CN112149911A

CN112149911A - Hypersensitive short satellite same-orbit multipoint target in-motion imaging task planning method

Info

Publication number: CN112149911A
Application number: CN202011055017.9A
Authority: CN
Inventors: 沈欣; 路泽忠; 陈亚欣
Original assignee: Wuhan University WHU
Current assignee: Wuhan University WHU
Priority date: 2020-09-30
Filing date: 2020-09-30
Publication date: 2020-12-29
Anticipated expiration: 2040-09-30
Also published as: CN112149911B

Abstract

The invention provides a method for planning imaging tasks in the same-orbit multipoint targets of a hypersensitive agile satellite. The method carries out preprocessing of multi-point target clustering and non-tracing band decomposition, and solves the imaging time windows of all band endpoints; constructing a strip imaging serial number sequence and a strip imaging direction sequence to obtain a strip endpoint imaging sequence, and calculating a strip endpoint imaging time sequence by using an imaging time normalization coefficient; constructing a decision variable, an imaging coverage benefit maximization and a task completion time minimization to construct an objective function, and constructing constraint conditions of an imaging time window and a posture conversion time according to a strip imaging serial number sequence, a strip imaging direction sequence and a strip endpoint imaging time sequence, and further constructing an imaging task planning mathematical model in the same-orbit multipoint target; and optimizing by improving the particle swarm optimization to obtain an optimal imaging task planning scheme. The method can fully utilize the attitude maneuvering capability of the agile satellite to realize the optimization of the imaging task scheme in the same-orbit multipoint target motion.

Description

Hypersensitive short satellite same-orbit multipoint target in-motion imaging task planning method

Technical Field

The invention relates to the technical field of satellite remote sensing, in particular to a method for planning imaging tasks in the same-orbit multipoint targets of a hypersensitive agile satellite.

Background

An Agile satellite (Agile EOS) refers to a satellite with a payload fixed on the satellite and 3 axial rotations of pitching, rolling and yawing of the whole satellite controlled by an attitude control system. The agile satellite can perform attitude maneuver in the interval of executing a plurality of imaging tasks, and performs push-broom imaging after the attitude is stable, so that imaging modes such as side-sway imaging, pitching imaging, single-track stereo, single-track strip splicing, same-track multi-point targets and the like can be realized. In recent years, with the development of satellite attitude control technology and sensor imaging technology, agile satellites have been provided with "ultra" agile in-motion imaging capability, that is, imaging of ground targets is completed simultaneously in the attitude maneuver process, such as french pleiades satellites and high-resolution multimode comprehensive imaging satellites. Besides the imaging mode of the agile satellite, the imaging agile satellite can realize the push-broom imaging capability of any track in motion, so that the observation of more targets is realized under the condition of single transit, and the working efficiency of the earth observation satellite is further improved.

The same-orbit multi-point target imaging is a typical working mode of an agile satellite, namely, imaging of a plurality of point targets which are distributed adjacently on a track of points under the satellite is completed through one-time transit. In the process of imaging multiple point targets, the traditional agile satellite utilizes the attitude mobility to realize attitude switching, and realizes push-broom imaging of multiple point targets in a push-broom mode parallel to the track of points under the satellite. However, because the imaging process of each target by the satellite is 'steady-state' push-broom, the imaging of the ground cannot be simultaneously finished in the attitude maneuver process, the speed of the imaging of the push-broom of the ground is determined by the orbital motion speed of the satellite, the maximum push-broom range of the push-broom is limited by 'hard' of the motion speed of points under the satellite, and when the number of the point targets is large, the imaging of all the point targets cannot be guaranteed; due to the adoption of the imaging agile satellite, the speed of ground push-broom imaging can be greatly improved through attitude maneuver due to the capabilities of side maneuver and side imaging, the maximum push-broom range can also break through the limitation of the speed of motion of points under the satellite, a single transit can be realized through imaging in the process of moving to image more point targets, and the working efficiency of the satellite is greatly improved.

The emergence of imaging modes in hypersensitive agility presents a new challenge to the planning of the same-track multipoint target imaging task. Firstly, the limitation of stable posture in the imaging process is broken through, so that the initial time of the imaging time of a ground target with fixed length in single push-broom is not fixed, and the posture constraint is more complex than that of the traditional agile satellite; secondly, due to the non-tracking characteristic of imaging in the process, the direction of the push-broom strip can not be parallel to the track of the subsatellite point, and the task decomposition mode is richer than that of the traditional agile satellite. Therefore, a need exists for a novel hypersensitive satellite-based imaging mission plan that is tailored to the user's observation needs.

Aiming at the imaging task requirement of the same-orbit multipoint target of the ultra-agile satellite and the defects of the existing method, a new technical scheme is urgently needed to be proposed in the field.

Disclosure of Invention

The invention provides a method for planning imaging tasks in the same-orbit multipoint targets of a hypersensitive short satellite aiming at the problems in the prior art.

The invention provides a hypersensitive agile satellite same-orbit multipoint target in-motion imaging task planning method, which comprises the following steps of:

step 1, inputting longitude and latitude coordinates of a plurality of point targets, converting the longitude and latitude coordinates of the point targets into plane coordinates of the point targets, dividing the plane coordinates of the point targets into a plurality of point target clusters through a density-based clustering method, constructing a minimum external rectangular coordinate point set of the clusters through a rotating and clamping method, carrying out strip decomposition on the minimum external rectangular coordinate point set of the clusters to obtain a strip number set, a strip endpoint number set and a strip endpoint longitude and latitude coordinate set, and solving imaging time windows of all strip endpoints;

step 2, constructing a strip imaging number sequence and a strip imaging direction sequence, calculating a strip endpoint imaging sequence according to the strip imaging number sequence and the strip imaging direction sequence, and calculating a strip endpoint imaging time sequence by using an imaging time normalization coefficient in combination with the strip endpoint imaging sequence;

step 3, constructing a decision variable according to a strip imaging number sequence, a strip imaging direction sequence and a strip endpoint imaging time sequence, constructing a target function through imaging coverage profit maximization and task completion time minimization, constructing a constraint condition through an imaging time window and posture conversion time, and further constructing an imaging task planning mathematical model in the same-orbit multipoint target;

step 4, combining a mathematical model of imaging task planning in the simultaneous multi-point target, and optimizing by improving a particle swarm optimization to obtain an imaging task planning scheme with the maximum observation yield and the shortest task completion time;

preferably, the step 1 of inputting the longitude and latitude coordinates of the plurality of point targets, and converting the longitude and latitude coordinates of the plurality of point targets into the plane coordinates of the plurality of point targets is as follows:

assume that the plurality of point targets are set as P ═ P₁,P₂,…,P_i,…,P_npThe longitude and latitude coordinates corresponding to each point target are { (B)₁,L₁),(B₂,L₂),...,(B_i,L_i),...,(B_np,L_np)}

Where np is the number of point targets, P_iIs the ith point target, (B)_i,L_i) As latitude and longitude coordinates of the ith point target, B_iIs the latitude, L, of the ith point target_iFor the longitude of the ith point target, i ∈ [1, np [ ]]；

Transforming the longitude and latitude coordinates of the point targets into plane coordinates of the point targets by using a Gaussian projection forward calculation formula: { (x)₁,y₁),(x₂,y₂),...,(x_i,y_i),...,(x_np,y_np)}

Where np is the number of point targets, (x)_i,y_i) Is the plane coordinate of the ith point target, x_iIs the X-axis coordinate, y, of the ith point target_iIs the Y-axis coordinate of the ith point target, i ∈ [1, np ∈ [ ]]；

The step 1 of dividing the plane coordinates of the point targets into a plurality of point target clusters by a density-based clustering method is as follows:

after converting longitude and latitude coordinates of the multi-point target into plane coordinates, clustering the point target by adopting a density-based clustering method, namely a DBSCAN method;

calculating the clustering radius with the satellite orbit height of h, the earth radius of R and the half field angle of V as follows:

wherein eps is a clustering radius, h is a satellite orbit height, R is an earth radius, V is a half field angle, and a density threshold is Minpts 2;

the plane distance between two adjacent point targets in the plane coordinates of the point targets is as follows:

i∈[1,np]

where np is the number of point targets, dis_i,i+1The plane distance between the ith point target and the (i +1) th point target, namely two adjacent point targets in the plane coordinate is calculated;

taking any point target in the P as the center of a circle and eps as the radius, and dividing dis in the plane coordinates of a plurality of point targets in the circle_i,i+1Clustering all point targets with the number of point targets more than or equal to 2 in a circle which is less than or equal to eps and less than or equal to Minpts, and gradually expanding the cluster to obtain a final point target cluster;

the point target cluster is:

wherein the content of the first and second substances,

indicates the inclusion of c in the ith cluster_lPoint targets, e represents the number of clusters of point targets, l ∈ [1, e ∈]， c_lRepresents the number of point targets in the ith cluster, and c₁+c₂+...+c_l+...+c_eEnsuring that e clustering clusters formed by clustering the point targets contain all the point targets;

step 1, constructing a minimum external rectangular coordinate set of the point target cluster by a rotating and shell-clamping method, wherein the minimum external rectangular coordinate set is as follows:

constructing a minimum external rectangle of the cluster based on the rotating and shell clamping idea;

constructed by using the idea of rotating the clamping shell

The minimum circumscribed rectangle of the area obtained by the method has the following vertex coordinate set: { Point_l ^a,Point_l ^b,Point_l ^c,Point_l ^d}；

Wherein Point_l ^aPoint, which represents the 1 st vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^a＝(x_l ^a,y_l ^a)， Point_l ^bPoint, the 2 nd vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^b＝(x_l ^b,y_l ^b)，Point_l ^cPoint representing the 3 rd vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^c＝(x_l ^c,y_l ^c)，Point_l ^dPoint representing the 4 th vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^d＝(x_l ^d,y_l ^d) The obtained minimum external rectangle vertex coordinate set of the cluster comprises the cluster;

step 1, obtaining a strip set, a strip endpoint set and a strip endpoint longitude and latitude coordinate set by using the minimum external rectangle coordinate point set of the cluster through a strip decomposition method, wherein the strip set, the strip endpoint set and the strip endpoint longitude and latitude coordinate set are as follows:

four vertex coordinates in combination with a minimum bounding rectangle Point_l ^a,Point_l ^b,Point_l ^c,Point_l ^dAnd then dividing the strips of the circumscribed rectangle along the long edge or the short edge according to the fixed imaging breadth, namely eps, so as to obtain four vertex coordinate sets of m strips in the cluster l as follows:

