satellite idle periods to achieve uninterrupted insertion of temporary tasks, ensuring task continuity through single-satellite independence and multi-satellite relay collaboration, and triggering global rescheduling only when necessary, balancing real-time response and system stability; 3) It innovatively integrates a multi-objective genetic algorithm, simultaneously optimizing task completion rate, relay count, and scheduling disturbance amount in the initial scheduling and rescheduling stages. Efficient solutions are achieved through chromosome encoding and task-level genetic operations, overcoming the bottleneck of poor adaptability to dynamic tasks in traditional static scheduling or single rescheduling modes. Ultimately, it achieves multi-objective collaborative optimization of task completion, resource utilization, and scheduling robustness in complex and ever-changing observation scenarios.

[0073] The following describes specific embodiments of the present invention.

[0074] A dynamic scheduling method for multi-satellite collaborative observation missions is proposed. Addressing the practical application requirements of both fixed and temporary tasks, an intelligent scheduling mechanism is designed that balances task completion, resource utilization, and minimizing scheduling disturbances. The specific implementation of this dynamic scheduling method mainly consists of the following four parts:

[0075] I. Observation Task Sub-Time Period

[0076] Assume the satellite set is Fixed task set is Each task With a defined observation duration ,satellite For the task The observable window is ,T This refers to the mission scheduling cycle. To improve resource utilization and enhance scheduling flexibility, an observation mission can be completed by multiple satellites in succession, provided that a certain overlap in observation time is met, thus ensuring the continuity and integrity of the observations.

[0077] 1. Division of observation task sub-time periods

[0078] For an observation task Its observation time requirement is Divide it into several consecutive k Individual time period:

[0079] in, These are the segmentation points of the divided time period. This is the minimum overlap observation time required for relay observation. The value of can be determined by combining the satellite platform's attitude maneuverability, sensor sampling rate, and data fusion algorithm requirements, and is usually a system preset constant.

[0080] To standardize scheduling granularity and improve algorithm efficiency, discrete time slices are defined as follows: The length of each time slice is the minimum overlap observation time required for relay observation. Then the task Observation time It can be divided into continuous A time slice.

[0081] 2. Observation mission relay

[0082] To ensure continuity and data consistency in tasks completed by multiple satellites, tasks can be completed collaboratively by multiple satellites within adjacent time periods. However, observation tasks completed in relay by multiple satellites require a period of overlap, and a minimum observation overlap time must be met. t Assigned to satellite The observation time segment is Then the satellite Start time of observation Must meet ,in Indicates satellite Observation mission t Number of time slices.

[0083] like Figure 1 The image shows two satellites. and Relay to complete the observation mission The diagram illustrates the observation task. The observation time consists of two parts:

[0084] in, It is a satellite The observation time It is a satellite The observation time It is the minimum overlap observation time required for relay observation.

[0085] II. Initial Scheduling of Fixed Tasks

[0086] In the fixed task scheduling phase, the goal is to generate an initial allocation scheme for all fixed tasks that is resource-efficient, has strong task continuity, and yields excellent scheduling results, without considering temporary tasks. This phase employs the multi-objective genetic algorithm NSGA-II (Non-dominated Sorting Genetic Algorithm II) for scheduling optimization.

[0087] 1. Chromosome coding

[0088] Each chromosome consists of multiple gene segments, and each gene segment corresponds to a task allocation scheme:

[0089] Gene segment = [Task ID, Satellite ID, Start Time Slice, End Time Slice, Relay Flag, Relay Satellite ID, Overlap Time Slice].

[0090] In the relay flag, 0 indicates that there is no relay, and 1 indicates that there is a relay. If the relay flag is 0, the last two fields are set to empty.

[0091] like Figure 2 The diagram shown illustrates chromosome coding. Represents chromosome gene segments, among which This indicates that the first gene segment is [Task ID=1, Satellite ID=3, Start Time Slice=L1, End Time Slice=L4, Relay Observation Exists, Relay Satellite ID=7, Overlapping Time Slice=L4], This indicates that the (n-2)th gene segment is [Task ID=j, Satellite ID=6, Start Time Slice=L9, End Time Slice=L15, No relay observation,].

[0092] 2. Design of Multi-Objective Optimization Functions

[0093] Objective 1: Maximize task completion rate

[0094] This objective aims to ensure that all scheduled tasks are completed as much as possible, avoiding task loss due to insufficient time windows or resource conflicts. The objective is expressed as follows:

[0095] in, Indicates task Whether the task was not successfully scheduled (1 for failure, 0 for success). This objective is a hard constraint; individuals who fail to complete the task will be penalized for not fulfilling their fitness goals.

