A method for selecting the landing area on the entire moon based on evidence weight and fractals

Abstract

The present application relates to a method for selecting the site of the landing area for the whole moon based on evidence weight and fractals, including: obtaining the network mean layer corresponding to each evidence factor based on grid division; calculating the prior probability of each evidence factor, as well as the weight and posterior probability in each unit grid based on binary variable representation; determining the first threshold based on fractal theory, and determining the distribution complexity of the impact craters in each unit grid according to the first threshold; correcting the posterior probability of each evidence factor based on the index value of the unit grid; obtaining the posterior probability map through color rendering, and combining the expert screening layer and the impact crater database to select the site of the landing area for the whole moon. This method changes the traditional way of manually determining suitable landing areas according to scientific needs under engineering constraints, greatly reduces the workload and human intervention, and does not require a lot of professional knowledge to better define the landing range, which has reference significance for the site selection of landing areas in lunar exploration missions.

Description

Technical Field

[0001] The present application belongs to the field of image processing technology, and specifically relates to a method for selecting a landing area for the entire moon based on evidence weight and fractals. Background Art

[0002] The moon is the only natural satellite of the earth and the closest celestial body to us except for meteors and artificial celestial bodies. It naturally became the focus of early deep space exploration. In addition, lunar exploration is conducive to the development of science and technology. The exploration of the moon can not only reveal the mysteries of science, but also drive the development of science and technology in other fields, such as launch vehicles, artificial intelligence, robots, remote control operations, hypersonic flight, etc., which can bring great benefits to the national economy.

[0003] At present, the main method for selecting the landing area for the whole moon at home and abroad is for experts to determine the appropriate landing area based on scientific needs and engineering constraints. Although this can delineate the scope of the landing area, it is a large workload and there are many human factors involved. Therefore, a lot of professional knowledge is usually required to better select the landing area.

[0004] In view of this, a full-moon landing area site selection method based on evidence weight and fractal is provided to reduce the workload and human intervention in the full-moon landing site selection process to a certain extent. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present application provides a method for selecting a landing area for the entire moon based on evidence weight and fractals.

[0007] (II) Technical solution

[0008] In order to achieve the above objectives, this application adopts the following technical solutions:

[0009] In the first aspect, the present application provides a method for selecting a landing area for the entire moon based on evidence weight and fractals, comprising:

[0010] S1. Divide the entire moon into grids, calculate the number of impact craters contained in each unit grid, and obtain the network mean layer corresponding to each evidence factor by calculating the grid mean;

[0011] The grid mean is the mean of the pixel points in the unit grid;

[0012] The evidence factors include lunar crust thickness, roughness map, slope, digital elevation model, gravity gradient, iron oxide content distribution and lunar soil optical maturity;

[0013] S2, using binary variables to represent each network mean layer and target layer, and calculating the prior probability O 先验 , and the weight of each evidence factor in each unit grid and the posterior probability P 后验 ;

[0014] The target layer includes 50 landing sites, including Apollo 11, 12, 13, 14, 15, 16, 17, Luna 16, 17, 20, 21, 23, 24, Surveyor 1, 2, 3, 5, 6, 7, Chang'E3, Ranger 6, 7, 8, 9, CRAIL and LADEE;

[0015] S3. Based on fractal theory, the total number A of unit grids containing more than S impact craters in all unit grids is counted, and a first threshold is determined by fitting a log(S)-log(A) double logarithmic graph, and the distribution complexity of the impact craters in each unit grid is determined according to the first threshold;

[0016] Where S = 0, 1, ..., l, and A = A 0 ,A 1 ,...,A l ;

[0017] S4, based on the index values ​​of the first unit grid and the second unit grid, reducing the posterior probability P of the first unit grid 后验 , increase the posterior probability P of the second unit grid 后验 , to obtain the modified posterior probability P 修正后验 ;

[0018] The first unit grid is a unit grid in which the number of impact craters S is greater than the inflection point value T;

[0019] The second unit grid is a unit grid with 0 impact craters;

[0020] The index value is a value set to distinguish different grids when dividing the grid based on the entire moon range;

[0021] S5, by modifying the posterior probability P 修正后验 Color rendering is performed to obtain a posterior probability map, which is then combined with the expert screening layer and the impact crater database to select the landing area for the entire moon.

