CN112149911B

CN112149911B - Ultra-agile satellite same-orbit multipoint target in-motion imaging task planning method

Info

Publication number: CN112149911B
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: 2024-04-02
Anticipated expiration: 2040-09-30
Also published as: CN112149911A

Abstract

The invention provides a planning method for imaging tasks in motion of a same-orbit multipoint target of a hypersensitive satellite. The method comprises the steps of carrying out pretreatment of multipoint target clustering and non-trace-along strip decomposition, and solving imaging time windows of all strip 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 decision variables according to the strip imaging number sequence, the strip imaging direction sequence and the strip endpoint imaging time sequence, constructing an objective function by maximizing imaging coverage benefit and minimizing task completion time, constructing constraint conditions by an imaging time window and gesture conversion time, and further constructing an in-motion imaging task planning mathematical model of the same-track multipoint target; and optimizing by improving a particle swarm algorithm to obtain an optimal imaging task planning scheme. The method can fully utilize the attitude maneuver capability of the agile satellite and realize the optimization of the imaging task scheme in the same orbit multipoint target motion.

Description

Ultra-agile 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 a super-agile satellite co-orbit multipoint target motion.

Background

Agile satellites (Agile EOS) refer to satellites with payloads fixed to the satellite that rely on a attitude control system to control the entire satellite to rotate in 3 axes of pitch, roll, yaw. The agile satellite can perform gesture maneuver in a gap for executing a plurality of imaging tasks, push-broom imaging is performed after the gesture is stable, imaging modes such as side swing imaging, pitching imaging, single-track three-dimensional, single-track strip splicing, co-track multipoint targets and the like can be realized, and compared with a non-agile satellite capable of imaging only the points under the satellites, the observation efficiency is greatly improved. In recent years, along with the development of satellite attitude control technology and sensor imaging technology, agile satellites have "super" agile in-motion imaging capability, that is, imaging of ground targets is completed simultaneously in the attitude maneuver process, such as french pliades satellites and high-resolution multimode comprehensive imaging satellites. Besides the imaging mode of the agile satellite, the imaging agile satellite in motion can also realize the push-broom imaging capability of any track, and under the condition of single passing, the observation of more targets is realized, so that the working efficiency of the earth observation satellite is further improved.

In-orbit multi-point target imaging is a typical working mode of agile satellites, namely, imaging a plurality of point targets distributed adjacently on a satellite-based point track is completed in one transit. In the process of multi-point target imaging, the traditional agile satellite realizes gesture switching by using gesture mechanical capability, and realizes push-broom imaging of a plurality of point targets in a push-broom mode parallel to the understar point track. However, since the satellite performs "steady-state" push-broom on each target in the imaging process, the ground imaging cannot be completed simultaneously in the attitude maneuver process, the speed of imaging the ground push-broom is determined by the speed of satellite orbit motion, the maximum push-broom range is constrained by the "hard" of the speed of motion of the points below the satellite, and when the number of the point targets is large, the imaging of all the point targets cannot be ensured; the speed of ground push-broom imaging can be greatly improved through attitude maneuver due to the mobility and imaging capability of the imaging agile satellite in motion, the maximum push-broom range can break through the limit of the movement speed of the points below the satellite, more point targets can be imaged through single transit in motion, and the working efficiency of the satellite is greatly improved.

The appearance of an imaging mode in hypersensitive movement provides a new challenge for the planning of the same-track multi-point target imaging task. Firstly, the limitation of stable posture in the imaging process is broken through, so that the imaging time of a single push-broom fixed-length ground target is not fixed at the initial moment, and the posture constraint is more complex than that of a traditional agile satellite; secondly, the non-tracking characteristic of imaging in motion ensures that the direction of the push-broom strip can not be parallel to the track of the point under the satellite, and the task decomposition mode is richer than that of the traditional agile satellite. Therefore, a new feature based on the hypersensitive satellite is needed, and an imaging mission plan of the hypersensitive satellite is formulated pertinently according to the observation requirement of a user.

Aiming at the requirements of imaging tasks in the same orbit and multipoint targets of the hypersensitive satellites and the defects of the existing method, a new technical scheme is needed to be provided in the field.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides a planning method for imaging tasks in the same orbit and multiple points of targets in motion of a hypersensitive satellite.

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

step 1, inputting longitude and latitude coordinates of a plurality of point targets, converting the longitude and latitude coordinates of the plurality of point targets into plane coordinates of the plurality of point targets, dividing the plane coordinates of the plurality of point targets into a plurality of point target clusters through a density-based clustering method, constructing a minimum circumscribed rectangular coordinate point set of the clusters through a rotating clamping method by the point target clusters, carrying out strip decomposition on the minimum circumscribed 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 numbering sequence and a strip imaging direction sequence, calculating a strip endpoint imaging sequence according to the strip imaging numbering sequence and the strip imaging direction sequence, and calculating a strip endpoint imaging time sequence by utilizing an imaging time normalization coefficient in combination with the strip endpoint imaging sequence;

Step 3, constructing decision variables according to the strip imaging numbering sequence, the strip imaging direction sequence and the strip endpoint imaging time sequence, constructing an objective function through imaging coverage gain maximization and task completion time minimization, constructing constraint conditions through imaging time windows and gesture conversion time, and further constructing an in-motion imaging task planning mathematical model of the same-track multipoint target;

step 4, combining with a simultaneous multipoint target imaging task planning mathematical model, and optimizing by improving a particle swarm algorithm to obtain an imaging task planning scheme with the largest observation income and the shortest task completion time;

preferably, in the step 1, longitude and latitude coordinates of a plurality of point targets are input, and the conversion of the longitude and latitude coordinates of the plurality of point targets into plane coordinates of the plurality of point targets is as follows:

assume that the multiple point target set is p= { P ₁ ,P ₂ ,…,P _i ,…,P _np The longitude and latitude coordinates corresponding to each point target are { (B) ₁ ,L ₁ ),(B ₂ ,L ₂ ),...,(B _i ,L _i ),...,(B _np ,L _np )}

Where np is the number of point targets, P _i For the ith point target, (B) _i ,L _i ) Is the longitude and latitude coordinates of the ith point target, B _i For the latitude of the ith point target, L _i Longitude for the ith point target, i e [1, np]；

Converting longitude and latitude coordinates of a plurality of point targets into plane coordinates of the plurality of point targets by using a Gaussian projection forward calculation formula: { (x) ₁ ,y ₁ ),(x ₂ ,y ₂ ),...,(x _i ,y _i ),...,(x _np ,y _np )}

Where np is the number of point targets, (x) _i ,y _i ) Plane coordinates, x, of the ith point target _i X-axis coordinates, y for the ith point target _i For the Y-axis coordinates of the ith point target, i ε [1, np ]]；

In the step 1, the dividing the plane coordinates of the plurality of 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 a multi-point target into plane coordinates, clustering the point targets by adopting a density-based clustering method, namely a DBSCAN method;

the clustering radius is calculated by the satellite orbit height h, the earth radius R and the half field angle V as follows:

wherein eps is a cluster radius, h is a satellite orbit height, R is an earth radius, V is a half field angle, and a density threshold value 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+1 The plane distance between the ith point target and the (i+1) th point target, namely, two adjacent point targets in plane coordinates;

taking any point target in P as a circle center, eps as a radius, and disco in plane coordinates of a plurality of point targets in the circle _i,i+1 Clustering all point targets with the points in the circle of which the numbers are less than or equal to eps and the points in the circle of which the numbers are more than or equal to 2, and gradually expanding the clustered clusters to obtain final point target clusters;

The point target cluster is as follows:

wherein,indicating that c is included in the first cluster _l The number of point targets, e, represents the number of point target clusters, l E [1, e]，c _l Representing the number of point targets in the first cluster, and c ₁ +c ₂ +...+c _l +...+c _e =np, ensuring that e clusters formed by clustering point targets contain all point targets;

in the step 1, the minimum circumscribed rectangular coordinate point set of the point target cluster constructed by the rotating shell clamping method is as follows:

based on the rotation clamping idea, constructing an external rectangle with the minimum area of the cluster;

construction by using the idea of rotating the cartridgeArea of (2)The minimum circumscribed rectangle, the minimum circumscribed rectangle vertex coordinate set of the obtained cluster is: { Point _l ^a ,Point _l ^b ,Point _l ^c ,Point _l ^d }；

