A satellite mission planning method, system and device for point group targets

Abstract

The present invention provides a satellite mission planning method, system, and device for point cluster targets. The method comprises: first, preprocessing and merging point target missions; then, determining a mission synthesis plan by extending satellite observation strips; and finally, outputting a final satellite scheduling plan by designing a satellite mission planning model to find the optimal solution. Based on the characteristics of dense point cluster target missions, the present invention implements integrated modeling and calculation by incorporating strip extension ratios into decision variables, achieving global optimization. Furthermore, the method fully considers the timeliness and profitability of the missions, enabling rapid and accurate planning of satellite point cluster target missions.

Description

Technical Field

[0001] The present invention relates to the fields of satellite data processing and satellite mission planning, and in particular to a satellite mission planning method, system and device oriented to point group targets. Background Art

[0002] Imaging satellite observation scheduling generally refers to a scheduling method that determines the satellite imaging plan while meeting various satellite imaging constraints and combining mission requirements to achieve maximum observation benefits.

[0003] With the advancement of satellite technology, the missions assigned to satellites are gradually expanding. For example, in scientific research and disaster relief operations, it is often necessary to obtain image information of a large number of point targets within a specific area. In such scenarios, the number and density of satellite missions are high, and the time windows between missions overlap significantly. Many missions exhibit strong temporal and spatial coupling. How to fully utilize satellite resources and maximize the benefits of these missions, tailored to the specific scenarios, has become a pressing issue.

[0004] Especially in scenarios with dense point clusters, satellites need mission-synthesis observations. Onboard remote sensors often have a limited field of view. In push-broom mode, a single imaging pass can cover a swath of ground area of a certain width. When two mission targets are close together and meet synthesis constraints, the imaging swath can be extended by extending the satellite's push-broom time. This allows for simultaneous imaging of multiple targets in a single observation, conserving satellite resources and increasing observation yield.

[0005] There are many advantages to using synthetic observations from satellites:

[0006] First, the attitude transition time between multiple observation activities is related to the satellite's roll rate and angle differences. Especially in dense point target scenarios, when the time windows between tasks overlap significantly, most tasks are forced to be abandoned due to insufficient transition time.

[0007] Secondly, because roll imaging can cause attitude instability in the satellite, the number of roll imaging missions a satellite can perform within each orbit is strictly limited. Without synthetic observations, the number of missions a satellite can observe within each orbit would be extremely limited.

[0008] In addition, using one imaging to complete observations for some adjacent missions is beneficial to reducing the number of satellite power-on and power-off and sway times, thereby protecting the service life of satellite resources. For some satellites with limited sway performance, their sway rate is slow, the conversion time is long, and the number of sway times in a single circle is small, it is even more necessary to use synthetic observations. Figure 1As shown in Figure 1, there are three dense point target observation tasks in this scenario: a, b, and c. As shown in Figure (a), if time windows are generated for each of the three tasks, the three time windows overlap, and the satellite can only complete at most one of the tasks. However, if the strip merging and extension method is used, the strip of target a is merged with the strip of target b, and the strip of target a is extended to target c. At this time, all three tasks can complete the observation through the new strip of target a, as shown in Figure (b).

[0009] At present, the following problems exist in imaging task synthesis planning for processing dense observation scenes:

[0010] First, the problem constraints and benefit calculations are more complex, and the spatial computational complexity of the problem solution is high and challenging. Under multi-satellite conditions, satellites and missions can be combined in a variety of ways. Due to the heterogeneity between satellite orbits, the strip geographic information and time window information for point targets vary. This also leads to differences in the point target merging relationships within different satellite strips, which are dynamically updated as the satellite-mission matching relationship changes. Therefore, the computational complexity of the problem solution becomes even greater. How to fully utilize the geographic location information of point targets, avoid redundant calculations, and dynamically and accurately determine the point target merging relationships, thereby conserving computing resources and improving algorithm efficiency, is a critical issue.

[0011] Second, there is a lack of a scheduling model that can uniformly optimize the point target merging relationship through decision variables in dense observation scenarios, comprehensively consider observation strip constraints and benefits, and guide algorithm design. Satellite mission planning models in dense observation scenarios need to determine the extension ratio of the mission strip and complete the point target benefit calculation based on the determined strip information. Traditional planning methods are mostly based on first classifying point targets and then calculating the multi-satellite mission scheduling model in stages. This model severs the relationship between point target merging and the overall task scheduling problem, which is not conducive to global optimization. Therefore, there is an urgent need for a model that can meet the scheduling needs of dense observation scenarios. Summary of the Invention

[0012] The object of the present invention is to provide a satellite mission planning method, system and device for point group targets to solve at least one of the above-mentioned technical problems existing in the prior art.

[0013] To solve the above technical problems, the present invention provides a satellite mission planning method for point group targets, comprising the following steps:

[0014] Step 1: For point target tasks, merge them through preprocessing to obtain a set of mergeable target tasks;

[0015] Step 2: Based on the set of mergeable target tasks in step 1, a task synthesis scheme is determined by extending the satellite observation strip to obtain a decision task set;

[0016] Step 3: Based on the decision task set obtained in step 2, the final satellite scheduling plan is obtained by finding the optimal solution through the satellite mission planning model, and the decision factors of the satellite mission planning model include the strip extension ratio.

[0017] Furthermore, the pretreatment method of step 1 is:

[0018] Step A: Input all point target tasks, filter out the default point target tasks, and output the point target task set;

[0019] Step B: setting empty mergeable target task sets according to the time windows of different satellites;

[0020] Step C: Taking a certain point target task as a benchmark, traverse the remaining point target tasks for comparison, and add the point target tasks that meet the merging conditions to the corresponding mergeable target task set;

[0021] Step D: Execute step C on the remaining point target tasks in the point target task set, and output a pre-processed mergeable target task set.

[0022] Furthermore, the merging condition of step C is that the targets are all visible on the same orbit of the same satellite, and the observation time window of the target is within the time window threshold range of the satellite.

[0023] Through preprocessing, the solution space of the problem is narrowed to improve the operating efficiency of the algorithm.

