An Agile Satellite Emergency Mission Planning Method

Abstract

The present invention discloses an agile satellite emergency task planning method, obtains an emergency task visible time window set; sorts the emergency tasks from high to low priority; selects the emergency task with the highest current priority, selects the start execution time within its visible time window, and inserts it into the current planning scheme to obtain the inserted planning scheme, and removes the emergency task from the emergency task list; determines whether there is a conflict in the inserted planning scheme, if not, the inserted planning scheme is used as the new current planning scheme; otherwise, a window sliding strategy is used to resolve the conflict and obtain a new current planning scheme; determines whether the new current planning scheme meets the constraint conditions, if so, the current planning scheme is updated to the new current planning scheme; otherwise, remains unchanged; loops through all emergency tasks and outputs the planning scheme. The present invention reduces the disturbance of the emergency task to the initial planning scheme, reduces the difficulty of satellite secondary scheduling, and improves the total mission revenue.

Description

Technical Field

[0001] The invention belongs to the technical field of emergency mission planning for agile satellites, and relates to an emergency mission planning method for an agile satellite, in particular to an emergency mission planning method for an agile satellite based on a time window sliding strategy. Background Art

[0002] With the rapid development and widespread application of agile imaging satellite technology, users have put forward higher and higher requirements for the emergency response and rapid observation of agile imaging satellites in emergency situations. Not only do they need to effectively observe routine tasks, but they also need to respond quickly to emergency tasks proposed by users. After the emergency task arrives, it is necessary to quickly and effectively adjust the initial planning scheme and generate a new planning scheme to complete the scheduling of the emergency task.

[0003] Emergency mission observation is the observation activity undertaken by earth imaging satellites to provide support for ground emergency operations. The mission time and observation targets cannot be determined in advance, and it has a strong timeliness. Compared with conventional tasks, its characteristics are mainly reflected in the three aspects of urgency, time concentration and space concentration. If the traditional planning method is used to completely re-plan the emergency mission, it will take a long time and cannot meet the requirements of quickly generating adjustment planning schemes. In addition, the generated new mission planning scheme may be quite different from the initial planning scheme, which increases the difficulty of satellite scheduling. At present, in the research on emergency mission planning, the adjustment algorithm based on rules and heuristic strategies and the intelligent optimization algorithm are mainly used to deal with the emergency mission planning problem. However, the existing technical research has ignored the problem of disturbance to the initial planning scheme. Summary of the invention

[0004] In view of the above-mentioned prior art, the technical problem to be solved by the present invention is to provide an agile satellite emergency task planning method based on a time window sliding strategy, so as to overcome the problem of large disturbance to the initial plan after the emergency task is inserted.

[0005] In order to solve the above technical problems, the present invention provides an agile satellite emergency mission planning method, comprising:

[0006] Step 1: Receive the emergency mission, calculate the satellite visibility to the emergency mission, and obtain the visible time window set ETS of the emergency mission;

[0007] Step 2: Sort all emergency tasks in descending order of priority to obtain an emergency task list;

[0008] Step 3: Select the emergency task T with the highest current priority in the task list i , in its visible time window [ws i ,we i] Select a certain time as the start execution time ts of the task i , the emergency task T i Insert the current plan into Schedule to get the inserted plan Schedule', and add the emergency task T i Take it off the emergency task list;

[0009] Step 4: Determine whether there is a time window conflict in the inserted plan Schedule'. If there is no conflict, use the inserted plan Schedule' as the new current plan Schedule new Otherwise, the window sliding strategy is used to resolve the conflict and obtain the new current planning scheme Schedule new ;

[0010] Step 5: Determine the new current plan schedule new Whether all constraints are met, if so, the current plan Schedule is updated to the new current plan Schedule new ; Otherwise, the current plan Schedule remains unchanged;

[0011] Step 6: Determine whether to traverse all emergency tasks. If so, output the current planning scheme Schedule, otherwise return to step 3.