[0096] Objective 2: Minimize the number of task relays

[0097] Since mission relay increases data fusion costs and mission management complexity, it is necessary to minimize the number of times a mission is executed by multiple satellites in relay. The objective is expressed as:

[0098] in, Indicates task Whether to use a relay scheduling method. By minimizing this objective, more tasks are encouraged to be executed entirely by a single satellite, thereby improving scheduling efficiency and data consistency.

[0099] 3. Decoding and Fitness Evaluation

[0100] During the decoding and fitness evaluation process, the structured gene segments corresponding to each task in the chromosome are first analyzed. Following the task scheduling order, each task is checked to see if it can be allocated consecutive time slices within the observable window of its designated satellite, and to verify whether there are conflicts with other tasks. Conflict detection is achieved by maintaining a two-dimensional satellite-time slice occupancy matrix. If the corresponding position in the matrix is ​​unoccupied and within the observable window during the specified time period of the task, the scheduling is successful; otherwise, the task scheduling fails, and the chromosome is marked as an infeasible solution. For tasks using relay, it is further necessary to verify whether the relay satellite meets the condition of minimum overlapping time slices with the preceding satellite within its corresponding time window. If the scheduling scheme corresponding to the chromosome satisfies all constraints, its fitness value on both objective functions is calculated; otherwise, it is marked as an infeasible solution and a penalty fitness is assigned.

[0101] 4. Genetic manipulation

[0102] Genetic operations consist of three steps: selection, crossover, and mutation. The selection operation employs a binary tournament method based on non-dominated ordering and crowding distance, prioritizing the retention of non-dominated solutions from the current population to ensure solution diversity and the correctness of the evolutionary direction. The crossover operation is performed at the task level, exchanging several task gene segments between two parent chromosomes to ensure each task occurs only once, while maintaining the structural consistency of scheduling parameters. The mutation operation applies slight perturbations to gene segments on a single chromosome, including adjusting the task's start time slot, changing observable satellites, modifying relay flags, or altering the length of overlapping time slots, to explore a better solution space. The design of genetic operations fully incorporates the structural characteristics of task scheduling, ensuring the feasibility of individuals and the continuous improvement of solutions during evolution.

[0103] III. Temporary Tasks and Air Conditioning

[0104] For sudden, temporary observation tasks, priority should be given to scheduling them by inserting them into the remaining available satellite resources in the current scheduling plan.

[0105] 1. Satellite idle time slice detection

[0106] Based on the initial fixed task scheduling scheme, the set of idle time slots for each satellite within the scheduling cycle is calculated, i.e., continuous time slices not occupied by any fixed task. If the length of a continuous idle time slice is ≥2 (i.e., it contains at least two time slices), it is considered a usable insertion interval. The minimum length of the continuous idle time slice can be adjusted according to the minimum observation duration requirement of the temporary task. For example, a minimum number of schedulable time slices N_min can be set, and when the number of idle time slices is ≥N_min, it is considered a usable insertion interval. A structured index is created for the continuous idle time slots of each satellite, stored as (satellite ID, start time slice, end time slice) triples, and sorted by start time to form a timeline tree structure.

[0107] 2. Strategy for inserting temporary tasks

[0108] For temporary tasks The required observation time is The number of observation time slices for the temporary task is Each satellite For temporary tasks The observable window is .

[0109] When scheduling temporary tasks, priority is given to completing them independently on a single satellite. The idle time list of each satellite is then searched to determine if a continuous idle time segment exists with a length not less than [a certain value]. Furthermore, this time period falls entirely within the observable time window of the satellite for the mission. If the conditions are met, the temporary mission will be directly inserted into this time segment of the satellite and observed using a single-satellite independent execution method.

[0110] If the idle time slots of all satellites are insufficient to independently complete the continuous time slices required for the mission, a multi-satellite relay strategy is adopted. Taking a two-satellite relay as an example, the first step is to select a pair of satellites with partial time slice overlap ( s 1, s 2) If each of the two satellites has a period of idle time that meets some of the observation requirements, and the two time slices overlap by at least one time slice, and the total number of observation time slices of the two satellites, after deducting the overlap area, can meet the overall observation requirements of the mission. In this case, the temporary task is completed by two satellites relaying each other. The relay time slices can be calculated in ways such as, but not limited to, the following: Let L1 be the number of time slices corresponding to the idle period of satellite s1, L2 be the number of time slices corresponding to the idle period of satellite s2, and Lₒᵥ be the number of overlapping time slices. Then the total number of effective time slices is...