Wherein, point _l ^a Point representing the 1 st vertex coordinate of the smallest bounding rectangle in the l cluster _l ^a ＝(x _l ^a ,y _l ^a )，Point _l ^b Point, which represents the 2 nd vertex coordinates of the smallest bounding rectangle in the l cluster _l ^b ＝(x _l ^b ,y _l ^b )，Point _l ^c 3 rd vertex coordinates Point representing the smallest bounding rectangle in the first cluster _l ^c ＝(x _l ^c ,y _l ^c )，Point _l ^d 4 th vertex coordinates Point representing the smallest bounding rectangle in the first cluster _l ^d ＝(x _l ^d ,y _l ^d ) The minimum circumscribed rectangle vertex coordinate set of the obtained cluster comprises the cluster;

step 1, a stripe set, a stripe endpoint set and a stripe endpoint longitude and latitude coordinate set are obtained by a stripe decomposition method through a minimum circumscribed rectangular coordinate point set of the cluster, wherein the stripe set, the stripe endpoint set and the stripe endpoint longitude and latitude coordinate set are as follows:

Four vertex coordinates { Point ] combined with minimum bounding rectangle _l ^a ,Point _l ^b ,Point _l ^c ,Point _l ^d Then dividing the circumscribed rectangle along long sides or along short sides according to fixed imaging breadth, namely eps, so as to obtain four vertex coordinate sets of m strips in the cluster l, wherein the four vertex coordinate sets are 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 ,Point _l,j ^c ,Point _l,j ^d ；...,；Point _l,m ^a ,Point _l,m ^b ,Point _l,m ^c ,Point _l,m ^d }；

the four vertex coordinate sets of m strips obtained through the division of the strips comprise the vertex coordinate set of the minimum circumscribed rectangle of the cluster;

wherein m represents the number of stripes, j represents the j-th stripe, 1.ltoreq.j.ltoreq.m, point _l,j ^a Point representing the 1 st vertex coordinate of the j-th stripe in the l-th cluster _l,j ^a ＝(x _l,j ^a ,y _l,j ^a )，Point _l,j ^b Point representing the 2 nd vertex coordinates of the jth stripe in the ith cluster _l,j ^b ＝(x _l,j ^b ,y _l,j ^b )，Point _l,j ^c Point representing the 3 rd vertex coordinates of the jth stripe in the ith cluster _l,j ^c ＝(x _l,j ^c ,y _l,j ^c )，Point _l,j ^d Point representing the 4 th vertex coordinate of the j-th stripe in the l-th cluster _l,j ^d ＝(x _l,j ^d ,y _l,j ^d )；

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

taking the midpoints of the 1 st and 2 nd vertex coordinates and the midpoints of the 3 rd and 4 th vertex coordinates of the strip as strip endpoints, respectively numbering 2*m strip endpoints in a cluster l as different continuous positive integers of [1,2 x m ], wherein the obtained strip endpoint numbering set is {1,2, …, k, …,2*m }, k represents the strip endpoint number, and k is more than or equal to 1 and less than or equal to 2*m; the set of stripe endpoint coordinates corresponding to 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 stripe, are:

Wherein m represents the number of stripesJ represents a band number, j is 1.ltoreq.j.ltoreq.m,the plane coordinates of the two strip end points of the j-th strip in the cluster l are respectively +.>

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

after the strip endpoint coordinate set corresponding to the cluster l is obtained, converting the strip endpoint coordinate set into a strip endpoint longitude and latitude coordinate set SendPoint by using a Gaussian projection back calculation formula _l Expressed as:

SendPoint _l ＝{endPoint _l,1 ,endPoint _l,2 ,...,endPoint _l,M ,...,endPoint _l,2*m }

wherein M represents the number of stripes, 2*m represents the number of stripe end points, M is not less than 1 and not more than 2, and endPoint is provided _l,M The longitude and latitude coordinates of the endpoint of the M-th strip in the cluster l are obtained;

after each cluster is striped, numbering all the stripes, and the stripe numbering sets S, s= {1,2, …, k, …, ns };

wherein k represents the number of stripes, k is not less than 1 and not more than ns, and ns represents the number of the final stripes;

SP represents the final set of stripe endpoint numbers, sp= {1,2,..n,..2, 2 x ns }, where n represents the stripe endpoint number, 1.ltoreq.n.ltoreq.2 x ns,2 x ns represents the final number of stripe endpoints; all stripe endPoint longitude and latitude coordinate set, sendpoint= { endPoint = { endPoint ₁ ,endPoint ₂ ,...,endPoint _n ,...,endPoint _2*ns }, wherein (EndPoint) _n And the longitude and latitude coordinates corresponding to the strip end points with the strip end point number of n are represented, wherein n is more than or equal to 1 and less than or equal to 2 x ns.

The imaging time window for all the strip endpoints is obtained in the step 1:

according to the strip endpoint number set SP and the corresponding strip endpoint longitude and latitude coordinate set SendPoint, a characteristic cone method (Shen Xin, li Deren, yao Huang) is adopted, an optical remote sensing satellite imaging window rapid forecasting method [ J ] facing imaging task planning]University of martial arts newspaper (information science edition), 2012,37 (12): 1468-1471) determining each of the strip end point imaging time windows to obtain each of the strip end points in the strip end point number set SP as an imaging time window [ TW ] _n,s ,TW _n,e ]；

Wherein [ TW ] _n,s ,TW _n,e ]TW representing the original imaging time window corresponding to the band endpoint with the band endpoint number n _n,s TW is the time window starting time _n,e The time window is the ending moment of the time window;

preferably, the imaging numbering sequence of the construction band in step 2 is:

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 _r Represents the sequence of the imaging numbers of the strips, r represents the r group permutation and combination of the imaging numbers of the strips, ns represents the number of the strips, ns-! Representing the number of all permutations and combinations of the strip imaging numbers

k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth stripe, s_ind _r,k Strip imaging number sequence S_ind representing the r-th permutation and combination _r Number of kth band, s_ind _r,k S, S represents the stripe number set in the step 1;

wherein S_ind _r The method can be determined according to the r group permutation combination r of the strip imaging number and the strip number set S obtained in the step 1 by adopting a recursion calling method;

the imaging direction sequence of the construction strip in the step 2 is as follows:

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Representing the number of all permutations and combinations of the strip imaging numbers

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

k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band, drt _r,k Strip imaging number sequence S_ind representing the r-th permutation and combination _r In the imaging direction corresponding to the number of the kth stripe, when drt _r,k When=0, the imaging number sequence s_ind of the stripe representing the r-th permutation and combination _r In the imaging direction of the kth stripe is forward push-broom, when drt _r,k When=1, the imaging number sequence s_ind of the stripe representing the r-th permutation and combination _r The imaging direction of the kth strip is reverse push-broom;

the imaging sequence of the calculated band endpoint in the step 2 is as follows:

First, a sequence S_ind of the strip imaging number is determined according to an r-th group of permutation and combination r of the strip imaging number _r After that, S_ind _r Each band corresponds to two band end points, and it cannot be determined that the band end points are imaged first, so that the band imaging number sequence S_ind combined according to the r group arrangement is needed _r Imaging direction d of each stripe in (a) _r,k Determining S_ind _r The imaging direction of each strip in (a) to thereby ultimately determine the strip endpoint 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}