[0024] Furthermore, the specific method of extending the satellite observation strip in step 2 is:

[0025] Step 1) Input the mergeable target set, divide it by satellite, output the point target task set corresponding to each satellite, further subdivide it by orbit, output the point target task set under different orbits of each satellite, and sort it in the order of time window;

[0026] Step 2) Based on the point target mission set under different orbits of each satellite, the extension ratio of the strip is determined one by one by the decision variables, and the advanced strip is obtained after the strip is updated;

[0027] Step 3) Based on the advanced strip, determine the extension direction of the advanced strip according to the satellite orbit direction, and obtain the longitude and latitude of the four endpoints of the advanced strip;

[0028] Step 4) Enter the latitude and longitude of the point target and verify whether the advanced strip covers the point target:

[0029] If the advanced strip covers the point target, determine the task synthesis plan;

[0030] If the advanced strip does not cover the point target, update the decision variable and execute step 2).

[0031] Furthermore, the method for verifying whether the advanced strip covers the point target may be the ray method. Of course, other determination methods known in the art may also be used to determine whether the advanced strip covers the point target.

[0032] Furthermore, the decision variables in step 2) include: imaging execution timing, data transmission execution timing and strip extension ratio.

[0033] The above decision variables directly determine the start time of satellite imaging and satellite data transmission tasks, thereby completing the matching of satellite tasks and satellite resources; the strip extension ratio determines the strip length and strip execution time in the scheme.

[0034] Furthermore, the mission planning model in step 3 includes the following assumptions:

[0035] Condition 1) In the task type, only point target tasks are considered, and the dynamic update of point target tasks and observation resources is not considered;

[0036] Condition 2) In the planning scheme, each point target task can only be executed once at most;

[0037] Condition 3) Discretize the time window into seconds. For a point target task, define each available execution time as a meta-task window. The decision timing set for each point target task is the sum of the task execution timings under all meta-task windows of this task.

[0038] Condition 4) For imaging and data transmission missions, the satellite can only perform one mission each at the same time without interruption;

[0039] Condition 5) The satellite's storage and power resources are certain to the satellite's mission.

[0040] By making assumptions, the problem space can be further simplified and computing time can be saved.

[0041] Furthermore, the mission planning model in step 3 also includes typical constraints:

[0042] Constraint 1: Execution uniqueness constraint means that for any decision task, at most one imaging execution timing and one data transmission execution timing can be selected.

[0043] Constraint 2, task timing logic constraints, means that for any decision task, the task data transmission start time cannot be earlier than the imaging start time. If the decision task is executed in real-time transmission mode, imaging and data transmission must start simultaneously; if the decision task is executed in record-playback mode, the data transmission start time must be later than the imaging end time.

[0044] Constraint 3: On-board conversion time constraints, including the constraints on the conversion time between on-board imaging tasks and the constraints on the conversion time between imaging tasks and data transmission tasks.

[0045] Constraint 4, ground station switching time constraint, refers to the switching time constraint of the satellite's data transmission task to the ground station: when the satellite is the same, there is no switching time for the data transmission task; when the satellites are different, there is a switching time for the data transmission task.

[0046] Constraint 5: Onboard resource constraints, including power constraints and fixed storage constraints: The power constraint means that the total power consumption of any mission cannot exceed the power threshold of the satellite and the single orbit on which it is located; the fixed storage constraint means that the total fixed storage occupied by missions that have been imaged but not transmitted at any time cannot exceed the onboard fixed storage threshold.

[0047] Through typical constraints, the scope of solution variables of the mission planning model is pointed out.

[0048] Furthermore, the mission planning model in step 3 also includes satellite mission execution benefits, and the satellite mission execution benefits refer to the sum of the benefits of completing the mission.

[0049] Through the benefits, the objective function of the task planning model is pointed out.

[0050] Furthermore, the method for finding the optimal solution of the task planning model in step 3 is:

[0051] Step a: Initialize the algorithm parameters and output the initial benefit value and initial solution through local optimal calculation: first, divide the decision task set into the set of tasks to be assigned to each satellite; then perform single-satellite task scheduling for each satellite; finally, let the current optimal value be the initial benefit value and the current optimal solution be the initial solution.

[0052] Step b: Initialize the weights of all operators and mark the call of task assignment as false.

[0053] Step c: Use each satellite's current observation sequence as the set of tasks to be scheduled for that satellite. Re-perform the neighborhood search, generate a new solution using the destruction operator, and store the corresponding task in the satellite's pending task list. If the call flag is true, proceed to step d; otherwise, proceed to step e.

[0054] Step d: Define the set of tasks to be assigned as the union of the current tasks to be assigned to each satellite, and assign the tasks to different satellites through the assignment operator, update the score of the assignment operator, and set the call flag to false.

[0055] In step e, based on the set of tasks to be assigned, the scheduling plan for each satellite is repaired. This involves selecting an insertion operator to insert the task into the schedule and outputting a new solution. If the new solution's benefit is better than the current optimal solution, the optimal solution is updated and step f is executed. Otherwise, step e is repeated until an optimal solution is found or the iteration threshold is reached. If the iteration threshold is reached, the call flag is set to true.

[0056] Step f: Input the result of step e, update the operator scores in each operator library, and then update the selection probability of the operator based on the operator scores.

[0057] Step g: If the termination condition is met, the historical optimal solution, i.e., the final satellite scheduling plan, is output and the planning ends; otherwise, the process returns to step c.

[0058] Furthermore, the selection of the operator in step f is performed by roulette according to the selection probability of the operator.

[0059] Furthermore, the termination condition in step g refers to reaching the maximum number of iterations N, and / or the current optimal solution has not been updated after n consecutive iterations.

[0060] Furthermore, the operator in step 3 is encapsulated in an operator library.

[0061] The operator library makes it easy to add and modify operators to improve the algorithm's solution efficiency when facing scene changes and improve the robustness of the algorithm.

[0062] Furthermore, the operators in step 3 include an allocation operator, a destruction operator, and a repair operator.