[0012] Furthermore, in step 4, the window sliding strategy is used to resolve the conflict and obtain the new current planning scheme Schedule new Specifically:

[0013] Step 4.1: Insert the emergency task as the focus task T v ;

[0014] Step 4.2: Determine the focus task T v Is the number of moves greater than the maximum number of moves M for a single task? T If yes, set the window sliding execution result flag to 0, compare the priorities of the current conflicting tasks, retain the tasks with higher priorities, and remove the tasks with lower priorities. If the priorities are the same, retain the tasks with shorter execution time and execute step 4.6; otherwise, execute step 4.3;

[0015] Step 4.3: Determine T v The adjacent task T on the left L and the adjacent task T on the right R The conflict relationship in the time window, when T v With T L or T RIf there is a time window conflict, execute step 4.4; otherwise, the window sliding strategy is successful, the window sliding execution result flag is set to 1, and execute step 6;

[0016] Step 4.4: When T v With T L When there is a time window conflict, determine T L The left margin and T v Whether the right margin of satisfies both 0 and T v With T R When there is a time window conflict, determine T R The right margin and T v Whether the left margin of is satisfied and is 0 at the same time, if it is satisfied, compare the priorities of the conflicting tasks, retain the tasks with higher priority, and remove the tasks with lower priority. If the priorities are the same, retain the tasks with shorter execution time, set the window sliding execution result flag to 0, and execute step 4.6; otherwise, execute step 4.5;

[0017] Step 4.5: When T v With T L When there is a time window conflict: If T v Priority over T L , and T L The left margin of is not 0, or T v The priority is lower than T L , but T v The right margin is 0, then T L Shift left, the task T will be moved L Treat as attention task T v , focus on task T v The number of moves of is reset to 1, and step 4.2 is executed; otherwise, T v Move right and focus on task T v The number of moves is increased by 1, and step 2 is executed; when T v With T R When there is a time window conflict: If T v Priority over T R , and T R The right margin is not 0, or T v The priority is lower than T R , but T v The left margin is 0, then T R Move right, task T will be moved R Treat as attention task T v , focus on task T v The number of moves of is reset to 1, and step 4.2 is executed; otherwise, T v Move left and focus on task T vAdd 1 to the number of moves and execute step 4.2;

[0018] Step 4.6: Return the window sliding execution result flag and the current planning scheme, and end the window sliding strategy.

[0019] Furthermore, the left margin and right margin describe the range within which the execution time window can slide, specifically:

[0020] Left margin yd l It is the smaller value between the difference Δt1 between the task execution start time and its earliest observable time and the difference Δt2 between the task execution start time and the execution end time of the task on its left, satisfying:

[0021] yd l =min(Δt1,Δt2)

[0022] Right margin yd r It is the smaller value between the difference Δt1 between the task execution end time and its latest observable time and the difference Δt2 between the task execution end time and the execution start time of the task on its right, satisfying:

[0023] yd r =min(Δt1,Δt2)

[0024] Furthermore, the constraints are as follows: the observation time of any target must be within its visible time window, the attitude angle of the satellite at any time must be less than the range of the maximum attitude angle, the minimum solar altitude angle requirement must be met when performing observation tasks, the on-board storage space of the satellite does not exceed the maximum allowable storage capacity, the power of the satellite at any time must be greater than the minimum power requirement, two adjacent tasks performed on the same satellite must have a sufficient time interval to ensure that the onboard remote sensor can complete the attitude conversion, the number of satellite power ons and off times in a single orbit cannot exceed the number of times the satellite is turned on and off in any orbital circle, and the cumulative duration of all tasks performed within the time period of any orbital cycle of the satellite resource operation shall not exceed the given maximum working time of a single circle.

[0025] Beneficial effects of the present invention: The present invention eliminates conflicts between tasks and reduces the impact on the initial planning scheme by sliding the execution time window of a small number of local tasks. With the goal of maximizing the total task benefit, the emergency tasks are inserted and planned one by one, and the planning of the emergency tasks is finally completed to generate a new adjusted planning scheme. The window sliding method is used to eliminate the time window conflicts between tasks that may be caused after the emergency tasks are inserted, ensuring that the emergency tasks with high priority are executed, while reducing the disturbance to the initial plan, reducing the difficulty of satellite secondary scheduling, and improving the total task benefit. Compared with the prior art, the beneficial effects of the present invention are:

[0026] 1. The present invention reduces the disturbance to the initial planning scheme while completing the insertion of high-priority tasks;