[0022] Optionally, the S1 includes:

[0023] S11, reading the pixels of each evidence layer into a two-dimensional matrix;

[0024] The evidence layer is a layer corresponding to each evidence factor;

[0025] S12, taking the common divisor of the number of rows and columns of the pixel matrix of each evidence layer as the size of the divided grid, and dividing the area within the entire moon into grids;

[0026] The pixel matrix is ​​a matrix formed by pixel points in a unit grid;

[0027] S13, if the pixel value of the evidence layer cannot be divided by the grid size, the edge pixel points of the evidence layer are cropped;

[0028] S14. Based on the function cal_avi, traverse the pixel points in each unit grid in each evidence layer, calculate the mean value of each unit grid pixel point, so as to obtain the grid mean layer corresponding to each evidence factor.

[0029] Optionally, S2 includes:

[0030] S21, representing the grid mean layer corresponding to each evidence factor as a 0-1 variable, and determining the unit grid where each landing point is located based on the coordinates of the landing point, and representing the determined unit grid with a binary variable;

[0031] S22, determine the prior probability P according to the total number of unit grids and the number of unit grids containing landing points 先验 , calculate the prior probability O by the weight of evidence method 先验 , and the weight of each evidence factor in each unit grid and the posterior probability P 后验 .

[0032] Optionally, the S21 includes:

[0033] S211, traversing the grid means in each grid mean layer, judging according to the second threshold, if the grid mean is greater than / less than the second threshold, setting the unit grid corresponding to the grid mean to "0 / 1" or "1 / 0";

[0034] The second threshold is a threshold determined based on expert knowledge;

[0035] S212, adjusting the coordinate range of the landing point from -90° to 90° to 0° to 180°, and adjusting -180° to 180° to 0° to 360°;

[0036] S213. Based on the coordinates of the landing point, traverse the divided grids to determine whether the unit grid contains the landing point; by setting the unit grid with the landing point to "1" and setting the unit grid without the landing point to "0", a landing point grid layer is obtained.

[0037] Optionally, the S22 includes:

[0038] S221, traverse the coordinates of the landing points represented by the binary variables, calculate the total number m of the unit grids represented as "1" in the landing point grid layer, and Calculate the prior probability P 先验 , where n is the number of total unit grids divided in the whole month range, through Calculate the prior probability O 先验 .

[0039] Optionally, the S22 further includes:

[0040] S222, traverse the divided grids, and determine the status of the grid mean layer and the landing point grid layer corresponding to the positions of different unit grids, specifically:

[0041] Count the number of unit grids that are set to "1" in both the grid mean layer and the landing point network layer, and record the number of corresponding unit grids as count_BD;

[0042] Count the number of unit grids that are set to "0" in both the grid mean layer and the landing point network layer, and record the number of corresponding unit grids as count_nBnD;

[0043] Count the number of unit grids that are set to "1" in the grid mean layer and are set to "0" in the landing point network layer, and record the number of corresponding unit grids as count_BnD;

[0044] Count the number of unit grids that are set to "0" in the grid mean layer and are set to "1" in the landing point network layer, and record the number of corresponding unit grids as count_nBD;

[0045] S223. When all the network layers of the landing point are set to "1", as well as Calculate the first weight w + ;

[0046] S224. When the network layer of the landing point is set to "0", as well as Calculate the second weight w - ;

[0047] S225, perform the above processing on the network mean layer corresponding to each evidence factor, and combine the first weight w corresponding to each evidence factor + and the second weight w - , get the weight of each evidence factor in each unit grid

[0048] Wherein, k represents the divided unit grid, k=1,2,...n; j represents the evidence factor, j=1,2,...7.