[0063] Furthermore, the allocation operator includes:

[0064] Allocation operator 1, random allocation, means randomly assigning tasks to a satellite;

[0065] Allocation operator 2, conflict allocation, is to allocate the task to the satellite with the shortest total overlap time between the task and the task time window in the satellite scheduling plan;

[0066] Allocation operator 3, maximum time window allocation, means allocating the task to the satellite with the longest time window for the task;

[0067] Allocation operator 4, maximum possible merge allocation, means allocating the task to the satellite with the most tasks that can be merged with the task in the satellite scheduling plan;

[0068] Allocation operator 5, experience allocation, refers to allocating tasks to the satellite that has the highest average benefit obtained by the corresponding satellite in the historical scheduling process.

[0069] Furthermore, the destruction operator includes:

[0070] Destruction operator 1, randomly delete n tasks;

[0071] Destruction operator 2: sort the tasks in ascending order of priority and delete the first n tasks;

[0072] Destruction operator 3: sort the tasks in the mergeable set in descending order by quantity and delete the first n tasks;

[0073] Destruction operator 4: Sort the tasks in descending order according to the conflict degree of their time windows and delete the first n tasks.

[0074] Furthermore, the repair operator includes:

[0075] Repair operator 1, sorts in descending order of priority, selects n tasks and inserts them into the scheduling plan;

[0076] Repair operator 2, sort in ascending order by the number of time windows, select n tasks and insert them into the scheduling plan;

[0077] Repair operator 3, insert the tasks with the smallest time window conflict into the scheduling plan one by one.

[0078] Furthermore, the priority may be a task execution priority, which is determined when the task is input. Of course, the priority may also be defined as a priority based on other criteria such as time sequence priority.

[0079] Furthermore, the conflict degree refers to the total length of overlapping time of time windows of different tasks. The longer the overlapping time, the greater the conflict degree.

[0080] The allocation operator, destruction operator and repair operator with score attributes update the operator scores by calling the operators during the iteration process, thereby reflecting the overall adaptive characteristics of the algorithm.

[0081] To adapt to multi-satellite scenarios, the algorithm adds a task allocation layer. If the solution benefit cannot be improved after multiple iterations during the local search process, the task allocation result is updated, and there is a chance to jump out of the local optimum, thereby achieving global optimization and retaining the algorithm's efficient solution capability to the greatest extent.

[0082] On the other hand, the present invention also discloses a satellite mission planning system for point group targets, which includes a mission receiving module, a mission processing module and a solution generating module.

[0083] The task receiving module is used to receive point target tasks and send them to the task processing module.

[0084] The task processing module is used to operate the point target task according to steps 1 to 3 of the satellite planning method for point group targets, and mainly includes a task preprocessing unit, a task synthesis unit and a task planning unit:

[0085] The task preprocessing unit receives point target tasks, merges them through preprocessing, and outputs a set of mergeable target tasks;

[0086] The task synthesis unit receives a set of mergeable target tasks, determines a task synthesis scheme by extending the satellite observation strip, and outputs a decision task set;

[0087] The task planning unit receives a set of decision tasks, designs a satellite task planning model to find an optimal solution, and outputs a final satellite scheduling solution.

[0088] The solution generation module is used to output the final satellite scheduling solution.

[0089] On the other hand, the present invention also provides a satellite mission planning device for point group targets, which mainly includes a processor, a memory and a bus. The memory stores instructions that can be read by the processor. The processor is used to call the instructions in the memory to execute the satellite mission planning method for point group targets. The bus connects the functional components to transmit information.

[0090] By adopting the above technical solution, the present invention has the following beneficial effects:

[0091] Based on the characteristics of dense point group target tasks, this solution implements integrated modeling and calculation by adding the strip extension ratio to the decision variables, thus avoiding the problem of being unable to escape local optimization in existing technologies. The preprocessing method based on point group target task planning provides a narrower problem-solving space and improves the efficiency of the algorithm. The calculation model provided for the merging and coverage of dense point group targets based on strip extension enables fast and accurate dynamic judgment of target merging. The meta-task processing mode provided implements unified modeling based on the characteristics of the task time window. The optimal solution algorithm provided fully considers the timeliness and profitability of the task. The destruction operator, repair operator, and task allocation operator provided are adaptively selected by the algorithm to solve the problem accurately and efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0093] Figure 1 Example diagram for satellite strip extension;

[0094] Figure 2 A flowchart of satellite mission planning for point group targets provided by an embodiment of the present invention;

[0095] Figure 3 Flowchart of the algorithm provided by the embodiment of the present invention;

[0096] Figure 4 A diagram of a satellite mission planning system for point group targets provided by an embodiment of the present invention.

[0097] Reference numerals:

[0098] (a)- Figure 1 Satellite strip extension before merging; (b)- Figure 1 Satellite strips are extended after merging. DETAILED DESCRIPTION

[0099] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0100] The terms "first," "second," and "third" and the like in the specification embodiments and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, including a series of steps or modules. Methods, systems, products, or devices are not necessarily limited to those steps or modules clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products, or devices. "And / or" is used to indicate that one or all of the two objects to which it is connected are selected.

[0101] In order to help those skilled in the art better understand the embodiments of the present application, four key points for solving the problems of the prior art are first described as follows:

[0102] 1) In order to reduce the amount of computation, a preprocessing method for dense point group target task planning is needed;

[0103] 2) In order to achieve fast and accurate dynamic judgment of target merging, it is necessary to design a dense point group target merging coverage model based on strip extension;

[0104] 3) In order to achieve fast and accurate dynamic planning of satellite missions, a mission planning model for imaging satellites that considers swath extension is needed;

[0105] 4) In order to improve the timeliness and profitability of tasks, it is necessary to study task planning algorithms for point group target merging.

[0106] The present invention will be further explained below with reference to specific embodiments.