[0111] If none of the above insertion strategies can be completed, then a coordinated rescheduling of fixed tasks and temporary tasks will be performed.

[0112] IV. Coordinated Rescheduling of Fixed and Temporary Tasks

[0113] When temporary tasks cannot be scheduled using the insertion strategy (i.e., all single-star and double-star relay insertion schemes are infeasible), cooperative rescheduling is used for dynamic scheduling of temporary tasks. Temporary and fixed tasks are included in the scheduling scope, and the overall scheduling scheme is re-optimized using the aforementioned multi-objective genetic algorithm. Known methods such as setting an upper limit on the number of iterations, using a fast non-dominated sorting method, or reusing the initial Pareto front solution set as the initial population can be used to ensure the real-time performance of the rescheduling.

[0114] To minimize the impact on fixed task scheduling, an optimization function is added:

[0115] Objective 3: Minimize scheduling disturbance

[0116] Compared to the initial fixed mission scheduling scheme, the goal is to minimize the mission start time or the number of changes to the executing satellites.

[0117] in, Represents a fixed set of tasks. Indicates task t Has the satellite being executed changed? Indicates task t Has the start time changed? For and 1 indicates a change, and 0 indicates no change. By minimizing this objective, the stability of the scheduling scheme is maintained, and the impact on the original schedule is reduced. This is understandable. This can be replaced with the offset of the task start time relative to the initial schedule. and The impact of different types of disturbances can be balanced by weighting coefficients, which can be calibrated in conjunction with domain knowledge.

[0118] The main process of the dynamic scheduling method for this multi-satellite collaborative observation mission is summarized as follows:

[0119] Step 1: Discretize the scheduling period into uniform observation time slices based on the minimum overlapping observation time of the tasks;

[0120] Step 2: Use a multi-objective genetic algorithm to perform initial scheduling of fixed tasks and generate a task allocation scheme;

[0121] Step 3: Based on the fixed task scheduling scheme, calculate the set of idle time periods for each satellite within the scheduling cycle;

[0122] Step 4: Determine if the temporary task can be completed using the satellite's remaining time slice. If it can, proceed to Step 5; otherwise, proceed to Step 6.

[0123] Step 5: Arrange temporary tasks accordingly, and then proceed to Step 7.

[0124] Step 6: Use a multi-objective genetic algorithm to reschedule temporary and fixed tasks. After completion, proceed to Step 7.

[0125] Step 7: Output the final dynamic scheduling result.

[0126] Figure 3 The specific process of the dynamic scheduling method based on insertion and rescheduling is demonstrated.

[0127] In summary, this invention addresses the coexistence of fixed and temporary tasks in multi-satellite collaborative observation missions. Considering task completion and scheduling efficiency as optimization objectives, it proposes a dynamic scheduling method for multi-satellite collaborative observation missions. This method is based on insertion and rescheduling, and integrates a multi-objective genetic algorithm to ensure efficient insertion of temporary tasks and collaborative completion of the overall mission while satisfying the execution of fixed tasks.

[0128] Key innovations of this invention include: discretizing the task scheduling cycle based on the minimum overlapping observation time, and uniformly managing satellite resources and time windows; prioritizing the use of remaining satellite observation time slots for scheduling temporary tasks, triggering collaborative rescheduling upon failure, minimizing interference with the original fixed task plan, while ensuring the completion of temporary tasks. Its intelligent scheduling mechanism balances task completion, resource utilization, and minimizing scheduling disturbances.

[0129] Compared to existing static scheduling methods, this invention possesses stronger dynamic response and resource optimization capabilities. The proposed unified time-slice modeling and multi-satellite relay observation mechanism effectively improve resource utilization; through the strategy of "prioritizing insertion and rescheduling as a fallback," efficient dynamic scheduling of temporary tasks is achieved; and the use of a multi-objective genetic algorithm to optimize task allocation not only balances completion rate and scheduling stability but also significantly reduces interference with the original plan, making it suitable for complex and ever-changing real-world observation scenarios.

[0130] This invention also provides a storage medium for storing a computer program, which, when executed, performs at least the methods described above.

[0131] This invention also provides a control device, including a processor and a storage medium for storing a computer program; wherein the processor executes the computer program by performing at least the method described above.

