CN115909083B - Satellite Earth Observation Discrete Points of Interest Clustering Planning Method and Device - Google Patents

Abstract

The invention discloses a satellite earth observation discrete point of interest clustering planning method and device, belonging to the technical field of satellite earth observation task planning. The method includes: obtaining information on the points of interest in the target area; wherein the information includes: longitude, latitude and interest value; clustering the points of interest based on the information, and obtaining the degree of interest of each category in the clustering results; according to The satellite orbit of the satellite Sat _k , obtain the stripe _kg of the satellite Sat _k in the target area, and obtain the decision of the stripe _kg by whether the satellite Sat _k observes the stripe _kg The value x _kg of the variable; based on the interest degree of each category and the value x _kg of the decision variable, maximize the sum of the interest degrees of all satellites Sat _k observing the target area to obtain the clustering planning result of the interest point. The invention not only reduces the calculation complexity of the cluster planning algorithm, but also can achieve higher satellite observation benefits.

Description

Satellite Earth Observation Discrete Points of Interest Clustering Planning Method and Device

technical field

The invention relates to a satellite earth observation discrete point of interest cluster planning method and device, and belongs to the technical field of satellite earth observation mission planning.

Background technique

In order to improve the utilization efficiency of effective satellite resources, it is necessary to arrange the earth observation activities of satellites reasonably, so as to form an efficient mission execution plan. This process needs to search the execution sequence of a large number of tasks under various constraints such as satellite platforms, which is a complex combinatorial optimization problem. The current satellite mission planning implementation requires that after receiving the mission request, the ground mission planning center will calculate according to various constraints such as mission characteristics, satellite visible time window and satellite platform load, and upload the planned result instructions to the satellite. The uploaded instructions execute Earth observation missions.

When the number of earth observation tasks is relatively large, such as a large number of discretely distributed interest points, the execution sequence composed of these tasks is quite large, and the optimization complexity will increase dramatically with the increase of interest points. Although merging POIs with similar geographic locations can reduce the number of tasks to be planned, how to select a feasible merging scheme will directly affect the effect of POI preprocessing and task planning.

Contents of the invention

The present invention proposes a satellite-to-earth observation discrete point of interest clustering planning method and device. The method completes iterative clustering according to the location information and importance of the point of interest, that is, the interest value, and finally multiple nearby points of interest will be stabilized in one The cluster representation of the circular area reflects the importance of the whole. On this basis, the task planning solution for all the clustered circular areas not only reduces the computational complexity of the planning algorithm, but also prioritizes observations in areas where key points of interest appear with a high probability to reach Higher overall payoff for observations.

To achieve the above object, the present invention adopts the following technical solutions:

A satellite earth observation discrete point of interest clustering planning method, the method comprising:

Acquiring information on points of interest in the target area; wherein the information includes: longitude, latitude and interest value;

Clustering the interest points based on the information, and obtaining the degree of interest of each category in the clustering results;

Obtain the stripe _g of the satellite Sat _k in the target area according to the satellite orbit of the satellite Sat _k , and obtain the stripe g of the stripe _g according to whether the satellite Sat _k observes the stripe _g The value of the decision variable x _kg ;

Based on the interest degrees of each category and the value x _kg of the decision variable, the sum of the interest degrees of all satellites Sat _k observing the target area is maximized to obtain the cluster planning result of the interest points.

Further, the clustering of interest points based on the information, and obtaining the degree of interest of each category in the clustering result includes: selecting any one of the interest points in the current set of interest points as the center of the circle;

With a set value as the radius, a circle C _i is generated, and all interest points located in the circle C _i are determined to generate a set of interest points

According to the collection of points of interest In the interest point information, update the center position of the circle _Ci , until the distance between the center positions of two adjacent updates is less than a threshold, based on the updated center position, obtain a subset of interest points;

Obtain the interest degree of the interest point subset based on the interest value of each interest point in the interest point subset;

deleting the subset of interest points from the current set of interest points, and judging whether the updated set of interest points is an empty set;

In the case that the updated set of interest points is not an empty set, return to the current set of interest points and select any one of the interest points as the center of the circle;

In the case that the updated set of interest points is an empty set, the interest degrees of each subset of interest points are output.

Further, according to the set of interest points The longitude and latitude of the point of interest in the middle, update the center position of the circle _Ci , including:

Based on the set of interest points The longitude and interest value of the interest point in the center, calculate the longitude of the new center position /> Among them, x _j represents the longitude of the jth interest point, w _j represents the interest value of the jth interest point, and n represents the set of interest points/> the number of points of interest in

Based on the set of interest points The latitude and interest value of the middle interest point, calculate the latitude of the new center position /> Among them, _yj represents the latitude of the jth interest point.

