CN113758467A - Remote sensing satellite region observation method based on region division and local grid nesting - Google Patents

Abstract

The invention discloses a remote sensing satellite region observation method based on region division and local grid nesting, which comprises the following steps: 1. dividing a target area into relatively independent local areas, wherein the areas can be overlapped but do not interfere with each other; 2. carrying out gridding dispersion of different granularities on a target area according to a local area; 3. the strips of grids at different levels may together constitute a feasible observation scheme. The method can be applied to the cooperative observation of one target area by a plurality of satellites, the target area is divided into local areas which are not influenced by each other and is processed independently, the balance between the consumption of computing resources and the optimality of a solution is realized, and therefore a better regional target cooperative observation scheme can be obtained by using proper computing resources.

Description

Remote sensing satellite region observation method based on region division and local grid nesting

Technical Field

The invention belongs to the field of remote sensing satellite task observation, and particularly relates to a remote sensing satellite region observation method based on region division and local grid nesting.

Background

A remote sensing satellite is an artificial satellite used as a remote sensing platform in the outer space of the earth. When the remote sensing satellite runs along the geosynchronous orbit, the remote sensing satellite can continuously observe some designated regions on the earth surface. The data acquired by the remote sensing satellite can be applied to the fields of agriculture, forestry, oceans, homeland, environmental protection, weather and the like. Only one strip area with limited length and width can be shot by a single satellite in transit once, and if the area to be observed is large, the whole area cannot be completely observed by the single satellite in transit for many times. In this case, multiple satellites must be mobilized for simultaneous planning to provide coordinated observations.

In order to better utilize the existing satellite observation resources, a reasonable plan should be formulated, and the observation attitude, the observation start time and the observation end time of the satellite when the satellite operates and reaches the target area need to be planned according to the target area information and the parameters of the satellite, so that the cooperative observation effect of a plurality of satellites can meet the observation requirement as much as possible. The method is a typical operational optimization problem, and has urgent practical requirements.

This is a highly coupled problem with computational geometry, where the target area is mechanically divided into several closely spaced parallel strips when studying the coverage observations of a single satellite at an early stage. However, because the orbits of different satellites are not uniform, the method is only suitable for observation coverage of a single satellite and is not suitable for observation coverage of multiple satellites. In the processing of multi-satellite region observation problem, the grid discretization method is a common means for region processing at present. However, the grid discretization method is difficult to directly determine the optimal grid granularity in use, and the determination of the grid granularity is empirically determined in most practical applications. When the granularity of grid division is larger, the strips constructed on the basis are sparse, and are not beneficial to the combination of schemes; when the granularity of grid division is small, a large number of grids can be generated, and the number of the generated strips is large, the arrangement on the positions is also very dense, and a large calculation load and a large search load can be caused.

Disclosure of Invention

The invention aims to solve the defects of the prior art, and provides a remote sensing satellite region observation method based on region division and local grid nesting, so that balance between calculation resource consumption and solution optimality can be achieved, observation strips of multiple satellites can be obtained through proper calculation resources, satellite resources can be fully utilized in an actual environment, and a proper multi-satellite cooperative observation scheme can be obtained.

In order to achieve the purpose, the invention adopts the following technical scheme:

the invention relates to a remote sensing satellite region observation method based on region division and local grid nesting, which is characterized in that R is used for representing a target region to be observed, and R | is used for representing the area of the target region R; with C ═ C₁,c₂,...,c_i,...,c_nDenotes the imaging windows of all satellites in time, i.e. the observation opportunities, c_iRepresents the ith observation opportunity, and n represents the number of observation opportunities;

with B_iIndicates the ith observation opportunity c_iAt the earliest observation start time of (1), using E_iIndicates the ith observation opportunity c_iAt the latest observation end time of l_iIndicates the ith observation opportunity c_iProjection straight line of corresponding orbit on the earth surface, using w_iIndicates the ith observation opportunity c_iThe maximum deflection angle of;

step 1, obtaining the ith observation opportunity c_iCorresponding track projection straight line l_iFirst observation start time B_iLatest observation end time E_iProjection of the corresponding orbit on the earth's surface l_iMaximum deflection angle w_i；

Step 2, calculating the ith observation opportunity c_iMaximum observation region of

Step 2.1, projecting_iAbove find B_iCorresponding position projection point P of time_iBAnd through P_iBPoint making and projection_iVertical straight line l_iB；