Therein, SP_ srt _r Stripe corresponding to the r-th group permutation r of the imaging number of the expression stripeWith endpoint imaging sequence, r denotes the r group permutation of the stripe imaging numbers, ns denotes the stripe number, ns ≡! Representing the number of all permutation and combination of the strip imaging numbers; k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band in (a); ST denotes the start imaging strip endpoint of the strip, ET denotes the end imaging strip endpoint of the strip;

wherein s_ srt _r,k,ST Sum s_ srt _r,k,ET Strip number imaging sequence S_ind corresponding to the r-th group permutation and combination r for representing strip imaging number _r Stripe endpoint pair of kth stripe in (c), s_ srt _r,k,ST Strip imaging number sequence S_ind corresponding to the r-th group permutation r of the strip imaging numbers _r Number of the starting imaging band end point of the kth band, s_ srt _r,k,ET Strip imaging sequence S_ind corresponding to r-th group permutation and combination r for representing strip imaging number _r The number of the end imaging stripe end point of the kth stripe;

if there are ns stripes, the number of stripe endpoints is 2 x ns, so the stripe endpoint 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}

therein, SP_ srt _r The imaging sequence of the strip endpoint corresponding to the r group of permutation and combination r of the strip imaging number is represented, r represents the r group of permutation and combination of the strip imaging number, ns represents the strip number, 2 x ns represents the strip endpoint number, ns-! Representing the number of all permutation and combination of the strip imaging numbers; k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band in (a); n represents SP_ srt _r Sp_ srr of the nth band end point _r,n Represent SP_ srt _r Numbering of the nth band end point, sp_ srt _r,n The SP represents the band endpoint number set in the step 1;

and 2, calculating a strip endpoint imaging time sequence as follows:

the imaging moment normalization coefficient is as follows:

t_nrm _r,n ∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

wherein t_ nrm _r,n The imaging sequence SP_ srt of the strip endpoint corresponding to the r-th group of permutation and combination r representing the imaging number of the strip _r The imaging time normalization coefficient of the nth band endpoint in (a), n representing the band endpoint observation sequence SP_ srt _r An nth stripe endpoint of (a);

wherein t is _r,n Normalizing the coefficient t_ nrm according to the imaging moment _r,n Calculating to obtain;

imaging sequence SP_ srt at the calculated band end point _r Imaging time t corresponding to each strip end point _r,n When it is, it includes: cutting an imaging time window and normalizing imaging time;

the imaging time window is cut:

the requirements for clipping imaging time windows of two adjacent strip endpoints are:

the starting time of the imaging time window of the end point observed later is not earlier than the starting time of the imaging time window of the end point observed earlier, and the ending time of the imaging time window of the end point observed earlier is not later than the ending time of the imaging time window of the end point observed later, specifically:

if T _r,n,s ＞T _r,n+1,s ITW is then _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,e ITW is then _r,n+1,e ＝T _r,n+1,e ,ITW _r,n+1,e ＝T _r,n+1,e

In the above, T _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of a time window corresponding to the nth observed strip endpoint; t (T) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r In the time window junction corresponding to the nth observed strip endpointA bundle time; ITW (International telecommunication System) _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of the cut time window corresponding to the nth observed strip endpoint; ITW (International telecommunication System) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The end time of the time window after cutting corresponding to the nth observed strip endpoint;

normalizing the imaging moment:

imaging sequence at the end of the band, SP_ srt _r Between any two adjacent strip endpoints, the imaging moment of the observed point can be restrained on the feasible range of the observed point, and a normalization coefficient is introduced to normalize the imaging moment, specifically:

if n=1, then t _r,1 ＝ITW _r,1,s +t_nrm _r,n *(ITW _r,1,e -ITW _r,1,s )

If n is not equal to 1

If t _r,n ＞ITW _r,n+1,s Then t _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,s Then t _r,n+1 ＝ITW _r,n+1,s +t_nrm _r,n *(ITW _r,n+1,e -ITW _r,n+1,s )

Wherein t is _r,1 Imaging sequence sp_ srt for a stripe endpoint _r In the 1 st observed band endpoint, imaging time t of the time window after clipping _r,n Imaging sequence sp_ srt for a stripe endpoint _r In the imaging time, t, of the nth observed strip endpoint _r,n+1 Imaging sequence sp_ srt for a stripe endpoint _r In the imaging instant of the (n+1) th observed band endpoint, ITW _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of the cut time window corresponding to the nth observed strip endpoint; ITW (International telecommunication System) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The end time of the time window after cutting corresponding to the nth observed strip endpoint;

preferably, in the step 3, the decision variables are constructed according to the sequence of the imaging number of the strip, the sequence of the imaging direction of the strip and the sequence of the imaging time of the end point of the strip:

The band imaging numbering sequence is:

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 _r Represents the sequence of the imaging numbers of the strips, r represents the r group permutation and combination of the imaging numbers of the strips, ns represents the number of the strips, ns-! Represents the number of all permutation and combination of the strip imaging number, and k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth stripe, s_ind _r,k Strip imaging number sequence S_ind representing the r-th permutation and combination _r Number of kth band, s_ind _r,k S, S represents the stripe number set in the step 1;

the sequence of the imaging directions of the strips is as follows:

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the strip imaging number, and k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band, drt _r,k Strip imaging number sequence S_ind representing the r-th permutation and combination _r In the imaging direction corresponding to the number of the kth stripe, when drt _r,k When=0, the imaging number sequence s_ind of the stripe representing the r-th permutation and combination _r In the imaging direction of the kth stripe is forward push-broom, when drt _r,k When=1, the imaging number sequence s_ind of the stripe representing the r-th permutation and combination _r The imaging direction of the kth strip is reverse push-broom;

The imaging time sequence of the strip endpoint is as follows:

t _r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end point, t _r,n The imaging sequence SP_ srt of the strip endpoint corresponding to the r-th group of permutation and combination r representing the imaging number of the strip _r Imaging moment of the nth band endpoint;

the constructed decision variables are:

r∈{1,2,...,ns！}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Representing the number of all permutation combinations of the strip imaging number, determining a strip imaging number sequence S_ind from the step 2 according to the r group permutation combination r of the strip imaging number _r ；

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

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

k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band, drt _r,k Strip imaging number sequence S_ind representing the r-th permutation and combination _r Imaging direction corresponding to the number of the kth stripe in the image data

t_nrm _r,n ∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

Wherein t_ nrm _r,n The imaging sequence SP_ srt of the strip endpoint corresponding to the r-th group of permutation and combination r representing the imaging number of the strip _r Normalized coefficient of imaging time of nth stripe end pointN represents the strip endpoint observation sequence sp_ srt _r An nth stripe endpoint of (a);

therein, SP_ srt _r The imaging sequence of the strip endpoint corresponding to the r group of permutation and combination r of the imaging number of the representing strip, SP_ srt _r The sequence S_ind of the strip imaging number can be determined according to the r group of permutation and combination r of the strip imaging number _r And a stripe imaging direction sequence drt _r,k Determined by step 2;

the imaging coverage benefit maximization in step 3 is as follows:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

representing the number of imaging coverage benefits, namely observation point targets, to be maximized;

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the strip imaging number, and k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band, c _r,k Representing S_ind _r Whether the kth band of (c) can complete imaging, c _r,k E {0,1}, if c _r,k =1, then s_ind _r The k-th strip of the plurality can complete imaging, otherwise, imaging can not be achieved; g _r,k Representing S_ind _r The number of points covered by the kth stripe;

c _r,k the judging basis of (a) comprises two aspects:

the time from the end point of the k-1 stripe to the start point of the k stripe meets the gesture conversion time constraint;

The time from the start point of the k strips to the end point thereof satisfies the gesture 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 ) C is _r,k ＝1

Otherwise c _r,k ＝0

And 3, the minimum task completion time is an objective function, which is as follows:

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

representing minimizing imaging task completion time;

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end, Δt _r,n Representing a strip endpoint imaging sequence SP_ srt _r At the observation time interval from the n-1 th band end point to the n-th band end point _r,n ＝t _r,n -t _r,n-1 ，t _r,n Imaging sequence sp_ srt for a stripe endpoint _r Imaging time of the upper nth band endpoint;

wherein c_tr _r,n Characterization of the stripe endpoint imaging sequence sp_ srt _r Whether the gesture conversion time constraint is satisfied from the n-1 th stripe end point to the n-th stripe end point, namely:

if delta t _r,n ≥t_tr _r,n Then c_tr _r,n ＝1

Otherwise c_tr _r,n ＝0

The imaging time window constraint in the step 3 is as follows:

t _r,n ∈[ITW _r,n-s ,ITW _r,n-e ]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! All permutation and combination for representing imaging numbers of stripN represents the number of the strip endpoint observation sequence sp_ srt _r The nth band end point, t _r,n ∈[ITW _r,n-s ,ITW _r,n-e ]Representing a strip endpoint imaging sequence SP_ srt _r The imaging instant of each of the strip endpoints must meet the cropped imaging time window constraint,

and 3, the gesture conversion time constraint is as follows:

Δt _r,n ≥t_tr _r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end, Δt _r,n ≥t_tr _r,n Representing a strip endpoint imaging sequence SP_ srt _r The imaging time interval between two adjacent strip endpoints needs to meet the gesture conversion time constraint between the two adjacent endpoints;

calculating the gesture conversion time between the two adjacent endpoints by adopting a continuous gesture adjustment method:

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth stripe endpoint, t_tr _r,n Representing a strip endpoint imaging sequence SP_ srt _r From the n-1 th stripe endpoint to the n-th stripe endpoint,is the momentum of the gesture machine, delta omega, along the rolling axis from the n-1 th strip end point to the n-th strip end point _n,n-1 From the n-1 th band end pointAttitude machine momentum delta kappa along pitch axis between nth strip end point _n,n-1 The momentum of the attitude machine along the yaw axis from the n-1 th strip end point to the n-th strip end point, respectively,>for satellite roll motorized speed,/->Is satellite pitching axis maneuvering speed,/->The three-axis maneuvering speeds of the satellite are respectively;

and 3, constructing the same-track multipoint target in-motion imaging task planning mathematical model in the step 3 through decision variables, objective functions and constraint conditions.

