Strip coverage and improved multi-satellite swarm coordination regional target observation scheduling method

Abstract

The application provides a strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling method, and relates to the field of multi-satellite cooperative observation.In the application, firstly, a task conflict matrix is constructed by obtaining a to-be-observed task list and satellite resources;secondly, a strip coverage scheme and strip priority are generated by discretizing a regional target based on the task conflict matrix;thirdly, a multi-satellite cooperative task allocation scheme is generated by using an adaptive ant colony task allocation algorithm;then, an optimization target is generated based on the strip priority and the multi-satellite cooperative task allocation scheme, and a scheduling decision model is constructed by combining energy constraints, time window constraints and task unique execution constraints;finally, the scheduling decision model is solved by using an improved branch and bound algorithm to obtain a global optimal solution for deciding whether each satellite observes a task at different time or not.The application significantly improves the task execution efficiency and resource utilization rate of multi-satellite cooperative observation by improving the ant colony algorithm and the dynamic strip coverage strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of multi-satellite cooperative observation, and in particular to a strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling method. BACKGROUND

[0002] With the rapid development of space technology, multi-satellite cooperative observation has been widely used in many fields. As the number of satellites increases and the task requirements become increasingly complex and diverse, how to achieve efficient task allocation and reasonable scheduling to fully utilize the advantages of multi-satellite systems has become a key problem to be solved.

[0003] Different types of satellites have different observation capabilities and resource limitations, while various observation tasks have different time windows, priorities and observation requirements, and the method selection of task allocation is also different, which makes task allocation and scheduling extremely challenging.

[0004] In related technologies, the research on multi-satellite cooperative regional target observation scheduling has realized a variety of classic observation scheduling methods, mainly including: regional coverage optimization method for multi-satellite cooperative observation, large-scale task multi-satellite cooperative scheduling method based on neural network and meta-heuristic algorithm, satellite regional target scheduling planning based on deep reinforcement adaptive large neighborhood search algorithm, and multi-satellite cooperative observation scheduling based on heuristic algorithm. Among them, the ant colony algorithm is a commonly used heuristic algorithm.

[0005] However, the traditional ant colony algorithm is prone to destroy the structure of high-quality solutions when jumping out of local optimum, thereby affecting the efficiency and rationality of multi-satellite cooperative observation. SUMMARY

[0006] (I) Technical problems solved

[0007] In view of the deficiencies of the prior art, the present application provides a strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling method, which solves the technical problem that the traditional ant colony algorithm is prone to destroy the structure of high-quality solutions when jumping out of local optimum, thereby affecting the efficiency and rationality of multi-satellite cooperative observation.

[0008] (II) Technical solutions

[0009] To achieve the above purposes, the present application is implemented by the following technical solutions:

[0010] A strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling method, comprising:

[0011] Obtaining the list of tasks to be observed and satellite resources of regional targets, and constructing a task conflict matrix;

[0012] discretize the area target based on the task conflict matrix, generate a strip coverage scheme of each satellite observing different tasks, and obtain a strip priority based on the number of strip coverage grids and the task priority;

[0013] generate a plurality of multi-satellite cooperative task allocation schemes by using an adaptive ant colony task allocation algorithm; the adaptive ant colony task allocation algorithm comprises a state transition mechanism based on disturbance of the strip priority and a pheromone update mechanism based on a double-threshold dynamic adjustment strategy;

[0014] generate an optimization target based on the strip priority and the multi-satellite cooperative task allocation scheme, and construct a scheduling decision model in combination with an energy constraint, a time window constraint and a task unique execution constraint;

[0015] solve the scheduling decision model by using an improved branch and bound algorithm, and obtain a global optimal solution for decision-making of whether each satellite observes a task at different time by introducing a dynamic relaxation factor.

[0016] Preferably, the method comprises:

[0017] a dynamic adjustment trigger function is predefined, and when it is detected that cloud coverage change of the area target exceeds a preset proportion or task priority fluctuation exceeds a preset threshold, an adaptive adjustment mechanism is triggered to update the global optimal solution by using a local rescheduling strategy or a pheromone dynamic resetting strategy after quantifying adjustment urgency of different task observation areas.

[0018] Preferably, the construction process of the task conflict matrix comprises:

[0019] obtain time overlap information based on observation time windows of different tasks;

[0020] obtain resource conflict information based on satellite resources required by different tasks;

[0021] calculate a task conflict matrix between any two tasks based on the time overlap information and the resource conflict information.

[0022] Preferably, the discretization of the area target based on the task conflict matrix, the generation of a strip coverage scheme of each satellite observing different tasks, and the obtaining of a strip priority based on the number of strip coverage grids and the task priority comprise:

[0023] initially divide the area target into coarse-grained grids according to a first interval;

[0024] Based on the task conflict matrix, a task conflict density of different task areas is calculated, if the task conflict density is higher than a preset density threshold, the coarse-grained grid is preferentially divided into a fine-grained grid according to a second interval, and a local strip is generated in combination with satellite transit information; otherwise, an extended strip is generated based on the coarse-grained grid or based on the coarse-grained grid and the fine-grained grid;

[0025] Each local strip and extended strip is traversed, and the number of grid covered by each strip and the task priority of the corresponding task are obtained to obtain the strip priority of each strip.

