Multi-star Earth Observation Mission Planning Method and System

Abstract

The present invention provides a multi-satellite earth observation mission planning method and system, which relates to the technical field of satellite mission scheduling. The present invention comprehensively considers the observation cost and observation benefit of the mission to screen out the most suitable observation time window for each observation mission, which can effectively reduce the conflict between the observation time windows of different observation missions in the same satellite orbit, and improve the overall utilization rate of satellite resources. The present invention also sequentially selects the tasks to be planned and inserts them into the task planning sequence of the currently planned satellite. When selecting tasks, it is necessary to comprehensively consider the task observation benefit and the satellite attitude adjustment time. When inserting, it is necessary to verify the angle constraint and the longest power-on time constraint to obtain a better planning result. The present invention also re-plans the tasks to be planned, so as to reduce task conflicts, improve observation benefits, and maximize the utilization of satellite resources.

Description

Technical Field

[0001] The present invention relates to the technical field of satellite mission scheduling, and particularly relates to a multi-satellite earth observation mission planning method and system. Background Art

[0002] Multi-satellite earth observation mission planning can be described as allocating reasonable satellite resources for each observation mission among multiple satellites and multiple observation missions under complex constraint conditions. In recent years, with the increasing maturity of satellite technology, satellite-based observations have been widely applied in many fields such as post-disaster assessment and hot-spot area monitoring. How to reasonably arrange satellite resources for observation tasks has become an urgent problem to be solved for the wide application of satellite technology.

[0003] Existing methods generally default that satellites use a single observation method to observe mission targets, and generally randomly select an observation time window for observation when the observation requirements are met.

[0004] However, the above existing methods have the problem of low utilization rate of satellite observation resources. Summary of the Invention

[0005] (1) Technical Problems to be Solved

[0006] Aiming at the deficiencies of the existing technology, the present invention provides a multi-satellite earth observation mission planning method and system, which solves the problem of low utilization rate of satellite observation resources existing in the existing methods.

[0007] (2) Technical Solutions

[0008] To achieve the above purposes, the present invention is realized through the following technical solutions:

[0009] In a first aspect, a multi-satellite earth observation mission planning method is provided, and the method includes:

[0010] Constructing a set of tasks to be planned corresponding to each satellite orbit based on an objective function;

[0011] Performing mission planning on the set of tasks to be planned for each satellite orbit to obtain a corresponding mission planning sequence;

[0012] Reprogramming the remaining tasks to be planned in each set of tasks to be planned.

[0013] Further, the constructing a set of tasks to be planned corresponding to each satellite orbit based on an objective function includes:

[0014] S101. Calculating each task to be planned T in the task set T = {T1, T2,..., T m ,..., T M}m Set of observation time windows For each observation time window in The corresponding objective function value 1 ≤ i ≤ I, 1 ≤ j ≤ J, 1 ≤ m ≤ M, 1 ≤ n ≤ N;

[0015] S102. Add the optimal objective function value The corresponding task to be scheduled T m To this time window Of the satellite cycle SC ij In the set SCA of tasks to be scheduled ij At the same time, remove T from T, and set the remaining time windows outside the optimal observation time window of T to invalid; m Delete T from T, and set the remaining time windows outside the optimal observation time window of T to invalid; m Set the remaining time windows outside the optimal observation time window to invalid;

[0016] S103. Repeat S101 - S102 until all tasks to be scheduled are assigned, obtaining the set SCA of tasks to be scheduled corresponding to each satellite cycle ij ;

[0017] Wherein,

[0018] S i Represents the i-th satellite, and there are I satellites in total;

[0019] SC = {SC1, SC2,..., SC i ,..., SC I} Represents the set of satellite cycles;

[0020] SC i = {SC i1 , SC i2 ,..., SC ij ,..., SC iJ} Represents the set of cycles of satellite S i ;

[0021] SC ij Represents the j-th cycle of satellite S i ;

[0022] T = {T1, T2,..., T m ,..., T M} Represents a total of M tasks to be scheduled;

[0023] T m Represents the m-th task to be scheduled;

[0024] Represents the set of observation time windows of the task to be scheduled T m , with a total of N;

[0025] Denote T m The nth observation time window of T in satellite S i in the jth orbit of S;

[0026] SCA ij Denote the set of tasks to be planned for the jth orbit of satellite S i as SCA.

[0027] Furthermore, the calculation method of the objective function value is as follows:

[0028]

[0029] where

[0030] denotes the sum of the observation task benefits of the remaining valid observation time windows that have time overlap with the observation time window ;

[0031] TP m Denote the observation benefit of T m as TP;

[0032] Denote the observation duration of T m in the nth observation time window as

[0033] ; m respectively denote the end and start observation times of T

[0034] SCT ij Denote the longest power-on time of satellite S i in the jth orbit as SCT.

[0035] Denote the influence degree of the observation of T on other observation tasks when observing T m ; the smaller, the better;

[0036] Denote the proportion of the resources consumed by satellite S when executing T m in this orbit to the satellite resources in this orbit; the smaller, the better.