[0027] 2. The present invention aims to maximize the total mission benefit, minimize the disturbance to the initial plan and minimize the imaging maneuver angle, comprehensively considers various factors and has higher practicality;

[0028] 3. The present invention also has the advantages of clear and understandable logic, high flexibility and strong robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Planning flow chart for emergency tasks based on window sliding strategy;

[0030] Figure 2 It is the time margin diagram of the window sliding strategy;

[0031] Figure 3 It is the flow chart of window sliding strategy;

[0032] Figure 4 Resolve the conflict graph for the window sliding strategy;

[0033] Figure 5 orbit information for Agile Imaging satellites;

[0034] Figure 6 Basic resource parameters for agile imaging satellites;

[0035] Figure 7 Simulate planning results for the number of perturbations of the initial solution for different task sizes;

[0036] Figure 8 Simulate planning results for total mission benefits at different mission sizes;

[0037] Fig. 9 Simulate planning results for total maneuver angles at different mission sizes;

[0038] Fig.10 Simulate planning results for runtimes at different task sizes;

[0039] Fig.11 Simulation planning results for different task types. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0041] Aiming at the problem of agile imaging satellite mission planning, which has a large search space and complex solutions, the needs and characteristics of emergency missions are analyzed, and a model for emergency mission planning of agile imaging satellites is established:

[0042] The present invention mainly considers the adjustment planning problem of the initial task sequence after the emergency task arrives before the task instruction is injected into the satellite. After the emergency task arrives, the visibility of each task must be calculated first to obtain the visible time window set of the task. This process can be achieved through STK; then, under the satellite usage constraints, the emergency task is planned to obtain the planning scheme after the emergency task is inserted. The planning scheme is the task sequence, which mainly includes the task objectives and the start and end time of the task, that is, which tasks the satellite is to observe, when each task starts and ends, and in what posture the task is to be observed.

[0043] The constraints of emergency mission planning are as follows: the observation time of any strip must be within its visible time window, the attitude angle of the satellite at any time must be less than the range of the maximum attitude angle, the minimum solar altitude angle requirement must be met when performing observation tasks to ensure effective observation of the onboard payload, the onboard storage space of the satellite does not exceed the maximum allowable storage capacity, the power of the satellite at any time must be greater than the minimum power requirement, two adjacent tasks performed on the same satellite must have a sufficient time interval to ensure that the onboard remote sensor can complete the attitude conversion, the number of satellite power ons and offs in a single orbit cannot exceed the number of times the satellite is turned on and off in any orbital circle, and the cumulative duration of all tasks performed in the time period of any orbital cycle of satellite resource operation shall not exceed the given maximum working time of a single circle.

[0044] The requirements for emergency mission planning are:

[0045] (1) Rapid response. For emergency tasks, the planning scheme should be adjusted as quickly as possible and a new planning scheme should be generated to meet the timeliness requirements.

[0046] (2) Imaging quality. When an imaging satellite observes a ground target, a large maneuvering angle will cause the captured image to tilt and distort. In order to reduce the impact of geometric distortion on image quality, the smaller the satellite maneuvering angle during imaging, the better.

[0047] (3) Minor changes to the initial plan. The planning of emergency tasks should change the initial plan as little as possible to reduce the difficulty of secondary satellite scheduling.

[0048] (4) Higher total benefits. Emergency tasks generally have higher priorities, that is, higher benefits. The adjusted planning scheme after adding emergency tasks should have higher benefits than the initial planning scheme.

[0049] According to the above constraints and requirements, the optimization objectives of the composite decision variable model for agile satellite emergency mission planning are as follows:

[0050] Maximize the total task benefit TB, that is, the sum of all task priorities in the planning scheme is maximized:

[0051]

[0052] Where n represents the total number of tasks in the plan, p i Indicates the priority of the i-th task.

[0053] Minimize the imaging maneuver angle SA, that is, the sum of the squares of the maneuver angles at the start and end of all tasks in the planning scheme is minimized:

[0054]

[0055] where angle s,i represents the maneuvering angle of the agile imaging satellite at the start time of the i-th task in the planning scheme Schedule, angle e,i It represents the maneuvering angle of the agile imaging satellite at the end time of the i-th task in the planning scheme Schedule.