[0026] Preferably, the state transition mechanism based on the strip priority disturbance is represented as:

[0027]

[0028]

[0029] Wherein, is the probability of the kth ant pairing satellite j and task i; represents the pheromone concentration between satellite j and task i; is a pheromone importance factor, is a heuristic information weight; is the strip priority of task i observed by satellite j; is a heuristic function between satellite j and task i, , , respectively, the energy level, attitude control margin, and storage capacity of satellite j; represents any satellite m in the satellite set S that can observe task i; represents the pheromone concentration between satellite m and task i; represents the heuristic function between satellite m and task i; represents the strip priority of task i observed by satellite m;

[0030] Preferably, the pheromone update mechanism based on the double-threshold dynamic adjustment strategy is represented as:

[0031]

[0032]

[0033] Wherein, when the global optimal multi-satellite cooperative task allocation scheme is not updated for g consecutive times in the iteration process, the update formula of is triggered; is the pheromone concentration left by the ant between satellite j and task i; represents an update symbol; The volatility coefficient is... This is the initial volatility coefficient; This represents the current iteration number. This represents the maximum number of iterations. It is an exponential function.

[0034] Preferably, the construction process of the scheduling decision model specifically includes:

[0035] Based on the pairing relationship between satellites and tasks in the multi-satellite collaborative task allocation scheme, the product of the power consumption parameter, execution time and decision variable used to characterize whether each satellite observes the corresponding task is calculated to generate an energy consumption sub-target.

[0036] Based on the product of the penalty coefficient for each satellite's unobserved corresponding task, the strip priority, and the decision variables, a task priority penalty sub-objective is generated.

[0037] The optimization objective is constructed by minimizing the sum of the energy consumption sub-objective and the mission priority penalty sub-objective; wherein the power consumption parameter is determined by the satellite's energy level.

[0038] Based on the product of the power consumption parameters, execution time, and decision variables of each satellite observation corresponding task, it is determined whether the product is less than or equal to the product of the energy level and the energy utilization efficiency coefficient, which serves as the energy constraint; wherein the energy utilization efficiency coefficient is determined by the satellite's attitude control margin.

[0039] Based on the decision variables, it is determined whether they satisfy the observability function, which serves as the time window constraint; wherein the observability function is determined by the latitude and longitude information of the regional target, the orbital parameters of the satellite, and the mission time window.

[0040] Based on the decision variables, it is determined whether each task satisfies the condition of being observed by at most one satellite once within the scheduling period, which serves as the unique execution constraint for the task.

[0041] A multi-satellite collaborative regional target observation scheduling system with strip coverage and improved ant colony features, comprising:

[0042] The acquisition module is used to acquire the list of tasks to be observed and satellite resources for regional targets, and to construct a task conflict matrix;

[0043] The processing module is used to discretize the regional targets based on the task conflict matrix, generate strip coverage schemes for different tasks of each satellite observation, and obtain strip priority based on the number of strip coverage grids and task priority.

[0044] The generation module is used to generate several multi-star collaborative task allocation schemes using an adaptive ant colony task allocation algorithm; the adaptive ant colony task allocation algorithm includes a state transition mechanism based on the stripe priority perturbation and a pheromone update mechanism based on a dual-threshold dynamic adjustment strategy.

[0045] The construction module is used to generate optimization objectives based on the strip priority and multi-star collaborative task allocation scheme, and to construct a scheduling decision model by combining energy constraints, time window constraints and task unique execution constraints.

[0046] The solution module is used to solve the scheduling decision model using an improved branch and bound algorithm. By introducing a dynamic relaxation factor, it obtains the global optimal solution for deciding whether each satellite should perform an observation mission at different times.

[0047] Preferably, the multi-satellite collaborative regional target observation and scheduling system includes:

[0048] The adjustment module is used to predefine a dynamic adjustment trigger function. When the cloud coverage change of the target area exceeds a preset ratio or the task priority fluctuation exceeds a preset threshold, the adaptive adjustment mechanism is triggered. After quantifying the adjustment urgency of different task observation areas, the global optimal solution is updated by adopting a local rescheduling strategy or a pheromone dynamic reset strategy.

[0049] A storage medium storing a computer program for strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling, wherein the computer program causes a computer to execute the multi-satellite cooperative regional target observation scheduling method as described above.

[0050] An electronic device, comprising:

[0051] One or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing multi-satellite collaborative regional target observation scheduling as described above.