[0037] Furthermore, perform task planning for the set of tasks to be planned for each satellite orbit to obtain the corresponding task planning sequence, including:

[0038] Perform task planning for the set of tasks to be planned for each satellite orbit in sequence, and the steps for performing task planning for the set of tasks to be planned for any satellite orbit are as follows:

[0039] S201. Determine S from the set of tasks to be planned SCA ij and determine the first task T i executed in the j - th loop, start and delete the first task T start from the set of tasks to be planned SCA ij ;

[0040] S202. Traverse each task to be planned T ij in SCA m ; If TS m −TE l ≥0, then add T m to the set of tasks to be arranged SCW ij ;

[0041] Among them,

[0042] T l represents the last task in the task planning sequence SCG ij ;

[0043] TE l represents the end observation time of T l , and when the best observation window of T l is its n - th observation time window,

[0044] S203. Determine whether SCW ij is an empty set;

[0045] If SCW ij is an empty set, perform task planning for the next satellite loop;

[0046] If SCW ij is not an empty set, select the task to be arranged T ij from SCW a ;

[0047] S204. If the task to be arranged T a simultaneously satisfies the angle adjustment constraint and the longest power - on time constraint, then add T a to the task planning sequence SCG ij , delete T a from SCA ij , clear SCW ij , and then return to S202;

[0048] Otherwise, delete T a from SCW ij , and then return to S203.

[0049] Furthermore, the said from the set of tasks to be planned SCA ijDetermine S i The first task T executed in the j-th cycle start , including:

[0050] Select the first task T using the following formula start :

[0051]

[0052] And the selected task T to be arranged from SCW ij , including: a , including:

[0053] Select the task T to be arranged from SCW using the following formula ij : a :

[0054]

[0055] Wherein,

[0056] TS m represents the start observation time of T m , and when the best observation window of T m is its n-th observation time window ,

[0057] TP m represents the obtained benefit after the completion of the task T to be planned m ;

[0058] λ is a balance coefficient used to balance the dimension.

[0059] Furthermore, the angle adjustment constraint is:

[0060] TE l +t la ≤TS a

[0061] Wherein,

[0062] represents S i from the execution of T l to the execution of T a the posture adjustment time between;

[0063] TA l , TA a respectively represent the best observation angles of T l and T a ;

[0064] represents S i unit time angle deflection rate;

[0065] TS a represents the start observation time at T a ;

[0066] And the longest power-on time constraint is:

[0067] TE a ≤TS start +SCT ij

[0068] Wherein,

[0069] TE a represents the end observation time at T a ;

[0070] TS start represents the start observation time of the first task T start ;

[0071] SCT ij represents the longest power-on time of S i in the jth cycle.

[0072] Furthermore, the replanning of the remaining to-be-planned tasks in each to-be-planned task set includes:

[0073] If the to-be-planned task set SCA ij is an empty set, no replanning is performed;

[0074] Otherwise, the to-be-planned task T ij is selected from SCA m' in sequence according to the order of the task start observation time for replanning;

[0075] And the replanning includes: insertion strategy, synthesis strategy, replacement strategy, recombination strategy, and the priority is insertion strategy > synthesis strategy > replacement strategy > recombination strategy.

[0076] Furthermore, the insertion strategy includes:

[0077] Select two adjacent planned tasks T ij and T f and T b in sequence according to the order of the task start observation time in the task planning sequence SCG

[0078] If both TE f +t fm' ≤TS m' , TE m' +t m'b ≤TS b , then insert the task T m' into SCG ijT in the middle f and T b in between, indicating S i Execute T successively at the j-th cycle f , T m' and T b , and delete T m' from SCA ij ;

[0079] Among them,

[0080] TE f represents the end observation time of T f ;

[0081] indicates S i from the execution of T f to the execution of T m' the posture adjustment time in between;

[0082] TA f , TA m' respectively represent the best observation angles of T f and T m' ;

[0083] indicates S i the angular deflection rate per unit time;

[0084] TS m' and TE m' respectively represent the start observation time and end observation time of T m' ;

[0085] indicates S i from the execution of T m' to the execution of T b the posture adjustment time in between;

[0086] TA m' , TA b respectively represent the best observation angles of T m' and T b ;

[0087] TS b represents the start observation time of T b ;

[0088] The synthesis strategy includes:

[0089] Select the planned task T ij in the task planning sequence SCG c in turn according to the order of the start observation time of the task, and SCG ijThe first task and the last task in it are not selected;

[0090] If T c and T m' observation time windows have an intersection, and at the same time satisfy TP d ≥TP c and TE f +t fd ≤TS d , TE d +t db ≤TS b , then use T d to replace T in SCG ij , and delete T c from SCA m' ; ij

[0091] Among them, T d represents the new task synthesized by T c and T m' . T f and T b represent the previous task and the next task adjacent to T ij in SCG; c

[0092] TP d = TP m' (1 - θ|TA m' - TA d |)+ TP c (1 - θ|TA c - TA d |) represents the benefit of T d ; c

[0093] TP c , TP m' represent the observation benefits of T c and T m' ;

[0094] TA d =(TA m' + TA c ) / 2, TA m' , TA c respectively represent the best observation angles of T d , T m' and T c ;

[0095] θ represents the loss rate of observation benefit per unit angle deviation between the actual observation angle and the best observation angle of the task when performing the observation task;

[0096] TE fIndicates the end observation time of T f ;

[0097] Indicates the posture adjustment time from the execution of T i to T f ; d ;

[0098] TA f Indicates the best observation angle of T f ;

[0099] Indicates the angular deflection rate per unit time of S i ;

[0100] TS d = min(TS c , TS m' ) and TE d = max(TE c , TE m' ) respectively indicate the start observation time and end observation time of T d ;

[0101] min(TS c , TS m' ) indicates taking the smaller value of the start observation times of T c and T m' ;

[0102] max(TE c , TE m' ) indicates taking the larger value of the end observation times of T c and T m' ;

[0103] Indicates the posture adjustment time from the execution of T i to T d ; b ;

[0104] TA b Indicates the best observation angle of T b ;

[0105] TS b Indicates the start observation time of T b ;

[0106] The replacement strategy includes:

[0107] Select the planned task T ij in the task planning sequence SCG c in the order of the start observation time of the task, and SCG ijThe first and last tasks in the game are not selected;