[0049] Optionally, the S22 further includes:

[0050] S226. Calculate the weight superposition and weight of each evidence factor in each unit grid pass Calculate Q by Calculate the posterior probability P 后验 ;

[0051] in,

[0052] Optionally, the S3 includes:

[0053] The intersection of two extended slope lines in the log(S)-log(A) double logarithmic graph is obtained, and the abscissa value of the intersection coordinate is used as the first threshold.

[0054] Optionally, after S3 and before S4, it also includes:

[0055] Based on the distribution complexity of impact craters in each unit grid, the unit grids are classified into first unit grids and second unit grids.

[0056] Optionally, the S4 includes:

[0057] S41, determining a first unit grid and a second unit grid according to index values ​​of the first unit grid and the second unit grid;

[0058] S42, based on Reduce the posterior probability P of the first unit grid 后验 ,based on Raise the posterior probability P of the second unit grid 后验 , correct the posterior probability P of each evidence factor in each unit grid 后验 , to obtain the modified posterior probability P 修正后验 ;

[0059] Among them, α 1 and α 2 is the absolute value of the slope obtained based on the log(S)-log(A) double logarithmic plot, α 1 >α 2 .

[0060] (III) Beneficial effects

[0061] The landing area site selection by the method described in this application has changed the traditional way in which experts manually determined suitable landing areas based on scientific needs under engineering constraints. It greatly reduces the workload and human intervention and does not require a lot of professional knowledge to better define the landing range. It has reference significance for the landing area site selection in lunar exploration missions. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The present application is described with the aid of the following drawings:

[0063] Figure 1 The figure is a schematic diagram of the calculation process of the full moon landing area site selection method based on evidence weight and fractals;

[0064] Figure 2 The overall process diagram of the moon landing site selection method based on evidence weight and fractals is shown;

[0065] Figure 3 It is a schematic diagram of the posterior probability results of two evidence factors;

[0066] Figure 4 The following is a schematic diagram of the posterior probability results of the five evidence factors;

[0067] Figure 5 The following is a schematic diagram of the posterior probability results of the seven evidence factors;

[0068] Figure 6 A schematic diagram of the simulation of the power of evidence;

[0069] Figure 7 This is the distribution map of impact craters on the whole moon;

[0070] Figure 8 is a posterior probability map based on reducing the posterior probability of the first unit grid;

[0071] Fig. 9 is a posterior probability map based on raising the posterior probability of the second unit grid. DETAILED DESCRIPTION

[0072] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below in conjunction with the accompanying drawings through specific implementation methods. It is understood that the specific embodiments described below are only used to explain the relevant inventions, rather than to limit the invention. It should also be noted that the embodiments and features in the embodiments of this application can be combined with each other in the absence of conflict; for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0073] Embodiment 1 provides a method for selecting a landing area for the entire moon based on evidence weight and fractal. Figure 1 As shown, specifically:

[0074] In specific applications, the evidence factors include lunar crust thickness, roughness map, slope, digital elevation model, gravity gradient, iron oxide content distribution and lunar soil optical maturity, and the target layers include the landing sites;

[0075] In specific applications, data preprocessing is used to complete the division of the full-month grid, the clipping of edge pixels of the evidence layer, the calculation of the network mean, and the binary variable representation of the evidence layer and the target layer.

[0076] In specific applications, the prior probability, the weight of each evidence factor in each unit grid and the posterior probability are calculated based on the evidence weight method;

[0077] In a specific application, based on fractal theory, by obtaining double logarithmic graph information, linear regression fitting is performed to calculate the first threshold;

[0078] In a specific application, based on the first threshold, the posterior probability of each evidence factor in each unit grid is corrected to obtain a corrected posterior probability;

[0079] In specific applications, the posterior probability map is obtained based on the modified posterior probability, and the posterior probability map is combined with the expert screening layer and crater database to select the landing area for the entire moon.