[0052] (III) Beneficial Effects

[0053] This invention provides a multi-satellite collaborative regional target observation scheduling method with strip coverage and improved ant colony techniques. Compared with existing technologies, it has the following advantages:

[0054] In this invention, firstly, a list of tasks to be observed and satellite resources are acquired, and a task conflict matrix is ​​constructed. Secondly, based on the task conflict matrix, regional targets are discretized to generate strip coverage schemes and strip priorities. Thirdly, an adaptive ant colony task allocation algorithm is used to generate a multi-satellite collaborative task allocation scheme. This algorithm is based on the state transition mechanism of the strip priority perturbation and the pheromone update mechanism based on a dual-threshold dynamic adjustment strategy. Next, an optimization objective is generated based on the strip priority and the multi-satellite collaborative task allocation scheme. Combining energy constraints, time window constraints, and unique task execution constraints, a scheduling decision model is constructed. Finally, an improved branch and bound algorithm is used to solve the scheduling decision model to obtain the globally optimal solution for deciding whether each satellite should conduct an observation task at different times. This invention significantly improves the task execution efficiency and resource utilization of multi-satellite collaborative observation by improving the ant colony algorithm and the dynamic strip coverage strategy. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A block diagram illustrating a multi-satellite collaborative regional target observation scheduling method with strip coverage and improved ant colony technology, provided in an embodiment of the present invention.

[0057] Figure 2 A block diagram illustrating another strip coverage and improved ant colony multi-satellite collaborative regional target observation scheduling method provided in this embodiment of the invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] This application provides a multi-satellite collaborative regional target observation scheduling method with strip coverage and improved ant colony, which solves the technical problem that traditional ant colony algorithms easily destroy the high-quality solution structure when escaping local optima, thus affecting the efficiency and rationality of multi-satellite collaborative observation.

[0060] The technical solution in this application is to solve the above-mentioned technical problems, and the general idea is as follows:

[0061] This invention proposes a multi-satellite collaborative observation task allocation and scheduling system based on stripe coverage and an improved ant colony algorithm, wherein:

[0062] A state transition probability formula is introduced into the adaptive ant colony allocation, which integrates satellite resources and mission priorities, and achieves dynamic control through a dual-threshold pheromone update mechanism.

[0063] A dynamic relaxation factor is introduced during the observation scheduling optimization process. The weight of the penalty term is adjusted to enhance the execution rigidity of high-priority tasks as the iteration progresses. This mechanism ensures that the scheduling scheme prioritizes coverage accuracy in high-priority areas while meeting satellite energy, orbit, and sensor constraints. Simultaneously, it extends the satellite's operational cycle through energy consumption optimization. Compared to traditional scheduling methods, this improves resource utilization and ensures the time window satisfaction rate of high-priority tasks.

[0064] Furthermore, by constructing an objective function that includes energy constraints and dynamically adjusting the stages to achieve hierarchical optimization through trigger functions and priority matrices, the shortcomings of the local optimum exit mechanism in the original technology are resolved, resource utilization is improved, and the response efficiency of high-priority tasks is ensured.

[0065] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0066] Example 1:

[0067] like Figure 1 As shown, this embodiment of the invention provides a multi-satellite collaborative regional target observation scheduling method with strip coverage and improved ant colony, including:

[0068] S1. Obtain the list of tasks to be observed and satellite resources for regional targets, and construct a task conflict matrix;

[0069] S2. Discretize the regional targets based on the task conflict matrix to generate strip coverage schemes for different tasks of each satellite observation, and obtain strip priority based on the number of strip coverage grids and task priority;

[0070] S3. Several multi-star collaborative task allocation schemes are generated using an adaptive ant colony task allocation algorithm; the adaptive ant colony task allocation algorithm includes a state transition mechanism based on the stripe priority perturbation and a pheromone update mechanism based on a dual-threshold dynamic adjustment strategy.

[0071] S4. Based on the strip priority and multi-star collaborative task allocation scheme, generate optimization objectives, and combine energy constraints, time window constraints and unique task execution constraints to construct a scheduling decision model;

[0072] S5. Solve the scheduling decision model using the improved branch and bound algorithm. By introducing a dynamic relaxation factor, obtain the global optimal solution for deciding whether each satellite should perform an observation task at different times.

[0073] The embodiments of the present invention significantly improve the task execution efficiency and resource utilization of multi-satellite collaborative observation by improving the ant colony algorithm and dynamic stripe coverage strategy.

[0074] In an alternative implementation, such as Figure 2 As shown, the multi-satellite collaborative regional target observation scheduling method further includes:

[0075] S6. A dynamic adjustment trigger function is predefined. When the cloud coverage change of the target area exceeds a preset ratio or the task priority fluctuation exceeds a preset threshold, an adaptive adjustment mechanism is triggered. After quantifying the adjustment urgency of different task observation areas, a local rescheduling strategy or a pheromone dynamic reset strategy is adopted to update the global optimal solution.

[0076] In dynamic adjustment scenarios, the embodiments of the present invention can complete the scheme reconstruction in a short time, improve the task execution rate and emergency response efficiency; through energy constraint optimization and priority weighting mechanism, not only is the satellite working cycle extended, but the coverage accuracy of high priority areas is also ensured, realizing the dual optimization of task allocation rationality and scheduling dynamic adaptability.

[0077] The following will detail each step of the above solution:

[0078] In step S1, the list of tasks to be observed and satellite resources of the regional targets are obtained, and a task conflict matrix is ​​constructed.

[0079] After the ground control center receives the user's request for observation of regional targets, this step generates a list of tasks to be observed for the regional targets. Where I represents the number of missions. Simultaneously, satellite data is acquired based on the regional target observation mission cycle. The transit area target information includes parameters such as the mission's spatial location, time window, priority, and sensor type, where N is the number of satellites.