[0107] like Figure 2 As shown, this embodiment provides a satellite mission planning method for point group targets, including the following steps:

[0108] Step 1: For point target tasks, merge them through preprocessing to obtain a set of mergeable target tasks;

[0109] Step 2: Based on the set of mergeable target tasks in step 1, a task synthesis scheme is determined by extending the satellite observation strip to obtain a decision task set;

[0110] Step 3: Based on the decision task set obtained in step 2, the final satellite scheduling plan is obtained by finding the optimal solution through the satellite mission planning model, and the decision factors of the satellite mission planning model include the strip extension ratio.

[0111] Furthermore, the pretreatment method of step 1 is:

[0112] Step A: Based on all point target tasks, data cleaning is performed by filtering out the default point target tasks to obtain a point target task set P;

[0113] Step B: setting empty mergeable target task sets according to the time windows of different satellites;

[0114] Step C: Taking a certain point target task as a benchmark, traverse the remaining point target tasks for comparison, and add the point target tasks that meet the merging conditions to the corresponding mergeable target task set;

[0115] Step D: Execute step C on the remaining point target tasks in the point target task set, and output a pre-processed mergeable target task set.

[0116] Furthermore, the merging condition of step C is that the targets are all visible on the same orbit of the same satellite, and the observation time window of the target is within the time window threshold range of the satellite.

[0117] Through preprocessing, the solution space of the problem is narrowed to improve the operating efficiency of the algorithm.

[0118] Furthermore, the specific method of extending the satellite observation strip in step 2 is:

[0119] Step 1) Based on the mergeable target set, divide it by satellite to obtain the point target task set corresponding to each satellite, and then subdivide it by orbit to obtain the point target task set under different orbits of each satellite, and sort them in the order of time window;

[0120] Step 2) Based on the point target mission set under different orbits of each satellite, the extension ratio of the strip is determined one by one by the decision variables, and the advanced strip is obtained after the strip is updated;

[0121] Step 3) Based on the advanced strip, determine the extension direction of the advanced strip according to the satellite orbit direction, and obtain the longitude and latitude of the four endpoints of the advanced strip;

[0122] Step 4) Enter the latitude and longitude of the point target and verify whether the advanced strip covers the point target:

[0123] If the advanced strip covers the point target, determine the task synthesis plan;

[0124] If the advanced strip does not cover the point target, update the decision variable and execute step 2).

[0125] Furthermore, the method for verifying whether the advanced strip covers the point target may be the ray method. Of course, other determination methods known in the art may also be used to determine whether the advanced strip covers the point target.

[0126] Furthermore, the decision variables in step 2) include: imaging execution timing, data transmission execution timing and strip extension ratio.

[0127] The above decision variables directly determine the start time of satellite imaging and satellite data transmission tasks, thereby completing the matching of satellite tasks and satellite resources; the strip extension ratio determines the strip length and strip execution time in the scheme.

[0128] Furthermore, the mission planning model in step 3 includes the following assumptions:

[0129] Condition 1) In the task type, only point target tasks are considered, and the dynamic update of point target tasks and observation resources is not considered;

[0130] Condition 2) In the planning scheme, each point target task can only be executed once at most;

[0131] Condition 3) Discretize the time window into seconds. For a point target task, define each available execution time as a meta-task window. The decision timing set for each point target task is the sum of the task execution timings under all meta-task windows of this task.

[0132] Condition 4) For imaging and data transmission missions, the satellite can only perform one mission each at the same time without interruption;

[0133] Condition 5) The satellite's storage and power resources are certain to the satellite's mission.

[0134] By making assumptions, the problem space can be further simplified and computing time can be saved.

[0135] Furthermore, the decision variables are defined as follows:

[0136] The point target task set is P, where the i-th task is p i ;

[0137] The satellite set is S, where the jth satellite is s j , s j The set of orbitals on is O j , where the kth orbital is o jk ;

[0138] s j On p i The meta-task window set is T ij , where the lth meta-task window is t ijl ;

[0139] According to the assumptions about the number of execution times of point target tasks, the execution time of point target tasks is selected in all meta-task time windows under all satellites. Therefore, the decision task set DT is defined, and the point target task set is P d , the i-th point target task p i (p i ∈P d ) corresponds to a decision task dt i ;

[0140] Based on the two types of processes, imaging and data transmission, corresponding to dt i , respectively define the imaging execution time as IEO i The execution time of the data transmission is DEO i , taking this as satellite resource, the multi-satellite joint scheduling problem is defined as follows:

[0141]

[0142]

[0143]

[0144] Among them, x ij Indicates the decision task at its imaging execution time IEO j Start execution below;

[0145] y ij Indicates the decision task at its data transmission execution time DEOj Start execution below;

[0146] The above decision variables directly determine the start time of the imaging and data transmission tasks, completing the matching of satellite tasks and their resources;

[0147] In addition, to meet the task requirements of stripe extension and merging, we define z i For the decision task dt i The strip extension ratio determines the strip length and strip execution time in the plan.

[0148] Therefore, by combining the above decision variables and the on-board ephemeris information, we can calculate the start and end time of each satellite mission and the start and end status of each strip, and finally obtain the satellite system planning scheme.

[0149] Furthermore, the mission planning model in step 3 also includes typical constraints:

[0150] Constraint 1: Execution uniqueness constraint means that for any decision task dt i , select at most one imaging execution time and one data transmission execution time:

[0151]

[0152]

[0153] Constraint 2, task timing logic constraint, refers to any decision task dt i , task data transmission start time b D,i Cannot be earlier than imaging start time b O,i If the execution mode of the decision task is real transmission mode, imaging and data transmission must start at the same time; if the execution mode of the decision task is record playback mode, the start time of data transmission must be later than the end time of imaging:

[0154]

[0155] Constraint 3: On-board conversion time constraint, including the conversion time ΔO(dt i ,dt i' ), the constraints of the conversion time ΔM between the imaging task and the data transmission task, b O(D),i For the decision task dt i The start and end time of the imaging task, e O(D),i For the decision task dt i The start and end time of the data transmission task:

[0156]

[0157]

[0158] Constraint 4, ground station switching time constraint, refers to the switching time constraint of the data transmission task from the satellite to the ground station: when the satellite is the same, there is no switching time for the data transmission task; when the satellites are different, there is a switching time ΔD for the data transmission task:

[0159]

[0160]

[0161] Constraint 5: Onboard resource constraints, including power constraints and storage constraints: Power constraints refer to the time it takes for any mission to execute i The total power consumption cannot exceed the satellite s(dt i ), where the monorail o(dt i ) power threshold Q; the storage constraint means that the total amount of storage occupied by the mission that has been imaged and not transmitted at any time cannot exceed the threshold M of the onboard storage, i(dt i ,t) and i(dt i ,t) respectively judge whether the task has completed imaging and data transmission at time t:

[0162]

[0163]

[0164] Through typical constraints, the scope of solution variables of the mission planning model is pointed out.