[0056] Minimize the disturbance SC to the initial plan, that is, the difference between the total number of tasks in the initial planning plan and the number of initial tasks retained in the new planning plan after emergency task planning is minimized:

[0057] SC=N Schedule -N Schedule∩Schedule'

[0058] Among them, Schedule' represents the new planning scheme after emergency task planning, and Schedule∩Schedule' represents the tasks in the initial planning scheme retained in the new planning scheme.

[0059] Combination Figure 1 The present invention provides an emergency task planning method based on a time window sliding strategy, comprising:

[0060] Step 1: After the emergency task arrives, pre-process it, calculate the visibility of the satellite to the emergency task, obtain the visible time window set ETS of the emergency task, and record the number of tasks num;

[0061] Step 2: Sort all emergency tasks according to priority and plan them one by one in order from high to low priority;

[0062] Step 3: Select the emergency task T with the highest current priority i , in its visible time window [ws i ,we i ] Select a certain time as the start execution time ts of the task i , the emergency task T i Insert into the current plan Schedule to obtain the inserted plan Schedule';

[0063] Step 4: Determine whether there is a conflict in the inserted plan Schedule'. If there is no conflict, replace the inserted plan with the current plan; otherwise, use the window sliding strategy to resolve the conflict and obtain the current plan.

[0064] Step 5: Determine whether the current planning scheme meets all constraints. If so, update the planning scheme; otherwise, keep the original planning scheme unchanged.

[0065] Step 6: Repeat steps 3 to 5 until all emergency tasks are traversed;

[0066] Step 7: Obtain the adjusted new planning scheme.

[0067] The specific operation of resolving the conflict of the window sliding strategy in step 4 is as follows:

[0068] The window sliding strategy mainly uses the idea of ​​recursion. Each time an emergency task T is inserted, only the inserted task and its two adjacent tasks T are considered. L and T R If the task T is moved during the conflict resolution process, L Or task T R , we continue to focus on the time window problem between the moving task and its adjacent tasks. And set the maximum number of moves for a single task M T , as the recursive end condition. For a task i in the task sequence, in order to describe the range in which its execution time window can slide left and right, the left margin yd of the task i is defined l and right margin yd r ,like Figure 2 As shown. Left margin yd l It is the smaller value between the difference Δt1 between the task execution start time and its earliest observable time and the difference Δt2 between the task execution start time and the execution end time of the task on its left. The calculation formula is as follows:

[0069] yd l =min(Δt1,Δt2)

[0070] Similarly, the right margin of the task can be defined.

[0071] The specific idea of ​​the window sliding strategy is:

[0072] (1) Treat the inserted emergency task as a focus task;

[0073] (2) For the task of interest, determine the conflict relationship between the task and its two adjacent tasks on the left and right in the time window. If there is a conflict (taking the left side as an example, the right side is the same), determine whether the left margin of the left task and the right margin of the task are both 0. If so, compare the priorities of the conflicting tasks, retain the task with a higher priority and eliminate the task with a lower priority. If the priorities of the two tasks are the same, retain the task with a shorter execution time; otherwise, proceed to the next step;

[0074] (3) Compare the priorities of the two conflicting tasks. If the priority of this task is higher than that of the left task and the left margin of the left task is not 0, or if the priority of this task is lower than that of the left task but the right margin of this task is 0, then the left task should be moved left to resolve the conflict; otherwise, the task should be moved right to resolve the conflict;

[0075] (4) The moved task is regarded as the focus task, and steps (2) to (3) are repeated until there is no conflict between all tasks in the entire planning scheme and the window sliding strategy is successful; or if a task reaches the maximum number of moves and the conflict is still not resolved, the priorities of the conflicting tasks are compared, and the tasks with higher priorities are retained and the tasks with lower priorities are eliminated.