{Point_l,1 ^a,Point_l,1 ^b,Point_l,1 ^c,Point_l,1 ^d；Point_l,2 ^a,Point_l,2 ^b,Point_l,2 ^c,Point_l,2 ^d；...,；Point_l,j ^a,Point_l,j ^b,Poi nt_l,j ^c,Point_l,j ^d；...,；Point_l,m ^a,Point_l,m ^b,Point_l,m ^c,Point_l,m ^d}；

four vertex coordinate sets of the m strips obtained by dividing the strips comprise a vertex coordinate set of a minimum circumscribed rectangle of the clustering cluster;

wherein m represents the number of strips, j represents the jth strip, j is more than or equal to 1 and less than or equal to m, and Point_l,j ^aThe 1 st vertex coordinate, Point, of the jth strip in the ith cluster_l,j ^a＝(x_l,j ^a,y_l,j ^a)，Point _l,j ^b2 nd vertex coordinate, Point, representing the jth stripe in the ith cluster_l,j ^b＝(x_l,j ^b,y_l,j ^b)，Point_l,j ^cDenotes the 3 rd vertex coordinate, Point, of the jth strip in the ith cluster_l,j ^c＝(x_l,j ^c,y_l,j ^c)，Point_l,j ^dThe 4 th vertex coordinate, Point, representing the jth strip in the ith cluster_l,j ^d＝(x_l,j ^d,y_l,j ^d)；

The m strips in the cluster l are respectively numbered as different continuous positive integers of [1, m ], the strip number set is {1,2, …, j, …, m }, j represents the strip number, and j is more than or equal to 1 and less than or equal to m;

taking the middle points of the 1 st vertex coordinate and the 2 nd vertex coordinate of the strip and the middle points of the 3 rd vertex coordinate and the 4 th vertex coordinate of the strip as strip end points, respectively numbering the 2 m strip end points in the clustering cluster l as different continuous positive integers of [1,2 m ], wherein the obtained strip end point number set is {1,2, …, k, …,2 m }, k represents the strip end point number, and k is more than or equal to 1 and less than or equal to 2 m; the strip endpoint coordinate set corresponding to the clustering cluster l, namely the midpoint of the 1 st and 2 nd vertex coordinates and the midpoint of the 3 rd and 4 th vertex coordinates of the strip are as follows:

wherein m represents the number of strips, j is more than or equal to 1 and less than or equal to m,

the two end point plane coordinates of the jth strip in the cluster l are respectively

Wherein m represents the number of strips, j represents the number of the strips, and j is more than or equal to 1 and less than or equal to m; the obtained 2 x m strip endpoint longitude and latitude coordinate sets comprise four vertex coordinate sets of m strips;

after the strip endpoint coordinate set corresponding to the clustering cluster l is obtained, the strip endpoint coordinate set is converted into a strip endpoint longitude and latitude coordinate set SendPoint by utilizing a Gaussian projection back calculation formula_lExpressed as:

SendPoint_l＝{endPoint_l,1,endPoint_l,2,...,endPoint_l,M,...,endPoint_l,2*m}

wherein m represents the number of bands, 2 m represents the number of band endsMesh, M represents the end point number of the strip, M is more than or equal to 1 and less than or equal to 2M, endPoint_l,MThe longitude and latitude coordinates of the M-th strip endpoint in the clustering cluster l are obtained;

after dividing each cluster into strips, numbering all the strips, wherein a strip number set S is {1,2, …, k, …, ns };

wherein k represents the number of the strips, k is more than or equal to 1 and less than or equal to ns, and ns represents the final number of the strips;

SP represents the final set of stripe end point numbers, SP ═ 1, 2., n., 2 × ns }, where n represents the stripe end point number, 1 ≦ n ≦ 2 × ns, and 2 × ns represents the final number of stripe end points; set of longitude and latitude coordinates of all strip endpoints, SendPoint ═ endPoint₁,endPoint₂,...,endPoint_n,...,endPoint_2*nsIn which endPoint_nAnd representing longitude and latitude coordinates corresponding to the strip end point with the strip end point number n, wherein n is more than or equal to 1 and less than or equal to 2 x ns.

Step 1, the imaging time window for solving the end points of all the strips is:

according to the strip endpoint number set SP and the corresponding strip endpoint longitude and latitude coordinate set SendPoint, a characteristic cone method (Shenxin, Lide Kernel, YaoShuang) is adopted, and an optical remote sensing satellite imaging window rapid prediction method facing imaging task planning [ J]Wuhan university newspaper (information science edition), 2012,37(12):1468-_n,s,TW_n,e]；

Wherein, [ TW_n,s,TW_n,e]Representing the original imaging time window, TW, corresponding to the strip end point with the strip end point number n_n,sFor the start time of the time window, TW_n,eIs the time of the end of the time window;

preferably, the imaging number sequence of the building strips in the step 2 is as follows:

S_ind_r＝{s_ind_r,1,s_ind_r,2,...,s_ind_r,k,...,s_ind_r,ns}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein, S _ ind_rRepresenting a sequence of slice imaging numbers, r representing an r-th group permutation combination of the slice imaging numbers, ns representing the number of slices, ns! Indicating the number of all permutation combinations of the strip imaging numbers

k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rThe k-th slice, s _ ind_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rNumber of the k-th slice in, s _ ind_r,kE is S, and S represents the strip number set in the step 1;

wherein, S _ ind_rDetermining by a recursive calling method according to the r-th group permutation and combination r of the strip imaging numbers and the strip number set S obtained in the step 1;

step 2, constructing a strip imaging direction sequence as follows:

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicating the number of all permutation combinations of the strip imaging numbers

Determining the imaging direction of the strip as a binary number;

k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rMiddle k-th strip, drt_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rImaging direction corresponding to the number of the kth strip, when drt_r,kWhen 0, the slice imaging number sequence S _ ind representing the r-th group arrangement combination_rIn the middle, the imaging direction of the k-th strip is forward push-scan when drt_r,kWhen 1, the slice imaging number sequence S _ ind indicating the r-th group arrangement combination_rIn the step (2), the imaging direction of the kth strip is reverse push-broom;

step 2, the imaging sequence of calculating the strip end point is as follows:

firstly, determining a strip imaging number sequence S _ ind according to the r-th group permutation and combination r of the strip imaging numbers_rThen, S _ ind_rEach stripe corresponds to two stripe end points, and it cannot be determined that the stripe end point is imaged preferentially, so that the stripe imaging number sequence S _ ind needs to be arranged and combined according to the r-th group_rIn the direction of imaging d of each strip_r,kDetermining S _ ind_rThe imaging direction of each strip, thereby finally determining the strip end point imaging sequence SP _ srt_r；

SP_srt_r＝{sp_srt_r,1,ST,sp_srt_r,1,ET,sp_srt_r,2,ST,sp_srt_r,1,ET,...,sp_srt_r,k,ST,sp_srt_r,k,ET ...,sp_srt_r,ns,ST,sp_srt_r,ns,ET}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Wherein, SP _ srt_rThe r-th group permutation and combination r of the strip imaging numbers represents the strip endpoint imaging sequence corresponding to the strip imaging number, r represents the r-th group permutation and combination of the strip imaging numbers, ns represents the strip number, ns! Representing the number of all permutation and combination of the strip imaging numbers; k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rThe kth band; ST denotes the starting imaged slice end point of the slice, ET denotes the ending imaged slice end point of the slice;

wherein, s _ srt_r,k,STAnd s _ srt_r,k,ETThe r-th group arrangement combination r representing the strip imaging number corresponds to the strip number imaging sequence S _ ind_rThe pair of end points of the k-th stripe, s _ srt_r,k,STThe r-th group arrangement combination r representing the strip imaging number corresponds to the strip imaging number sequence S _ ind_rNumber of start imaging strip end point of kth strip, s _ srt_r,k,ETThe r-th group arrangement combination r representing the strip imaging number corresponds to the strip imaging sequence S _ ind_rThe number of end imaging strip end points of the kth strip;

if there are ns strips, the number of strip ends is 2 × ns, so the strip end imaging sequence can also be expressed as:

SP_srt_r＝{sp_srt_r,1,sp_srt_r,2,sp_srt_r,3,sp_srt_r,4,...,sp_srt_r,n,sp_srt_r,n+1...,sp_srt_r,2*ns-1,sp_srt_r,2*ns}