[0080] The landing area site selection method described in Example 1 has changed the traditional way in which experts manually determined suitable landing areas based on scientific needs under engineering constraints, greatly reduced the workload and human intervention, and can better define the landing range without a lot of professional knowledge, which has reference significance for the landing area site selection in lunar exploration missions.

[0081] Embodiment 2 provides a method for selecting a landing area for the entire moon based on evidence weight and fractal. Figure 2 As shown, the specific steps are as follows:

[0082] S1. Divide the entire moon into grids, calculate the number of impact craters contained in each unit grid, and obtain the network mean layer corresponding to each evidence factor by calculating the grid mean;

[0083] In this embodiment, the full moon range used is 20037.4 km*40074.4 km, which is divided into 72*144 grids, that is, the size of each grid is set to 278.3 km*278.3 km.

[0084] In this embodiment, the grid mean is the mean of the pixels within the unit grid, and the evidence factors include lunar crust thickness, roughness map, slope, digital elevation model, gravity gradient, iron oxide content distribution, and lunar soil optical maturity;

[0085] Based on the aforementioned step S1, in the specific processing process, first, the pixels of each evidence layer need to be read into a two-dimensional matrix, wherein the evidence layer is the layer corresponding to each evidence factor.

[0086] Second, the common divisor of the number of rows and columns of the pixel matrix of each evidence layer needs to be used as the size of the divided grid, and the area within the entire moon needs to be gridded. Specifically, the pixel matrix is ​​the matrix formed by the pixel points in the unit grid;

[0087] Third, if the pixel value of the evidence layer cannot be divided by the grid size, the edge pixels of the evidence layer are cropped;

[0088] For example, the pixel points of the evidence layer corresponding to the evidence factor lunar crust thickness are 721*1441, which is cropped to 720*1440 to be divisible by the common divisor 72*144; and the pixel matrix of the evidence layer corresponding to the evidence factor slope is 46080*92160, which is divisible by the common divisor 72*144, so there is no need to perform edge pixel cropping operations on it.

[0089] Fourth, based on the function cal_avi, it is necessary to traverse the pixels in each unit grid in each evidence layer and calculate the mean value of each unit grid pixel to obtain the grid mean layer corresponding to each evidence factor;

[0090] In the specific processing process, a function cal_avi is written to calculate the mean value of pixels in the unit grid. According to the size of the divided grid, the unit grid moves from left to right and from top to bottom in turn, traversing the pixels in the unit grid, so as to calculate the mean value of the pixels in each unit grid. In addition, the corresponding parameter of each evidence factor is input into the function cal_avi for the same processing, and the grid mean layer of each evidence layer is obtained.

[0091] During the specific processing, the unit grid moves from left to right and from top to bottom in turn, and the specific process of selecting the selected pixel point set in the unit grid is: when 0<i<number of grid rows, 0<j<number of grid columns, it is judged that (number of pixel rows / number of grid rows)*i<pixel matrix row index<(number of pixel rows / number of grid rows)*(i+1) and (number of pixel columns / number of grid columns)*j<pixel matrix row index<(number of pixel columns / number of grid columns)*(j+1).

[0092] S2, using binary variables to represent each network mean layer and target layer, and calculating the prior probability O 先验 , and the weight of each evidence factor in each unit grid and the posterior probability P 后验 ;

[0093] In this embodiment, the target layer includes 50 landing sites, including Apollo 11, 12, 13, 14, 15, 16, 17, Luna 16, 17, 20, 21, 23, 24, Surveyor 1, 2, 3, 5, 6, 7, Chang'E3, Ranger 6, 7, 8, 9, CRAIL and LADEE;

[0094] Based on the aforementioned step S2, in the specific processing process, it is necessary to represent the grid mean layer corresponding to each evidence factor as a 0-1 variable, and determine the unit grid where each landing point is located based on the coordinates of the landing point, and use a binary variable to represent the determined unit grid. The above steps are described in detail below:

[0095] First, it is necessary to traverse the grid means in each grid mean layer and make a judgment based on the second threshold. If the grid mean is greater than / less than the second threshold, the unit grid corresponding to the grid mean is set to "0 / 1" or "1 / 0". The second threshold is a threshold determined based on expert knowledge.