[0076] like Figure 3 As shown, the main steps of the window sliding strategy are as follows:

[0077] Step 1: Insert the emergency task as the focus task T v ;

[0078] Step 2: Determine the focus task T v Is the number of moves greater than the maximum number of moves M for a single task? T If yes, set the window sliding execution result flag to 0, compare the priorities of the current conflicting tasks, retain the tasks with higher priorities and remove the tasks with lower priorities. If the priorities are the same, retain the tasks with shorter execution time and execute step 6; otherwise, execute step 3;

[0079] Step 3: Focus on task T v , determine the task T v The two tasks T adjacent to it L and T R The conflict relationship in the time window is based on the left task T L As an example, the right side is the same. If T v With T L If there is no conflict, the window sliding strategy is successful, the window sliding execution result flag is set to 1, and step 6 is executed; if T v With T L There are conflicts, such as Figure 4 As shown, execute step 4;

[0080] Step 4: Determine the left task T L The left margin and the task T v Are the right margins of both are 0 at the same time? If so, compare the priorities of the conflicting tasks, keep the tasks with higher priority and remove the tasks with lower priority. If the priorities are the same, keep the tasks with shorter execution time, set the window sliding execution result flag to 0, and execute step 6; otherwise, execute step 5;

[0081] Step 5: Compare the priorities of the two conflicting tasks. If the task T v Priority is higher than the task T on the left L , and the left task T L The left margin of the task is not 0, or the task T v Although the priority is lower than the left task T L , but the task T v The right margin is 0, then the left task T L Shift left, the task T will be moved L Treat as attention task T v , focus on task T v The number of moves is reset to 1 and step 2 is executed; otherwise, the task T v Move right and focus on task T v Add 1 to the number of moves and execute step 2;

[0082] Step 6: Return the window sliding execution result flag (flag=1 indicates that the window sliding strategy resolves the conflict successfully, and flag=0 indicates that the window sliding strategy resolves the conflict unsuccessfully) and the planning scheme, and end the window sliding strategy.

[0083] To test the feasibility of the algorithm in application, software simulation is performed:

[0084] By setting simulation conditions such as satellite orbit information and basic resource parameters, simulations are carried out under different task scales and task types to verify that the emergency task planning algorithm based on the time window sliding strategy can reduce the disturbance to the initial plan while ensuring the insertion of high-priority emergency tasks, thereby improving the total task benefit and verifying the practicality and effectiveness of the invention.

[0085] The simulation scene is set up to calculate the orbital parameters of the agile satellite through STK software. The orbital information of the agile imaging satellite is as follows: Figure 5 As shown in Figure 2, the basic resource constraints of agile imaging satellites are as follows: Figure 6 In order to verify the effectiveness of the algorithm, the following two groups of simulations are set up, and the existing emergency task planning algorithm based on task replacement is used as a comparative example (HDAA) of this embodiment (TWSA):

[0086] (1) Different task scales: Assuming that the number of tasks in the initial planning scheme is represented by m and the number of emergency tasks is represented by n, the task scale of the emergency task planning problem of agile imaging satellite can be expressed as m:n. The task scales are set to 20:5, 20:10, 25:5, 25:10, 25:15, 30:5, 30:10, 30:15, 30:20, 35:5, 35:10, 35:15, 35:20, 40:5, 40:10, 40:15 and 40:20, and 17 groups of task scales are simulated and represented by letters A to Q respectively. The tasks in the initial planning scheme and the emergency tasks are both strip targets. The priority of the tasks in the initial planning scheme is 1 to 5, and the priority of the emergency tasks is 6 to 10. The larger the value, the higher the priority.

[0087] (2) Different task types: Under the condition of a task scale of 40:20, emergency tasks were set for simulation of strip targets and regional targets respectively, and the task priority was set the same as the previous set of simulations.

[0088] Figure 7 to Figure 11 The feasibility and effectiveness of the algorithm were verified, indicating that an emergency task planning method based on a window sliding strategy has the characteristics of small disturbance to the initial plan, high total task benefit and fast running speed.