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein, SP _ srt_rThe r-th group permutation and combination r of the strip imaging numbers represents the strip end point imaging sequence corresponding to the strip end point imaging sequence, r represents the r-th group permutation and combination of the strip imaging numbers, ns represents the strip number, 2 x ns represents the strip end point number, ns! Representing the number of all permutation and combination of the strip imaging numbers; k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rThe kth band; n represents SP _ srt_rThe nth stripe endpoint, sp _ srr_r,nRepresentation SP _ srt_rNumber of nth stripe endpoint, sp _ srt_r,nE, SP represents the strip endpoint number set in the step 1;

step 2, calculating the imaging time sequence of the strip end points as follows:

the imaging time normalization coefficient is as follows:

t_nrm_r,n∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein, t _ nrm_r,nStrip end point imaging sequence SP _ srt corresponding to r-th group permutation and combination r representing strip imaging number_rThe imaging time normalization coefficient of the nth strip end point is shown, and n represents the observation sequence SP _ srt of the strip end point_rThe nth stripe endpoint;

wherein, t_r,nNormalizing the coefficient t _ nrm according to the imaging time_r,nCalculating to obtain;

imaging sequence SP _ srt in calculating strip end point_rImaging time t corresponding to each strip end point_r,nAnd then, includes: cutting an imaging time window and normalizing imaging time;

and (3) cutting the imaging time window:

the requirements for clipping the imaging time windows of two adjacent strip end points are as follows:

the starting time of the imaging time window of the endpoint to be observed later is not earlier than the starting time of the imaging time window of the endpoint to be observed first, and the ending time of the imaging time window of the endpoint to be observed first is not later than the ending time of the imaging time window of the endpoint to be observed later, specifically:

if T_r,n,s＞T_r,n+1,sThen ITW_r,n+1,s＝T_r,n,s,ITW_r,n,s＝T_r,n,s

If T_r,n,e＞T_r,n+1,eThen ITW_r,n+1,e＝T_r,n+1,e,ITW_r,n+1,e＝T_r,n+1,e

In the above formula, T_r,n,sImaging sequence representing end points of a strip SP _ srt_rThe start time of the time window corresponding to the nth observed strip endpoint; t is_r,n,eImaging sequence representing end points of a strip SP _ srt_rThe end time of the time window corresponding to the nth observed strip endpoint; ITW_r,n,sImaging sequence representing end points of a strip SP _ srt_rThe starting time of the cut time window corresponding to the nth observed strip end point; ITW_r,n,eImaging sequence representing end points of a strip SP _ srt_rThe end time of the clipped time window corresponding to the nth observed strip end point;

the imaging time is normalized:

imaging sequence at the end of the strip, SP _ srt_rBetween any two adjacent end points of the strip, the imaging time of the point observed first can restrict the feasible range of the point observed later, and a normalization coefficient is introduced to normalize the imaging time, specifically:

if n is 1, then t_r,1＝ITW_r,1,s+t_nrm_r,n*(ITW_r,1,e-ITW_r,1,s)

If n ≠ 1, then

If t_r,n＞ITW_r,n+1,sThen t is_r,n+1＝t_r,n+t_nrm_r,n*(ITW_r,n+1,e-t_r,n)

If t_r,n≤ITW_r,n+1,sThen t is_r,n+1＝ITW_r,n+1,s+t_nrm_r,n*(ITW_r,n+1,e-ITW_r,n+1,s)

Wherein, t_r,1Imaging sequence SP _ srt for strip endpoint_rThe imaging time, t, of the trimmed time window corresponding to the 1 st observed strip end point_r,nImaging sequence SP _ srt for strip endpoint_rAt the imaging time of the nth observed strip end point, t_r,n+1Imaging sequence SP _ srt for strip endpoint_rThe imaging time of the (n +1) th observed band end point, ITW_r,n,sImaging sequence representing end points of a strip SP _ srt_rThe starting time of the cut time window corresponding to the nth observed strip end point; ITW_r,n,eImaging sequence representing end points of a strip SP _ srt_rThe end time of the clipped time window corresponding to the nth observed strip end point;

preferably, the step 3 of constructing decision variables according to the strip imaging number sequence, the strip imaging direction sequence, and the strip endpoint imaging time sequence is:

the strip imaging number sequence is as follows:

S_ind_r＝{s_ind_r,1,s_ind_r,2,...,s_ind_r,k,...,s_ind_r,ns}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein, S _ ind_rRepresenting a sequence of slice imaging numbers, r representing an r-th group permutation combination of the slice imaging numbers, ns representing the number of slices, ns! Representing the number of all permutation combinations of the band imaging numbers, and k representing the band imaging number sequence S _ ind of the r-th permutation combination_rThe k-th slice, s _ ind_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rNumber of k-th stripe in, s _ ind_r,kE is S, and S represents the strip number set in the step 1;

the swath imaging direction sequence is:

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Representing the number of all permutation combinations of the band imaging numbers, and k representing the band imaging number sequence S _ ind of the r-th permutation combination_rMiddle k-th strip, drt_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rImaging direction corresponding to the number of the kth strip, when drt_r,kWhen 0, the slice imaging number sequence S _ ind representing the r-th group arrangement combination_rIn the middle, the imaging direction of the k-th strip is forward push-scan when drt_r,kWhen 1, the slice imaging number sequence S _ ind indicating the r-th group arrangement combination_rIn the step (2), the imaging direction of the kth strip is reverse push-broom;

the sequence of imaging time instants of the strip end points is as follows:

t_r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rMiddle nth stripe end point, t_r,nStrip end point imaging sequence SP _ srt corresponding to r-th group permutation and combination r representing strip imaging number_rThe imaging time of the nth strip end point;

the constructed decision variables are:

r∈{1,2,...,ns！}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Representing the number of all permutation combinations of the slice imaging numbers, the slice imaging number sequence S _ ind is determined by step 2 according to the r-th group permutation combination r of the slice imaging numbers_r；

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Determining the imaging direction of the strip as a binary number;

k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rMiddle k-th strip, drt_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rImaging direction corresponding to the number of the k-th strip

t_nrm_r,n∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein, SP _ srt_rThe strip end point imaging sequence corresponding to the r-th group permutation and combination r representing the strip imaging number, SP _ srt_rThe sequence of slice imaging numbers S _ ind can be determined according to the r-th group permutation and combination r of the slice imaging numbers_rAnd a sequence of strip imaging directions drt_r,kDetermined by step 2;

step 3, maximizing the imaging coverage yield:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

representing the number of the maximized imaging coverage gains, namely observation point targets;

wherein r isR-th group arrangement combination representing slice imaging number, ns representing the number of slices, ns! Representing the number of all permutation combinations of the band imaging numbers, and k representing the band imaging number sequence S _ ind of the r-th permutation combination_rThe k-th strip of (C)_r,kRepresents S _ ind_rWhether the k-th band can be imaged, c_r,kE.g. {0,1}, if c_r,kIf 1, then S _ ind_rThe imaging can be finished in the kth strip, otherwise, the imaging cannot be realized; g_r,kRepresents S _ ind_rThe number of points covered by the kth strip;

c_r,kthe judgment basis comprises two aspects:

the time from the end point of the k-1 strip to the start point of the k strip meets the posture conversion time constraint;

the time from the starting point to the end point of the k strips meets the posture conversion time constraint;

namely:

if (t)_r,2k+1-t_r,2k≥t_tr_r,2k+1)∩(t_r,2k-t_r,2k-1≥t_tr_r,2k) Then c is_r,k＝1

Otherwise c_r,k＝0

Step 3, the minimized task completion time is an objective function:

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

indicating a minimized imaging task completion time;

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rMiddle nth stripe end, Δ t_r,nImaging sequence representing end points of a strip SP _ srt_rLast observation time interval, Δ t, from the n-1 th strip end point to the n-th strip end point_r,n＝t_r,n-t_r,n-1，t_r,nImaging sequence SP _ srt for strip endpoint_rImaging time of the upper nth strip end point;

wherein, c _ tr_r,nCharacterization strip endpoint imaging sequence SP _ srt_rWhether the above from the n-1 th to nth stripe endpoints satisfy the pose transition time constraint, i.e.:

if Δ t_r,n≥t_tr_r,nThen c _ tr_r,n＝1

Otherwise c _ tr_r,n＝0

Step 3, the imaging time window constraint is as follows:

t_r,n∈[ITW_r,n-s,ITW_r,n-e]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rMiddle nth stripe end point, t_r,n∈[ITW_r,n-s,ITW_r,n-e]Imaging sequence representing end points of a strip SP _ srt_rThe imaging instant of each strip end point must satisfy the cropped imaging time window constraint,

and 3, constraining the posture conversion time as follows:

Δt_r,n≥t_tr_r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rMiddle nth stripe end, Δ t_r,n≥t_tr_r,nImaging sequence representing end points of a strip SP _ srt_rThe imaging time interval between two adjacent strip end points needs to meet the posture conversion time constraint between the two adjacent end points;

and calculating the posture conversion time between the two adjacent end points by adopting a continuous posture adjusting method:

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rThe nth stripe end point, t _ tr_r,nImaging sequence representing end points of a strip SP _ srt_rThe fastest gesture transition time from the nth-1 to nth stripe end point,

for the attitude momentum along the roll axis between the n-1 th and nth strip end points, Δ ω_n,n-1Attitude momentum along pitch axis, Δ κ, from nth-1 to nth strip end_n,n-1Respectively the attitude momentum along the yaw axis from the n-1 th strip end point to the n-th strip end point,

the maneuvering speed of the rolling axis of the satellite,

The maneuvering speed of the satellite pitching axis,

Respectively the three-axis maneuvering speed of the satellite;

and (3) finishing the construction of the same-orbit multipoint target in-motion imaging task planning mathematical model in the step (3) through decision variables, an objective function and constraint conditions.