[0165] Furthermore, the mission planning model in step 3 also includes the satellite mission performance benefit, which refers to the mission performance benefit pr i The sum of check(p i ,ST) function is used to determine the task p i Whether it is included in the stripe ST selected by the scheduling scheme, if included, it is set to 1, otherwise it is set to 0:

[0166]

[0167] Through the benefits, the objective function of the task planning model is pointed out.

[0168] Furthermore, if Figure 3 As shown, in step 3, the final satellite scheduling solution is output by calculating the optimal solution of the mission planning model (for example, the adaptive large neighborhood search algorithm was first used for vehicle routing problems with time windows. The satellite scheduling problem also has time-dependent characteristics, and the adaptive large neighborhood algorithm also provides a custom heuristic operator to ensure the efficiency of the algorithm.). The detailed steps are as follows:

[0169] Step a: Initialize the algorithm parameters and output the initial benefit value and initial solution through local optimal calculation (such as greedy algorithm): first, divide the decision task set into the set of tasks to be assigned to each satellite; then perform single-satellite task scheduling for each satellite; finally, let the current optimal value be the initial benefit value and the current optimal solution be the initial solution.

[0170] Step b: Initialize the weights of all operators and set the call mark η of the task assignment to false, that is, η←fa l se.

[0171] Step c: Use each satellite's current observation sequence as the set of tasks to be scheduled for that satellite. Re-perform the neighborhood search, generate a new solution using the destruction operator, and store the corresponding task in the satellite's pending task list. If n is true, proceed to step d; otherwise, proceed to step e.

[0172] Step d: Define the set of tasks to be assigned as the union of the current tasks to be assigned to each satellite, and assign the tasks to different satellites through the assignment operator, update the score of the assignment operator, and set η←fa l se.

[0173] In step e, the scheduling plan for each satellite is repaired based on the set of tasks to be assigned. This involves selecting an insertion operator to insert the tasks into the scheduling plan and outputting a new solution. If the new solution's profit value is superior to the current optimal solution, as determined by simulated annealing or other criteria known in the art, the optimal solution is updated and step f is executed. Otherwise, step e is repeated until an optimal solution is obtained or the iteration threshold is reached. If the iteration threshold is reached, set η←true.

[0174] Step f: Input the result of step e, update the operator scores in each operator library, and then update the selection probability of the operator based on the operator scores.

[0175] Step g: If the termination condition is met, the historical optimal solution, i.e., the final satellite scheduling plan, is output and the planning ends; otherwise, the process returns to step c.

[0176] Of course, in other embodiments, other algorithms known in the art may be selected to achieve the same technical effect.

[0177] Furthermore, the selection of the operator in step f is performed by roulette according to the selection probability of the operator.

[0178] Furthermore, the termination condition in step g refers to reaching the maximum number of iterations N, and / or the current optimal solution has not been updated after n consecutive iterations.

[0179] Likewise, in other embodiments, other methods known in the art may be used to select operators.

[0180] Furthermore, the operator in step 3 is encapsulated in an operator library.

[0181] The operator library makes it easy to add and modify operators to improve the algorithm's solution efficiency when facing scene changes and improve the robustness of the algorithm.

[0182] Furthermore, the operators in step 3 include an allocation operator, a destruction operator, and a repair operator.

[0183] Furthermore, the allocation operator includes:

[0184] Allocation operator 1, random allocation, means randomly assigning tasks to a satellite;

[0185] Allocation operator 2, conflict allocation, is to allocate the task to the satellite with the shortest total overlap time between the task and the task time window in the satellite scheduling plan;

[0186] Allocation operator 3, maximum time window allocation, means allocating the task to the satellite with the longest time window for the task;

[0187] Allocation operator 4, maximum possible merge allocation, means allocating the task to the satellite with the most tasks that can be merged with the task in the satellite scheduling plan;

[0188] Allocation operator 5, experience allocation, refers to allocating tasks to the satellite that has the highest average benefit obtained by the corresponding satellite in the historical scheduling process.

[0189] Furthermore, the destruction operator includes:

[0190] Destruction operator 1, randomly delete n tasks;

[0191] Destruction operator 2: sort the tasks in ascending order of task execution priority and delete the first n tasks;

[0192] Destruction operator 3: sort the tasks in the mergeable set in descending order by quantity and delete the first n tasks;

[0193] Destruction operator 4: Sort the tasks in descending order according to the conflict degree of their time windows and delete the first n tasks.

[0194] Furthermore, the repair operator includes:

[0195] Repair operator 1, sorts in descending order of priority, selects n tasks and inserts them into the scheduling plan;

[0196] Repair operator 2, sort in ascending order by the number of time windows, select n tasks and insert them into the scheduling plan;

[0197] Repair operator 3, insert the tasks with the smallest time window conflict into the scheduling plan one by one.

[0198] Furthermore, the priority may be a task execution priority, which is determined when the task is input. Of course, the priority may also be defined as a priority based on other criteria such as time sequence priority.

[0199] Furthermore, the conflict degree refers to the total length of overlapping time of time windows of different tasks. The longer the overlapping time, the greater the conflict degree.

[0200] The allocation operator, destruction operator and repair operator with score attributes update the operator scores by calling the operators during the iteration process, thereby reflecting the overall adaptive characteristics of the algorithm.