Further, the acquisition of the value x _kg of the decision variable whether to observe the stripe _g through the satellite Sat _k includes:

In the case that the satellite Sat _k observes the stripe _g , the value of the decision variable x _kg =1;

In the case that the satellite Sat _k does not observe the stripe _g , the value of the decision variable x _kg =0.

Further, based on the interest degrees of each category and the value x _kg , maximizing the sum of the interest degrees of all satellites Sat _k observing the target area, to obtain the cluster planning results of interest points, including:

Based on the degree of interest and the value x _kg , construct an objective function to be optimized Wherein, S represents the quantity of the satellite Sat _k , G represents the quantity of the strip stripe _g , _{and v} represents the degree of interest of the corresponding circle C _i ;

setting constraints on the objective function;

The particle swarm optimization algorithm is used to optimize the objective function to be optimized to obtain the cluster planning result of the interest points.

Further, the constraints of the objective function include:

The observation quantity constraint of the satellite Sat _k ; wherein, the observation quantity constraint represents that within the planning time interval, each circular area can only be observed once by the satellite Sat _k at most;

and,

The imaging time constraint of the satellite Sat _k ; wherein, the imaging time constraint indicates that the satellite Sat _k operates within the entire scheduling time range, and the observation time of any stripe stripe _g must meet the imaging time requirement;

and,

The roll angle constraint of the satellite Sat _k ; wherein, the roll angle constraint indicates that in the planning time interval, the satellite Sat _k can only select a strip under a roll angle when passing through a single clustering circle ;

and,

The storage capacity constraint of the satellite Sat _k .

Further, the optimization of the objective function to be optimized by the particle swarm optimization algorithm is used to obtain the clustering planning results of the points of interest, including:

Initialize the satellite Sat _k and the strip stripe _g ;

The number of satellites, the number of corresponding target areas, and related constraints are calculated as system variables, and the variables that violate the constraints are corrected to calculate the maximum interest of the system as the current objective function value;

Determine the optimal combination pbest of the current satellite and the strip according to the objective function;

The optimal combination gbest of satellites and stripes selected from the optimal combination pbest of all satellites and stripes iterated;

Determine whether the maximum number of iterations has been reached;

If the maximum number of iterations is not reached, return to the calculation of the number of satellites, the number of corresponding target areas, and related constraints as system variables, and correct the variables that violate the constraints to calculate the maximum weight sum of the system as the current objective function value;

When the maximum number of iterations is reached, the optimal combination gbest of satellites and strips is taken as the result of earth observation planning.

A satellite earth observation discrete point of interest cluster planning device, characterized in that the device comprises:

An information acquisition module, configured to acquire information on points of interest in the target area; wherein the information includes: longitude, latitude, and interest value;

A point of interest clustering module, configured to cluster the points of interest based on the information, and obtain the degree of interest of each category in the clustering result;

The variable calculation module is used to obtain the stripe _g of the satellite Sat _k in the target area according to the satellite orbit of the satellite Sat _k , and obtain a decision by whether the satellite Sat _k observes the stripe stripe _g the value of the variable x _kg ;

The cluster planning module is configured to maximize the sum of the interest degrees of all satellites Sat _k observing the target area based on the interest degrees of each category and the value x _kg , so as to obtain the cluster planning results of interest points.

A storage medium, wherein a computer program is stored in the storage medium, wherein the computer program is configured to execute the above-mentioned satellite earth observation discrete point of interest clustering planning method during operation.

An electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the above-mentioned satellite earth observation discrete point of interest clustering planning method.

Compared with the prior art, the present invention has the advantages of:

1. After this clustering process, the regional distribution of discrete interest points is more distinct, and the number of clusters is much smaller than the number of interest points, which makes the calculation efficiency of satellite observation task planning higher.

2. The clustering process adjusts the center of the cluster according to the importance of each interest point, so that the more important interest points are located in the center of the cluster area, and these more important interest points are more likely to be observed.

3. The clustering method requires few parameters, does not need to determine the number of clusters in advance, has high processing efficiency, and can quickly converge to a large number of discrete ground interest points.

Description of drawings

Fig. 1 is a flow chart of the satellite earth observation discrete point of interest clustering planning method of the present invention.

Fig. 2 is a block diagram of the satellite earth observation discrete point of interest cluster planning device of the present invention.

Detailed ways

In order to make the above features and advantages of the present invention more comprehensible, the technical solutions of the present invention will be further described below through specific examples.

The purpose of the present invention is to propose a method for clustering planning of satellite-to-earth observation discrete interest points, which improves the observation efficiency of a large number of satellite-to-earth discrete interest points.