Preferably, in step 4, the imaging task planning scheme with the largest observation benefit and the shortest task completion time obtained by optimizing the improved particle swarm algorithm is as follows:

step 4.1, randomly generating an initial population:

generating initial populations with the number 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 constructed as described in step 3, i.e. r _iter,u 、drt _r,k,iter,u 、t_nrm _r,n,iter,u The initial value of each particle is a random value in the respective value range, and the initial speed of the particle is set to be zero;

wherein ITER represents an ith iteration process for improving the particle swarm optimization algorithm, and ITER epsilon {1,2,.. The term "ITER,. The term" ITER }, u represents the ith particle in the ith iteration process, and u epsilon {1,2,. The term "u,. The term" Z }; r is (r) _iter,u An r group permutation and combination of the strip imaging numbers corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,u Band imaging of the r-th group of permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithmNumbering sequence S_ind _r The imaging direction corresponding to the number of the kth strip; t_ nrm _r,n,iter,u Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u The imaging moment normalization coefficient of the nth strip endpoint;

therein, SP_ srt _r,iter,u The imaging sequence of the strip endpoint corresponding to the r group of permutation and combination r of the strip imaging number corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is represented;

step 4.2, calculating the observation benefits and task completion time of each particle in the current population, and obtaining the particle with the optimal particle fitness value in the current population, wherein the fitness value consists of the observation benefits and the task completion time;

in the calculation of the current population, the observed benefits and task completion time of each particle are as follows:

According to r _iter,u 、drt _r,k,iter,u 、t_nrm _r,n,iter,u Restoring a strip endpoint imaging sequence SP_ srt corresponding to the ith particle in the ith iteration process of the improved particle swarm optimization algorithm by utilizing the step 2 _r,iter,u And imaging time t corresponding to each strip endpoint _r,n,iter,u The method comprises the steps of carrying out a first treatment on the surface of the With the transition time t_tr of the posture between two adjacent strip end points _r,n,iter,u Obtaining a strip endpoint observation sequence capable of completing push-broom for judgment basis, and obtaining observation benefits and task completion time;

wherein ITER represents an ith iteration process for improving the particle swarm optimization algorithm, and ITER epsilon {1,2,.. The term "ITER,. The term" ITER }, u represents the ith particle in the ith iteration process, and u epsilon {1,2,. The term "u,. The term" Z }; r is (r) _iter,u An r group permutation and combination of the strip imaging numbers corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,u Band imaging number sequence S_ind representing the r-th group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u Numbering of the kth band in (b)Corresponding imaging direction, S_ind _r,iter,u Band imaging number sequence of the r group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm, t_ nrm _r,n,iter,u Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u The imaging time normalization coefficient of the nth band endpoint of (a), SP_ srt _r,iter,u An r-th group permutation and combination r-corresponding strip endpoint imaging sequence of a strip imaging number corresponding to a u-th particle in an ith iteration process of an improved particle swarm optimization algorithm is represented, and t _r,n,iter,u The imaging moment of the nth strip endpoint in the strip endpoint imaging sequence corresponding to the r group permutation and combination r of the strip imaging number corresponding to the ith particle in the ith iteration process of the improved particle swarm optimization algorithm is represented;

the particles with optimal particle fitness values (observation benefits and task completion time) in the current population are:

sequencing the fitness value of each particle in the current population to obtain a particle best with the optimal fitness value in the particle population, namely comparing the observation benefits of all the particles, taking the particle with the largest observation benefit as the optimal particle, and if the situation that the observation benefits are the same occurs, taking the particle with the least task completion time as the particle with the optimal fitness value of the current population;

wherein best represents particles with optimal fitness value in particle swarm in the ith iteration process of improved particle swarm optimization algorithm, and Fit _iter,best Representing the fitness value corresponding to the optimal particle best in the ith iteration process of the improved particle swarm optimization algorithm;

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

randomly selecting two particles o and q, comparing the observation benefits of the two particles with the task completion time, wherein the particle with the largest observation benefit in the two particles is the better particle, the particle with the smallest observation benefit is the inferior particle, and if the situation that the observation benefits are the same occurs, the particle with the smallest task completion time is the better particle;

if the preferred particle is o and the inferior particle is q, the preferred particle o is P _win And the average position P of all particles _center As inferior particles P _lose Updating the inferior particles P in the direction of evolution of the current position _lose Speed and position of (2);

wherein r is _iter,q And t_ nrm _r,n,iter,q Particle position and velocity update strategy with real PSO, drt _r,k,iter,q A particle position and speed updating strategy of binary PSO is adopted;

wherein, ITER represents an ith iteration process for improving the particle swarm optimization algorithm, and ITER is {1,2,.. The term "ITER, & gt, ITER }, q represents the q-th particle in the ith iteration process, and q is {1,2,. & gt, u, & gt, Z }; r is (r) _iter,q An (r) group permutation and combination of the strip imaging numbers corresponding to the (q) th particle in the (er) th iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,q Band imaging number sequence S_ind representing the r-th group permutation and combination of band imaging numbers corresponding to the q-th particles in the ith iteration process of improved particle swarm optimization algorithm _r,iter,q The imaging direction corresponding to the number of the kth strip; t_ nrm _r,n,iter,q Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the q-th particles in the ith iteration process of improved particle swarm optimization algorithm _r,iter,q The imaging moment normalization coefficient of the nth strip endpoint;

wherein r is _iter,q And t_ nrm _r,n,iter,q Particle location and velocity update strategy using real PSO is improved PSO particle location and velocity update strategy:

v(iter+1)＝wv(iter)+c ₁ *(P _win x(iter)-P _lose x(iter))+c ₂ *(P _center x(iter)-P _lose x(iter))

P _lose x(iter+1)＝P _lose x(iter)+v(iter+1)

wherein w represents inertial weight, and the value range is [0,1]，c ₁ ，c ₂ Represents an acceleration factor, v (iter), v (iter+1) are particles, respectivelyCurrent speed and new speed, P _win x(iter),P _lose x (iter) represents the current position of the preferred particle and the current position of the inferior particle, P _lose x (iter+1) represents the new position of the inferior particle, P _center x (iter) represents the current average position of all particles.