[0108] If both TP c ≤TP m' ,TE f +t fm' ≤TS m' ,TE m' +t m'b ≤TS b , then use T m' Replace SCG ij T c , and T m' From SCA ij Delete T c Join SCA ij middle;

[0109] Among them, T f ,T b T m At SCG ij The previous task and the next task in ;

[0110] TP c ,TP m' Respectively represent T c and T m' of income;

[0111] TE f Indicates T f The end time of observation;

[0112] Indicates S i From the execution of T f to T m' The time between posture adjustments,

[0113] TA f ,TA m' Respectively represent T f and T m' The best viewing angle,

[0114] Indicates S i Angular deflection rate per unit time;

[0115] TS m' and TE m' Respectively represent T m' The start and end observation times of

[0116] Indicates S i From the execution of T m' to T bThe posture adjustment time between

[0117] TA m' and TA b respectively represent the best observation angles of T m' and T b ;

[0118] TS b represents the start observation time of T b ;

[0119] The recombination strategy includes:

[0120] If the task T to be planned m' does not meet the insertion, synthesis, and replacement strategies, then the observation time window on the current satellite circle is abandoned for observing T m' , and T m' is deleted from SCA ij , and the observation time window of T m' on other satellite circles where replanning has not been performed is selected, and then the three strategies of insertion, synthesis, and replacement are used for planning.

[0121] In a second aspect, a multi-satellite earth observation mission planning system is provided. The system includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0122] (III) Beneficial effects

[0123] The present invention provides a multi-satellite earth observation mission planning method and system. Compared with the prior art, the following beneficial effects are achieved:

[0124] The present invention comprehensively considers the observation cost and observation benefit of the mission to screen out the most suitable observation time window for each observation mission, which can effectively reduce the conflict between the observation time windows of different observation missions on the same satellite circle and improve the overall utilization rate of satellite resources. The present invention also sequentially selects the tasks to be planned and inserts them into the task planning sequence of the currently planned satellite. When selecting tasks, the observation benefit of the task and the satellite posture adjustment time need to be comprehensively considered. When inserting, the angle constraint and the longest power-on time constraint need to be verified to obtain a better planning result. The present invention also re-plans the tasks to be planned to reduce task conflicts, improve the observation benefit, and maximize the utilization of satellite resources. Description of the drawings

[0125] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0126] Figure 1 is a flow chart of an embodiment of the present invention;

[0127] Figure 2 A flowchart of constructing a set of tasks to be planned corresponding to each satellite circle based on an objective function according to an embodiment of the present invention;

[0128] Figure 3 The present invention is a flowchart of performing task planning for a set of tasks to be planned for each satellite orbit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0129] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0130] The embodiments of the present application solve the problem of low satellite observation resource utilization in existing methods by providing a multi-satellite earth observation mission planning method and system.

[0131] The technical solution in the embodiment of the present application is to solve the above technical problems, and the overall idea is as follows:

[0132] In traditional research, it is generally assumed that satellites use a single observation method to observe mission targets. Single observation means that satellite sensors can only observe one task at a certain angle. Under this observation method, for two observation tasks with overlapping time windows, the observation cannot be completed at the same time, and only one of the tasks can be selected for observation. This will reduce the number of observation tasks and waste satellite observation resources.

[0133] Secondly, for an observation mission, it usually contains multiple observation time windows on different satellite orbits. In traditional satellite mission planning, an observation time window is generally randomly selected for observation when the observation requirements are met. Although this selection method is simple and convenient, the randomly selected observation time window may have a high conflict rate with the observation time windows of other missions within the satellite circle to which it belongs, resulting in many observation time windows in the satellite circle being unusable, reducing the overall utilization of satellite resources.

[0134] Finally, in the traditional satellite mission planning model, it is generally assumed that satellite resources are sufficient, and only reasonable satellite resources need to be allocated for each mission for observation. However, in practical applications, satellite resources are often scarce and cannot meet the needs of all observation missions simultaneously. Therefore, it is necessary to comprehensively consider the observation benefits and costs of missions and selectively select missions for observation to maximize the utilization of satellite resources. The traditional satellite mission planning model cannot solve the satellite mission planning problem under the condition of insufficient satellite resources.

[0135] Based on the above defects:

[0136] (1) The present invention designs a time window screening method that comprehensively considers the observation costs and benefits of missions to screen out the most suitable observation time window for each observation mission, which can effectively reduce the conflicts between the observation time windows of different observation missions in the same satellite orbit and improve the overall utilization rate of satellite resources. In the traditional mode, generally a random observation time window is selected under the condition of meeting the constraints for observation. Although this method has a fast selection speed, the conflict degree between the observation time windows of different missions is high, and the utilization rate of satellite resources is poor.

[0137] (2) The present invention designs a dynamic sequence construction method for generating an initial planning scheme. The basic idea is to sequentially select the missions to be planned and insert them into the mission planning sequence of the currently planned satellite. When selecting a mission, it is necessary to comprehensively consider the mission observation benefits and the satellite attitude adjustment time. When inserting, it is necessary to verify the angle constraint and the longest power-on time constraint. The dynamic programming method can make a choice from a global perspective in the process of continuous change of mission resources, so as to obtain a better planning result. The traditional planning method generally selects missions for planning in a random or fixed manner and does not change with the change of the missions to be planned and the remaining satellite resources during the planning process. Although it saves time, it is easy to fall into a local optimum, with a high mission conflict degree, resulting in a poor final planning result.