[0080] It should be noted that the definition of "regional target" above is: a target is an area to be observed submitted by the user. If it can be covered by an imaging strip of the satellite, the target is called a point target; otherwise, it is called a regional target. The definition of "passage" above is: the time period during which the satellite flies over a regional target.

[0081] The next step is to construct the task conflict matrix. The relevant steps are as follows:

[0082] S10. Obtain time overlap information based on the observation time windows of different tasks;

[0083] S20. Obtain resource conflict information based on the satellite resources requested for different missions;

[0084] S30. Based on the time overlap information and resource conflict information, calculate the task conflict matrix between any two tasks.

[0085] Specifically, the task conflict matrix is ​​represented as C, where any element This indicates that task i and task k have a time or resource conflict; otherwise, the value is zero. The calculation formula is:

[0086]

[0087] in, Represents the observation time window for task i. Observation time window of task k Whether they overlap, a value of 1 indicates overlap, and a value of 0 indicates no overlap; Indicates the satellite resources that task i needs to request. Satellite resources required for mission k Whether there is a conflict, a value of 1 indicates a conflict, and a value of 0 indicates no conflict.

[0088] In step S2, the regional targets are discretized based on the task conflict matrix to generate strip coverage schemes for different tasks of each satellite observation, and strip priority is obtained based on the number of strip coverage grids and task priority.

[0089] In an optional implementation, this step specifically includes:

[0090] S21. The target area is initially divided into a coarse-grained grid according to the first interval.

[0091] This can be done according to the first interval. Based on the latitude and longitude coordinates of the regional targets, they are divided into uniform coarse-grained grids, with each coarse-grained grid corresponding to a geographic region unit.

[0092] S22. Based on the task conflict matrix, calculate the task conflict density of different task regions. If the task conflict density is higher than a preset density threshold, preferentially divide the coarse-grained grid into a fine-grained grid according to the second interval, and generate local stripes in combination with satellite transit information; otherwise, generate extended stripes based only on the coarse-grained grid or based on the coarse-grained grid and the fine-grained grid.

[0093] The task conflict density in the different task regions mentioned above can be expressed as:

[0094]

[0095] Understandably, when the task conflict density is high, the strip coverage scheme directly generated based on the above coarse-grained grid may result in insufficient coverage accuracy. Therefore, a density threshold of 0.6 is set here. If the value is greater than 0.6, then the second interval will be used first. The coarse-grained grid is further divided into fine-grained grids, and local stripes are generated by combining satellite overpass information; if When the value is ≤0.6, the extended strips can be generated based on the coarse-grained grid only or on both the coarse-grained and fine-grained grids as needed to improve coverage diversity.

[0096] Specifically, the generated stripes can be further filtered quickly to exclude those that violate satellite energy requirements (where the remaining power is less than 1.15 times the mission power consumption) and those involving multiple sensors operating simultaneously. It should be noted that the aforementioned value of 1.15 is an energy margin coefficient, representing 115% of the actual mission power consumption used to determine whether the satellite's current remaining energy budget is exceeded. In other embodiments, this can be adjusted by the technician according to the actual situation.

[0097] S23. Traverse each local stripe and extended stripe, and obtain the stripe priority of each stripe by matching the number of grids covered by each stripe with the task priority of its corresponding task.

[0098] For any given band, the above band priority can be expressed as:

[0099]

[0100] in, Specifically, it indicates the priority corresponding to the stripe of observation task i for satellite j.

[0101] In step S3, an adaptive ant colony task allocation algorithm is used to generate several multi-star collaborative task allocation schemes; the adaptive ant colony task allocation algorithm includes a state transition mechanism based on the strip priority perturbation and a pheromone update mechanism based on a dual-threshold dynamic adjustment strategy.

[0102] This step uses the Max-MinAntSystem (MMAS) framework and is implemented through the following sub-steps:

[0103] 1) Initialize the pheromone matrix and set the upper and lower limits of pheromone concentration;

[0104] 2) Each ant selects a task-satellite pair based on the state transition probability;

[0105] 3) After all ants have completed their path search, only the globally optimal path is allowed to update the pheromone;

[0106] 4) If no better solution is found after g consecutive iterations, a local search mechanism is triggered, randomly perturbing the pheromone matrix to escape the local optimum.

[0107] Compared with related technologies, the embodiments of the present invention have optimized at least the state transition probability of the algorithm and how to escape local optima. The two improvements correspond to the state transition mechanism based on the strip priority perturbation and the pheromone update mechanism based on the dual threshold dynamic adjustment strategy, respectively.

[0108] Specifically:

[0109] The state transition mechanism based on the strip priority perturbation is expressed as follows:

[0110]

[0111]

[0112] in, Let $\mathbf$ be the probability that the $k$-th ant will pair satellite $j$ with task $i$. This indicates the pheromone concentration between satellite j and mission i; For pheromone importance factors, Weights based on heuristic information; For satellite j, the strip priority of observation mission i; Let j be the heuristic function between satellite j and mission i. , , These are the energy level, attitude control margin, and storage capacity of satellite j, respectively. Let m represent any one satellite in the set S of all satellites that can observe mission i; This indicates the pheromone concentration between satellite m and mission i; A heuristic function representing the relationship between satellite m and mission i; This indicates the strip priority of satellite m observation task i.