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

repeating the optimization iteration step 4.2 and the optimization iteration step 4.3 for a plurality of times, and comparing the fitness value corresponding to the optimal particle in the previous iteration process and the current iteration process, namely comparing Fit _iter-1,best And Fit _iter,best If Fit _iter-1,best Is better than Fit _iter,best If Fit, the fitness value corresponding to the optimal particle in the previous iteration process is a better imaging task scheme _iter,best Is better than Fit _iter-1,best The fitness value corresponding to the optimal particle in the current iteration process is a better imaging task scheme;

repeating the optimization iteration until the iteration termination condition so as to obtain particles with optimal particle swarm fitness values, wherein the corresponding imaging scheme is the optimal imaging task scheme Fit _*,best ；

Wherein best represents particles with optimal fitness value in particle swarm in the ith iteration process of improved particle swarm optimization algorithm, and Fit _iter,best Representing the fitness value corresponding to the optimal particle best in the ith iteration process of the improved particle swarm optimization algorithm; fit _iter-1,best Representing the fitness value corresponding to the optimal particle best in the ith-1 iteration process of the improved particle swarm optimization algorithm; fit _*,best And 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:

the r group of imaging numbers of the strips is used for arranging and combining r and the imaging direction drt of the strips _r,k Normalized coefficient t_ nrm of imaging moment _r,n Avoidance as decision variableWhen the imaging moment is not directly used as an optimization model decision variable, the evaluation and calculation process of the imaging scheme is complex, and the optimization solving efficiency is low; constructing a task planning model with maximum imaging coverage benefit and minimum task completion time as optimization targets, and realizing benefit priority and efficiency; further, the improved PSO algorithm is utilized to carry out optimization solution on the decision variables, so that the precocity of the standard PSO is overcome, and the requirement of the binary decision variable optimization solution can be met. The method can fully utilize the attitude maneuver capability 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 characterized by comprising the steps of observing a satellite imaging stripe decomposition mode for the traditional agility to the ground;

fig. 2: a satellite imaging strip decomposition mode is observed for hypersensitive earth;

fig. 3: the method is a schematic diagram of the point target in the embodiment of the invention;

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

fig. 5: the method is a schematic diagram of the band decomposition result in the embodiment of the invention;

fig. 6: the push-broom direction in the strip is numbered schematically;

fig. 7: imaging a time window overlapping schematic diagram for the embodiment of the invention;

Fig. 8: clipping a schematic diagram for an imaging time window according to the embodiment of the invention;

fig. 9: the method is a schematic diagram for restraining the front and rear imaging time;

fig. 10: the method is a schematic diagram of a decision variable reduction imaging moment flow in the embodiment of the invention;

fig. 11: imaging a schematic diagram of a band atomic task set for an embodiment of the invention;

fig. 12: the method is a schematic diagram of a decision variable reduction imaging moment process in the embodiment of the invention;

fig. 13: schematic diagram of a push-broom sequence of a strip endpoint in the embodiment of the invention;

fig. 14: imaging time windows and normalized imaging time diagrams cut for embodiments of the present invention;

fig. 15: the method is a schematic diagram of the relationship between the point targets and the undersea point tracks in the embodiment of the invention;

fig. 16: the embodiment of the invention points to a target pretreatment result schematic diagram;

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

fig. 18: the step flow chart of the embodiment of the 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 that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

The following describes a specific embodiment of the present invention with reference to fig. 1 to 18 as a method for planning an imaging task in motion of a multi-point target in a same orbit of a hypersensitive satellite, comprising the following steps:

in the step 1, longitude and latitude coordinates of a plurality of point targets are input, and the longitude and latitude coordinates of the plurality of point targets are converted into plane coordinates of the plurality of point targets as follows:

as shown in fig. 1, the stripe push direction of the conventional agile satellite must be parallel to the satellite's understar trajectory, so 2 stripes are required to achieve coverage of all point targets. For a "super" agile satellite, with imaging capability in motion, the push-broom direction may not be parallel to the sub-satellite point trajectory, as shown in fig. 2, and one stripe may be implemented to cover all point targets.

i∈[1,np]

the point target cluster is as follows:

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

Generating by clusteringIs a cluster of (a)For example, a non-trace strip decomposition process based on a rotating stuck shell is described:

construction by using the idea of rotating the cartridgeThe minimum circumscribed rectangle of the area of (2) and the obtained minimum circumscribed rectangle vertex coordinate set of the cluster is as follows: { Point _l ^a ,Point _l ^b ,Point _l ^c ,Point _l ^d }；

Wherein m represents the number of stripes, j represents the number of stripes, 1.ltoreq.j.ltoreq.m,the plane coordinates of the two strip end points of the j-th strip in the cluster l are respectively +.>

As shown in fig. 3-5, examples of point target preprocessing via DBSCAN clustering and rotational stuck-shell non-trace strip decomposition are presented. Fig. 3 is a point target set p= { P ₁ ,p ₂ ,…,p ₁₇ -a }; FIG. 4 is a clustering result, wherein the target point set is clustered into 5 clusters; fig. 5 shows the stripe set s= { a, B, C, D, E, F }, stripe endpoint set sp= { a, B, C, D, E, F, g, h, i, j, k, l }, which are formed after preprocessing.

The imaging time window for all the strip endpoints is obtained in the step 1:

imaging time window forecasting, namely determining the starting point and the end point of a time period in which a satellite can image each target in a future time period, is the basis of agile satellite imaging mission planning. The target can be observed in the optical imaging satellite orbit motion process, and the real-time maximum observable range is an area with 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 the maximum observation range is related to the maximum attitude maneuver angle and the field angle of the agile satellite. For a target, the time period in which the target is continuously observed is the time window of the target. In view of the fact that most of the currently mainstream agile remote sensing satellites adopt linear array sensors, the maximum observable range of the satellites can be described by adopting characteristic cones, when the imaging time window of each strip endpoint is calculated, papers (Shen Xin, li Deren, yao Huang) can be referred to, an imaging task planning-oriented optical remote sensing satellite imaging window rapid forecasting method [ J ]. University of armed university (information science edition), 2012,37 (12): 1468-1471), and the method is not repeated.

According to the strip endpoint number set SP and the corresponding strip endpoint longitude and latitude coordinate set SendPoint, determining imaging time windows of all strip endpoints by adopting a characteristic cone method to obtain the imaging time windows of all strip endpoints in the strip endpoint number set SP as [ TW ] _n,s ,TW _n,e ]；

the imaging numbering sequence of the construction strip 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 imaging number sequence of the strip is S_ind _r R is determined by the r-th group permutation r of the stripe imaging numbers, r is {1,2, …, ns! Between }Each positive integer corresponds to a set of sequences of stripe imaging numbers. Taking 4 imaging strips generated by decomposition as an example, the 4 imaging strips have different strip imaging numbering sequences in 24, as shown in table 2, the numbers 0 and 23 are completely reversed, and the preservation of the coding neighborhood structure in the evolutionary algorithm is facilitated.

TABLE 1 stripe imaging numbering sequence corresponding to the r th group permutation and combination r of stripe imaging numbering

The reduction of the band imaging numbering sequence S_ind from r is given below _r The pseudo code of (2) is as follows:

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

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

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

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Therein, SP_ srt _r The imaging sequence of the strip endpoint corresponding to the r-th group permutation and combination r of the imaging number of the strip is represented, r represents the r-th group permutation and combination of the imaging number of the strip, ns represents the number of the strips, ns-! Representing the number of all permutation and combination of the strip imaging numbers; k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band in (a); ST denotes the start imaging strip endpoint of the strip, ET denotes the end imaging strip endpoint of the strip;

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

if there are three imaging strips, the strip number set s= {1,2,3}, the strip endpoint number set sp= {1,2,3,4,5,6}, the two strip endpoints corresponding to the strip with number 1 are 1 and 2, the two strip endpoints corresponding to the strip with number 2 are 3 and 4, and the two strip endpoints corresponding to the strip with number 3 are 5 and 6. After the strip imaging sequence is determined according to the r-th group permutation r of the strip imaging numbers, if r=1, there is s_ind ₁ = {1,3,2}, in case the stripe imaging direction is not determined, the determined stripe endpoint imaging sequence is: {1or2, 5or6,3or 4}, the imaging sequence of the end point of the strip can be determined finally only if the imaging direction of the strip is determined, if drt in the imaging direction sequence of the strip _1,1 ＝0，drt _1,2 ＝1,drt _1,3 =1, then the final imaging sequence of the strip end point sp_ srt ₁ ＝{1,2,6,5,4,3}；

And 2, calculating a strip endpoint imaging time sequence as follows:

the imaging moment normalization coefficient is as follows:

t_nrm _r,n ∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the imaging time window is cut:

stripe endpoint imaging sequence sp_ srt _r Can be represented by r and drt _r,k It was determined that, however, due to the fact that the imaging time windows of the strip end points may overlap each other, as shown in fig. 7, the imaging time windows of the three strip end points observed in succession are respectively [ T ] _{r,1_s} ,T _{r,1_e} ]、[T _{r,2_s} ,T _{r,2_e} ]、[T _{r,3_s} ,T _{r,3_e} ]Obviously in SP_ srt _r In the case of determination, [ T ] _{r,3_e} ,T _{r,1_e} ]、[T _{r,3_e} ,T _{r,2_e} ]、[T _{r,3_s} ,T _{r,2_s} ]The three time periods are invalid time ranges. To avoid searching for dead space, the sequence SP_ srt is imaged according to the end point of the strip _r The imaging time windows for all the strip endpoints are cropped.