Step 2.2, projecting_iAbove find E_iCorresponding position of timeShadow point P_iEOver P_iEPoint making and projection_iVertical straight line l_iE；

Step 2.3, calculate the ith observation opportunity c_iThe corresponding satellite has a leftward deflection angle of w_iThe farthest point P on the left side of the observation range_iLOver P_iLPoint making and projection_iParallel straight lines l_iL；

Step 2.4, calculate the ith observation opportunity c_iThe corresponding satellite has a right deflection angle of w_iThe right farthest point P of the observation range_iROver P_iRPoint making and projection_iParallel straight lines l_iR；

Step 2.5, four straight lines l_iB,l_iE,l_iLAnd l_iRThe closed area is the ith observation opportunity c_iMaximum observation region LR of_ci；

Step 3, dividing the target region R into n local regions according to the maximum observation region; wherein, according to the maximum observation region LR_ciDividing to obtain an ith local area;

step 4, enabling K to represent a grid level, initializing K to be 1, and performing unified primary division on the target region R to obtain a primary grid of the target region R and using the primary grid as a K-th grid;

step 5, constructing the ith observation opportunity c_iSet of alternate stripes on K-th grid

Wherein the content of the first and second substances,

indicates the ith observation opportunity c_iThe jth candidate strip constructed on the kth layer of mesh,

indicates the ith observation opportunity c_iThe number of candidate strips constructed on the K-th layer;

step 6, nesting the local area, and constructing a new-level alternative strip set:

step 6.1, changing i to 1; let K_maxRepresenting the maximum nesting level number;

step 6.2, setting the ith observation opportunity c_iOf the final nesting level K_i∈[1,K_max]A random integer of (1);

step 6.3, if K_iIf 1, step 6.4 is executed; otherwise, for the ith local area

Performing nested division:

step 6.3.1, let K equal to 1, find that all four vertices in the kth-level mesh are located in the ith local area

An inner cell;

step 6.3.2, dividing the found cells into 4 small cells with the side length being 1/2 of the original side length;

step 6.3.3, assign K +1 to K, repeat step 6.3.2 until K ═ K_iUntil the end;

step 6.4, assigning i +1 to i, and if i is less than n, repeating the step 6.3; otherwise, it is indicated in the ith local area

Upper division to obtain K_iLayer gridding, and turning to step 7;

step 7, constructing an alternative strip set for each observation opportunity according to the corresponding local area:

step 7.1, changing i to 1;

step 7.2, at the ith observation opportunity c_iCorresponding local area

According to the K obtained by final division_iThe layers are meshed to form strips, and the strips are put into an alternative strip set

Performing the following steps;

step 7.3, from the set of alternative bands

Randomly selecting one stripe;

and 7.4, assigning i +1 to i, and returning to the step 7.2 until i is equal to n, so that the n strips form an observation scheme.

Compared with the prior art, the invention has the beneficial effects that:

the invention adopts the concept of 'divide-and-conquer' to put forward the concept of the local area, namely, the whole target area is divided into a plurality of mutually independent and non-interfering local areas according to the maximum observation range of the satellite observation opportunity; different local areas can be dispersed by using grids with different granularities, and suitable strips are independently constructed; because the grids of all local regions are nested based on the common primary grids, the grids with different granularities can be fused with each other, and the strips constructed on the grids of different levels can exist at the same time to jointly form a complete observation scheme, thereby solving the problem of difficult granularity selection of the regional target grids, improving the flexibility of regional processing and being beneficial to obtaining a better multi-satellite cooperative observation scheme by using proper computing resources.

Drawings

FIG. 1 is a flow chart of a remote sensing satellite region observation method based on region division and local grid nesting according to the present invention;

FIG. 2 is a multi-satellite regional target collaborative viewing map of the present invention;

FIG. 3 is a fragmentary, partially sectioned view of the present invention;

FIG. 4 is a partial area grid nesting diagram of the present invention;

FIG. 5 is a diagram of an observation scheme of the present invention.