[0096] Second, the coordinate range of the landing point needs to be adjusted from -90° to 90° to 0° to 180°, and from -180° to 180° to 0° to 360°;

[0097] Third, based on the coordinates of the landing point, it is necessary to traverse the divided grids to determine whether the unit grid contains the landing point; by setting the unit grid with the landing point to "1" and the unit grid without the landing point to "0", the landing point grid layer is obtained;

[0098] And, in the specific processing process, the specific process of traversing the divided grids and judging whether the unit grid contains a landing point is: when 0<i<number of grid rows, 0<j<number of grid columns, judge that (180 / number of grid rows)*i<x coordinate of the landing point<(180 / number of grid rows)*(i+1) and (360 / number of grid columns)*j<y coordinate of the landing point<(360 / number of grid columns)*(j+1).

[0099] Based on the aforementioned step S2, in the specific processing process, it is also necessary to determine the prior probability P according to the number of total unit grids and the number of unit grids containing landing points. 先验 , calculate the prior probability O by the weight of evidence method 先验 , and the weight of each evidence factor in each unit grid and the posterior probability P 后验 , the above steps are described in detail below:

[0100] For better understanding, in a possible implementation process, the prior probability P is determined based on the number of total unit grids and the number of unit grids containing landing points. 先验 , and the prior probability O calculated by the weight of evidence method 先验 The specific steps are as follows:

[0101] Traverse the coordinates of the landing points represented by the binary variables, calculate the total number m of the unit grids represented as "1" in the landing point grid layer, and use Calculate the prior probability P 先验 , where n is the number of total unit grids divided in the whole month range, through Calculate the prior probability O 先验 .

[0102] For better understanding, in one possible implementation, the weight of each evidence factor in each unit grid is calculated by the evidence weight method. The specific steps are as follows:

[0103] First, traverse the divided grids to determine the grid mean layer and the state of the landing point grid layer corresponding to the different unit grids, specifically:

[0104] Count the number of unit grids that are set to "1" in both the grid mean layer and the landing point network layer, and record the number of corresponding unit grids as count_BD;

[0105] Count the number of unit grids that are set to "0" in both the grid mean layer and the landing point network layer, and record the number of corresponding unit grids as count_nBnD;

[0106] Count the number of unit grids that are set to "1" in the grid mean layer and are set to "0" in the landing point network layer, and record the number of corresponding unit grids as count_BnD;

[0107] Count the number of unit grids that are set to "0" in the grid mean layer and are set to "1" in the landing point network layer, and record the number of corresponding unit grids as count_nBD;

[0108] Second, when the network layer of the landing point is set to "1", as well as Calculate the first weight w + ;

[0109] Third, when the network layer of the landing point is set to "0", as well as Calculate the second weight w - ;

[0110] Fourth, perform the above processing on the network mean layer corresponding to each evidence factor, and combine the first weight w corresponding to each evidence factor + and the second weight w - , get the weight of each evidence factor in each unit grid

[0111] Wherein, k represents the divided unit grid, k=1,2,...n; j represents the evidence factor, j=1,2,...7.

[0112] For a better understanding, in a possible implementation process, the posterior probability P of each evidence factor in each unit grid is calculated by the evidence weight method. 后验 The specific steps are as follows:

[0113] Calculate the weight superposition and sum of each evidence factor in each unit grid pass Calculate Q by Calculate the posterior probability P 后验 ;in,

[0114] In addition, in this embodiment, the posterior probability P obtained by calculation 后验 The larger it is, the more suitable it is for landing.