[0138] (3) The present invention designs a replanning method for replanning the missions to be planned, which includes four strategies: insertion, synthesis, replacement, and recombination. It can effectively help the satellite selectively select missions with higher observation benefits for priority observation under the condition of insufficient resources, and at the same time can perform synthetic observation on missions with overlapping time windows to reduce mission conflicts, improve observation benefits, and maximize the utilization of satellite resources. The traditional satellite mission planning model cannot solve the satellite mission planning problem under the condition of insufficient satellite resources.

[0139] To better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0140] Embodiment 1:

[0141] As Figure 1 shown, the present invention provides a multi-satellite earth observation mission planning method, which is executed by a computer. The method includes:

[0142] Constructing a set of tasks to be planned corresponding to each satellite orbit based on an objective function;

[0143] Performing mission planning on the set of tasks to be planned for each satellite orbit to obtain a corresponding mission planning sequence;

[0144] Re-planning the remaining tasks to be planned in each set of tasks to be planned.

[0145] The beneficial effects of this embodiment are:

[0146] The present invention comprehensively considers the observation cost and observation benefit of the mission to screen out the most suitable observation time window for each observation task, which can effectively reduce the conflict between the observation time windows of different observation tasks in the same satellite orbit and improve the overall utilization rate of satellite resources. The present invention also sequentially selects the tasks to be planned and inserts them into the mission planning sequence of the currently planned satellite. When selecting tasks, it is necessary to comprehensively consider the mission observation benefit and the satellite attitude adjustment time. When inserting, it is necessary to verify the angle constraint and the longest power-on time constraint to obtain a better planning result. The present invention also re-plans the tasks to be planned to reduce task conflicts, improve observation benefits, and make the most of satellite resources.

[0147] The implementation process of the embodiment of the present invention will be described in detail below:

[0148] In this embodiment, it is defined that:

[0149] S = {S1, S2,... S i ,..., S I} represents the satellite set, with a total of I satellites;

[0150] S i (1 ≤ i ≤ I) represents the i-th satellite;

[0151] SC = {SC1, SC2,..., SC i ,..., SC I} represents the set of satellite orbits;

[0152] SC i = {SC i1 , SC i2 ,..., SC ij ,..., SC iJ} represents the set of orbits of satellite S i ;

[0153] SC ij (1 ≤ i ≤ I, 1 ≤ j ≤ J) represents the j-th cycle of satellite S i ;

[0154] SCT ij (1 ≤ i ≤ I, 1 ≤ j ≤ J) represents the longest on-time of satellite S i in the j-th cycle;

[0155] T = {T1, T2,..., T m ,..., T M} represents all tasks to be planned (observation tasks), a total of M tasks;

[0156] T m represents the m-th task to be planned;

[0157] For each task T to be planned m , it has the opportunity to be observed in different satellite cycles, and each observation opportunity is called an observation time window;

[0158] represents the set of observation time windows of T m , a total of N observation time windows;

[0159] represents that the n-th observation time window of T m is in the j-th cycle of S i ;

[0160] represents the best observation angle of T m in the n-th observation time window;

[0161] represents the start observation time of T m in the n-th observation time window;

[0162] represents the end observation time of T m in the n-th observation time window;

[0163] TP m (1 ≤ m ≤ M) represents the benefit obtained after completing the task T to be planned m ;

[0164] represents the angular deflection rate per unit time of satellite S i ;

[0165] In this embodiment, it mainly includes the following three steps:

[0166] (1) Observation time window screening: For each task to be planned, it has the opportunity to be observed in different satellite orbits. Each observation opportunity is called an observation time window. Selecting the most suitable observation time window from numerous time windows to observe the task helps avoid task conflicts and improve the overall utilization rate of satellite resources. That is, a set of tasks to be planned is constructed for each satellite orbit.

[0167] (2) Task planning: After constructing a set of tasks to be planned for each satellite orbit, perform the task planning operation to determine the task planning sequence for each satellite orbit.

[0168] (3) Task replanning: After task planning, due to limited satellite resources (the power-on time of the satellite in each orbit), there may be some observation tasks that are not planned. For these tasks, we use four strategies: insertion, synthesis, replacement, and recombination for replanning to increase the number of task observations and improve the utilization rate of satellite resources.

[0169] Therefore, in specific implementation, it includes the following steps:

[0170] S1. Based on the objective function, construct a set of tasks to be planned corresponding to each satellite orbit.

[0171] As Figure 2 shown below, a feasible method to construct a set of tasks to be planned is given:

[0172] S101. Calculate the objective function values corresponding to each observation time window in the observation time window set m ,...,T M} in the task set T = {T1, T2,..., T m} of each task to be planned T where each observation time window in

[0173] Among them, 1 ≤ i ≤ I, 1 ≤ j ≤ J, 1 ≤ m ≤ M, 1 ≤ n ≤ N.

[0174] In specific implementation, the observation time window screening mainly selects the most suitable observation time window for the observation task by comprehensively considering the observation cost and observation benefit. Therefore, the calculation formula of the objective function value can be:

[0175]

[0176] Among them,

[0177] represents the sum of the observation task benefits of the remaining valid observation time windows that have time overlap with the observation time window ;

[0178] An observation task has multiple observation time windows. After the optimal observation time window is selected for this task, the remaining observation time windows of this task are set to invalid. When calculating if there is an intersection between an invalid time window and a time window, the observation benefit of the task corresponding to this invalid time window is not calculated into .

[0179] TP m represents the observation benefit of T m ;

[0180] represents the observation duration of T m in the nth observation time window.

[0181] respectively represent the end and start observation times of T m in the nth observation time window.

[0182] SCT ij represents the longest power-on time of satellite S i in the jth orbit.