The satellite earth observation discrete point of interest clustering planning method proposed by the present invention, as shown in FIG. 1 , includes the following steps.

(1) Input the location and interest degree information of discrete interest points.

Assume that the set of all ground interest points is P=P _i , 1≤i≤n, where n represents the total number of points in the set. Each point is expressed as P _i ={xi _, y _i ,w _i }, 1≤i≤n, i represents the serial number of the point, x _i represents the longitude, y _i represents the latitude, and w _i represents the interest in the point degree. The set of clustering results is initialized as C={}.

(2) Use the sliding clustering method to process the interest points.

(2-1) Select any point P _i in the set P as the center of the circle.

(2-2) With r as the radius, generate a circle C _i and determine all interest points located in C _i

(2-3) According to The coordinates of the midpoint and the value of interest are calculated to update the position of the center of the circle, and the new coordinates of the center of the circle P _i ′ are where n is the total number of points in C _i , _xj represents the longitude of the jth interest point, _yj represents the latitude of the jth interest point, _wj represents the degree of interest of the jth interest point, and n represents the interest point collection /> The number of points of interest in ;.

(2-4) With P _i ' as the center of the circle, repeat steps (2-2) and (2-3) until the distance between two consecutive calculations of P _i ' is less than the threshold d and stop the loop. At this time, the center of C _i is the latest P _i ′, and the interest value of C _i is Update the set of clustering results C=C∪C _i .

(2-5) Remove all points in C _i from the set of interest points P, and update the set of ground interest points Repeat steps (2-1) to (2-4) until P is empty.

(3) Plan the clustering results

(3-1) Assuming that the total scheduling time range is [T _S , T _E ], T _S and _TE represent the start time and end time of the analysis time period respectively; assuming that the satellite set Sat _k , 1≤k≤S , S represents the number of satellites, the roll angle β of the satellite moving along the fixed orbit is variable in [-Β, Β], the change step is Δβ, and within the time range of [T _S , T _E ], the roll angle β constant.

Assume that the satellite Sat _k has a set of strips _g on the ground, 1≤g≤G, G is the number of all possible target areas of the satellite Sat _k to the clustering result set C, and the observation start time of the satellite Sat _k corresponding to the strip _g The end time is ws _kg and we _kg , the maximum storage capacity is E _k , and e _k is the storage capacity consumed by satellite Sat _k per unit time.

The above steps (2-4) return the set of clustering circles C={C ₁ ,C ₂ ,...,C _N }, N is the number of all circles in the clustering result C, and the radius of the circles is r . C corresponds to the interest degree set v={v ₁ ,v ₂ ,...,v _N }, where v _i corresponds to the interest degree of circle C _i , 1≤i≤N.

(3-2) Define decision variables

Express the objective function as the sum of interest degrees corresponding to the circles observed by the satellite, and maximize it:

S is the number of satellites, G is the number of all possible target areas of the satellite Sat _k in the region R, v _i is the interest value of the corresponding circle C _i .

(3-3) The above optimization process needs to meet the following constraints:

Constraint 1: Within the planning time interval, each circular area is required to be observed by satellite at most once, namely:

Constraint 2: The satellite operates within the entire scheduling time range, and the observation time of any strip must meet the imaging time requirement.

T _S ≤ ws _kg ≤ we _kg ≤ T _E , 1 ≤ k ≤ S, 1 ≤ g ≤ G (3-3)

Constraint 3: In the planning time interval, satellite Sat _k can only select a strip under one roll angle when it passes through a single circle.

Constraint 4: Satellite storage capacity

(3-4) In view of the above objective functions and related constraints, this design uses Particle Swarm Optimization (PSO) to optimize the problem. The main process is as follows:

(3-4-1) Initialize satellites and strips, set target area scheduling time [T _S , T _E ], visible time window of selected satellites, particle swarm algorithm parameters, etc.

(3-4-2) Input the number of satellites, the number of corresponding strips, and related constraints into the calculation model as system variables, and correct the variables that violate the constraints (for example, each satellite can only select one roll angle under the same mission. The corresponding strip below), calculate the maximum degree of interest of the system and Weight as the current objective function value.

(3-4-3) Determine the current optimal combination pbest of satellites and stripes according to the objective function, that is, select the current optimal combination from random combination schemes.

(3-4-4) Determine the optimal combination gbest of satellites and stripes selected from all pbest sets in the iteration. Take gbest as the global optimum of the particle swarm, that is, the optimal combination of selected satellites and strips.

(3-4-5) Judging whether the maximum number of iterations (Max) is reached, if so, execute step (3-4-6), if not, return to step (3-4-2).