[0115] In this embodiment, Figure 3 It is a schematic diagram of the posterior probability results of two evidence factors. Figure 4 is a schematic diagram of the posterior probability results of the five evidence factors, and Figure 5 Schematic diagram of the posterior probability results of seven evidence factors.

[0116] In this embodiment, Figure 6 This is a schematic diagram of the evidence weight simulation, with 5 evidence layers as an example. The entire moon is divided into 3*3 squares, and the black dots are the landing points. For the middle unit grid, since the unit grid contains the landing point, the evidence factor of the corresponding position of the unit grid indicates that the landing is favorable.

[0117] S3. Based on fractal theory, the total number A of unit grids containing more than S impact craters in all unit grids is counted, and a first threshold is determined by fitting a log(S)-log(A) double logarithmic graph, and the distribution complexity of the impact craters in each unit grid is determined according to the first threshold;

[0118] Where S = 0, 1, ..., l, and A = A 0 ,A 1 ,...,A l ;

[0119] In this embodiment, Figure 7 This is a distribution map of impact craters across the moon.

[0120] In the specific processing process, the divided grids are traversed to determine the total number of cells S in the unit grid that is greater than the number of craters. Specific process: when 0<i<number of grid rows, 0<j<number of grid columns, determine that (180 / number of grid rows)*i<x coordinate of the center of the crater<(180 / number of grid rows)*(i+1) and (360 / number of grid columns)*j<y coordinate of the center of the crater<(360 / number of grid columns)*(j+1).

[0121] Based on the aforementioned step S3, in the specific processing process, it is necessary to obtain the intersection of the two slope extension lines in the log(S)-log(A) double logarithmic graph, and use the horizontal coordinate value of the intersection coordinate as the first threshold.

[0122] In addition, in a specific application, after step S3 and before step S4, the unit grids need to be classified into first unit grids and second unit grids based on the distribution complexity of the impact craters in each unit grid.

[0123] S4, based on the index values ​​of the first unit grid and the second unit grid, reducing the posterior probability P of the first unit grid 后验 , increase the posterior probability P of the second unit grid 后验 , to obtain the modified posterior probability P 修正后验 ;

[0124] In this embodiment, the first unit grid is a unit grid in which the number of impact craters S is greater than the inflection point value T, and the second unit grid is a unit grid in which the number of impact craters is 0;

[0125] In this embodiment, the index value is a value set to distinguish different grids when the grid is divided based on the entire month range;

[0126] Based on the aforementioned step S4, in a specific processing process, first, the first unit grid and the second unit grid need to be determined according to the index values ​​of the first unit grid and the second unit grid;

[0127] Second, based on Reduce the posterior probability P of the first unit grid 后验 ,based on Raise the posterior probability P of the second unit grid 后验 , correct the posterior probability P of each evidence factor in each unit grid 后验 , to obtain the modified posterior probability P 修正后验 ;

[0128] Among them, α 1 and α 2is the absolute value of the slope obtained based on the log(S)-log(A) double logarithmic plot, α 1 >α 2 .

[0129] In this embodiment, Figure 8 is the posterior probability map based on reducing the posterior probability of the first unit grid, and Fig. 9 is a posterior probability map based on raising the posterior probability of the second unit grid.

[0130] For example, taking the full moon divided into 72*144 squares as an example, the maximum number of impact craters in the unit grid is calculated to be 17, and the minimum is 0. The probability of the cell with the number of impact craters equal to 0 increases, and the probability of the cell with the number of impact craters greater than the threshold decreases. The index values ​​of the unit grids where these situations are located are obtained, and then the calculated posterior probability P is changed accordingly. 后验 .