[0183] represents the degree of influence of the observation of T on other observation tasks. The smaller it is, the smaller the influence and the better the observation benefit can be obtained. m ;

[0184] Since the power-on time of the satellite is limited in each orbit, represents the proportion of resources consumed by the satellite after executing T m in this orbit to the satellite resources in this orbit. The smaller it is, the more resources the satellite has left to execute other tasks after observing T m .

[0185] Therefore, the smaller it is, the more suitable it is to use its nth observation time window to observe the observation task T m .

[0186] S102. Add the optimal objective function value corresponding to the task to be planned T m to the set of tasks to be planned SCA of the satellite orbit SC ij where this time window is located. At the same time, delete T ij from T, and set the remaining time windows outside the optimal observation time window of T m to invalid. m

[0187] Among them, the set of tasks to be planned SCAij Denote all the tasks to be planned for the \(j\)-th orbit of satellite \(S\). i

[0188] S103. Repeat S101 - S102 to allocate the next \(T\) m until all the tasks to be planned in \(T\) are allocated to a satellite \(S\) i in the \(j\)-th orbit.

[0189] Thus, the set of tasks to be planned corresponding to each satellite in each orbit can be obtained.

[0190] S2. Perform task planning for the set of tasks to be planned \(S_{CA}\) ij (1 ≤ \(i\) ≤ \(I\), 1 ≤ \(j\) ≤ \(J\)) of each satellite orbit \(SC\) ij to obtain the corresponding task planning sequence \(SC_G\). ij .

[0191] The order of task planning for each orbit does not matter. It can be planned in ascending order of \(i\) and \(j\), which is simple and convenient. Therefore, as Figure 3 shown below, a feasible method for obtaining any task planning sequence \(SC_G\) is given: ij

[0192] S201. First task selection: Determine the first task \(T\) ij executed by satellite \(S\) i in the \(j\)-th orbit from the set of tasks to be planned \(S_{CA}\) start , and delete the first task \(T\) start from the set of tasks to be planned \(S_{CA}\) ij .

[0193] Specifically, the following formula is used to determine the first task \(T\) start :

[0194]

[0195] where

[0196] \(T_S\) m represents the start observation time of \(T\) m , and when the best observation window of \(T\) m is its \(n\)-th observation time window ,

[0197] \(T_P\) m represents the revenue obtained after completing the task to be planned \(T\) m ;

[0198] \(\lambda\) is a balance coefficient used to balance dimensions;

[0199] ​​Generally, it is defaulted that the startup time of the satellite in a certain orbit is the start observation time of the satellite performing the first task in that orbit.

[0200] S202. Establish the set of tasks to be scheduled: Traverse each task to be planned T ij in SCA m . If it satisfies TS m - TE l ≥0, that is, the start time of the task to be planned is later than the end time of the last task in the planned sequence, then add T m to the set of tasks to be scheduled SCW ij .

[0201] Among them,

[0202] T l represents the last task in the task planning sequence SCG ij , that is, the last task arranged in the current satellite orbit;

[0203] TE l represents the end observation time of T l , and when the best observation window of T l is its nth observation time window,

[0204] S203. Selection of tasks to be scheduled: Judge whether SCW ij is an empty set;

[0205] If SCW ij is an empty set, it means that there are no more observation tasks to be scheduled in the current satellite orbit, and start to plan tasks for the next satellite orbit;

[0206] If SCW ij is not an empty set, then select the task to be scheduled T ij from SCW a .

[0207] Specifically, the task to be scheduled T ij can be selected from SCW a using the following formula:

[0208]

[0209] S204. If the task to be scheduled T a simultaneously satisfies the angle adjustment constraint and the longest startup time constraint, then add T a to the task planning sequence SCG ij , delete T a from SCA ij , clear SCW ij , and then return to S202;

[0210] Otherwise, delete T a from SCW ij and then return to S203 to determine whether SCW ij is an empty set at this time.

[0211] The angle adjustment constraint is:

[0212] TE l + t la ≤ TS a

[0213] Where

[0214] represents the posture adjustment time between the execution of T i and the execution of T l to the execution of T a ;

[0215] TA l , TA a respectively represent the best observation angles of T l and T a ;

[0216] represents the angular deflection rate per unit time of S i ;

[0217] TS a represents the start observation time of T a ;

[0218] And the longest power-on time constraint is:

[0219] TE a ≤ TS start + SCT ij

[0220] Where

[0221] TE a represents the end observation time of T a ;

[0222] TS start represents the start observation time of the first task T start ;

[0223] SCT ij represents the longest power-on time of S i in the jth cycle.

[0224] So far, the task planning sequence SCG ij of a satellite cycle can be determined. After repeating the above steps, the task planning sequences SCG of all satellite cycles can be obtainedij 。

[0225] S3. Re-plan the remaining to-be-planned tasks in each to-be-planned task set.

[0226] The following gives a feasible re-planning method:

[0227] Traverse the to-be-planned task set SCA of each satellite cycle according to the principle that i and j increase from small to large ij , if SCA ij is an empty set, it means that all the to-be-planned tasks in this cycle have been planned and no re-planning is required.

[0228] Otherwise, select the to-be-planned task T ij from SCA m' in sequence according to the order of the start observation time of the tasks for re-planning. The re-planning consists of four strategies: insertion, synthesis, replacement, and recombination.

[0229] Among them,

[0230] (1) The insertion strategy means: on the basis of not affecting the planned tasks, insert the to-be-planned task between two planned tasks in the task planning sequence.