[0113] It should be noted that the state transition mechanism proposed in this embodiment of the invention integrates satellite resource availability and priority weights, enabling ants to prioritize task-satellite matching pairs with high resource matching degree and high priority during the search, thereby improving the rationality of the allocation scheme.

[0114] The pheromone update mechanism based on the dual-threshold dynamic adjustment strategy is expressed as follows:

[0115]

[0116]

[0117] Specifically, when the globally optimal multi-star collaborative task allocation scheme is not updated for g consecutive times during the iteration process, it is triggered. The update formula; The concentration of pheromones left by the ant between satellite j and mission i; Indicates the update symbol; The volatility coefficient is... This is the initial volatility coefficient; This represents the current iteration number. This represents the maximum number of iterations. It is an exponential function.

[0118] It should be noted that the "dual thresholds" refer to the trigger threshold for changes in the command and control coefficient and the dynamic attenuation threshold. The trigger threshold corresponds to parameter g mentioned above, and the dynamic attenuation threshold corresponds to the parameter g mentioned above. Clearly, the pheromone update mechanism provided in this embodiment of the invention maintains high exploration capability in the early stages of the algorithm and enhances development capability in the later stages by using a volatile coefficient that decreases with the number of iterations, thus avoiding premature entrapment in local optima.

[0119] In step S4, an optimization objective is generated based on the strip priority and multi-star collaborative task allocation scheme. Combined with energy constraints, time window constraints and unique task execution constraints, a scheduling decision model is constructed.

[0120] In an optional implementation, the process of constructing the scheduling decision model specifically includes:

[0121] S100. Based on the pairing relationship between satellites and tasks in the multi-satellite collaborative task allocation scheme, solve for the product of the power consumption parameter, execution time, and decision variables used to characterize whether each satellite observes the task at different times for each satellite to observe the corresponding task, and generate an energy consumption sub-objective; generate a task priority penalty sub-objective based on the product of the penalty coefficient, strip priority, and decision variables for each satellite not observing the corresponding task; minimize the sum of the energy consumption sub-objective and the task priority penalty sub-objective to construct the optimization objective; wherein the power consumption parameter is determined by the energy level of the satellite.

[0122] S200. Based on the product of the power consumption parameters, execution time, and decision variables of each satellite observation corresponding task, determine whether it is less than or equal to the product of the energy level and the energy utilization efficiency coefficient, and use it as the energy constraint; wherein the energy utilization efficiency coefficient is determined by the satellite's attitude control margin.

[0123] S300. Based on the decision variables, determine whether they satisfy the observability function, which serves as the time window constraint; wherein the observability function is determined by the latitude and longitude information of the regional target, the orbital parameters of the satellite, and the mission time window.

[0124] S400. Based on the decision variables, determine whether each task satisfies the condition of being observed by at most one satellite once within the scheduling period, as the unique execution constraint of the task.

[0125] Specifically, the scheduling decision model includes:

[0126] Objective function:

[0127]

[0128] Where min is the minimization function; t represents time; and T represents the scheduling cycle length. For the power consumption parameters of satellite j for observation mission i, The execution duration of satellite j's observation mission i; Let be the decision variable to be solved to characterize whether satellite j observes mission i at time t. Let 1 represent observation and 0 represent no observation. is the penalty coefficient when satellite j does not observe mission i.

[0129] This objective function balances energy consumption with task priority penalties to ensure that high-priority stripes are executed first, while optimizing energy allocation.

[0130] constraint:

[0131] The energy constraint expression is:

[0132]

[0133] in, The energy level of satellite j; Let be the energy efficiency coefficient of satellite j, and be the attitude control margin of satellite j. Positive correlation, used to reflect the impact of attitude adjustment on energy consumption.

[0134] Time window constraints are achieved through observability functions. Implementation, represented as:

[0135]

[0136] in, This represents the position vector of satellite j at time t; Let be the ground normal vector of the region corresponding to task i. The minimum included angle threshold required for satellite j observation; Let be the start and end times of the observation time window for task i. These are the start time and the end time, respectively.

[0137] The unique execution constraint for a task is represented as:

[0138]

[0139] This constraint prevents the same task from being executed simultaneously by multiple satellites in a single scheduling cycle.

[0140] In step S5, the improved branch and bound algorithm is used to solve the scheduling decision model. By introducing a dynamic relaxation factor, the global optimal solution for deciding whether each satellite should perform an observation task at different times is obtained.

[0141] This step introduces a dynamic relaxation factor when solving the model using an improved branch-and-bound algorithm:

[0142]

[0143] in, Let be the relaxation factor at time t, and be the initial relaxation factor. , is the value of the relaxation factor at the initial moment (or baseline state), which serves as the basis for dynamic adjustment; T is an upper limit of a time scale, which in the satellite mission scheduling scenario is the total duration of the entire mission planning (i.e., the length of the scheduling cycle).