Preferably, the imaging task planning scheme with the maximum observation yield and the shortest task completion time obtained by improving the particle swarm optimization in the step 4 is as follows:

step 4.1, randomly generating an initial population:

generating initial populations with the quantity Z according to a predefined population scale Z, wherein each particle represents an imaging task scheme, and the maximum iteration number is ITER;

wherein the position of the particle represents the decision variable r constructed as described in step 3_iter,u、drt_r,k,iter,u、t_nrm_r,n,iter,uThe initial value of each particle is a random value in the value range of each particle, and the initial speed of each particle is set to be zero;

wherein, ITER represents a second ITER iteration process of the improved particle swarm optimization algorithm, ITER belongs to {1, 2.,. ITER,. and ITER }, u represents a u-th particle in the second ITER iteration process, and u belongs to {1, 2.,. u.,. and Z }; r is_iter,uThe r group of arrangement combinations of the strip imaging numbers corresponding to the u particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,uThe strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_rImaging direction corresponding to the number of the kth strip; t _ nrm_r,n,iter,uRepresenting the strip endpoint imaging sequence SP _ srt corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uNormalizing the coefficient of the imaging time of the nth strip endpoint;

wherein, SP _ srt_r,iter,uRepresenting a strip endpoint imaging sequence corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm;

step 4.2, calculating the observation income and the task completion time of each particle in the current population, and obtaining the particles with the optimal particle fitness value in the current population, wherein the fitness value is formed by the observation income and the task completion time;

the calculation of the observation yield and the task completion time of each particle in the current population is as follows:

according to r_iter,u、drt_r,k,iter,u、t_nrm_r,n,iter,uAnd restoring a strip endpoint imaging sequence SP _ srt corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm by utilizing the step 2_r,iter,uAnd the imaging time t corresponding to the end point of each strip_r,n,iter,u(ii) a With the posture conversion time t _ tr between two adjacent strip end points_r,n,iter,uObtaining a strip endpoint observation sequence capable of completing push-broom according to the judgment basis, and obtaining observation income and task completion time;

wherein, ITER represents a second ITER iteration process of the improved particle swarm optimization algorithm, ITER belongs to {1, 2.,. ITER,. and ITER }, u represents a u-th particle in the second ITER iteration process, and u belongs to {1, 2.,. u.,. and Z }; r is_iter,uThe r group of arrangement combinations of the strip imaging numbers corresponding to the u particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,uThe strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uImaging direction, S _ ind, corresponding to the number of the kth stripe_r,iter,uA strip imaging number sequence of the r-th group arrangement combination of the strip imaging numbers corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm, t _ nrm_r,n,iter,uRepresenting the strip endpoint imaging sequence SP _ srt corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uNormalization coefficient of imaging time of the nth strip end point, SP _ srt_r,iter,uRepresenting the strip endpoint imaging sequence corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm, t_r,n,iter,uRepresenting the imaging time of the nth strip endpoint in the strip endpoint imaging sequence corresponding to the r-th group arrangement combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm;

the particles with the optimal particle fitness value (observation income and task completion time) in the current population are obtained as follows:

sorting the fitness value of each particle in the current population to obtain a particle best with the optimal fitness value in the particle swarm, namely comparing the observation gains of all the particles, taking the particle with the maximum observation gain as the optimal particle, and if the observation gains are the same, taking the particle with the minimum task completion time as the particle with the optimal fitness value of the current population;

wherein best represents the particles with the optimal fitness value in the particle group in the iter iteration process of the improved particle swarm optimization algorithm, Fit_iter,bestRepresenting the fitness value corresponding to the best particle best in the iter iteration process of the improved particle swarm optimization algorithm;

and 4.3, updating the speed and the position of the particles:

randomly selecting two particles of o and q, and comparing the observation gains and the task completion time of the two particles, wherein the particle with the largest observation gain in the two particles is a better particle, the particle with the smallest observation gain is a worse particle, and if the observation gains are the same, the particle with the smallest task completion time is a better particle;

if the preferred particle is o and the inferior particle is q, the preferred particle is o, i.e., P_winAnd the average position P of all particles_centerAs inferior particles P_loseThe current position of the evolution direction, update the inferior particle P_loseSpeed and position of;

wherein r is_iter,qAnd t _ nrm_r,n,iter,qParticle position and velocity update strategy with real PSO, drt_r,k,iter,qUpdating a strategy by adopting the position and the speed of the particles of the binary PSO;

wherein, ITER represents a second ITER iteration process of the improved particle swarm optimization algorithm, ITER belongs to {1, 2.,. ITER,. and ITER }, q represents a q-th particle in the second ITER iteration process, and q belongs to {1, 2.,. u.,. and Z }; r is_iter,qThe r group of arrangement combinations of the strip imaging numbers corresponding to the q particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,qRepresenting improved particle swarm optimizationAnd in the iterative process of the iter time of the algorithm, the strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the q-th particle_r,iter,qImaging direction corresponding to the number of the kth strip; t _ nrm_r,n,iter,qRepresenting the strip endpoint imaging sequence SP _ srt corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the q-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,qNormalizing the coefficient of the imaging time of the nth strip endpoint;

wherein r is_iter,qAnd t _ nrm_r,n,iter,qThe particle position and velocity update strategy with real PSO is an improved PSO particle position and velocity update strategy:

v(iter+1)＝wv(iter)+c₁*(P_winx(iter)-P_losex(iter))+c₂*(P_centerx(iter)-P_losex(iter))

P_losex(iter+1)＝P_losex(iter)+v(iter+1)

wherein w represents the inertial weight and has a value range of [0,1 ]]，c₁，c₂Represents the acceleration factor, v (iter), v (iter +1) being the current and new velocities of the particles, P_winx(iter),P_losex (iter) represents the current position of the superior particle and the current position of the inferior particle, P_losex (iter +1) represents the new position of the inferior particle, P_centerx (iter) represents the current average position of all particles.

And 4.4, performing optimization iteration to obtain an imaging task planning scheme with the maximum observation income and the shortest task completion time:

repeating the optimization iteration step 4.2 and step 4.3 for multiple times, and comparing the fitness values corresponding to the optimal particles in the last iteration process and the current iteration process, namely comparing Fit_iter-1,bestAnd Fit_iter,bestIf Fit_iter-1,bestIs superior to Fit_iter,bestIf the fitness value corresponding to the optimal particle in the last iteration process is the optimal imaging task scheme, if Fit is adopted, the optimal particle is selected as the optimal particle_iter,bestIs superior to Fit_iter-1,bestThen the fitness corresponding to the optimal particle in the current iteration processAn imaging task scenario with a superior value;

repeating optimization iteration until the iteration termination condition, thereby obtaining particles with the optimal particle swarm fitness value, wherein the corresponding imaging scheme is the optimal imaging task scheme Fit_*,best；

Wherein best represents the particles with the optimal fitness value in the particle group in the iter iteration process of the improved particle swarm optimization algorithm, Fit_iter,bestRepresenting the fitness value corresponding to the best particle best in the iter iteration process of the improved particle swarm optimization algorithm; fit_iter-1,bestRepresenting the fitness value corresponding to the best particle best in the iter-1 iteration process of the improved particle swarm optimization algorithm; fit_*,bestAnd when the iteration termination condition is reached, improving the fitness value corresponding to the optimal particle best in the first iteration process of the particle swarm optimization algorithm, namely, the optimal imaging task scheme.

Compared with the prior art, the invention has the following advantages:

arranging and combining r and strip imaging direction drt in the r-th group of strip imaging numbers_r,kAnd the imaging time normalization coefficient t _ nrm_r,nThe method is used as a decision variable, and the problems that the evaluation and calculation process of the imaging scheme is complex and the optimization and solution efficiency is low when the imaging time is directly used as the optimization model decision variable are avoided; constructing a task planning model with the maximum imaging coverage benefit and the minimum task completion time as an optimization target, realizing priority of benefit and giving consideration to efficiency; furthermore, the decision variable is optimized and solved by using the improved PSO algorithm, so that the prematurity of the standard PSO is overcome, and the requirement for optimization and solution of the binary decision variable can be met. The method can fully utilize the attitude mobility of the agile satellite and realize the optimization of the imaging task scheme in the same-orbit multipoint target motion.

Drawings

FIG. 1: the method is a strip decomposition mode for imaging of a traditional agile earth observation satellite;

FIG. 2: a strip decomposition mode for ultra-agile earth observation satellite imaging;

FIG. 3: is a schematic diagram of a point target of an embodiment of the present invention;

FIG. 4: the point target clustering result is a schematic diagram in the embodiment of the invention;

FIG. 5: the result of the stripe decomposition in the embodiment of the present invention is shown schematically;

FIG. 6: the number of the strip inner pushing and sweeping directions is schematically shown in the embodiment of the invention;

FIG. 7: an image time window overlapping schematic diagram according to an embodiment of the invention;

FIG. 8: cutting a schematic diagram of an imaging time window according to the embodiment of the invention;

FIG. 9: a constraint schematic diagram of the imaging time before and after the embodiment of the invention;

FIG. 10: the method comprises the steps of (1) a schematic flow chart of decision variable reduction imaging time in the embodiment of the invention;

FIG. 11: an imaging strip atomic task set schematic diagram according to an embodiment of the invention;

FIG. 12: a schematic process diagram of the decision variable reduction imaging moment of the embodiment of the invention;

FIG. 13: a schematic diagram of a strip end point push-scan sequence according to an embodiment of the present invention;

FIG. 14: a schematic diagram of an imaging time window and a normalized imaging time clipped according to an embodiment of the invention;

FIG. 15: the relationship between the point target and the track of the subsatellite point in the embodiment of the invention is shown schematically;

FIG. 16: the point target preprocessing result of the embodiment of the invention is shown schematically;

FIG. 17: the strip endpoint push-broom sequence diagram of the optimal scheme of the embodiment of the invention;

FIG. 18: a flowchart of the steps of an embodiment of the present invention.

Detailed Description

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

The following describes an imaging task planning method in the same orbit multi-point target of the hypersensitive short satellite according to an embodiment of the present invention with reference to fig. 1 to 18, including the following steps:

step 1, inputting longitude and latitude coordinates of a plurality of point targets, and converting the longitude and latitude coordinates of the plurality of point targets into plane coordinates of the plurality of point targets:

as shown in fig. 1, the strip sweeping direction of the conventional agile satellite must be parallel to the track of the satellite subsatellite point, so that 2 strips are required to achieve coverage of all point targets. For the ultra-agile satellite, the imaging capability is provided in motion, the push-broom direction can not be parallel to the track of the point under the satellite, and as shown in fig. 2, one strip can be realized to cover all point targets.

i∈[1,np]

the point target cluster is:

wherein the content of the first and second substances,

and clustering all point targets into e cluster clusters through point target clustering, and constructing a non-edge editing strip decomposition method based on the rotary card shell on the basis. When non-tracing strip decomposition is carried out, m cluster clusters generated by clustering are covered by a plurality of parallel rectangular strips. If the number of the divided strips is minimum, a circumscribed rectangle with the minimum area of the clustering cluster needs to be determined.