[0231] The following gives the steps of a feasible insertion strategy:

[0232] Select two adjacent planned tasks T ij and T f in the task planning sequence SCG b in sequence according to the order of the start observation time of the tasks;

[0233] If both TE f +t fm' ≤TS m' , TE m' +t m'b ≤TS b are satisfied, then insert the task T m' into SCG ij between T f and T b , indicating that S i executes T f , T m' and T b in sequence in the jth cycle, and delete T m' from SCA ij ;

[0234] Among them,

[0235] TE f represents the end observation time of T f ;

[0236] Represent S i From the execution of T f To the execution of T m' The posture adjustment time in between;

[0237] TA f and TA m' respectively represent the best observation angles of T f and T m' ;

[0238] Represent S i The angular deflection rate per unit time;

[0239] TS m' and TE m' respectively represent the start observation time and end observation time of T m' ;

[0240] Represent S i From the execution of T m' To the execution of T b The posture adjustment time in between;

[0241] TA m' and TA b respectively represent the best observation angles of T m' and T b ;

[0242] TS b represents the start observation time of T b .

[0243] (2) The synthesis strategy means: On the basis of not affecting the planned tasks, synthesizing the to-be-planned tasks with the planned tasks in the task planning sequence for observation.

[0244] The following gives the steps of a feasible synthesis strategy:

[0245] Select the planned task T ij in the task planning sequence SCG c in turn according to the order of the task start observation time, and do not select the first task and the last task in SCG ij ;

[0246] If the observation time windows of T c and T m' have an intersection, and at the same time satisfy TP d ≥TP c and TE f +t fd ≤TS d, TE d +t db ≤TS b , then use T d to replace T in SCG ij and delete T from SCA c ; m' from SCA ij ;

[0247] where T d represents the new task synthesized by T c and T m' , T f and T b represent the previous task and the next task adjacent to T ij in SCG c ;

[0248] TP d = TP m' (1 - θ|TA m' - TA d |) + TP c (1 - θ|TA c - TA d |) represents the benefit of T d ;

[0249] TP c , TP m' represent the observed benefits of T c and T m' ;

[0250] TA d = (TA m' + TA c ) / 2, TA m' , TA c respectively represent the best observation angles of T d , T m' and T c ;

[0251] θ represents the loss rate of the observation benefit per unit angle deviation between the actual observation angle and the best observation angle when performing the observation task;

[0252] TE f represents the end observation time of T f ;

[0253] represents the posture adjustment time of S i from the execution of T f to T d ;

[0254] TA f represents Tf Optimal observation angle;

[0255] Denote S i Angle deflection rate per unit time;

[0256] TS d = min(TS c , TS m' ) and TE d = max(TE c , TE m' ) respectively denote the start observation time and end observation time of T d ;

[0257] min(TS c , TS m' ) means taking the smaller value of the start observation times of T c and T m' ;

[0258] max(TE c , TE m' ) means taking the larger value of the end observation times of T c and T m' ;

[0259] Denote S i The posture adjustment time from the execution of T d to T b ;

[0260] TA b Denote the optimal observation angle of T b ;

[0261] TS b Denote the start observation time of T b ;

[0262] (3) The replacement strategy means: on the basis of not affecting the planned tasks, replacing the to-be-planned task with the planned task in the task planning sequence.

[0263] The following gives the steps of a feasible replacement strategy:

[0264] Select the planned task T ij in the task planning sequence SCG c in turn according to the order of the start observation times of the tasks, and do not select the first task and the last task in SCG ij ;

[0265] If TP c ≤ TP m' , TEf +t fm' ≤TS m' ,TE m' +t m'b ≤TS b , then use T m' Replace SCG ij T c , and T m' From SCA ij Delete T c Join SCA ij middle;

[0266] Among them, T f ,T b T m At SCG ij The previous task and the next task in ;

[0267] TP c ,TP m' Respectively represent T c and T m' of income;

[0268] TE f Indicates T f The end time of observation;

[0269] Indicates S i From the execution of T f to T m' The time between posture adjustments,

[0270] TA f ,TA m' Respectively represent T f and T m' The best viewing angle,

[0271] Indicates S i Angular deflection rate per unit time;

[0272] TS m' and TE m' Respectively represent T m' The start and end observation times of

[0273] Indicates S i From the execution of T m' to T b The time between posture adjustments,

[0274] TA m' , T.A. b Respectively represent T m'and T b The best observation angle of;

[0275] TS b Indicates T b The start observation time of.

[0276] (4) The recombination strategy refers to: Selecting the observation time window of the task to be planned in other satellite orbits, and using the three strategies of insertion, synthesis, and replacement for planning.

[0277] The following gives the steps of a feasible recombination strategy:

[0278] If the task T to be planned m' Does not meet the insertion, synthesis, and replacement strategies, then abandon the use of the observation time window in the current satellite orbit to observe T m' Perform the observation, delete T m' From SCA ij Delete it, select the observation time window of T m' On other satellite orbits that have not performed replanning, and then use the three strategies of insertion, synthesis, and replacement for planning.

[0279] Embodiment 2:

[0280] A multi-satellite earth observation task planning system, the system includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:

[0281] Construct a set of tasks to be planned corresponding to each satellite orbit based on the objective function;

[0282] Perform task planning for the set of tasks to be planned for each satellite orbit to obtain the corresponding task planning sequence;

[0283] Perform replanning on the remaining tasks to be planned in each set of tasks to be planned.

[0284] It can be understood that the multi-satellite earth observation task planning system provided by the embodiments of the present invention corresponds to the above multi-satellite earth observation task planning method. For the explanations, examples, beneficial effects, etc. of the relevant content, reference can be made to the corresponding content in the multi-satellite earth observation task planning method, and details are not described here.