[0144] It should be noted that the branch-and-bound algorithm proposed in this embodiment adjusts the weight of the penalty term, enhancing the execution rigidity of high-priority tasks as the iteration progresses. Specifically, the penalty for "time window and accuracy of high-priority tasks" is increased more severely and faster, forming execution rigidity; adaptive pressure is applied to energy / attitude / storage violations, driving energy conservation and feasibility. This mechanism enables the scheduling scheme to prioritize coverage accuracy of high-priority areas while meeting satellite energy, orbit, and sensor constraints, and extends the satellite's working cycle through energy consumption optimization. Compared with traditional scheduling methods, this can improve resource utilization and ensure the time window satisfaction rate of high-priority tasks.

[0145] In step S6, a dynamic adjustment trigger function is predefined. When the cloud coverage change of the target area exceeds a preset ratio or the task priority fluctuation exceeds a preset threshold, an adaptive adjustment mechanism is triggered. After quantifying the adjustment urgency of different task observation areas, a local rescheduling strategy or a pheromone dynamic reset strategy is adopted to update the global optimal solution.

[0146] Considering that task priorities change during observation and are not static, for example, changes may arise from: 1) sudden events (earthquakes, floods, forest fires, etc.), causing a sudden increase in task priority; 2) users issuing urgent observation requests, leading to an increase in priority; 3) task cancellation or postponement: originally planned high-priority tasks are cancelled, and their priority drops to zero. Furthermore, considering that changes in cloud cover over the target area may also affect the task observation process, this embodiment of the invention further performs dynamic adjustment and feedback operations.

[0147] Specifically:

[0148] The system needs to monitor environmental changes and mission fluctuations in real time while the satellite is in orbit. When a change in regional target cloud coverage exceeds 30% or a mission priority fluctuation exceeds a preset threshold, an adaptive adjustment mechanism is triggered. The predefined dynamic adjustment trigger function is as follows:

[0149]

[0150] in, This represents the current change in cloud coverage. This serves as the initial cloud coverage baseline. This represents the fluctuation in task priority. This is the initial priority baseline value. And, when... ( When the threshold is set to 0.3 (the default value is 0.3), the adaptive adjustment mechanism is triggered.

[0151] During the adjustment process, an adjustment priority matrix is ​​first constructed based on the strip coverage scheme and the multi-satellite collaborative task allocation scheme. The priority of the mission area observed by satellite j is as follows:

[0152]

[0153] in, The fluctuation of the strip priority of satellite j observation task i (i.e., the fluctuation of task priority). The cloud cover impact factor is 0.3 when the cloud cover in the mission area observed by satellite j exceeds 70%, and 1 otherwise.

[0154] Understandably, the above matrix quantifies the urgency of adjustments for each strip by integrating priority fluctuations and the impact of cloud coverage.

[0155] Furthermore, after quantifying the urgency of adjustments for different bands, the adjustment strategy adopts a tiered optimization mechanism, specifically including:

[0156] (1) For high-priority adjustment items ( ), execute the local rescheduling strategy:

[0157]

[0158] As mentioned above, Let be the decision variable to be solved, characterizing whether or not satellite j observes mission i at time t. Correspondingly, in the above formula... These are the adjusted scheduling decision variables; For adjustment coefficient ( ), These are the decision variables for the new candidate scheduling scheme.

[0159] Furthermore, the adjustments need to be verified by energy constraints and time window constraints, which will not be elaborated here.

[0160] (2) For low-priority adjustment items ( A dynamic pheromone reset strategy is adopted:

[0161]

[0162] As mentioned above, This represents the pheromone concentration between satellite j and mission i. Correspondingly, in the above formula... The adjusted pheromone concentration between satellite j and mission i. Reset coefficients This operation guides subsequent ant colony searches towards the redistribution of high-priority tasks by reducing the pheromone intensity of non-urgent adjustment items, while avoiding the destruction of high-quality solution structures.

[0163] Furthermore, the dynamic adjustment process also needs to be linked with the optimization of the neighborhood solution space. When the adjustment causes a conflict in the task time window, it can be overcome by executing the time window shifting algorithm, which will not be elaborated here.

[0164] Thus, this embodiment of the invention completes the entire process of the multi-satellite collaborative regional target observation scheduling method with strip coverage and improved ant colony.

[0165] Example 2:

[0166] This invention provides a multi-satellite collaborative regional target observation scheduling system with strip coverage and improved ant colony, comprising:

[0167] The acquisition module is used to acquire the list of tasks to be observed and satellite resources for regional targets, and to construct a task conflict matrix;

[0168] The processing module is used to discretize the regional targets based on the task conflict matrix, generate strip coverage schemes for different tasks of each satellite observation, and obtain strip priority based on the number of strip coverage grids and task priority.

[0169] The generation module is used to generate several multi-star collaborative task allocation schemes using an adaptive ant colony task allocation algorithm; the adaptive ant colony task allocation algorithm includes a state transition mechanism based on the stripe priority perturbation and a pheromone update mechanism based on a dual-threshold dynamic adjustment strategy.

[0170] The construction module is used to generate optimization objectives based on the strip priority and multi-star collaborative task allocation scheme, and to construct a scheduling decision model by combining energy constraints, time window constraints and task unique execution constraints.

[0171] The solution module is used to solve the scheduling decision model using an improved branch and bound algorithm. By introducing a dynamic relaxation factor, it obtains the global optimal solution for deciding whether each satellite should perform an observation mission at different times.