A cluster generated by clustering

For example, a non-tracking stripe decomposition process based on a rotating card shell is illustrated:

constructed by using the idea of rotating the clamping shell

wherein m represents the number of strips, j represents the jth strip, j is more than or equal to 1 and less than or equal to m, and Point _l,j ^a1 st vertex coordinate, Point, representing the jth stripe in the ith cluster_l,j ^a＝(x_l,j ^a,y_l,j ^a)，Point _l,j ^b2 nd vertex coordinate, Point, representing the jth stripe in the ith cluster_l,j ^b＝(x_l,j ^b,y_l,j ^b)，Point_l,j ^cDenotes the 3 rd vertex coordinate, Point, of the jth strip in the ith cluster_l,j ^c＝(x_l,j ^c,y_l,j ^c)，Point_l,j ^dThe 4 th vertex coordinate, Point, representing the jth strip in the ith cluster_l,j ^d＝(x_l,j ^d,y_l,j ^d)；

SendPoint_l＝{endPoint_l,1,endPoint_l,2,...,endPoint_l,M,...,endPoint_l,2*m}

wherein M represents the number of strips, 2M represents the number of strip end points, M is more than or equal to 1 and less than or equal to 2M, endPoint_l,MThe longitude and latitude coordinates of the M-th strip endpoint in the clustering cluster l are obtained;

As shown in fig. 3-5, which are examples of point target preprocessing via DBSCAN clustering and rotating stuck shell non-tracing strip decomposition. Fig. 3 shows a point target set P ═ { P ═ P₁,p₂,....,p₁₇}; FIG. 4 shows a clustering result, clustering a set of target points into 5 clusters; fig. 5 shows a set of bands S ═ { a, B, C, D, E, F }, and a set of band endpoints SP ═ a, B, C, D, E, F, g, h, i, j, k, l }, which are formed after the preprocessing.

the imaging time window forecast determines the starting point and the ending point of a time period in which the satellite can image each target in a future time period, and is the basis of the imaging task planning of the agile satellite. The target can be observed in the process of orbital motion of the optical imaging satellite, and the real-time maximum observable range is an area with a fixed size. The maximum observation range of a traditional non-agile satellite is determined by the field angle of the traditional non-agile satellite, and for the agile satellite, the maximum observation range is related to the maximum attitude maneuver angle and the field angle. For an object, the time period for which the object is observed may be the time window for the object. In view of the fact that most of the existing mainstream agile remote sensing satellites adopt linear array sensors and can adopt a 'characteristic cone' to describe the maximum observable range of the satellite, when an imaging time window of each strip end point is calculated, reference can be made to a paper (Shenxin, Lideren, Yaoheng, an optical remote sensing satellite imaging window rapid forecasting method facing imaging task planning [ J ]. Wuhan university academic newspaper (information science edition), 2012,37(12): 1468-.

Determining the imaging time window of each strip endpoint by adopting a characteristic cone method according to the strip endpoint number set SP and the corresponding strip endpoint longitude and latitude coordinate set SendPoint to obtain the imaging time window [ TW ] of each strip endpoint in the strip endpoint number set SP_n,s,TW_n,e]；

the imaging number sequence of the constructed strips in the step 2 is as follows:

S_ind_r＝{s_ind_r,1,s_ind_r,2,...,s_ind_r,k,...,s_ind_r,ns}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

the sequence of the strip imaging numbers is S _ ind_rDetermined by the r-th group permutation and combination r of the strip imaging numbers, r is {1,2, …, ns! Positive integers between, each positive integer corresponding to a set of slice imaging numbering sequences. Taking 4 imaging strips generated by decomposition as an example, the 4 imaging strips have different strip imaging number sequences in 24, as shown in table 2, the number 0 and the number 23 are completely in reverse order, which is beneficial to maintaining the coding neighborhood structure in the evolutionary algorithm.

TABLE 1 sequence of slice imaging numbers corresponding to the r-th group permutation and combination r of slice imaging numbers

The reduction of the slice imaging number sequence S _ ind by r is given below_rThe pseudo-code of (1) is as follows:

step 2, constructing a strip imaging direction sequence as follows:

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Determining the imaging direction of the strip as a binary number;

as shown in fig. 6, in determining the imaging direction of the swath, a binary number of 0 represents a forward swipe in the swath imaging direction and a binary number of 1 represents a reverse swipe in the swath imaging direction.

Step 2, the imaging sequence of calculating the strip end point is as follows:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein, SP _ srt_rRepresenting the strip end point imaging sequence corresponding to the r-th group permutation and combination r of the strip imaging numbers, r representing the r-th group permutation and combination of the strip imaging numbers, ns representing the strip number, 2 × ns representing the stripNumber of tape ends, ns! Representing the number of all permutation and combination of the strip imaging numbers; k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rThe kth band; n represents SP _ srt_rThe nth stripe endpoint, sp _ srr_r,nRepresentation SP _ srt_rNumber of nth stripe endpoint, sp _ srt_r,nE, SP represents the strip endpoint number set in the step 1;

if there are three imaging slices, the slice number set S is {1,2,3}, the corresponding slice endpoint number set SP is {1,2,3,4,5,6}, the two slice endpoints corresponding to the slice numbered 1 are 1 and 2, the two slice endpoints corresponding to the slice numbered 2 are 3 and 4, and the two slice endpoints corresponding to the slice numbered 3 are 5 and 6. After the slice imaging sequence is determined according to the r-th group permutation and combination r of the slice imaging numbers, if r is 1, then there is S _ ind₁In the case where the swath imaging direction is not determined, the determined swath end imaging sequence is: {1or2,1or2,5or6,5or6,3or4,3or4}, the imaging sequence of the strip end point can be finally determined only when the strip imaging direction is determined, and if drt in the strip imaging direction sequence is determined_1,1＝0，drt_1,2＝1,drt _1,31, the final imaging sequence SP _ srt of the strip end point₁＝{1,2,6,5,4,3}；

the imaging time normalization coefficient is as follows:

t_nrm_r,n∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

imaging sequence SP _ sr in calculating strip end pointt_rImaging time t corresponding to each strip end point_r,nAnd then, includes: cutting an imaging time window and normalizing imaging time;

and (3) cutting the imaging time window:

strip endpoint imaging sequence SP _ srt_rCan be formed from r and drt_r,kIt is determined that, however, since the imaging time windows of the strip end points may overlap with each other, as shown in fig. 7, the imaging time windows of three strip end points observed consecutively are [ T [ ], respectively_{r,1_s},T_{r,1_e}]、 [T_{r,2_s},T_{r,2_e}]、[T_{r,3_s},T_{r,3_e}]Apparently at SP _ srt_rIn the case of confirmation, [ T_{r,3_e},T_{r,1_e}]、[T_{r,3_e},T_{r,2_e}]、[T_{r,3_s},T _{r,2_s}]Three time periods are invalid time ranges. To avoid searching for invalid spaces, the imaging sequence SP _ srt is based on the end-of-band_rCropping of the imaging time window for all strip endpoints.

if T_r,n,s＞T_r,n+1,sThen ITW_r,n+1,s＝T_r,n,s,ITW_r,n,s＝T_r,n,s

If T_r,n,e＞T_r,n+1,eThen ITW_r,n+1,e＝T_r,n+1,e,ITW_r,n,e＝T_r,n+1,e

according to the overlapping relationship between two adjacent strip endpoint imaging time windows, a cutting schematic diagram of a specific imaging time window is shown in fig. 8, wherein the left side represents an original imaging time window, and the right side represents a cut imaging time window.

The imaging time is normalized:

after the time window is cut, the strip end point imaging sequence SP _ srt_rIn any two adjacent strip end points, the imaging time of the point observed first can constrain the feasible range of the point observed later, and if the point is directly searched in the imaging time window after cutting, a large number of invalid solutions can be generated, and the model solving efficiency is influenced. As shown in FIG. 9, when the imaging time t is determined for the 1 st strip end point_r,1Then [ ITW ] in the imaging time window of the 2 nd strip end_{r,2_s}，t_r,1]For invalid search spaces, need to be at [ t ]_r,1，ITW_{r,2_e}]And searching for feasible solutions. In this case, the imaging time of each endpoint is not only constrained by the imaging time window after cropping, but also must satisfy the observation point sequence constraint, and the upper and lower bounds of the search cannot be fixed.

For the problem, after finishing the imaging time window cutting, an imaging time normalization coefficient t _ nrm is introduced_r,nNormalizing the interval of the imaging moments of two consecutive end points to [0,1 ]]And the upper and lower boundaries of the decision variables are fixed, so that the optimization algorithm is convenient to solve.

if n is 1, then t_r,1＝ITW_r,1,s+t_nrm_r,n*(ITW_r,1,e-ITW_r,1,s)

If n ≠ 1, then

If t_r,n＞ITW_r,n+1,sThen t is_r,n+1＝t_r,n+t_nrm_r,n*(ITW_r,n+1,e-t_r,n)

in the model solving process, when the observation yield is calculated, the normalization coefficient is restored to the corresponding imaging time by using the formula, and meanwhile, the restored imaging time is ensured to meet the clipped imaging time window constraint of each end point.

The method comprises the following steps of constructing a strip imaging number sequence and a strip imaging direction sequence by a specific example, calculating a strip endpoint imaging sequence according to the strip imaging number sequence and the strip imaging direction sequence, and calculating a strip endpoint imaging time sequence by utilizing an imaging time normalization coefficient by combining the strip endpoint imaging sequence:

triplet [ r, drt ]_r,1,drt_r,2,…,drt_r,k,…,drt_r,ns,t_nrm_r,1,t_nrm_r,2,...,t_nrm_r,m,…,t_nrm_r,2*ns]2 x ns imaging moments t corresponding to the strip end points can be restored according to the triplets_r,1,t_r,2,…,t_r,m,…,t_r,2*ns]Wherein r, drt_r,kImaging sequence for determining strip end point SP _ srt_r，t_nrm_r,mFor determining the observation sequence SP _ srt_rAt the moment of imaging of the end points of the respective strips.

The specific flow is shown in fig. 10.

First, according to r and drt_r,kDetermining observation sequence SP _ srt for a strap endpoint_r；

Second, according to the determined SP _ srt_rCutting the imaging time window of each strip end point;

thirdly, according to t _ nrm_r,nThe normalization operation is carried out to restore SP _ srt_rThe imaging time corresponding to the end point of each strip.