[0285] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0286] (1) The present invention designs a time window screening method, which comprehensively considers the observation cost and observation benefit of tasks to screen the most suitable observation time window for each observation task, can effectively reduce the conflict between the observation time windows of different observation tasks in the same satellite orbit, and improve the overall utilization rate of satellite resources. In the traditional mode, generally, a random observation time window is selected under the condition of meeting the constraints for observation. Although this method has a fast selection speed, the conflict degree between the observation time windows of different tasks is high, and the utilization rate of satellite resources is poor.

[0287] (2) The present invention designs a dynamic sequence construction method for generating an initial planning scheme. The basic idea is to sequentially select the tasks to be planned and insert them into the task planning sequence of the currently planned satellite. When selecting tasks, it is necessary to comprehensively consider the task observation benefit and the satellite attitude adjustment time. When inserting, it is necessary to verify the angle constraint and the longest power-on time constraint. The dynamic programming method can make a choice from a global perspective in the face of the continuous change of task resources, so as to obtain a better planning result. In the traditional planning method, generally, tasks are selected for planning in a random or fixed manner, and it will not change with the change of the tasks to be planned and the remaining satellite resources during the planning process. Although it saves time, it is easy to fall into a local optimum, with a high task conflict degree, resulting in a poor final planning result.

[0288] (3) The present invention designs a replanning method for replanning the tasks to be planned. It includes four strategies: insertion, synthesis, replacement, and recombination, which can effectively help the satellite selectively select tasks with higher observation benefits for priority observation when resources are insufficient. At the same time, it can perform synthetic observation on tasks with time window intersections to reduce task conflicts, improve observation benefits, and maximize the utilization of satellite resources. The traditional satellite task planning model cannot solve the satellite task planning problem in the case of insufficient satellite resources.

[0289] It should be noted that through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments. In this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of another identical element in the process, method, article or device including the said element.

[0290] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-satellite earth observation mission planning method, characterized in that The method includes: Constructing a set of tasks to be planned corresponding to each satellite orbit based on an objective function; Performing task planning for the set of tasks to be planned for each satellite orbit to obtain a corresponding task planning sequence; Re-planning the remaining tasks to be planned in each set of tasks to be planned; The constructing a set of tasks to be planned corresponding to each satellite orbit based on an objective function includes: S101. Calculate the set of observation time windows for each to-be-planned task \(T\) in the task set \(T=\{T_1, T_2, \ldots, T\) m , \ldots, T\) M}\), and for each observation time window m in the set of observation time windows , calculate the corresponding objective function value . ​​​​ where 1≤i≤I, 1≤j≤J, 1≤m≤M, 1≤n≤N; S102. Add the to-be-planned task T corresponding to the optimal objective function value to the satellite cycle SC m where this time window is located, and add it to the to-be-planned task set SCA ij of SC. Meanwhile, delete T ij from T, and set the remaining time windows outside the optimal observation time window of T m to be invalid; m ​ S103. Repeat the execution of S101 to S102 until all the tasks to be planned are assigned, and obtain the set SCA of tasks to be planned corresponding to each satellite orbit ij ; where S i represents the i-th satellite, and there are a total of I satellites; SC = {SC1, SC2,..., SC i ,..., SC I} represents the set of orbits of the satellite; SC i = {SC i1 , SC i2 ,..., SC ij ,..., SC iJ} represents the set of orbital passes of satellite S i ; SC ij represents the j-th revolution of satellite S i ; T = {T1, T2,..., T m ,..., T M} represents that there are a total of M tasks to be planned; T m represents the m-th task to be scheduled; Denote the set of observation time windows for the task T to be planned, with a total of N; m ​ Denote T m The nth observation time window of i in the jth revolution of satellite S; SCA ij Denote the set of tasks to be scheduled for the j-th orbit of satellite S i ; The performing task planning for the set of tasks to be planned for each satellite orbit to obtain a corresponding task planning sequence includes: Performing task planning for the set of tasks to be planned for each satellite orbit in sequence, and the steps of performing task planning for the set of tasks to be planned for any satellite orbit are as follows: S201. Determine S from the set of tasks to be planned SCA ij among them i the first task T executed in the j-th round start , and remove the first task T start from the set of tasks to be planned SCA ij therefrom; S202. Traverse SCA ij for each to-be-planned task T m in it; if it satisfies TS m -TE l ≥0, then add T m to the to-be-scheduled task set SCW ij ; where T l Indicates the last task in the mission planning sequence SCG ij ; TE l Indicates the end observation time of T l and when the best observation window of T l is its nth observation time window, S203. Determine whether the SCW ij is an empty set; If SCW ij is an empty set, mission planning is performed for the next satellite cycle; If SCW ij is not an empty set, then select the task T ij to be scheduled from SCW a ; S204. If the task T to be scheduled a simultaneously satisfies the angle adjustment constraint and the longest startup time constraint, then add T a to the task planning sequence SCG ij delete T a from SCA ij empty SCW ij and then return to S202; Otherwise, delete T a from SCW ij , and then return to S203.

2. The multi-satellite earth observation mission planning method according to claim 1, characterized in that, The objective function value is calculated as follows: where represents the sum of the observation task benefits of the remaining valid observation time windows that have time overlap with the observation time window; TP m Represents the observed benefit of T m ; Denote T m the observation duration in the nth observation time window respectively represent T m at the end and start of the nth observation time window SCT ij Indicates the satellite S i The longest power-on time in the j-th cycle; Indicating at the impact degree of the observation of T m on other observation tasks, the smaller the better; Indicates that the satellite has completed T in this orbit m The proportion of resources consumed by the satellite in this orbit, the smaller the better.