[0132] This invention also provides a processor that executes a computer program, at least performing the methods described above.

[0133] The storage medium can be implemented by any type of non-volatile storage device, or a combination thereof. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc or CD-ROM; magnetic surface memory can be disk storage or magnetic tape storage. The storage media described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0134] In the several embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0135] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0136] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0137] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0138] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0139] The methods disclosed in the several method embodiments provided by this invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0140] The features disclosed in the several product embodiments provided by this invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0141] The features disclosed in the several method or device embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0142] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various equivalent substitutions or obvious modifications can be made without departing from the concept of the present invention, and all such modifications, achieving the same performance or application, should be considered within the scope of protection of the present invention.

Claims

1. A dynamic scheduling method for multi-satellite collaborative observation missions, characterized in that, Includes the following steps: S1: Task time slice discretization: The scheduling cycle is divided into discrete observation time slices according to the minimum overlapping observation time, so as to manage satellite resources and time windows in a unified manner; S2: Initial scheduling of fixed tasks: A multi-objective optimization algorithm is used to initially schedule fixed tasks, generating a task allocation scheme that takes into account both task completion rate and minimizing the number of relays. S3: Idle Resource Detection: Based on the results of fixed task scheduling, calculate the set of consecutive idle time slices for each satellite and build a structured index; S4: Dynamic scheduling of temporary tasks: Prioritize the use of interleaved scheduling: search for continuous time slots in satellite idle time slots that meet the observation requirements of temporary tasks, and execute them independently on a single satellite; If a single satellite cannot complete the task, a multi-satellite relay strategy is employed: tasks are allocated during overlapping idle periods of multiple satellites to ensure that the minimum overlap time slice requirement is met. When the insertion strategy fails, a cooperative rescheduling is triggered: the temporary task is merged with the fixed task, and the scheduling scheme is re-optimized through a multi-objective optimization algorithm, while minimizing the disturbance to the fixed task.

2. The dynamic scheduling method for multi-satellite collaborative observation missions as described in claim 1, characterized in that, In step S1: The observation task is divided into consecutive sub-time periods, the length of which is equal to the minimum overlapping observation time. When multiple satellites work in relay to perform a mission, the observation periods between satellites must overlap by at least one time slice to ensure data continuity and consistency.

3. The dynamic scheduling method for multi-satellite collaborative observation missions as described in claim 1 or 2, characterized in that, In step S2: The multi-objective optimization algorithm is a multi-objective genetic algorithm that uses chromosome encoding. Each chromosome contains multiple gene segments, and each gene segment corresponds to a task allocation scheme. The encoding parameters include task identifier, satellite identifier, start time slice, end time slice, relay flag, and associated relay satellite identifier.

4. The dynamic scheduling method for multi-satellite collaborative observation missions as described in claim 3, characterized in that, In step S2: The optimization objectives of multi-objective genetic algorithms include: Maximize the completion rate of fixed tasks and impose a fitness penalty on unscheduled tasks; Minimize the number of task relays and reduce data fusion costs; Genetic operations include selection based on non-dominated sorting and crowding distance, task-level gene segment crossover, and parameter perturbation variation.

5. The dynamic scheduling method for multi-satellite collaborative observation missions as described in any one of claims 1 to 2, characterized in that, In step S3: The structured index stores consecutive idle time periods in triplets of satellite identifier, start time slice, and end time slice, and sorts them by start time to form a timeline tree structure for fast retrieval of available gaps.

6. The dynamic scheduling method for multi-satellite collaborative observation missions as described in any one of claims 1 to 2, characterized in that, In step S4, during the air conditioning settings: The multi-star relay strategy is implemented as follows: Two satellites with overlapping idle time slots are selected, and a portion of the observation time slots are allocated to each. The total number of time slots in the overlapping area, after deducting the overlapping portion, must meet the total observation requirements of the temporary mission.

7. The dynamic scheduling method for multi-satellite collaborative observation missions as described in any one of claims 1 to 2, characterized in that, In the coordinated rescheduling of step S4: The new optimization objective is to minimize the scheduling disturbance, which is achieved in the following way: The number of satellite changes and start time changes for fixed tasks are statistically analyzed, and the weighted sum is used as a disturbance penalty to maintain the stability of the scheduling scheme.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the dynamic scheduling method for multi-satellite collaborative observation tasks as described in any one of claims 1 to 7.

Images