Taking four imaging strips as an example, the specific situation of the imaging moment restored by the decision variable is explained. As shown in fig. 11, S ═ { a, B, C, D } and SP ═ {1,2, …,8} obtained by the preprocessing in step 1. The values of the three sets of decision variables are shown in table 2.

TABLE 2 three sets of decision variable values

The specific process of the decision variable triplet restoration imaging time is shown in fig. 12. The observation sequence of the end point of the strip, the clipping of the imaging time window and the normalization of the imaging time are shown in fig. 13 and 14.

in the existing imaging task planning model construction method, the imaging time is usually directly used as a decision variable, the quantitative relation between the decision variable and a target function and a constraint condition is established, and further the quantitative relation is establishedAnd solving the task planning model. However, the observation time of the imaging strip at the beginning and the observation time at the end are directly used as the decision variables of the optimization model, which leads to the problem of low efficiency of the model optimization solution. Therefore, a strip push-sweep sequence number r and a strip inner push-sweep direction number drt are introduced in the model construction process_r,kImaging time normalization coefficient t _ nrm_r,nAnd three types of variables are used as model decision variables. Wherein, r and drt_r,kDetermining a sequence of strip end-point observations, t _ nrm_r,nAnd determining the time interval between the imaging moments of the two adjacent strip end points. The three types of decision variables can form a one-to-one mapping relation with a group of imaging moments, and consumption of attitude maneuver constraint calculation is reduced.

Step 3, constructing decision variables according to the strip imaging number sequence, the strip imaging direction sequence and the strip endpoint imaging time sequence, wherein the decision variables are as follows:

the strip imaging number sequence is as follows:

S_ind_r＝{s_ind_r,1,s_ind_r,2,...,s_ind_r,k,...,s_ind_r,ns}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

the swath imaging direction sequence is:

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Representing the number of all permutation and combination of the strip imaging numbers, k representing the r-th groupPermutation and combination of strip imaging number sequence S _ ind_rMiddle k-th strip, drt_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rImaging direction corresponding to the number of the kth strip, when drt_r,kWhen 0, the slice imaging number sequence S _ ind representing the r-th group arrangement combination_rIn the middle, the imaging direction of the k-th strip is forward push-scan when drt_r,kWhen 1, the slice imaging number sequence S _ ind indicating the r-th group arrangement combination_rIn the step (2), the imaging direction of the kth strip is reverse push-broom;

the sequence of imaging time instants of the strip end points is as follows:

t_r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the constructed decision variables are:

r∈{1,2,...,ns！}

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Determining the imaging direction of the strip as a binary number;

t_nrm_r,n∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

step 3, maximizing the imaging coverage yield:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Representing the number of all permutation combinations of the band imaging numbers, and k representing the band imaging number sequence S _ ind of the r-th permutation combination_rThe k-th strip of (C)_r,kRepresents S _ ind_rWhether the k-th band can be imaged, c_r,kE.g. {0,1}, if c_r,kIf 1, then S _ ind_rThe imaging can be finished in the kth strip, otherwise, the imaging cannot be realized; g_r,kRepresents S _ ind_rThe number of points covered by the kth strip;

c_r,kthe judgment basis comprises two aspects:

namely:

Otherwise c_r,k＝0

Step 3, the minimized task completion time is an objective function:

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

indicating a minimized imaging task completion time;

if Δ t_r,n≥t_tr_r,nThen c _ tr_r,n＝1

Otherwise c _ tr_r,n＝0

Step 3, the imaging time window constraint is as follows:

t_r,n∈[ITW_r,n-s,ITW_r,n-e]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

and 3, constraining the posture conversion time as follows:

Δt_r,n≥t_tr_r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the maneuvering speed of the rolling axis of the satellite,

The maneuvering speed of the satellite pitching axis,

Respectively the three-axis maneuvering speed of the satellite;

Step 4, combining with a same-orbit multipoint target in-motion imaging task planning mathematical model, and optimizing by improving a particle swarm optimization to obtain an imaging task planning scheme with the maximum observation gain and the shortest task completion time;

the meta-heuristic algorithm is the most widely applied method in solving the agile satellite imaging task planning model, wherein the PSO and the improved variable algorithm thereof are one of the most commonly used methods for solving the imaging task planning model at present. In standard PSO, each particle moves to a new position based on a new velocity and current position. However, standard particle swarm optimization tends to fall into local optimality.

Aiming at the real number decision variables and the binary decision variables existing in the process of solving the constructed model, the method aims at the real number decision variables (r, t _ nrm)_r,m) A new PSO speed and position updating method is adopted. And randomly selecting two particles for comparison, and taking the current position of the better particle and the average position of all the particles as the evolution direction of the current position of the worse particle. The particle velocity and position are updated as:

P_losex(iter+1)＝P_losex(iter)+v(iter+1)

wherein iter represents the iter iteration process of the improved particle swarm optimization algorithm, w represents the inertia weight, and the value range is [0,1 ]]，c₁，c₂Represents the acceleration factor, v (iter), v (iter +1) being the current and new velocities of the particles, P_winx(iter),P_losex (iter) represents the current position of the superior particle and the current position of the inferior particle, P_losex (iter +1) represents the new position of the inferior particle, P_centerx (iter) represents the current average position of all particles.

The method is directed to binary decision variables (drt)_r,k) The speed and position update strategy of the BPSO algorithm is adopted.

Performing multiple iterations by using the improved PSO algorithm to determine an identical-orbit multipoint target imaging scheme of the ultra-agile satellite;

the method comprises the following specific steps:

step 4.1, randomly generating an initial population:

wherein, the granuleThe position of the child represents the decision variable r constructed as described in step 3_iter,u、drt_r,k,iter,u、t_nrm_r,n,iter,uThe initial value of each particle is a random value in the value range of each particle, and the initial speed of each particle is set to be zero;

according to r_iter,u、drt_r,k,iter,u、t_nrm_r,n,iter,uAnd restoring a strip endpoint imaging sequence SP _ srt corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm by utilizing the step 2_r,iter,uAnd the imaging time t corresponding to the end point of each strip_r,n,iter,u(ii) a To be adjacent toAttitude transition time t _ tr between two strip endpoints_r,n,iter,uObtaining a strip endpoint observation sequence capable of completing push-broom according to the judgment basis, and obtaining observation income and task completion time;

and 4.3, updating the speed and the position of the particles:

wherein, ITER represents a second ITER iteration process of the improved particle swarm optimization algorithm, ITER belongs to {1, 2.,. ITER,. and ITER }, q represents a q-th particle in the second ITER iteration process, and q belongs to {1, 2.,. u.,. and Z }; r is_iter,qThe r group of arrangement combinations of the strip imaging numbers corresponding to the q particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,qThe strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the q-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,qImaging direction corresponding to the number of the kth strip; t _ nrm_r,n,iter,qAn r group arrangement group of strip imaging numbers corresponding to the q particles in the iter iteration process of the improved particle swarm optimization algorithmResultant r corresponding strip endpoint imaging sequence SP _ srt_r,iter,qNormalizing the coefficient of the imaging time of the nth strip endpoint;

P_losex(iter+1)＝P_losex(iter)+v(iter+1)

repeating the optimization iteration step 4.2 and step 4.3 for multiple times, and comparing the fitness values corresponding to the optimal particles in the last iteration process and the current iteration process, namely comparing Fit_iter-1,bestAnd Fit_iter,bestIf Fit_iter-1,bestIs superior to Fit_iter,bestIf the fitness value corresponding to the optimal particle in the last iteration process is the optimal imaging task scheme, if Fit is adopted, the optimal particle is selected as the optimal particle_iter,bestIs superior to Fit_iter-1,bestIf so, the fitness value corresponding to the optimal particle in the current iteration process is a better imaging task scheme;

Wherein best represents the iter iteration of the improved particle swarm optimization algorithmFit, the particle with the best fitness value in the particle group in the generation process_iter,bestRepresenting the fitness value corresponding to the best particle best in the iter iteration process of the improved particle swarm optimization algorithm; fit_iter-1,bestRepresenting the fitness value corresponding to the best particle best in the iter-1 iteration process of the improved particle swarm optimization algorithm; fit_*,bestAnd when the iteration termination condition is reached, improving the fitness value corresponding to the optimal particle best in the first iteration process of the particle swarm optimization algorithm, namely, the optimal imaging task scheme.

In specific implementation, the above process can adopt computer software technology to realize automatic operation process. The system device of the corresponding operation flow should also be within the protection scope of the present invention.

TABLE 3 satellite parameters

TABLE 4 Point target coordinates

For reference, an example is provided, in which the satellite orbit-related parameters and the satellite attitude mobility are shown in table 3, the point targets are shown in table 4, and the position relationship between the point targets and the subsatellite point trajectory is shown in fig. 15.

as shown in fig. 16, the result of point object clustering and non-tracing decomposition is obtained, the set of slice numbers S ═ { a, B, …, K }, the set of slice end points SP ═ 1,2, …,22}, and the original imaging time windows of all the slice end points;

TABLE 5 original imaging time Window

Strip endpoint numbering	Raw imaging time window
		1	[2000-3-22-11-59-10,2000-3-22-12-1-52]
2	[2000-3-22-11-59-10,2000-3-22-12-1-49]
		3	[2000-3-22-11-59-10,2000-3-22-12-1-48]
4	[2000-3-22-11-59-10,2000-3-22-12-1-51]
		5	[2000-3-22-11-59-10,2000-3-22-12-1-50]
6	[2000-3-22-11-59-10,2000-3-22-12-1-46]
		7	[2000-3-22-11-59-10,2000-3-22-12-1-45]
8	[2000-3-22-11-59-10,2000-3-22-12-1-48]
		9	[2000-3-22-11-59-10,2000-3-22-12-1-44]
10	[2000-3-22-11-59-10,2000-3-22-12-1-46]
		11	[2000-3-22-11-59-10,2000-3-22-12-1-43]
12	[2000-3-22-11-59-10,2000-3-22-12-1-44]
		13	[2000-3-22-11-59-10,2000-3-22-12-1-43]
14	[2000-3-22-11-59-10,2000-3-22-12-1-41]
		15	[2000-3-22-11-59-10,2000-3-22-12-1-40]
16	[2000-3-22-11-59-10,2000-3-22-12-1-41]
		17	[2000-3-22-11-59-10,2000-3-22-12-1-50]
18	[2000-3-22-11-59-10,2000-3-22-12-1-48]
		19	[2000-3-22-11-59-10,2000-3-22-12-1-47]
20	[2000-3-22-11-59-10,2000-3-22-12-1-48]
		21	[2000-3-22-11-59-10,2000-3-22-12-1-44]
22	[2000-3-22-11-59-10,2000-3-22-12-1-43]

step 4.1, randomly generating an initial population:

1) let's assume random r in the current population_iter,u＝0、drt_r,k,iter,u＝{0,0,0,0,0,0,0,0,0,0,0}、 t_nrm_r,n,iter,uWhere {0.1,0.2,0.5,0.4,0.3,0.5,0.2,0.6,0.1,0.2,0.1,0.2,0.3,0.4,0.5,0.6,0.1,0.2,0.3,0.1,0.2,0.8}, then r can be determined according to r_iter,u、drt_r,k,iter,uDetermining a strip endpoint imaging sequence SP _ srt corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uIs 1-2-3-4-5-6-7-8-9-10-11-12-13-14-15-16-17-18-19-20-21-22.