(3-4-6) Output the global optimal solution gbest, which is the objective function Weight.

To sum up, the present invention adjusts the center of clustering according to the importance of each interest point, so that the regional distribution of discrete interest points is more distinct, and by maximizing the sum of the corresponding interest degrees corresponding to the clustering results, it not only reduces planning The computational complexity of the algorithm can also prioritize the observation of areas where key points of interest appear with a high probability, so as to achieve higher overall observation income.

The invention also discloses a satellite earth observation discrete point of interest cluster planning device, which can be computer equipment or be set in the computer equipment. As shown in FIG. 2 , it includes: an information acquisition module 201 , an interest point clustering module 202 , a variable calculation module 203 and a cluster planning module 204 .

An information acquisition module 201, configured to acquire information on points of interest in the target area; wherein the information includes: longitude, latitude, and interest value;

Points of interest clustering module 202, configured to cluster the points of interest based on the information, and obtain the degree of interest of each category in the clustering results;

The variable calculation module 203 is used to obtain the stripe _kg of the satellite Sat _k in the target area according to the satellite orbit of the satellite Sat _k , and obtain the stripe kg according to whether the satellite Sat _k observes the stripe _kg The value of the decision variable x _kg ;

The cluster planning module 204 is configured to maximize the sum of the interest degrees of all satellites Sat _k observing the target area based on the interest degrees of each category and the value x _kg , so as to obtain the cluster planning results of interest points.

For the specific implementation process and beneficial effects of the device modules, please refer to the description of the above-mentioned method embodiments, and details will not be repeated here.

In an exemplary embodiment, there is also provided a computer device, the computer device includes a memory and a processor, a computer program is stored in the memory, and the computer program is loaded and executed by the processor, so as to realize the above-mentioned A clustering planning method for satellite Earth observation discrete points of interest.

In an exemplary embodiment, there is also provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned method for clustering planning of satellite earth observation discrete points of interest is implemented.

In an exemplary embodiment, a computer program product is also provided. When the computer program product is run on a computer device, the computer device is made to execute the above-mentioned method for clustering and planning satellite earth observation discrete points of interest.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limiting them; although the application has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: it can still Modifications are made to the technical solutions described in the above embodiments, or equivalent replacements are made to some of the technical features; these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims (8)