3. The multi-star earth observation mission planning method according to claim 1, characterized in that The determination of S from the set of tasks to be planned SCA ij in S i The first task T executed in the j-th loop start , including: Select the first task T using the following formula start :[[]]END]] And select the task T to be arranged from the SCW ij including: a select the task T to be arranged from the SCW Select the task T to be scheduled from the SCW using the following formula ij a :​ where TS m Indicates the start observation time of T m and when the best observation window of T m is its nth observation time window when TP m Represents the task T to be planned m The income obtained upon completion; λ is a balance coefficient used to balance dimensions.

4. The multi-satellite earth observation mission planning method according to claim 1, characterized in that The angle adjustment constraint is: TE l +t la ≤TS a where Denote S i From the execution of T l To the execution of T a The posture adjustment time in between; TA l and TA a respectively represent the best observation angles of T l and T a ; Represents S i Angular deflection rate per unit time; TS a Indicates the start observation time at T a ; and the longest power-on time constraint is: TE a ≤TS start +SCT ij where TE a Indicates the end observation time of T a ; TS start Indicates the start observation time of the first task T start ; SCT ij Indicates S i The longest power-on time in the j-th cycle.

5. A multi-star earth observation mission planning method according to claim 1, characterized in that, The re-planning the remaining tasks to be planned in each set of tasks to be planned includes: If the set of tasks to be planned SCA ij is an empty set, no replanning is performed; Otherwise, select the task to be planned T from SCA in the order of the start observation time of the tasks ij in sequence; m' Perform replanning; and the re-planning includes: an insertion strategy, a synthesis strategy, a replacement strategy, and a recombination strategy, and the priority is insertion strategy > synthesis strategy > replacement strategy > recombination strategy.

6. The multi-satellite earth observation mission planning method according to claim 5, characterized in that The insertion strategy includes: Select two adjacent planned tasks T ij and T f in the mission planning sequence SCG in order of the start observation time of the task b ; If both TE f +t fm' ≤TS m' and TE m' +t m'b ≤TS b are satisfied, then task T m' is inserted into SCG ij between T f and T b , indicating that S i executes T f , T m' and T b in sequence in the j-th cycle, and T m' is deleted from SCA ij ; where TE f Indicates the end observation time of T f ; Denote S i From the execution of T f To the execution of T m' The posture adjustment time in between; TA f and TA m' respectively represent the optimal observation angles of T f and T m' ; Indicates S i Angular deflection rate per unit time; TS m' and TE m' respectively represent the start observation time and the end observation time of T m' ; Denote S i From the execution of T m' To the execution of T b The posture adjustment time therebetween; TA m' and TA b respectively represent the best observation angles of T m' and T b ; TS b Indicates the start observation time of T b ; The synthesis strategy includes: Select the planned task T in the task planning sequence SCG in order of the start observation time of the task ij in sequence, and c do not select the first task and the last task in SCG ij ; If T c and T m' observation time windows have an intersection and simultaneously satisfy TP d ≥TP c 、TE f +t fd ≤TS d 、TE d +t db ≤TS b , then replace T d in SCG ij with T c , and delete T m' from SCA ij ; Among them, T d represents the new task synthesized by T c and T m' ; T f and T b represent the previous task and the next task adjacent to T ij in SCG c respectively; TP d = TP m' (1 - θ|TA m' - TA d |) + TP c (1 - θ|TA c - TA d (1 - θ|TA d - TA|) represents the return of T TP c ,TP m' represents T c and T m' 's observed benefits; TA d =(TA m' +TA c ) / 2, TA m' , TA c respectively represent the optimal observation angles of T d , T m' and T c ; θ represents the loss rate of observation benefit per unit angle deviation between the actual observation angle and the optimal observation angle of the task when performing the observation task; TE f Indicates the end observation time of T f ; Denote S i From the execution of T f To T d The posture adjustment time in between; TA f represents the best viewing angle of T f ; Denote S i Angular deflection rate per unit time; TS d = min(TS c , TS m' ) and TE d = max(TE c , TE m' ) respectively represent the start observation time and the end observation time of T d ; min(TS c ,TS m' ) represents taking the smaller value of the start observation times of T c and T m' ; max(TE c ,TE m' ) represents taking the larger value of the end observation times of T c and T m' ; Denote S i From the execution of T d To T b The posture adjustment time in between; TA b represents the optimal viewing angle of T b ; TS b Indicates the start observation time of T b ; The replacement strategy includes: Select the planned task T in the task planning sequence SCG in the order of the start observation time of the task ij in sequence, and c do not select the first task and the last task in SCG ij ; If both TP c ≤TP m' , TE f +t fm' ≤TS m' , TE m' +t m'b ≤TS b , then replace T m' in SCG ij with T c , delete T m' from SCA ij , and add T c to SCA ij ; Among them, T f , T b are respectively the previous task and the next task of T m in SCG ij ; TP c ,TP m' respectively represent the returns of T c and T m' ; TE f Indicates the end observation time of T f ; Denote S i From the execution of T f To T m' The posture adjustment time between TA f and TA m' respectively represent the best observation angles of T f and T m' . Indicating S i Angular deflection rate per unit time; TS m' and TE m' respectively represent the start observation time and the end observation time of T m' ; Denote S i From the execution of T m' To T b The posture adjustment time in between, TA m' and TA b respectively represent the best observation angles of T m' and T b ; TS b Indicates the start observation time of T b ; The recombination strategy includes: If the task T to be planned m' does not meet the insertion, synthesis, and replacement strategies, then give up using the observation time window in the current satellite cycle for T m' for observation, delete T m' from SCA ij , select the observation time window of T m' on other satellite cycles that have not performed replanning, and then use the three strategies of insertion, synthesis, and replacement for planning.

7. A multi-satellite earth observation mission planning system, the system comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-6 above.

Images