If T _r,n,e ＞T _r,n+1,e ITW is then _r,n+1,e ＝T _r,n+1,e ,ITW _r,n,e ＝T _r,n+1,e

In the above, T _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of a time window corresponding to the nth observed strip endpoint; t (T) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The time window end time corresponding to the nth observed strip endpoint; ITW (International telecommunication System) _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of the cut time window corresponding to the nth observed strip endpoint; ITW (International telecommunication System) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The end time of the time window after cutting corresponding to the nth observed strip endpoint;

according to the overlapping relation of the imaging time windows of the two adjacent strip endpoints, a specific imaging time window clipping schematic diagram is shown in fig. 8, wherein the left side represents the original imaging time window, and the right side represents the imaging time window after clipping.

Normalizing the imaging moment:

after the time window is cut, the strip endpoint imaging sequence SP_ srt _r In any two upper adjacent strip endpoints, the imaging moment of the observed point can have constraint on the feasible range of the observed point, and if the observed 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 affected. As shown in fig. 9, when the imaging time of the 1 st strip end point is determined to be t _r,1 Then in the imaging time window of the 2 nd band endpoint [ ITW _{r,2_s} ，t _r,1 ]To be an ineffective search space, it is necessary to use the search space at [ t ] _r,1 ，ITW _{r,2_e} ]And searching for a feasible solution. In this case, the imaging time of each endpoint is not only constrained by its post-cropping imaging time window, but also must satisfy the observation sequence constraints, the last one of its searches,The lower bound cannot be fixed.

For this problem, after the imaging time window clipping is completed, an imaging time normalization factor t_ nrm is introduced _r,n Normalize the interval between consecutive two endpoint imaging instants to [0,1 ]]And the upper and lower bounds of the decision variables are fixed, so that the optimization algorithm is convenient to solve.

if n=1, then t _r,1 ＝ITW _r,1,s +t_nrm _r,n *(ITW _r,1,e -ITW _r,1,s )

If n is not equal to 1

in the model solving process, when the observation benefits are calculated, the normalization coefficient is restored to the corresponding imaging time by utilizing the above method, and meanwhile, the restored imaging time is ensured to meet the imaging time window constraint of each endpoint after clipping.

Building a band imaging numbering sequence and a band imaging direction sequence by a specific example, calculating a band endpoint imaging sequence according to the band imaging numbering sequence and the band imaging direction sequence, and calculating a band endpoint imaging time sequence process by using an imaging time normalization coefficient in combination with the band 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 ]According to the triplet, 2 x ns imaging moments t corresponding to the strip end points can be restored _r,1 ,t _r,2 ,…,t _r,m ,…,t _r,2*ns ]Wherein r, drt _r,k For determining the strip endpoint imaging sequence SP_ srt _r ，t_nrm _r,m For determining the observation sequence SP_ srt _r Imaging times of the respective strip end points.

The specific flow is shown in fig. 10.

First, according to r and drt _r,k Observed sequence SP_ srt for determining stripe endpoints _r ；

Second, according to the determined SP_ srt _r Cutting operation is carried out, and the imaging time window of each strip endpoint is cut;

third, according to t_ nrm _r,n Performing normalization operation to recover SP_ srt _r Imaging times corresponding to the respective strip end points.

Taking four imaging strips as an example, the specific situation of decision variable to restore the imaging moment is described. As shown in fig. 11, s= { a, B, C, D } and sp= {1,2, …,8} obtained by the pretreatment 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 concrete process of the decision variable triplet to restore the imaging moment is shown in fig. 12. The observation sequence of the strip end point, the imaging time window clipping and the imaging time normalization result are shown in fig. 13 and 14.

In the existing imaging task planning model construction method, imaging time is generally directly taken as a decision variable, and a quantitative relation between the decision variable and an objective function and constraint conditions is established, so that a task planning model is solved. However, directly taking the starting and ending observation moments of the imaging strip as decision variables of the optimization model can cause the problem of low model optimization solving efficiency. Therefore, the strip push-broom sequence number r and the strip push-broom direction number drt are introduced in the model construction process _r,k Normalized coefficient t_ nrm of imaging time _r,n Three types of variables are used as model decision variables. Wherein r and drt _r,k Determining the observed sequence of the end point of the strip, t_ nrm _r,n The time interval between imaging instants of the end points of two adjacent strips is determined. The three types of decision variables can form a one-to-one mapping relation with a group of imaging moments, so that the consumption of attitude maneuver constraint calculation is reduced.

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

the band imaging numbering sequence is:

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 _r Represents the sequence of the imaging numbers of the strips, r represents the r group permutation and combination of the imaging numbers of the strips, ns represents the number of the strips, ns-! Represents the number of all permutation and combination of the strip imaging number, and k represents the permutation and combination of the r group Strip imaging numbering sequence S_ind _r The kth stripe, s_ind _r,k Strip imaging number sequence S_ind representing the r-th permutation and combination _r Number of kth band, s_ind _r,k S, S represents the stripe number set in the step 1;

the sequence of the imaging directions of the strips is as follows:

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

the imaging time sequence of the strip endpoint is as follows:

t _r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the constructed decision variables are:

r∈{1,2,...,ns！}

wherein r represents the r group of permutation and combination of the imaging numbers of the stripsNs denotes the number of stripes, ns-! Representing the number of all permutation combinations of the strip imaging number, determining a strip imaging number sequence S_ind from the step 2 according to the r group permutation combination r of the strip imaging number _r ；

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

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

t_nrm _r,n ∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the imaging coverage benefit maximization in step 3 is as follows:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

c _r,k the judging basis of (a) comprises two aspects:

namely:

Otherwise c _r,k ＝0

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

Representation minimization intoLike task completion time;

if delta t _r,n ≥t_tr _r,n Then c_tr _r,n ＝1

Otherwise c_tr _r,n ＝0

The imaging time window constraint in the step 3 is as follows:

t _r,n ∈[ITW _r,n-s ,ITW _r,n-e ]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end point, t _r,n ∈[ITW _r,n-s ,ITW _r,n-e ]Representing a strip endpoint imaging sequence SP_ srt _r The imaging instant of each of the strip endpoints must meet the cropped imaging time window constraint,

and 3, the gesture conversion time constraint is as follows:

Δt _r,n ≥t_tr _r,n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

Wherein r represents the r group permutation and combination of the imaging numbers of the strips, ns represents the number of the strips, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end, Δt _r,n ≥t_tr _r,n Representing a strip endpoint imaging sequence SP_ srt _r The imaging time interval between two adjacent strip endpoints needs to meet the gesture conversion time constraint between the two adjacent endpoints;

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth stripe endpoint, t_tr _r,n Representing a strip endpoint imaging sequence SP_ srt _r From the n-1 th stripe endpoint to the n-th stripe endpoint,is the momentum of the gesture machine, delta omega, along the rolling axis from the n-1 th strip end point to the n-th strip end point _n,n-1 Attitude machine momentum delta kappa along pitch axis from n-1 th strip end point to n-th strip end point _n,n-1 The momentum of the attitude machine along the yaw axis from the n-1 th strip end point to the n-th strip end point, respectively, >For satellite roll motorized speed,/->Is satellite pitching axis maneuvering speed,/->The three-axis maneuvering speeds of the satellite are respectively;

Step 4, combining with an in-motion imaging task planning mathematical model of the same-track multipoint target, and optimizing by improving a particle swarm algorithm to obtain an imaging task planning scheme with the largest observation income and the shortest task completion time;

the meta heuristic algorithm is the most widely applied method in solving the imaging mission planning model of the agile satellite, wherein the PSO and the improved variant algorithm thereof are one of the most commonly used methods for solving the imaging mission planning model at present. In a standard PSO, each particle moves to a new location according to the new velocity and the current location. However, standard particle swarm algorithms tend to fall into local optima.