[0131] S5, by modifying the posterior probability P 修正后验 Perform color rendering to obtain a posterior probability map, combine the posterior probability map with the expert screening layer and the impact crater database to select the site of the landing area across the moon;

[0132] Based on step S5, in a possible implementation process, in order to make the result more intuitive, the calculated modified posterior probability P 修正后验 Color rendering is performed, specifically: rendering the attached figure in blue to obtain a posterior probability map, in which the darker the blue, the more suitable it is for landing.

[0133] For example, Figure 8 and Fig. 9 This is the result of superimposing the posterior probability map with the expert screening layer and the crater database. It can be clearly seen that the pre-selected landing points selected by experts are mainly distributed in the unit grids with higher corrected posterior probabilities, and the crater distribution complexity in areas with higher posterior probabilities is relatively low.

[0134] The landing area site selection method described in Example 2 has changed the traditional way in which experts manually determined suitable landing areas based on scientific needs under engineering constraints, greatly reduced the workload and human intervention, and can better define the landing range without a lot of professional knowledge, which has reference significance for the landing area site selection in lunar exploration missions.

[0135] It should be noted that in the claims, any figure marks between brackets should not be understood as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "one" or "an" preceding a component does not exclude the presence of multiple such components. In addition, it should be noted that in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" and the like refers to the specific features, structures, materials or characteristics described in conjunction with the embodiment or example included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0136] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0137] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A method for selecting a landing area for the entire moon based on evidence weight and fractals, characterized in that: The site selection method includes: S1. Divide the entire moon into grids, calculate the number of impact craters contained in each unit grid, and obtain the network mean layer corresponding to each evidence factor by calculating the grid mean; The grid mean is the mean of the pixel points in the unit grid; The evidence factors include lunar crust thickness, roughness map, slope, digital elevation model, gravity gradient, iron oxide content distribution and lunar soil optical maturity; S2, using binary variables to represent each network mean layer and target layer, and calculating the prior probability O 先验 , and the weight of each evidence factor in each unit grid and the posterior probability P 后验 ; Wherein, k represents the index value of the divided unit grid, k = 1, 2, ... n; j represents the evidence factor, j = 1, 2, ... 7; The target layer contains 50 landing sites, including Apollo 11, 12, 13, 14, 15, 16, 17, Luna 16, 17, 20, 21, 23, 24, Surveyor 1, 2, 3, 5, 6, 7, Chang'E3, Ranger 6, 7, 8, 9, CRAIL and LADEE. S3. Based on fractal theory, the total number A of unit grids containing more than S impact craters in all unit grids is counted, and a first threshold is determined by fitting a log(S)-log(A) double logarithmic graph, and the distribution complexity of the impact craters in each unit grid is determined according to the first threshold; Where S = 0, 1, ..., l, and A = A0, A1, ..., A l ; S4, based on the index values ​​of the first unit grid and the second unit grid, reducing the posterior probability P of the first unit grid 后验 , increase the posterior probability P of the second unit grid 后验 , to obtain the modified posterior probability P 修正后验 ; The first unit grid is a unit grid in which the number S of impact craters is greater than a first threshold; The second unit grid is a unit grid with 0 impact craters; The index value is a value set to distinguish different grids when dividing the grid based on the entire moon range; S5, by modifying the posterior probability P 修正后验 Perform color rendering to obtain a posterior probability map, combine the posterior probability map with the expert screening layer and the impact crater database to select the site of the landing area across the moon; Wherein, the S3 includes: The intersection of two extended slope lines in the log(S)-log(A) double logarithmic graph is obtained, and the abscissa value of the intersection coordinate is used as the first threshold.

2. The site selection method according to claim 1, characterized in that: The S1 includes: S11, reading the pixels of each evidence layer into a two-dimensional matrix; The evidence layer is a layer corresponding to each evidence factor; S12, taking the common divisor of the number of rows and columns of the pixel matrix of each evidence layer as the size of the divided grid, and dividing the area within the entire moon into grids; The pixel matrix is ​​a matrix formed by pixel points in a unit grid; S13, if the pixel value of the evidence layer cannot be divided by the grid size, the edge pixel points of the evidence layer are cropped; S14. Based on the function cal_avi, traverse the pixel points in each unit grid in each evidence layer, calculate the mean value of each unit grid pixel point, so as to obtain the grid mean layer corresponding to each evidence factor.