2) Determining a strip endpoint imaging sequence SP _ srt corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uThereafter, the imaging time window cropping of each strip end was performed, and the results are shown in table 6.

TABLE 6 cropped imaging time Window

Strip end point imaging sequence	Post-crop imaging time window
		1	[2000-3-22-11-59-10,2000-3-22-12-1-40]
2	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		3	[2000-3-22-11-59-10,2000-3-22-12-1-40]
4	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		5	[2000-3-22-11-59-10,2000-3-22-12-1-40]
6	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		7	[2000-3-22-11-59-10,2000-3-22-12-1-40]
8	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		9	[2000-3-22-11-59-10,2000-3-22-12-1-40]
10	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		11	[2000-3-22-11-59-10,2000-3-22-12-1-40]
12	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		13	[2000-3-22-11-59-10,2000-3-22-12-1-40]
14	[2000-3-22-11-59-10,2000-3-22-12-1-40]
		15	[2000-3-22-11-59-10,2000-3-22-12-1-40]
16	[2000-3-22-11-59-10,2000-3-22-12-1-41]
		17	[2000-3-22-11-59-10,2000-3-22-12-1-43]
18	[2000-3-22-11-59-10,2000-3-22-12-1-43]
		19	[2000-3-22-11-59-10,2000-3-22-12-1-43]
20	[2000-3-22-11-59-10,2000-3-22-12-1-43]
		21	[2000-3-22-11-59-10,2000-3-22-12-1-43]
22	[2000-3-22-11-59-10,2000-3-22-12-1-43]

3) According to t _ nrm_r,n,iter,uReduced observation sequence SP _ srt_r,iter,uThe imaging timings of the end points of the respective bands are shown in table 7.

TABLE 7 imaging moments for imaging moment normalization coefficient reduction

Strip end point imaging sequence	Imaging time
		1	[2000-3-22-11-59-25]
2	[2000-3-22-11-59-52]
		3	[2000-3-22-12-00-46]
4	[2000-3-22-12-01-7.6]
		5	[2000-3-22-12-01-17.32]
6	[2000-3-22-12-01-28.66]
		7	[2000-3-22-12-01-30.928]
8	[2000-3-22-12-01-36.3682]
		9	[2000-3-22-12-01-36.7314]
10	[2000-3-22-12-01-37.3851]
		11	[2000-3-22-12-01-37.6466]
12	[2000-3-22-12-01-38.1173]
		13	[2000-3-22-12-01-38.6821]
14	[2000-3-22-12-01-39.2093]
		15	[2000-3-22-12-01-39.6047]
16	[2000-3-22-12-01-40.4419]
		17	[2000-3-22-12-01-40.6978]
18	[2000-3-22-12-01-41.1582]
		19	[2000-3-22-12-01-42.0107]
20	[2000-3-22-12-01-42.1091]
		21	[2000-3-22-12-01-42.2873]
22	[2000-3-22-12-01-42.8575]

4) With the posture conversion time t _ tr between two adjacent strip end points_r,n,iter,uObtaining a strip endpoint observation sequence capable of completing push-broom according to the judgment basis, and obtaining observation income and task completion time;

and 4.3, updating the speed and the position of the particles:

the imaging sequence of the end points of the strip and the imaging time of each end point of the strip corresponding to the optimal result obtained by the final optimization solution are shown in table 8, wherein the imaging sequence of the end points of the strip is shown in fig. 17, and the arrow indicates the imaging direction of the strip.

Table 8 strip end point imaging sequence and each strip end point imaging time corresponding to optimal result

Referring to fig. 17 and table 8, full imaging of all point targets can be achieved, with a task completion time of 11.243s being optimal.

The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.

Claims

1. A hypersensitive agile satellite co-orbit multipoint target in-motion imaging task planning method is characterized by comprising the following steps of:

and 4, combining a mathematical model of imaging task planning in the simultaneous multi-point target, and optimizing by improving a particle swarm optimization to obtain an imaging task planning scheme with the maximum observation yield and the shortest task completion time.

2. The method for planning an imaging task in the same orbit multi-point target of the hypersensitive agile satellite according to claim 1, wherein:

Where np is the number of point targets, P_iIs the ith point target, (B)_i,L_i) Is the ith pointLatitude and longitude coordinates of the target, B_iIs the latitude, L, of the ith point target_iFor the longitude of the ith point target, i ∈ [1, np [ ]]；

taking any point target in P as the center of a circle and eps as the center of a circleRadius, dis in plane coordinates of multiple point objects within a circle_i,i+1Clustering all point targets with the number of point targets more than or equal to 2 in a circle which is less than or equal to eps and less than or equal to Minpts, and gradually expanding the cluster to obtain a final point target cluster;

the point target cluster is:

wherein the content of the first and second substances,

indicates the inclusion of c in the ith cluster_lPoint targets, e represents the number of clusters of point targets, l ∈ [1, e ∈]，c_lRepresents the number of point targets in the ith cluster, and c₁+c₂+...+c_l+...+c_eEnsuring that e clustering clusters formed by clustering the point targets contain all the point targets;

constructed by using the idea of rotating the clamping shell

The minimum circumscribed rectangle of the area obtained by the method has the following vertex coordinate set: { Point_l ^a，Point_l ^b，Point_l ^c，Point_l ^d}；

Wherein Point_l ^aPoint, which represents the 1 st vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^a＝(x_l ^a,y_l ^a)，Point_l ^bPoint, the 2 nd vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^b＝(x_l ^b,y_l ^b)，Point_l ^cPoint representing the 3 rd vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^c＝(x_l ^c,y_l ^c)，Point_l ^dPoint representing the 4 th vertex coordinate of the minimum bounding rectangle in the l-th cluster_l ^d＝(x_l ^d,y_l ^d) The obtained minimum external rectangle vertex coordinate set of the cluster comprises the cluster;

{Point_l，1 ^a，Point_l，1 ^b，Point_l，1 ^c，Point_l，1 ^d；Point_l，2 ^a，Point_l，2 ^b，Point_l，2 ^c，Point_l，2 ^d；...；Point_l，j ^a，Point_l，j ^b，Point_l，j ^c，Point_l，j ^d；...；Point_l，m ^a，Point_l，m ^b，Point_l，m ^c，Point_l，m ^d}；

where m denotes the number of stripes, j denotes the jth stripeZone, j is more than or equal to 1 and less than or equal to m, Point_l,j ^aThe 1 st vertex coordinate, Point, of the jth strip in the ith cluster_l,j ^a＝(x_l,j ^a,y_l,j ^a)，Point_l,j ^b2 nd vertex coordinate, Point, representing the jth stripe in the ith cluster_l,j ^b＝(x_l,j ^b,y_l,j ^b)，Point_l,j ^cDenotes the 3 rd vertex coordinate, Point, of the jth strip in the ith cluster_l,j ^c＝(x_l,j ^c,y_l,j ^c)，Point_l,j ^dThe 4 th vertex coordinate, Point, representing the jth strip in the ith cluster_l,j ^d＝(x_l,j ^d,y_l,j ^d)；

SendPoint_l＝{endPoint_l,1,endPoint_l,2,...,endPoint_l,M,...,endPoint_l,2*m}

SP represents the final set of stripe end point numbers, SP ═ 1, 2., n., 2 × ns }, where n represents the stripe end point number, 1 ≦ n ≦ 2 × ns, and 2 × ns represents the final number of stripe end points; set of longitude and latitude coordinates of all strip endpoints, SendPoint ═ endPoint₁,endPoint₂,...,endPoint_n,...,endPoint_2*nsIn which endPoint_nRepresenting longitude and latitude coordinates corresponding to the strip end point with the strip end point number n, wherein n is more than or equal to 1 and less than or equal to 2 x ns;

according to the strip endpoint number set SP and the corresponding strip endpoint longitude and latitude coordinate set SendPoint, a characteristic cone method (Shenxin, Lide Kernel, YaoShuang) is adopted, and an optical remote sensing satellite imaging window rapid prediction method facing imaging task planning [ J]Wuhan university studyThe newspaper (information science edition), 2012,37(12):1468-_n,s,TW_n,e]；

Wherein, [ TW_n,s,TW_n,e]Representing the original imaging time window, TW, corresponding to the strip end point with the strip end point number n_n,sFor the start time of the time window, TW_n,eIs the time window end time.