Claims

1. An agile satellite emergency mission planning method, characterized in that: include: Step 1: Receive the emergency mission, calculate the satellite visibility to the emergency mission, and obtain the visible time window set ETS of the emergency mission; Step 2: Sort all emergency tasks in descending order of priority to obtain an emergency task list; Step 3: Select the emergency task T with the highest current priority in the task list i , in its visible time window [ws i ,we i ] Select a certain time as the start execution time ts of the task i , the emergency task T i Insert the current plan into Schedule to get the inserted plan Schedule', and add the emergency task T i Remove it from the emergency task list; Step 4: Determine whether there is a time window conflict in the inserted plan Schedule'. If there is no conflict, use the inserted plan Schedule' as the new current plan Schedule new Otherwise, the window sliding strategy is used to resolve the conflict and obtain the new current planning scheme Schedule new ; Step 4.1: Insert the emergency task as the focus task T v ; Step 4.2: Determine the focus task T v Is the number of moves greater than the maximum number of moves M for a single task? T If yes, set the window sliding execution result flag to 0, compare the priorities of the current conflicting tasks, retain the tasks with higher priorities, and remove the tasks with lower priorities. If the priorities are the same, retain the tasks with shorter execution time and execute step 4.6; otherwise, execute step 4.3; Step 4.3: Determine T v The adjacent task T on the left L and the adjacent task T on the right R The conflict relationship in the time window, when T v With T L or T R If there is a time window conflict, execute step 4.4; Otherwise, the window sliding strategy succeeds, the window sliding execution result flag is set to 1, and step 6 is executed; Step 4.4: When T v With T L When there is a time window conflict, determine T L The left margin and T v Whether the right margin of satisfies both 0 and T v With T R When there is a time window conflict, determine T R The right margin and T v Whether the left margin of is satisfied and is 0 at the same time, if it is satisfied, compare the priorities of the conflicting tasks, retain the tasks with higher priority, and remove the tasks with lower priority. If the priorities are the same, retain the tasks with shorter execution time, set the window sliding execution result flag to 0, and execute step 4.6; otherwise, execute step 4.5; Step 4.5: When T v With T L When there is a time window conflict: If T v Priority over T L , and T L The left margin of is not 0, or T v The priority is lower than T L , but T v The right margin is 0, then T L Shift left, the task T will be moved L Treat as attention task T v , focus on task T v The number of moves of is reset to 1, and step 4.2 is executed; otherwise, T v Move right and focus on task T v The number of moves is increased by 1, and step 2 is executed; when T v With T R When there is a time window conflict: If T v Priority over T R , and T R The right margin is not 0, or T v The priority is lower than T R , but T v The left margin is 0, then T R Move right, task T will be moved R Treat as attention task T v , focus on task T v The number of moves of is reset to 1, and step 4.2 is executed; otherwise, T v Move left and focus on task T v Add 1 to the number of moves and execute step 4.2; Step 4.6: Return the window sliding execution result flag and the current planning scheme, and end the window sliding strategy; Step 5: Determine the new current plan schedule new Whether all constraints are met, if so, the current plan Schedule is updated to the new current plan Schedule new ; Otherwise, the current plan Schedule remains unchanged; Step 6: Determine whether to traverse all emergency tasks. If so, output the current planning scheme Schedule, otherwise return to step 3.

2. The method for agile satellite emergency mission planning according to claim 1, characterized in that: The left margin and right margin describe the range within which the execution time window can slide, specifically: Left margin yd l It is the smaller value between the difference Δt1 between the task execution start time and its earliest observable time and the difference Δt2 between the task execution start time and the execution end time of the task on its left, satisfying: yd l =min(Δt1,Δt2) Right margin yd r It is the smaller value between the difference Δt1 between the task execution end time and its latest observable time and the difference Δt2 between the task execution end time and the execution start time of the task on its right, satisfying: yd r =min(Δt1,Δt2)。 3. The method for planning an agile satellite emergency mission according to claim 1, characterized in that: The constraints are specifically as follows: the observation time of any target must be within its visible time window, the attitude angle of the satellite at any time must be less than the range of the maximum attitude angle, the minimum solar altitude angle requirement must be met when performing observation tasks, the on-board storage space of the satellite does not exceed the maximum allowable storage capacity, the power of the satellite at any time must be greater than the minimum power requirement, two adjacent tasks performed on the same satellite must have a sufficient time interval to ensure that the onboard remote sensor can complete the attitude conversion, the number of satellite power ons and off times in a single orbit cannot exceed the number in any orbital circle, and the cumulative duration of all tasks performed within the time period of any orbital cycle of the satellite resource operation shall not exceed the given maximum working time of a single circle.

Images