Detailed Description

In this embodiment, as shown in fig. 2, a remote sensing satellite region observation method based on region division and local grid nesting is applied to cooperatively observing a target by multiple satellitesIn the planning of the region, the optimization goal is to make the coverage rate of the observation scheme to the region target as large as possible. Using R to represent a target region to be observed, and using R to represent the area size of the target region R; a single observation opportunity can cover only a small portion of the target area, and therefore requires coordinated observation by multiple observation opportunities, C ═ C₁,c₂,...,c_i,...,c_nAnd all the satellites in time sequence pass through an imaging window which is over the area to be observed, namely an observation opportunity. c. C_iRepresents the ith observation opportunity, and n represents the number of observation opportunities;

with B_iIndicates the ith observation opportunity c_iAt the earliest observation start time of (1), using E_iIndicates the ith observation opportunity c_iAt the latest observation end time of l_iIndicates the ith observation opportunity c_iApproximate straight line of projection of corresponding orbit on the earth's surface, using w_iIndicates the ith observation opportunity c_iAt the observation opportunity c_iThe angle of deflection of the inner satellite to both sides on the plane perpendicular to the flight orbit must not be greater than w_i；

As shown in fig. 1, the remote sensing satellite region observation method is performed according to the following steps:

step 1, obtaining the ith observation opportunity c_iCorresponding track projection straight line l_iFirst observation start time B_iLatest observation end time E_iProjection of the corresponding orbit on the earth's surface l_iMaximum deflection angle w_i；

Step 2, calculating the ith observation opportunity c_iMaximum observation region of

Step 2.1, as shown in FIG. 3, projecting on a projection l_iAbove find B_iCorresponding position projection point P of time_iBAnd through P_iBPoint making and projection_iVertical straight line l_iB；

Step 2.2, projecting_iAbove find E_iCorresponding position of timeProjection point P_iEOver P_iEPoint making and projection_iVertical straight line l_iE；

Step 2.3, calculate the ith observation opportunity c_iThe corresponding satellite has a leftward deflection angle of w_iThe farthest point P on the left side of the observation range_iLOver P_iLPoint making and projection_iParallel straight lines l_iL；

Step 2.4, calculate the ith observation opportunity c_iThe corresponding satellite has a right deflection angle of w_iThe right farthest point P of the observation range_iROver P_iRPoint making and projection_iParallel straight lines l_iR；

Step 2.5, four straight lines l_iB,l_iE,l_iLAnd l_iRThe closed area is the ith observation opportunity c_iMaximum observation region of

Step 3, dividing the target region R into n local regions according to the maximum observation region; wherein, according to the maximum observation area

Dividing to obtain an ith local area; there may be overlap between different local regions, but not influence each other. The division of the local regions is abstract, so that there may be overlap between different local regions, but there is no mutual influence. If opportunity c₁And opportunity c₂Corresponding local area

And

there is an overlap for a local area

Making the division from K to K +1 layers only implies the opportunity c₁Has the structure thatCondition of K +1 layer of stripes, for opportunity c₂No influence is caused.

Step 4, enabling K to represent a grid level, initializing K to be 1, and performing unified primary division on the target region R to obtain a primary grid of the target region R and using the primary grid as a K-th grid;

and step 5, constructing the ith observation opportunity c by referring to a patent' planning method for multi-imaging satellite area coverage tasks considering sensor side sway_iSet of alternate stripes on K-th grid

Wherein the content of the first and second substances,

indicates the ith observation opportunity c_iThe jth candidate strip constructed on the kth layer of mesh,

indicates the ith observation opportunity c_iThe number of candidate strips constructed on the K-th layer;

step 6, as shown in fig. 4, nesting the local regions for constructing a new level of candidate stripe sets:

step 6.1, changing i to 1; let K_maxRepresenting the maximum nesting level number;

step 6.2, setting the ith observation opportunity c_iOf the final nesting level K_i∈[1,K_max]A random integer of (1);

step 6.3, if K_iIf 1, step 6.4 is executed; otherwise, for the ith local area

Performing nested division:

step 6.3.1, let K equal to 1, find that all four vertices in the kth-level mesh are located in the ith local area

An inner cell;

step 6.3.2, dividing the found cells into 4 small cells with the side length being 1/2 of the original side length;

step 6.3.3, assign K +1 to K, repeat step 6.3.2 until K ═ K_iUntil the end;

step 6.4, assigning i +1 to i, and if i is less than n, repeating the step 6.3; otherwise, it is indicated in the ith local area

Upper division to obtain K_iLayer gridding, and turning to step 7;

step 7, constructing an alternative strip set for each observation opportunity according to the grids after the corresponding local region nesting; in the same step 5, the grid-based strip construction method can refer to a patent "planning method for coverage of multi-imaging satellite region considering sensor sidesway" (i.e.:

step 7.1, changing i to 1;

step 7.2, at the ith observation opportunity c_iCorresponding local area

According to the K obtained by final division_iThe layers are meshed to form strips, and the strips are put into an alternative strip set

Performing the following steps;

step 7.3, from the set of alternative bands

Randomly selecting one stripe;

step 7.4, i +1 is assigned to i, and step 7.2 is returned until i ═ n, so that an observation scheme is formed by n strips, as shown in fig. 5.