3. The site selection method according to claim 1, characterized in that: The S2 includes: S21, representing the grid mean layer corresponding to each evidence factor as a 0-1 variable, and determining the unit grid where each landing point is located based on the coordinates of the landing point, and representing the determined unit grid with a binary variable; S22, determine the prior probability P according to the total number of unit grids and the number of unit grids containing landing points 先验 , calculate the prior probability O by the weight of evidence method 先验 , and the weight of each evidence factor in each unit grid and the posterior probability P 后验 .

4. The site selection method according to claim 3, characterized in that: The S21 includes: S211, traversing the grid mean in each grid mean layer, judging according to the second threshold, if the grid mean is greater than / less than the second threshold, setting the unit grid corresponding to the grid mean to "0 / 1" or "1 / 0"; The second threshold is a threshold determined based on expert knowledge; S212, adjusting the coordinate range of the landing point from -90° to 90° to 0° to 180°, and adjusting -180° to 180° to 0° to 360°; S213. Based on the coordinates of the landing point, traverse the divided grids to determine whether the unit grid contains the landing point; by setting the unit grid with the landing point to "1" and the unit grid without the landing point to "0", a landing point grid layer is obtained.

5. The site selection method according to claim 3, characterized in that: The S22 includes: S221, traverse the coordinates of the landing points represented by the binary variables, calculate the total number m of the unit grids represented by "1" in the landing point grid layer, and Calculate the prior probability P 先验 , where n is the number of total unit grids divided in the whole month range, through Calculate the prior probability O 先验 .

6. The site selection method according to claim 5, characterized in that: The S22 further includes: S222, traverse the divided grids, and determine the status of the grid mean layer and the landing point grid layer corresponding to the positions of different unit grids, specifically: Count the number of unit grids that are set to "1" in both the grid mean layer and the landing point network layer, and record the number of corresponding unit grids as count_BD; Count the number of unit grids that are set to "0" in both the grid mean layer and the landing point network layer, and record the number of corresponding unit grids as count_nBnD; Count the number of unit grids that are set to "1" in the grid mean layer and are set to "0" in the landing point network layer, and record the number of corresponding unit grids as count_BnD; Count the number of unit grids that are set to "0" in the grid mean layer and are set to "1" in the landing point network layer, and record the number of corresponding unit grids as count_nBD; S223. When all the network layers of the landing point are set to "1", as well as Calculate the first weight w S224. When the network layer of the landing point is set to "0", as well as Calculate the second weight w - ; S225, perform the above processing on the network mean layer corresponding to each evidence factor, and combine the first weight w corresponding to each evidence factor + and the second weight w _ , get the weight of each evidence factor in each unit grid 7. The site selection method according to claim 6, characterized in that: The S22 further includes: S226. Calculate the weight superposition and weight of each evidence factor in each unit grid pass Calculate Q by Calculate the posterior probability P 后验 ; in, 8. The site selection method according to claim 1, characterized in that: After S3 and before S4, it also includes: Based on the distribution complexity of impact craters in each unit grid, the unit grids are classified into first unit grids and second unit grids.

9. The site selection method according to claim 1, characterized in that: The S4 includes: S41, determining a first unit grid and a second unit grid according to index values ​​of the first unit grid and the second unit grid; S42, based on Reduce the posterior probability P of the first unit grid 后验 ,based on Raise the posterior probability P of the second unit grid 后验 , correct the posterior probability P of each evidence factor in each unit grid 后验 , to obtain the modified posterior probability P 修正后验 ; Wherein, α1 and α2 are absolute values ​​of slopes obtained based on a log(S)-log(A) double logarithmic graph, and α1>α2.

Images