3. The method for planning an imaging task in the same orbit multi-point target of the hypersensitive agile satellite according to claim 1, wherein:

S_ind_r＝{s_ind_r,1,s_ind_r,2,...,s_ind_r,k,...,s_ind_r,ns}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

step 2, constructing a strip imaging direction sequence as follows:

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

drt_r,k∈{0,1}

k belongs to {1, 2.,. ns } which is a binary number, and the imaging direction of the strip is determined;

step 2, the imaging sequence of calculating the strip end point is as follows:

SP_srt_r＝{sp_srt_r,1,ST,sp_srt_r,1,ET,sp_srt_r,2,ST,sp_srt_r,1,ET,...,sp_srt_r,k,ST,sp_srt_r,k,ET...,sp_srt_r,ns,ST,sp_srt_r,ns,ET}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Wherein, SP _ srt_rThe r-th group permutation and combination r of the strip imaging numbers is represented, r represents the r-th group permutation and combination of the strip imaging numbers, ns represents the strip number,ns! Representing the number of all permutation and combination of the strip imaging numbers; k represents the r-th group permutation and combination of the strip imaging number sequence S _ ind_rThe kth band; ST denotes the starting imaged slice end point of the slice, ET denotes the ending imaged slice end point of the slice;

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the imaging time normalization coefficient is as follows:

t_nrm_r,n∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

and (3) cutting the imaging time window:

if T_r,n,s＞T_r,n+1,sThen ITW_r,n+1,s＝T_r,n,s,ITW_r,n,s＝T_r,n,s

If T_r,n,e＞T_r,n+1,eThen ITW_r,n+1,e＝T_r,n+1,e,ITW_r,n+1,e＝T_r,n+1,e

In the above formula, T_r,n,sImaging sequence representing end points of a strip SP _ srt_rThe start time of the time window corresponding to the nth observed strip endpoint; t is_r,n,eImaging sequence representing end points of a strip SP _ srt_rThe end time of the time window corresponding to the nth observed strip endpoint; ITW_r,n,sImaging sequence representing end points of a strip SP _ srt_rIn the n-th observed band end point corresponding to the cutA time window start time; ITW_r,n,eImaging sequence representing end points of a strip SP _ srt_rThe end time of the clipped time window corresponding to the nth observed strip end point;

the imaging time is normalized:

if n is 1, then t_r,1＝ITW_r,1,s+t_nrm_r,n*(ITW_r,1,e-ITW_r,1,s)

If n ≠ 1, then

If t_r,n＞ITW_r,n+1,sThen t is_r,n+1＝t_r,n+t_nrm_r,n*(ITW_r,n+1,e-t_r,n)

Wherein, t_r,1Imaging sequence SP _ srt for strip endpoint_rThe imaging time, t, of the trimmed time window corresponding to the 1 st observed strip end point_r,nImaging sequence SP _ srt for strip endpoint_rAt the imaging time of the nth observed strip end point, t_r,n+1Imaging sequence SP _ srt for strip endpoint_rThe imaging time of the (n +1) th observed band end point, ITW_r,n,sImaging sequence representing end points of a strip SP _ srt_rThe starting time of the cut time window corresponding to the nth observed strip end point; ITW_r,n,eImaging sequence representing end points of a strip SP _ srt_rAnd (5) ending the cut time window corresponding to the nth observed strip endpoint.

4. The method for planning an imaging task in the same orbit multi-point target of the hypersensitive agile satellite according to claim 1, wherein:

the strip imaging number sequence is as follows:

S_ind_r＝{s_ind_r,1,s_ind_r,2,...,s_ind_r,k,...,s_ind_r,ns}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

wherein, S _ ind_rRepresenting a sequence of slice imaging numbers, r representing an r-th group permutation combination of the slice imaging numbers, ns representing the number of slices, ns! Representing the number of all permutation combinations of the band imaging numbers, and k representing the band imaging number sequence S _ ind of the r-th permutation combination_rThe k-th slice, s _ ind_r,kStrip imaging number sequence S _ ind representing r-th group permutation and combination_rNumber of the k-th slice in, s _ ind_r,kE is S, and S represents the strip number set in the step 1;

the swath imaging direction sequence is:

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

the sequence of imaging time instants of the strip end points is as follows:

t_r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the constructed decision variables are:

r∈{1,2,...,ns！}

drt_r,k∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

drt_r,k∈{0,1}

t_nrm_r,n∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

Wherein, t _ nrm_r,nStrip end point imaging sequence SP _ srt corresponding to r-th group permutation and combination r representing strip imaging number_rThe n-th one ofThe imaging time normalization coefficient of the strip end point, n represents the observation sequence SP _ srt of the strip end point_rThe nth stripe endpoint;

step 3, maximizing the imaging coverage yield:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

c_r,kthe judgment basis comprises two aspects:

namely:

if (t)_r,2k+1-t_r,2k≥t-tr_r,2k+1)∩(t_r,2k-t_r,2k-1≥t_tr_r,2k) Then c is_r,k＝1

Otherwise c_r,k＝0

Step 3, the minimized task completion time is an objective function:

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

indicating a minimized imaging task completion time;

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rMiddle nth stripe end, Δ t_r,nImaging sequence representing end points of a strip SP _ srt_rLast observation time interval, Δ t, from the n-1 th strip end point to the n-th strip end point_r，n＝t_r，n-t_r，n-1，t_r，nImaging sequence SP _ srt for strip endpoint_rImaging time of the upper nth strip end point;

if Δ t_r,n≥t-tr_r,nThen c _ tr_r,n＝1

Otherwise c _ tr_r,n＝0

Step 3, the imaging time window constraint is as follows:

t_r,n∈[ITW_r,n-s,ITW_r,n-e]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

and 3, constraining the posture conversion time as follows:

Δt_r,n≥t_tr_r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein r represents the r-th group arrangement combination of the slice imaging numbers, ns represents the number of slices, ns! Indicates the number of all permutation and combination of the strip imaging numbers, and n indicates the strip endpoint observation sequence SP _ srt_rMiddle nth stripe end, Δ t_r,n≥t-tr_r,nImaging sequence representing end points of a strip SP _ srt_rThe imaging time interval between two adjacent strip end points needs to meet the posture conversion time constraint between the two adjacent end points;

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the maneuvering speed of the rolling axis of the satellite,

The maneuvering speed of the satellite pitching axis,

Respectively the three-axis maneuvering speed of the satellite;

5. The method for planning an imaging task in the same orbit multi-point target of the hypersensitive agile satellite according to claim 1, wherein:

and 4, optimizing by improving the particle swarm optimization to obtain an imaging task planning scheme with the maximum observation yield and the shortest task completion time, wherein the imaging task planning scheme comprises the following steps:

step 4.1, randomly generating an initial population:

wherein iter represents the iter iteration process of the improved particle swarm optimization algorithm,ITER ∈ {1, 2., ITER }, u denotes the u-th particle in the ITER iteration, u ∈ {1, 2., u., Z }; r is_iter,uThe r group of arrangement combinations of the strip imaging numbers corresponding to the u particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,uThe strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_rImaging direction corresponding to the number of the kth strip; t _ nrm_r,n,iter,uRepresenting the strip endpoint imaging sequence SP _ srt corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uNormalizing the coefficient of the imaging time of the nth strip endpoint;

wherein, ITER represents the iterative process of the second ITER of the improved particle swarm optimization algorithm, ITER belongs to {1, 2.,. ITER.,. ITER }, and u represents the particle of the u th in the iterative process of the second ITERChild, u ∈ {1,2,. u,. Z }; r is_iter,uThe r group of arrangement combinations of the strip imaging numbers corresponding to the u particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,uThe strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uImaging direction, S _ ind, corresponding to the number of the kth stripe_r,iter,uA strip imaging number sequence of the r-th group arrangement combination of the strip imaging numbers corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm, t _ nrm_r,n,iter,uRepresenting the strip endpoint imaging sequence SP _ srt corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,uNormalization coefficient of imaging time of the nth strip end point, SP _ srt_r,iter,uRepresenting the strip endpoint imaging sequence corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm, t_r,n,iter,uRepresenting the imaging time of the nth strip endpoint in the strip endpoint imaging sequence corresponding to the r-th group arrangement combination r of the strip imaging number corresponding to the u-th particle in the iter iteration process of the improved particle swarm optimization algorithm;

and 4.3, updating the speed and the position of the particles:

wherein, ITER represents a second ITER iteration process of the improved particle swarm optimization algorithm, ITER belongs to {1, 2.,. ITER,. and ITER }, q represents a q-th particle in the second ITER iteration process, and q belongs to {1, 2.,. u.,. and Z }; r is_iter,qThe r group of arrangement combinations of the strip imaging numbers corresponding to the q particles in the iter iteration process of the improved particle swarm optimization algorithm are represented; drt_r,k,iter,qThe strip imaging number sequence S _ ind of the r-th group arrangement combination of the strip imaging numbers corresponding to the q-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,qImaging direction corresponding to the number of the kth strip; t _ nrm_r,n,iter,qRepresenting the strip endpoint imaging sequence SP _ srt corresponding to the r-th group permutation and combination r of the strip imaging number corresponding to the q-th particle in the iter iteration process of the improved particle swarm optimization algorithm_r,iter,qNormalizing the coefficient of the imaging time of the nth strip endpoint;

P_losex(iter+1)＝P_losex(iter)+v(iter+1)

wherein w represents the inertial weight and has a value range of [0,1 ]]，c₁，c₂Represents the acceleration factor, v (iter), v (iter +1) being the current and new velocities of the particles, P_winx(iter),P_losex (iter) represents the current position of the superior particle and the current position of the inferior particle, P_losex (iter +1) represents the new position of the inferior particle, P_centerx (iter) represents the current average position of all particles;

Wherein best represents the particles with the optimal fitness value in the particle group in the iter iteration process of the improved particle swarm optimization algorithm, Fit_iter,bestRepresenting the fitness value corresponding to the best particle best in the iter iteration process of the improved particle swarm optimization algorithm; fit_iter-1,bestRepresenting the fitness value corresponding to the best particle best in the iter-1 iteration process of the improved particle swarm optimization algorithm; fit_*,bestMeans for improving the particle swarm optimization algorithm when the iteration termination condition is reachedAnd in the secondary iteration process, the fitness value corresponding to the optimal particle best is the optimal imaging task scheme.