Claims (1)

1. A remote sensing satellite region observation method based on region division and local grid nesting is characterized in that R is used for representing a target region to be observed, and R is used for representing the target region to be observedThe area of the target region R; with C ═ C₁,c₂,...,c_i,...,c_nDenotes the imaging windows of all satellites in time, i.e. the observation opportunities, c_iRepresents the ith observation opportunity, and n represents the number of observation opportunities;

with B_iIndicates the ith observation opportunity c_iAt the earliest observation start time of (1), using E_iIndicates the ith observation opportunity c_iAt the latest observation end time of l_iIndicates the ith observation opportunity c_iProjection straight line of corresponding orbit on the earth surface, using w_iIndicates the ith observation opportunity c_iThe maximum deflection angle of;

step 1, obtaining the ith observation opportunity c_iCorresponding track projection straight line l_iFirst observation start time B_iLatest observation end time E_iProjection of the corresponding orbit on the earth's surface l_iMaximum deflection angle w_i；

Step 2, calculating the ith observation opportunity c_iMaximum observation region of

Step 2.1, projecting_iAbove find B_iCorresponding position projection point P of time_iBAnd through P_iBPoint making and projection_iVertical straight line l_iB；

Step 2.2, projecting_iAbove find E_iCorresponding position projection point P of time_iEOver P_iEPoint making and projection_iVertical straight line l_iE；

Step 2.3, calculate the ith observation opportunity c_iThe corresponding satellite has a leftward deflection angle of w_iThe farthest point P on the left side of the observation range_iLOver P_iLPoint making and projection_iParallel straight lines l_iL；

Step 2.4, calculate the ith observation opportunity c_iThe corresponding satellite has a right deflection angle of w_iThe right farthest point P of the observation range_iROver P_iRPoint making and projection_iParallel straight lines l_iR；

Step 2.5, four straight lines l_iB,l_iE,l_iLAnd l_iRThe closed area is the ith observation opportunity c_iMaximum observation region of

Step 3, dividing the target region R into n local regions according to the maximum observation region; wherein, according to the maximum observation area

Dividing to obtain an ith local area;

step 4, enabling K to represent a grid level, initializing K to be 1, and performing unified primary division on the target region R to obtain a primary grid of the target region R and using the primary grid as a K-th grid;

step 5, constructing the ith observation opportunity c_iSet of alternate stripes on K-th grid

Wherein the content of the first and second substances,

indicates the ith observation opportunity c_iThe jth candidate strip constructed on the kth layer of mesh,

indicates the ith observation opportunity c_iThe number of candidate strips constructed on the K-th layer;

step 6, nesting the local area, and constructing a new-level alternative strip set:

step 6.1, changing i to 1; let K_maxRepresenting the maximum nesting level number;

step 6.2, setting the ith observation opportunity c_iOf the final nesting level K_i∈[1,K_max]A random integer of (1);

step 6.3, if K_iIf 1, step 6.4 is executed; otherwise, for the ith local area

Performing nested division:

step 6.3.1, let K equal to 1, find that all four vertices in the kth-level mesh are located in the ith local area

An inner cell;

step 6.3.2, dividing the found cells into 4 small cells with the side length being 1/2 of the original side length;

step 6.3.3, assign K +1 to K, repeat step 6.3.2 until K ═ K_iUntil the end;

step 6.4, assigning i +1 to i, and if i is less than n, repeating the step 6.3; otherwise, it is indicated in the ith local area

Upper division to obtain K_iLayer gridding, and turning to step 7;

step 7, constructing an alternative strip set for each observation opportunity according to the corresponding local area:

step 7.1, changing i to 1;

step 7.2, at the ith observation opportunity c_iCorresponding local area

According to the K obtained by final division_iThe layers are meshed to form strips, and the strips are put into an alternative strip set

Performing the following steps;

step 7.3, from the set of alternative bands

Randomly selecting one stripe;

and 7.4, assigning i +1 to i, and returning to the step 7.2 until i is equal to n, so that the n strips form an observation scheme.

Images