[0201] The classic adaptive large neighborhood algorithm uses a dual loop: an inner loop performs a local search process involving destruction and repair operations, and an outer loop updates the optimal solution using simulated annealing. To adapt to multi-satellite scenarios, the algorithm provided in this embodiment incorporates a task allocation layer. If the local search process fails to improve the solution's benefit after multiple iterations, the task allocation result is updated, potentially allowing for an escape from the local optimum. This allows for global optimization while maximizing the algorithm's efficient solution capabilities.

[0202] On the other hand, this embodiment also provides a satellite mission planning system for point group targets, including a mission receiving module, a mission processing module and a solution generating module. Figure 4 shown.

[0203] The task receiving module is used to receive point target tasks and send them to the task processing module.

[0204] The task processing module is used to operate the point target task according to steps 1 to 3 of the satellite planning method for point group targets, and mainly includes a task preprocessing unit, a task synthesis unit and a task planning unit.

[0205] The task preprocessing unit receives point target tasks, merges them through preprocessing, and outputs a mergeable target set. The preprocessing method is as follows:

[0206] Step A: Input all point target tasks, perform data cleaning by filtering out the default point target tasks, and output the point target task set P;

[0207] Step B: setting empty mergeable target task sets according to the time windows of different satellites;

[0208] Step C: Taking a certain point target task as a benchmark, traverse the remaining point target tasks for comparison, and add the point target tasks that meet the merging conditions to the corresponding mergible target task set, where the merging conditions are that the targets are all visible on the same orbit of the same satellite and the observation time window of the target is within the time window threshold range of the satellite;

[0209] Step D: Execute step C on the remaining point target tasks in the point target task set, and output a pre-processed mergeable target task set;

[0210] Through preprocessing, the solution space of the problem is narrowed to improve the operating efficiency of the algorithm.

[0211] The task synthesis unit receives a set of mergeable targets, determines a task synthesis scheme by extending the satellite observation strip, and outputs a decision task set, wherein the specific method of extending the satellite observation strip is:

[0212] Step 1) Input the mergeable target set, divide it by satellite, output the point target task set corresponding to each satellite, further subdivide it by orbit, output the point target task set under different orbits of each satellite, and sort it in the order of time window;

[0213] Step 2) Based on the point target mission set under different orbits of each satellite, the extension ratio of the strip is determined one by one by the decision variables, and the advanced strip is obtained after the strip is updated;

[0214] Step 3) Based on the advanced strip, determine the extension direction of the advanced strip according to the satellite orbit direction, and obtain the longitude and latitude of the four endpoints of the advanced strip;

[0215] Step 4) Enter the latitude and longitude of the point target and verify whether the advanced strip covers the point target:

[0216] If the advanced strip covers the point target, determine the task synthesis plan;

[0217] If the advanced strip does not cover the point target, update the decision variable and execute step 2).

[0218] The mission planning unit receives a set of decision tasks, designs a satellite mission planning model to find the optimal solution, and outputs a final satellite scheduling plan;

[0219] The assumptions of the mission planning model include:

[0220] Condition 1) In the task type, only point target tasks are considered, and the dynamic update of point target tasks and observation resources is not considered;

[0221] Condition 2) In the planning scheme, each point target task can only be executed once at most;

[0222] Condition 3) Discretize the time window into seconds. For a point target task, define each available execution time as a meta-task window. The decision timing set for each point target task is the sum of the task execution timings under all meta-task windows of this task.

[0223] Condition 4) For imaging and data transmission missions, the satellite can only perform one mission each at the same time without interruption;

[0224] Condition 5) The satellite's storage and power resources are certain to the satellite's mission.

[0225] By making assumptions, the problem space can be further simplified and computing time can be saved.

[0226] The decision variables of the mission planning model are defined as follows:

[0227] The point target task set is P, where the i-th task is p i ;

[0228] The satellite set is S, where the jth satellite is s j , s j The set of orbitals on is O j , where the kth orbital is o jk ;

[0229] s j On p i The meta-task window set is T ij , where the lth meta-task window is t ijl ;

[0230] According to the assumptions about the number of execution times of point target tasks, the execution time of point target tasks is selected in all meta-task time windows under all satellites. Therefore, the decision task set DT is defined, and the point target task set is P d , the i-th point target task p i (p i ∈P d ) corresponds to a decision task dt i ;

[0231] Based on the two types of processes, imaging and data transmission, corresponding to dt i , respectively define the imaging execution time as IEO i The execution time of the data transmission is DEO i , taking this as satellite resource, the multi-satellite joint scheduling problem is defined as follows:

[0232]

[0233]

[0234]

[0235] Among them, x ij Indicates the decision task at its imaging execution time IEO j Start execution below;

[0236] y ij Indicates the decision task at its data transmission execution time DEO j Start execution below;

[0237] The above decision variables directly determine the start time of the imaging and data transmission tasks, completing the matching of satellite tasks and their resources;

[0238] Define z i For the decision task dt i The strip extension ratio determines the strip length and strip execution time in the plan.

[0239] Therefore, by combining the above decision variables with the on-board ephemeris information, we can calculate the start and end time of each satellite mission and the start and end status of each strip, and finally obtain the satellite system planning scheme, laying the foundation for the construction of the mission planning model.

[0240] Typical constraints for a mission planning model are as follows:

[0241] Constraint 1: Execution uniqueness constraint means that for any decision task dt i , select at most one imaging execution time and one data transmission execution time:

[0242]

[0243]

[0244] Constraint 2, task timing logic constraint, refers to any decision task dt i , task data transmission start time b D,i Cannot be earlier than imaging start time b O,i If the execution mode of the decision task is real transmission mode, imaging and data transmission must start at the same time; if the execution mode of the decision task is record playback mode, the start time of data transmission must be later than the end time of imaging:

[0245]

[0246] Constraint 3: On-board conversion time constraint, including the conversion time ΔO(dt i ,dt i' ), the constraints of the conversion time ΔM between the imaging task and the data transmission task, b O(D),i For the decision task dt i The start and end time of the imaging task, e O(D),i For the decision task dt i The start and end time of the data transmission task:

[0247]

[0248]

[0249] Constraint 4, ground station switching time constraint, refers to the switching time constraint of the data transmission task from the satellite to the ground station: when the satellite is the same, there is no switching time for the data transmission task; when the satellites are different, there is a switching time ΔD for the data transmission task:

[0250]

[0251]

[0252] Constraint 5: Onboard resource constraints, including power constraints and storage constraints: Power constraints refer to the time it takes for any mission to execute i The total power consumption cannot exceed the satellite s(dt i ), where the monorail o(dt i ) power threshold Q; the storage constraint means that the total amount of storage occupied by the mission that has been imaged and not transmitted at any time cannot exceed the threshold M of the onboard storage, i(dt i ,t) and i(dt i ,t) respectively judge whether the task has completed imaging and data transmission at time t:

[0253]

[0254]

[0255] Through typical constraints, the scope of solution variables of the mission planning model is pointed out.