For two types of real and binary decision variables when solving the constructed model, the method aims at the real decision variable (r, t_ nrm _r,m ) A new PSO speed position updating method is adopted. And randomly selecting two particles for comparison, and taking the current position of the superior particle and the average position of all particles as the evolution direction of the current position of the inferior particle. Particle velocity and position updates are:

P _lose x(iter+1)＝P _lose x(iter)+v(iter+1)

Wherein iter represents an ith iteration process for improving a particle swarm optimization algorithm, w represents inertia weight, and the value range is 0,1]，c ₁ ，c ₂ Representing acceleration factors, v (iter), v (iter+1) being the current and new velocities, P, of the particles, respectively _win x(iter),P _lose x (iter) represents the current position of the preferred particle and the current position of the inferior particle, P _lose x (iter+1) represents the new position of the inferior particle, P _center x (iter) represents the current average position of all particles.

The method aims at binary class decision variables (drt _r,k ) By BPSO algorithmSpeed and location update policies.

Performing multiple iterations by using the improved PSO algorithm to determine the same-orbit multipoint target imaging scheme of the hypersensitive satellite;

the method comprises the following specific steps:

step 4.1, randomly generating an initial population:

wherein ITER represents an ith iteration process for improving the particle swarm optimization algorithm, and ITER epsilon {1,2,.. The term "ITER,. The term" ITER }, u represents the ith particle in the ith iteration process, and u epsilon {1,2,. The term "u,. The term" Z }; r is (r) _iter,u An r group permutation and combination of the strip imaging numbers corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,u Band imaging number sequence S_ind representing the r-th group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r The imaging direction corresponding to the number of the kth strip; t_ nrm _r,n,iter,u Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u The imaging moment normalization coefficient of the nth strip endpoint;

wherein ITER represents an ith iteration process for improving the particle swarm optimization algorithm, and ITER epsilon {1,2,.. The term "ITER,. The term" ITER }, u represents the ith particle in the ith iteration process, and u epsilon {1,2,. The term "u,. The term" Z }; r is (r) _iter,u An r group permutation and combination of the strip imaging numbers corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,u Band imaging number sequence S_ind representing the r-th group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u Imaging direction corresponding to the number of the kth stripe, S_ind _r,iter,u Band imaging number sequence of the r group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm, t_ nrm _r,n,iter,u Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u The imaging time normalization coefficient of the nth band endpoint of (a), SP_ srt _r,iter,u An r-th group permutation and combination r-corresponding strip endpoint imaging sequence of a strip imaging number corresponding to a u-th particle in an ith iteration process of an improved particle swarm optimization algorithm is represented, and t _r,n,iter,u Banding corresponding to the ith particle in the ith iteration process of improved particle swarm optimization algorithmImaging moment of an nth strip endpoint in the strip endpoint imaging sequence corresponding to an r group of permutation and combination r of image numbers;

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

P _lose x(iter+1)＝P _lose x(iter)+v(iter+1)

wherein w represents inertial weight, and the value range is [0,1]，c ₁ ，c ₂ Representing acceleration factors, v (iter), v (iter+1) being the current and new velocities, P, of the particles, respectively _win x(iter),P _lose x (iter) represents the current position of the preferred particle and the current position of the inferior particle, P _lose x (iter+1) represents the new position of the inferior particle, P _center x (iter) represents the current average position of all particles.

repeating the optimization iteration step 4.2 and the optimization iteration step 4.3 for a plurality of times, and comparing the fitness value corresponding to the optimal particle in the previous iteration process and the current iteration process, namely comparing Fit _iter-1,best And Fit _iter,best If Fit _iter-1,best Is better than Fit _iter,best The optimal grain in the last iteration processIf the sub-corresponding fitness value is the optimal imaging task scheme, if Fit _iter,best Is better than Fit _iter-1,best The fitness value corresponding to the optimal particle in the current iteration process is a better imaging task scheme;

In specific implementation, the above process may be implemented by using a computer software technology. The system device of the corresponding operation flow is also within the protection scope of the invention.

TABLE 3 satellite parameters

TABLE 4 Point target coordinates

For ease of reference, an example is now presented in which satellite orbit-related parameters and satellite attitude maneuver capabilities are shown in Table 3, point targets are shown in Table 4, and the relationship of the point targets to the undersea point trajectory is shown in FIG. 15.

as shown in fig. 16, the stripe number set s= { a, B, …, K }, the stripe endpoint number set sp= {1,2, …,22}, and the original imaging time window of all stripe endpoints are obtained by the point target clustering and non-trace decomposition;

table 5 raw imaging time window

Band end point 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 r be random out 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,u = {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 the following can be used _iter,u 、drt _r,k,iter,u Determining a strip endpoint imaging sequence SP_ srt corresponding to a ith particle in an ith iteration process of an improved particle swarm optimization algorithm _r,iter,u 1-2-3-4-5-6-7-8-9-10-11-12-13-14-15-16-17-18-19-20-21-22.

2) Strip endpoint imaging sequence SP_ srt corresponding to the ith particle in the process of determining the ith iteration of the improved particle swarm optimization algorithm _r,iter,u Thereafter, imaging time window clipping of each band endpoint was performed, and the results are shown in table 6.

TABLE 6 imaging time window after clipping

Strip endpoint imaging sequence	Imaging time window after clipping
		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,u Restored observation sequence sp_ srt _r,iter,u The imaging times of the various strip endpoints are shown in table 7.

Table 7 imaging time of imaging time normalization coefficient restoration

Strip endpoint 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 transition time t_tr of the posture between two adjacent strip end points _r,n,iter,u Obtaining a strip endpoint observation sequence capable of completing push-broom for judgment basis, and obtaining observation benefits and task completion time;

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

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

Table 8 shows the imaging sequence of the strip end points corresponding to the optimal result and the imaging time of each strip end point

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 offered by way of example only to illustrate the spirit of the invention. Those skilled in the art may make various modifications or additions to the described embodiments or substitutions, without departing from the spirit of the invention or exceeding the scope of the invention as defined in the accompanying claims.

Claims

1. The method for planning the imaging task in the same orbit multipoint target motion of the hypersensitive satellite is characterized by comprising the following steps of:

and 4, combining the simultaneous multipoint target in-motion imaging task planning mathematical model, and optimizing by improving a particle swarm algorithm to obtain the imaging task planning scheme with the maximum observation benefit and the shortest task completion time.

2. The method for planning the imaging mission in the same orbit and multiple points of targets on the move of the hypersensitive satellite according to claim 1 is characterized in that:

i∈[1,np]

the point target cluster is 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 ,Poin t _l,j ^c ,Point _l,j ^d ；...,；Point _l,m ^a ,Point _l,m ^b ,Point _l,m ^c ,Point _l,m ^d }；

SP represents the final set of stripe endpoint numbers, sp= {1,2,..n,..2, 2 x ns }, where n represents the stripe endpoint number, 1.ltoreq.n.ltoreq.2 x ns,2 x ns represents the final number of stripe endpoints; all stripe endPoint longitude and latitude coordinate set, sendpoint= { endPoint = { endPoint ₁ ,endPoint ₂ ,...,endPoint _n ,...,endPoint _2*ns }, wherein (EndPoint) _n Representing longitude and latitude coordinates corresponding to a strip endpoint with a strip endpoint number of n, wherein n is more than or equal to 1 and less than or equal to 2 x ns;

the imaging time window for all the strip endpoints is obtained in the step 1:

set SP and pairing according to stripe endpoint numbersDetermining imaging time windows of all strip endpoints by adopting a characteristic cone method according to a corresponding strip endpoint longitude and latitude coordinate set SendPoint to obtain the imaging time windows of all strip endpoints in a strip endpoint number set SP as [ TW ] _n,s ,TW _n,e ]；

Wherein [ TW ] _n,s ,TW _n,e ]TW representing the original imaging time window corresponding to the band endpoint with the band endpoint number n _n,s TW is the time window starting time _n,e Is the time window ending moment.