[0172] In an optional implementation, the multi-satellite collaborative regional target observation scheduling system further includes:

[0173] The adjustment module is used to predefine a dynamic adjustment trigger function. When the cloud coverage change of the target area exceeds a preset ratio or the task priority fluctuation exceeds a preset threshold, the adaptive adjustment mechanism is triggered. After quantifying the adjustment urgency of different task observation areas, the global optimal solution is updated by adopting a local rescheduling strategy or a pheromone dynamic reset strategy.

[0174] Example 3:

[0175] This invention provides a storage medium storing a computer program for strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling, wherein the computer program causes a computer to execute the multi-satellite cooperative regional target observation scheduling method as described in Embodiment 1.

[0176] Example 4:

[0177] This invention provides an electronic device, comprising:

[0178] One or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing a multi-satellite cooperative regional target observation scheduling method as described in Example 1.

[0179] It is understood that the strip coverage and improved ant colony multi-satellite collaborative regional target observation scheduling system, storage medium and electronic device provided in the embodiments of the present invention correspond to the strip coverage and improved ant colony multi-satellite collaborative regional target observation scheduling method provided in the embodiments of the present invention. The explanation, examples and beneficial effects of the relevant contents can be referred to the corresponding parts of the method, and will not be repeated here.

[0180] In summary, compared with existing technologies, it has the following beneficial effects:

[0181] 1. The embodiments of the present invention significantly improve the task execution efficiency and resource utilization of multi-satellite collaborative observation by improving the ant colony algorithm and dynamic stripe coverage strategy.

[0182] 2. In dynamic adjustment scenarios, the embodiments of the present invention can complete the scheme reconstruction in a short time, improve the task execution rate and emergency response efficiency; through energy constraint optimization and priority weighting mechanism, not only is the satellite working cycle extended, but the coverage accuracy of high priority areas is also ensured, realizing the dual optimization of task allocation rationality and scheduling dynamic adaptability.

[0183] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0184] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for scheduling multi-satellite collaborative regional target observation using strip coverage and improved ant colony methods, characterized in that, include: Obtain the list of tasks to be observed and satellite resources for regional targets, and construct a task conflict matrix; Based on the task conflict matrix, the regional targets are discretized to generate strip coverage schemes for different tasks of each satellite observation, and strip priority is obtained based on the number of strip coverage grids and task priority. Several multi-star collaborative task allocation schemes are generated using an adaptive ant colony task allocation algorithm. The adaptive ant colony task allocation algorithm includes a state transition mechanism based on stripe priority perturbation and a pheromone update mechanism based on a dual-threshold dynamic adjustment strategy. Based on the strip priority and multi-star collaborative task allocation scheme, an optimization target is generated, and a scheduling decision model is constructed by combining energy constraints, time window constraints and unique task execution constraints. The scheduling decision model is solved using an improved branch and bound algorithm. By introducing a dynamic relaxation factor, the global optimal solution for deciding whether each satellite should perform an observation task at different times is obtained. The state transition mechanism based on strip priority perturbation is expressed as follows: in, Let $\mathbf$ be the probability that the $k$-th ant will pair satellite $j$ with task $i$. This indicates the pheromone concentration between satellite j and mission i; For pheromone importance factors, Weights for heuristic information; For satellite j, the strip priority of observation mission i; Let j be the heuristic function between satellite j and mission i. , , These are the energy level, attitude control margin, and storage capacity of satellite j, respectively. Let m represent any one satellite in the set S of all satellites that can observe mission i; This indicates the pheromone concentration between satellite m and mission i; A heuristic function representing the relationship between satellite m and mission i; Indicates the strip priority of observation task i for satellite m; and / or The pheromone update mechanism based on the dual-threshold dynamic adjustment strategy is expressed as follows: Specifically, when the globally optimal multi-star collaborative task allocation scheme is not updated for g consecutive times during the iteration process, it is triggered. The update formula; The concentration of pheromones left by the ant between satellite j and mission i; Indicates the update symbol; The volatility coefficient is... This is the initial volatility coefficient; This represents the current iteration number. This represents the maximum number of iterations. It is an exponential function.

2. The multi-satellite collaborative regional target observation scheduling method as described in claim 1, characterized in that, include: A predefined dynamic adjustment trigger function is used to trigger an adaptive adjustment mechanism when the cloud coverage change of the target area exceeds a preset ratio or the task priority fluctuation exceeds a preset threshold. After quantifying the urgency of adjustment in different task observation areas, a local rescheduling strategy or a pheromone dynamic reset strategy is adopted to update the global optimal solution.

3. The multi-satellite collaborative regional target observation scheduling method as described in claim 1, characterized in that, The process of constructing the task conflict matrix includes: Based on the observation time windows of different tasks, obtain time overlap information; Based on the satellite resources requested by different missions, obtain resource conflict information; Based on the time overlap information and resource conflict information, calculate the task conflict matrix between any two tasks.