1. A satellite earth observation discrete point of interest clustering planning method, is characterized in that, said method comprises: Acquiring information on points of interest in the target area; wherein the information includes: longitude, latitude and interest value; Clustering the interest points based on the information, and obtaining the degree of interest of each category in the clustering result; wherein, the clustering of the interest points based on the information, and obtaining the degree of interest of each category in the clustering result include: In the current set of interest points, select any one of the interest points as the center of the circle; With a set value as the radius, a circle C _i is generated, and all interest points located in the circle C _i are determined to generate a set of interest points According to the collection of points of interest In the interest point information, update the center position of the circle _Ci , until the distance between the center positions of two adjacent updates is less than a threshold, based on the updated center position, obtain a subset of interest points; Obtain the interest degree of the interest point subset based on the interest value of each interest point in the interest point subset; deleting the subset of interest points from the current set of interest points, and judging whether the updated set of interest points is an empty set; In the case that the updated set of interest points is not an empty set, return to the current set of interest points and select any one of the interest points as the center of the circle; In the case that the updated set of interest points is an empty set, output the degree of interest of each subset of interest points; Obtain the stripe _g of the satellite Sat _k in the target area according to the satellite orbit of the satellite Sat _k , and obtain the stripe g of the stripe _g according to whether the satellite Sat _k observes the stripe _g The value of the decision variable x _kg ; Based on the interest of each category and the value x _kg of the decision variable, maximize the sum of the interest of all satellites Sat _k observing the target area to obtain the clustering planning result of the interest point; wherein, based on the The degree of interest of the category and the value x _kg of the decision variable maximize the sum of the degree of interest when all satellites Sat _k observe the target area, to obtain the clustering planning results of interest points, including: Based on the degree of interest and the value x _kg , construct an objective function to be optimized Wherein, S represents the quantity of the satellite Sat _k , G represents the quantity of the strip stripe _g , _{and v} represents the degree of interest of the corresponding circle C _i ; setting constraints on the objective function; The particle swarm optimization algorithm is used to optimize the objective function to be optimized to obtain the cluster planning result of the interest points. 2. The method according to claim 1, characterized in that according to the collection of interest points The longitude and latitude of the point of interest in the middle, update the center position of the circle _Ci , including: Based on the set of interest points The longitude and interest value of the interest point in the middle, calculate the longitude of the new center position Among them, x _j represents the longitude of the jth interest point, w _j represents the interest value of the jth interest point, and n represents the set of interest points/> the number of points of interest in Based on the set of interest points The latitude and interest value of the middle interest point, calculate the latitude of the new center position Among them, _yj represents the latitude of the jth interest point. 3. The method according to claim 1, characterized in that, whether the satellite Sat _k observes the stripe _g or not, obtaining the value x _kg of the decision variable comprises: In the case that the satellite Sat _k observes the stripe _g , the value of the decision variable x _kg =1; In the case that the satellite Sat _k does not observe the stripe _g , the value of the decision variable x _kg =0. 4. The method according to claim 1, wherein the constraints of the objective function include: The observation quantity constraint of the satellite Sat _k ; wherein, the observation quantity constraint represents that within the planning time interval, each circular area can only be observed once by the satellite Sat _k at most; and, The imaging time constraint of the satellite Sat _k ; wherein, the imaging time constraint indicates that the satellite Sat _k operates within the entire scheduling time range, and the observation time of any stripe stripe _g must meet the imaging time requirement; and, The roll angle constraint of the satellite Sat _k ; wherein, the roll angle constraint indicates that in the planning time interval, the satellite Sat _k can only select a strip under a roll angle when passing through a single clustering circle ; and, The storage capacity constraint of the satellite Sat _k . 5. The method according to claim 1, wherein said utilizing particle swarm optimization algorithm to optimize said objective function to be optimized, to obtain the point of interest cluster planning result, comprising: Initialize the satellite Sat _k and the strip stripe _g ; The number of satellites, the number of corresponding target areas, and related constraints are calculated as system variables, and the variables that violate the constraints are corrected to calculate the maximum interest of the system as the current objective function value; Determine the optimal combination pbest of the current satellite and the strip according to the objective function; Select the optimal combination gbest of satellites and strips from all iterations of the optimal combination pbest of satellites and strips; Determine whether the maximum number of iterations has been reached; If the maximum number of iterations is not reached, return to the calculation of the number of satellites, the number of corresponding target areas, and related constraints as system variables, and correct the variables that violate the constraints to calculate the maximum weight sum of the system as the current objective function value; When the maximum number of iterations is reached, the optimal combination gbest of satellites and strips is taken as the result of earth observation planning. 6. A satellite earth observation discrete point of interest cluster planning device, characterized in that the device comprises: An information acquisition module, configured to acquire information on points of interest in the target area; wherein the information includes: longitude, latitude, and interest value; The point of interest clustering module is used to cluster the points of interest based on the information, and obtain the degree of interest of each category in the clustering result; wherein, the clustering of the points of interest based on the information is used to obtain the interest of each category in the clustering result interests, including: In the current set of interest points, select any one of the interest points as the center of the circle; With a set value as the radius, a circle C _i is generated, and all interest points located in the circle C _i are determined to generate a set of interest points According to the collection of points of interest In the interest point information, update the center position of the circle _Ci , until the distance between the center positions of two adjacent updates is less than a threshold, based on the updated center position, obtain a subset of interest points; Obtain the interest degree of the interest point subset based on the interest value of each interest point in the interest point subset; deleting the subset of interest points from the current set of interest points, and judging whether the updated set of interest points is an empty set; In the case that the updated set of interest points is not an empty set, return to the current set of interest points and select any one of the interest points as the center of the circle; In the case that the updated set of interest points is an empty set, output the degree of interest of each subset of interest points; The variable calculation module is used to obtain the stripe _g of the satellite Sat _k in the target area according to the satellite orbit of the satellite Sat _k , and obtain a decision by whether the satellite Sat _k observes the stripe stripe _g the value of the variable x _kg ; The cluster planning module is used to maximize the sum of the interest degrees of all satellites Sat _k observing the target area based on the interest degrees of each category and the value x _kg , so as to obtain the cluster planning results of the points of interest; wherein, the Based on the interest degree of each category and the value x _kg of the decision variable, maximize the sum of interest degrees when all satellites Sat _k observe the target area, to obtain the cluster planning results of interest points, including: Based on the degree of interest and the value x _kg , construct an objective function to be optimized Wherein, S represents the quantity of the satellite Sat _k , G represents the quantity of the strip stripe _g , _{and v} represents the degree of interest of the corresponding circle C _i ; setting constraints on the objective function; The particle swarm optimization algorithm is used to optimize the objective function to be optimized to obtain the cluster planning result of the interest points. 7. A storage medium, wherein a computer program is stored in the storage medium, wherein the computer program is configured to execute the method according to any one of claims 1-5 when running. 8. An electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1-5.