[0256] In the mission planning model, the satellite mission performance benefit refers to the mission performance benefit pr i The sum of check(p i ,ST) function is used to determine the task p i Whether it is included in the stripe ST selected by the scheduling scheme, if included, it is set to 1, otherwise it is set to 0:

[0257]

[0258] Through the benefits, the objective function of the task planning model is pointed out.

[0259] The mission planning model uses an improved adaptive large neighborhood search algorithm to find the optimal solution and output the final satellite scheduling plan. The detailed steps are as follows:

[0260] Step a: Initialize the algorithm parameters and use the greedy algorithm to find the local optimal solution to output the initial benefit value and initial solution: first, divide the decision task set into the set of tasks to be assigned to each satellite; then perform single-satellite task scheduling for each satellite; finally, let the current optimal value be the initial benefit value and the current optimal solution be the initial solution.

[0261] Step b: Initialize the weights of all operators and set the call mark η of the task assignment to false, that is, η←fa l se.

[0262] Step c: Use each satellite's current observation sequence as the set of tasks to be scheduled for that satellite. Re-perform the neighborhood search, generate a new solution using the destruction operator, and store the corresponding task in the satellite's pending task list. If n is true, proceed to step d; otherwise, proceed to step e.

[0263] Step d: Define the set of tasks to be assigned as the union of the current tasks to be assigned to each satellite, and assign the tasks to different satellites through the assignment operator, update the score of the assignment operator, and set η←fa l se.

[0264] In step e, the scheduling plan for each satellite is repaired based on the set of tasks to be assigned. This involves selecting an insertion operator to insert the tasks into the scheduling plan and outputting a new solution. If the new solution's profit value is superior to the current optimal solution, as determined by simulated annealing or other criteria known in the art, the optimal solution is updated and step f is executed. Otherwise, step e is repeated until an optimal solution is obtained or the iteration threshold is reached. If the iteration threshold is reached, set η←true.

[0265] Step f: Input the result of step e, update the operator scores in each operator library, and then update the selection probability of the operator based on the operator scores.

[0266] In step g, if the termination condition is met, the historical optimal solution (i.e., the final satellite scheduling solution) is output, and planning ends. Otherwise, the process returns to step c. The termination condition is when the maximum number of iterations, N, is reached and / or the current optimal solution remains unupdated after n consecutive iterations.

[0267] The solution generation module is used to output the final satellite scheduling solution.

[0268] In another embodiment, the solution of the present invention can also be implemented by a satellite mission planning device for point group targets, which mainly includes a processor, a memory and a bus;

[0269] The memory stores instructions that can be read by the processor, and the size is flexibly configured according to storage needs;

[0270] The bus connects the functional components of the computer to transmit information;

[0271] The processor is used to call the instructions in the memory to execute a satellite mission planning method oriented to point group targets.

[0272] In another embodiment, this solution can be implemented by a device, which may include corresponding modules for performing each or several steps in each of the above embodiments. The modules may be one or more hardware modules specifically configured to perform the corresponding steps, or implemented by a processor configured to perform the corresponding steps, or stored in a computer-readable medium for implementation by the processor, or implemented by some combination thereof.

[0273] The processor performs the various methods and processes described above. For example, the method implementation in this solution can be implemented as a software program, which is tangibly contained in a machine-readable medium, such as a memory. In some embodiments, part or all of the software program can be loaded and / or installed via a memory and / or a communication interface. When the software program is loaded into the memory and executed by the processor, one or more steps in the method described above can be performed. Alternatively, in other embodiments, the processor can be configured to perform one of the above methods by any other appropriate means (e.g., by means of firmware).

[0274] The device can be implemented using a bus architecture. The bus architecture can include any number of interconnecting buses and bridges, depending on the specific application and overall design constraints of the hardware. The bus connects various circuits including one or more processors, memories, and / or hardware modules. The bus can also connect various other circuits such as peripherals, voltage regulators, power management circuits, external antennas, etc.

[0275] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0276] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A satellite mission planning method for point group targets, characterized in that: The steps include: Step 1: For point target tasks, merge them through preprocessing to obtain a set of mergeable target tasks; Step 2: Based on the set of mergeable target tasks, a task synthesis scheme is determined by extending the satellite observation strip to obtain a decision task set; Step 3: Based on the decision task set, a final satellite scheduling solution is obtained by finding an optimal solution through a satellite mission planning model, where the decision factors of the satellite mission planning model include a strip extension ratio; The specific method for finding the optimal solution of the satellite mission planning model in step 3 is: Step a: Initialize the algorithm parameters and output the initial benefit value and initial solution through local optimal calculation: first, divide the decision task set into the set of tasks to be assigned to each satellite; then perform single-satellite task scheduling for each satellite; finally, set the current optimal value as the initial benefit value and the current optimal solution as the initial solution; Step b: Initialize the weights of all operators and mark the call of task assignment as false; Step c: Use the current observation sequence of each satellite as the set of tasks to be scheduled for that satellite, re-perform the neighborhood search, generate a new solution by destroying the operator, and store the corresponding task in the list of tasks to be assigned for the satellite. If the call flag is true at this time, execute step d; otherwise, execute step e. Step d: Define the set of tasks to be assigned as the union of the current tasks to be assigned to each satellite, and assign the tasks to different satellites through the assignment operator, update the score of the assignment operator, and set the call flag to false; Step e: Repair the scheduling plan of each satellite based on the set of tasks to be assigned, that is, select the insertion operator to insert the task into the scheduling plan and output the new solution; if the benefit value of the new solution is better than the current optimal solution, update the optimal solution and execute step f; otherwise, repeat step e until the optimal solution is generated or the iteration threshold is reached. If the iteration threshold is reached, set the call flag to true; Step f: Input the result of step e, update the operator score in each operator library, and then update the selection probability of the operator based on the operator score; Step g: If the termination condition is met, the historical optimal solution, i.e., the final satellite scheduling plan, is output and the planning ends; otherwise, the process returns to step c.