3. The method for planning the imaging mission in the same orbit and multiple points of targets on the move of the hypersensitive satellite according to claim 1 is characterized in that:

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}

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

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

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

Therein, SP_ srt _r The imaging sequence of the strip endpoint corresponding to the r-th group permutation and combination r of the imaging number of the strip is represented, r represents the r-th group permutation and combination of the imaging number of the strip, and ns represents the number of the strips Ns-! Representing the number of all permutation and combination of the strip imaging numbers; k represents the strip imaging number sequence S_ind of the r group permutation and combination _r The kth band in (a); ST denotes the start imaging strip endpoint of the strip, ET denotes the end imaging strip endpoint of the strip;

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

and 2, calculating a strip endpoint imaging time sequence as follows:

the imaging moment normalization coefficient is as follows:

t_nrm _r,n ∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

the imaging time window is cut:

In the above, T _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of a time window corresponding to the nth observed strip endpoint; t (T) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The time window end time corresponding to the nth observed strip endpoint; ITW (International telecommunication System) _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the n observed strip end points correspond to the cutIs a time window start time of (1); ITW (International telecommunication System) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The end time of the time window after cutting corresponding to the nth observed strip endpoint;

normalizing the imaging moment:

if n=1, then t _r,1 ＝ITW _r,1,s +t_nrm _r,n *(ITW _r,1,e -ITW _r,1,s )

If n is not equal to 1

If t _r,n ≤ITW _r,n+1,s Then t _r,n+1 ＝ITW _r,n+1,s +t_nrm _r,n *(ITW _r,n+1,e -ITW _r,n+1,s ) Wherein t is _r,1 Imaging sequence sp_ srt for a stripe endpoint _r In the 1 st observed band endpoint, imaging time t of the time window after clipping _r,n Imaging sequence sp_ srt for a stripe endpoint _r In the imaging time, t, of the nth observed strip endpoint _r,n+1 Imaging sequence sp_ srt for a stripe endpoint _r In the imaging instant of the (n+1) th observed band endpoint, ITW _r,n,s Representing a strip endpoint imaging sequence SP_ srt _r In the method, the starting moment of the cut time window corresponding to the nth observed strip endpoint; ITW (International telecommunication System) _r,n,e Representing a strip endpoint imaging sequence SP_ srt _r The end time of the time window after clipping corresponding to the nth observed strip endpoint.

4. The method for planning the imaging mission in the same orbit and multiple points of targets on the move of the hypersensitive satellite according to claim 1 is characterized in that:

the band imaging numbering sequence is:

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 imaging directions of the strips is as follows:

drt _r,k ∈{0,1}

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

the imaging time sequence of the strip endpoint is as follows:

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 for a binary number;

t_nrm _r,n ∈[0,1]

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

Wherein t_ nrm _r,n The strip end corresponding to the r-th group of permutation and combination r for representing the strip imaging numberPoint imaging sequence SP_ srt _r The imaging time normalization coefficient of the nth band endpoint in (a), n representing the band endpoint observation sequence SP_ srt _r An nth stripe endpoint of (a);

the imaging coverage benefit maximization in step 3 is as follows:

r∈{1,2,...,ns！}

k∈{1,2,...,ns}

c _r,k the judging basis of (a) 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 ) C is _r,k ＝1

Otherwise c _r,k ＝0

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

representing minimizing imaging task completion time;

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end, vt _r,n Representing a strip endpoint imaging sequence SP_ srt _r Upper observation time interval from the n-1 th band end to the n-th band end, vt _r,n ＝t _r,n -t _r,n-1 ，t _r,n Imaging sequence sp_ srt for a stripe endpoint _r Imaging time of the upper nth band endpoint;

if Vt is _r,n ≥t_tr _r,n Then c_tr _r,n ＝1

Otherwise c_tr _r,n ＝0

The imaging time window constraint in the step 3 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, the gesture conversion time constraint is as follows:

Vt _r，n ≥t_tr _r，n

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth band end, vt _r,n ≥t_tr _r,n Representing a strip endpoint imaging sequence SP_ srt _r The imaging time interval between two adjacent strip endpoints needs to meet the gesture conversion time constraint between the two adjacent endpoints;

r∈{1,2,...,ns！}

n∈{1,2,...,2*ns}

where r represents the r group permutation and combination of the imaging numbers of the stripes, ns represents the number of stripes, ns-! Represents the number of all permutation and combination of the imaging numbers of the strip, and n represents the strip endpoint observation sequence SP_ srt _r The nth stripe endpoint, t_tr _r,n Representing a strip endpoint imaging sequence SP_ srt _r From the n-1 th stripe endpoint to the n-th stripe endpoint,is the momentum of the gesture machine, delta omega, along the rolling axis from the n-1 th strip end point to the n-th strip end point _n,n-1 Attitude machine momentum delta kappa along pitch axis from n-1 th strip end point to n-th strip end point _n,n-1 The momentum of the attitude machine along the yaw axis from the n-1 th strip end point to the n-th strip end point, respectively,>for satellite roll motorized speed,/->Is satellite pitching axis maneuvering speed,/->The three-axis maneuvering speeds of the satellite are respectively;

5. The method for planning the imaging mission in the same orbit and multiple points of targets on the move of the hypersensitive satellite according to claim 1 is characterized in that:

and step 4, optimizing the imaging task planning scheme with the largest observation benefit and the shortest task completion time through an improved particle swarm algorithm, wherein the imaging task planning scheme comprises the following steps:

step 4.1, randomly generating an initial population:

wherein iter representsAn ith iteration process of the improved particle swarm optimization algorithm, wherein the item epsilon {1, 2..the item, & gt, ITER }, u represents the ith particle in the ith iteration process, and u epsilon {1, 2..the item, & gt, Z }; r is (r) _iter,u An r group permutation and combination of the strip imaging numbers corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,u Band imaging number sequence S_ind representing the r-th group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r The imaging direction corresponding to the number of the kth strip; t_ nrm _r,n,iter,u Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u The imaging moment normalization coefficient of the nth strip endpoint;

Wherein iter represents an iter-th iterative process that improves the particle swarm optimization algorithm, iter e {1,2,., iter., ITER, u represents the u-th particle in the ith iteration process, u e {1,2,., u., Z }; r is (r) _iter,u An r group permutation and combination of the strip imaging numbers corresponding to the u-th particle in the ith iteration process of the improved particle swarm optimization algorithm is shown; drt _r,k,iter,u Band imaging number sequence S_ind representing the r-th group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u Imaging direction corresponding to the number of the kth stripe, S_ind _r,iter,u Band imaging number sequence of the r group permutation and combination of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm, t_ nrm _r,n,iter,u Band endpoint imaging sequence SP_ srt corresponding to the r-th group permutation and combination r of band imaging numbers corresponding to the u-th particle in the ith iteration process of improved particle swarm optimization algorithm _r,iter,u The imaging time normalization coefficient of the nth band endpoint of (a), SP_ srt _r,iter,u An r-th group permutation and combination r-corresponding strip endpoint imaging sequence of a strip imaging number corresponding to a u-th particle in an ith iteration process of an improved particle swarm optimization algorithm is represented, and t _r,n,iter,u The imaging moment of the nth strip endpoint in the strip endpoint imaging sequence corresponding to the r group permutation and combination r of the strip imaging number corresponding to the ith particle in the ith iteration process of the improved particle swarm optimization algorithm is represented;

the particles with the optimal particle fitness value in the current population are obtained by the following steps:

wherein best represents particles with optimal fitness value in particle swarm in the ith iteration process of improved particle swarm optimization algorithm, and Fit _iter,best Representing the fitness of the optimal particle best in the ith iteration process of the improved particle swarm optimization algorithmA value;

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

P _lose x(iter+1)＝P _lose x(iter)+v(iter+1)

wherein w represents inertial weight, and the value range is [0,1]，c ₁ ，c ₂ Representing acceleration factors, v (iter), v (iter+1) being the current and new velocities, P, of the particles, respectively _win x(iter),P _lose x (iter) represents the current position of the preferred particle and the current position of the inferior particle, P _lose x (iter+1) represents the new position of the inferior particle, P _center x (iter) represents the current average position of all particles;

Wherein best represents particles with optimal fitness value in particle swarm in the ith iteration process of improved particle swarm optimization algorithm, and Fit _iter,best Representing the fitness value corresponding to the optimal particle best in the ith iteration process of the improved particle swarm optimization algorithm; fit _iter-1,best Representing the fitness value corresponding to the optimal particle best in the ith-1 iteration process of the improved particle swarm optimization algorithm; fit _*,best Indicating that the iteration termination condition is reached, the particle swarm is improvedAnd (3) the fitness value corresponding to the optimal particle best in the iterative process of the first time of the optimization algorithm, namely an optimal imaging task scheme.