4. The multi-satellite collaborative regional target observation scheduling method as described in claim 1, characterized in that, The process of discretizing the regional targets based on the task conflict matrix to generate strip coverage schemes for different tasks observed by each satellite, and obtaining strip priority based on the number of strip coverage grids and task priority, includes: The target area is initially divided into a coarse-grained grid according to the first interval; Based on the task conflict matrix, the task conflict density of different task regions is calculated. If the task conflict density is higher than a preset density threshold, the coarse-grained grid is preferentially divided into a fine-grained grid according to the second interval, and a local strip is generated in combination with the satellite transit information; otherwise, an extended strip is generated only based on the coarse-grained grid or based on the coarse-grained grid and the fine-grained grid. Traverse each local stripe and extended stripe, and obtain the stripe priority of each stripe by matching the number of grids covered by each stripe with the task priority of its corresponding task.

5. The multi-satellite collaborative regional target observation scheduling method as described in claim 1, characterized in that, The construction process of the scheduling decision model specifically includes: Based on the pairing relationship between satellites and tasks in the multi-satellite collaborative task allocation scheme, the product of the power consumption parameter, execution time and decision variable used to characterize whether each satellite observes the corresponding task is calculated to generate an energy consumption sub-target. Based on the product of the penalty coefficient for each satellite's unobserved corresponding task, the strip priority, and the decision variables, a task priority penalty sub-objective is generated. The optimization objective is constructed by minimizing the sum of the energy consumption sub-objective and the mission priority penalty sub-objective; wherein the power consumption parameter is determined by the satellite's energy level. Based on the product of the power consumption parameters, execution time, and decision variables of each satellite observation corresponding task, it is determined whether the product is less than or equal to the product of the energy level and the energy utilization efficiency coefficient, which serves as the energy constraint; wherein the energy utilization efficiency coefficient is determined by the satellite's attitude control margin. Based on the decision variables, it is determined whether they satisfy the observability function, which serves as the time window constraint; wherein the observability function is determined by the latitude and longitude information of the regional target, the orbital parameters of the satellite, and the mission time window. Based on the decision variables, it is determined whether each task satisfies the condition of being observed by at most one satellite once within the scheduling period, which serves as the unique execution constraint for the task.

6. A multi-satellite collaborative regional target observation and scheduling system with strip coverage and improved ant colony, characterized in that, include: The acquisition module is used to acquire the list of tasks to be observed and satellite resources for regional targets, and to construct a task conflict matrix; The processing module is used to discretize the regional targets based on the task conflict matrix, generate strip coverage schemes for different tasks of each satellite observation, and obtain strip priority based on the number of strip coverage grids and task priority. The generation module is used to generate several multi-star collaborative task allocation schemes using an adaptive ant colony task allocation algorithm. The adaptive ant colony task allocation algorithm includes a state transition mechanism based on stripe priority perturbation and a pheromone update mechanism based on a dual-threshold dynamic adjustment strategy. The construction module is used to generate optimization objectives based on the strip priority and multi-star collaborative task allocation scheme, and to construct a scheduling decision model by combining energy constraints, time window constraints and task unique execution constraints. The solution module is used to solve the scheduling decision model using an improved branch and bound algorithm. By introducing a dynamic relaxation factor, it obtains the global optimal solution for deciding whether each satellite should perform an observation task at different times. The state transition mechanism based on strip priority perturbation is expressed as follows: in, Let $\mathbf$ be the probability that the $k$-th ant will pair satellite $j$ with task $i$. This indicates the pheromone concentration between satellite j and mission i; For pheromone importance factors, Weights for heuristic information; For satellite j, the strip priority of observation mission i; Let j be the heuristic function between satellite j and mission i. , , These are the energy level, attitude control margin, and storage capacity of satellite j, respectively. Let m represent any one satellite in the set S of all satellites that can observe mission i; This indicates the pheromone concentration between satellite m and mission i; A heuristic function representing the relationship between satellite m and mission i; Indicates the strip priority of observation task i for satellite m; and / or The pheromone update mechanism based on the dual-threshold dynamic adjustment strategy is expressed as follows: Specifically, when the globally optimal multi-star collaborative task allocation scheme is not updated for g consecutive times during the iteration process, it is triggered. The update formula; The concentration of pheromones left by the ant between satellite j and mission i; Indicates the update symbol; The volatility coefficient is... This is the initial volatility coefficient; This represents the current iteration number. This represents the maximum number of iterations. It is an exponential function.

7. The multi-satellite collaborative regional target observation and scheduling system as described in claim 6, characterized in that, include: The adjustment module is used to predefine a dynamic adjustment trigger function. When the cloud coverage change of the target area exceeds a preset ratio or the task priority fluctuation exceeds a preset threshold, the adaptive adjustment mechanism is triggered. After quantifying the adjustment urgency of different task observation areas, the global optimal solution is updated by adopting a local rescheduling strategy or a pheromone dynamic reset strategy.

8. A storage medium, characterized in that, It stores a computer program for strip coverage and improved ant colony multi-satellite cooperative regional target observation scheduling, wherein the computer program causes the computer to execute the multi-satellite cooperative regional target observation scheduling method as described in any one of claims 1 to 5.

9. An electronic device, characterized in that, include: One or more processors; Memory; And one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing the multi-satellite collaborative regional target observation scheduling method as described in any one of claims 1 to 5.

Images