2. The method according to claim 1, wherein The pretreatment method of step 1 is: Step A: Input all point target tasks, filter out the default point target tasks, and output the point target task set; Step B: setting empty mergeable target task sets according to the time windows of different satellites; Step C: Taking a certain point target task in the point target task set as a benchmark, traverse the remaining point target tasks for comparison, and add the point target tasks that meet the merging conditions to the corresponding mergeable target task set; Step D: Execute step C on the remaining point target tasks in the point target task set, and output a pre-processed mergeable target task set.

3. The method according to claim 1, wherein The specific method of extending the satellite observation strip in step 2 is: Step 1) Input the mergeable target set, divide it by satellite, output the point target task set corresponding to each satellite, further subdivide it by orbit, output the point target task set under different orbits of each satellite, and sort it in the order of time window; Step 2) Based on the point target mission set under different orbits of each satellite, the extension ratio of the strip is determined one by one by the decision variables, and the advanced strip is obtained after the strip is updated; Step 3) Based on the advanced strip, determine the extension direction of the advanced strip according to the satellite orbit direction, and obtain the longitude and latitude of the four endpoints of the advanced strip; Step 4) Enter the latitude and longitude of the point target and verify whether the advanced strip covers the point target: If the advanced strip covers the point target, determine the task synthesis plan; If the advanced strip does not cover the point target, update the decision variable and execute step 2).

4. The method according to claim 3, wherein The decision variables in step 2) include: imaging execution timing, data transmission execution timing and strip extension ratio; Imaging execution timing determines the start time of the satellite imaging mission; The data transmission execution timing determines the start time of the satellite data transmission task; The stripe extension ratio determines the stripe length and the execution time of the stripe in the solution.

5. The method according to claim 1, wherein The mission planning model in step 3 also includes typical constraints: Constraint 1: Execution uniqueness constraint means that for any decision task, at most one imaging execution time and one data transmission execution time can be selected; Constraint 2, task timing logic constraint, means that for any decision task, the task data transmission start time cannot be earlier than the imaging start time; Constraint 3: Onboard conversion time constraints, including the conversion time constraints between onboard imaging tasks and the conversion time constraints between imaging tasks and data transmission tasks; Constraint 4: Ground station switching time constraint, which refers to the switching time constraint of satellite-to-ground station data transmission tasks; Constraint 5: Onboard resource constraints, including power constraints and fixed storage constraints: The power constraint means that the total power consumption of any mission cannot exceed the power threshold of the satellite and the single orbit on which it is located; the fixed storage constraint means that the total fixed storage occupied by missions that have been imaged but not transmitted at any time cannot exceed the onboard fixed storage threshold.

6. The method according to claim 1, wherein The operators in step b include an allocation operator, a destruction operator, and a repair operator.

7. The method according to claim 6, wherein The specific manner of allocating the operator includes one of the following manners or any combination thereof: Allocation operator 1, random allocation, means randomly assigning tasks to a satellite; Allocation operator 2, conflict allocation, is to allocate the task to the satellite with the shortest total overlap time between the task and the task time window in the satellite scheduling plan; Allocation operator 3, maximum time window allocation, means allocating the task to the satellite with the longest time window for the task; Allocation operator 4, maximum possible merge allocation, means allocating the task to the satellite with the most tasks that can be merged with the task in the satellite scheduling plan; Allocation operator 5, experience allocation, refers to allocating tasks to the satellite with the highest average benefit obtained by the corresponding satellite in the historical scheduling process; The specific method of destroying the operator includes one of the following methods or any combination thereof: Destruction operator 1, randomly delete n tasks; Destruction operator 2: sort the tasks in ascending order of priority and delete the first n tasks; Destruction operator 3: sort the tasks in the mergeable set in descending order by quantity and delete the first n tasks; Destruction operator 4: sort the tasks in descending order of the conflict degree of their time windows and delete the first n tasks; The specific method of the repair operator includes one of the following methods or any combination thereof: Repair operator 1, sorts in descending order of priority, selects n tasks and inserts them into the scheduling plan; Repair operator 2, sort in ascending order by the number of time windows, select n tasks and insert them into the scheduling plan; Repair operator 3, insert the tasks with the smallest time window conflict into the scheduling plan one by one.

8. A satellite mission planning system for point group targets, characterized in that: Including task receiving module, task processing module and solution generation module: The task receiving module is used to receive point target tasks and send them to the task processing module; The task processing module is used to operate the point target task according to any one of the methods described in claims 1 to 7, and includes a task preprocessing unit, a task synthesis unit, and a task planning unit: The task preprocessing unit receives point target tasks, merges them through preprocessing, and outputs a set of mergeable target tasks; The task synthesis unit receives a set of mergeable target tasks, determines a task synthesis scheme by extending the satellite observation strip, and outputs a decision task set; The mission planning unit receives a set of decision tasks, designs a satellite mission planning model to find the optimal solution, and outputs a final satellite scheduling plan; The solution generation module is used to output the final satellite scheduling solution.

9. A satellite mission planning device for point group targets, characterized in that: The method comprises a processor, a memory and a bus, wherein the memory stores instructions readable by the processor, the processor is used to call the instructions in the memory to execute the method according to any one of claims 1 to 7, and the bus connects the functional components to transmit information.

Images