CN-122022420-A - Unmanned aerial vehicle selection decision-making method and system for basic mapping

CN122022420ACN 122022420 ACN122022420 ACN 122022420ACN-122022420-A

Abstract

The invention discloses an unmanned aerial vehicle model selection decision-making method and system for basic mapping, and relates to the technical field of unmanned aerial vehicle model selection; the method comprises the steps of grading and assigning each qualitative index, respectively standardizing positive, negative and interval indexes, carrying out atmospheric density correction on a height index to form a standardized positive matrix, introducing scene constraint factors to construct an improved probability matrix, calculating information entropy and improved difference coefficient, fusing priori weights to obtain a self-adaptive entropy weight vector, combining the standardized matrix and the weights to construct a weighted decision matrix, determining ideal solutions and negative ideal solutions, calculating group utility values, individual regrets and compromise indexes, and sequencing, carrying out advantage and stability inspection on the first scheme, and outputting a final selection result. The invention effectively improves the scientificity and accuracy of selection, reduces the purchasing cost and improves the utilization rate of the unmanned aerial vehicle.

Inventors

PENG JIANQIU
HU YUNHUA
LIU BIN
LI YALING
WANG WANGANG
SUN ZHENGCHAO
QIAN YUCHEN
LIU DONGLI
LIU HANG

Assignees

自然资源部第六地形测量队(自然资源部地下管线勘测工程院、四川省第三测绘工程院)

Dates

Publication Date: 20260512
Application Date: 20260416

Claims (10)

1. The unmanned aerial vehicle model selection decision-making method for basic mapping is characterized by comprising the following steps of: Step S1, constructing an index system consisting of a flight demand index, an unmanned aerial vehicle and a load parameter index, an environment demand index and other factor indexes according to basic mapping business requirements; Step S2, carrying out grading assignment on each qualitative index, respectively adopting corresponding standardization treatment on each positive index, each negative index and each interval type index to obtain a standardized evaluation value, carrying out atmospheric density correction on each height type index to obtain a standardized evaluation value, and constructing a standardized positive matrix by the standardized evaluation values of all the indexes; s3, introducing scene reference constraint factors to construct an improved probability distribution matrix, calculating information entropy values of all indexes, calculating improved difference coefficients of all indexes through difference adjustment factors, fusing expert priori weights of all indexes to obtain self-adaptive entropy value weights of all indexes, and forming weight vectors by the self-adaptive entropy value weights of all indexes; S4, multiplying each element in the standardized forward matrix by a corresponding self-adaptive entropy weight to obtain a weighted standardized decision matrix, determining an ideal solution and a negative ideal solution of each index, calculating a group utility value and an individual regressive value of each candidate scheme, calculating a compromise index of each candidate scheme through a self-adaptive decision coefficient, and sequencing the candidate schemes from small to large according to the compromise index; And S5, checking the acceptable advantage condition and the acceptable stability condition of the first-ranked candidate solution, and outputting a final type selection decision result according to the checking result.
2. The method of claim 1, wherein in step S1, the flight demand class indicators include minimum personnel configuration, complexity of ancillary equipment, operating condition level, and deployment time, the unmanned aerial vehicle and load parameter class indicators include cruise speed, endurance time, maximum takeoff weight, lift-off altitude, controlled distance, size, and sag height, the environmental demand class indicators include maximum takeoff altitude, minimum environmental temperature, maximum environmental temperature, optimal operational humidity range, wind resistance level, and weather condition level, and the other factor class indicators include market price, handling capability level, insurance claim level, after-market service level, and application scenario coverage.
3. The method according to claim 2, wherein said step S2 comprises the steps of: Step S2.1, carrying out grading assignment on each qualitative index according to a preset grading scale, and mapping an assignment range to a range from 0 to 1; S2.2, adopting a standardized formula for taking the ratio of the original index value to the ideal reference value and a minimum value for each positive index, and adopting a standardized formula for taking the ratio of the ideal reference value to the original index value and a minimum value for each negative index; S2.3, adopting three-section standardization processing for each section type index, wherein when the original index value is lower than the lower limit of the optimal section, the standardization evaluation value is a ratio of subtracting the difference between the lower limit of the optimal section and the original index value and dividing the difference between the lower limit of the optimal section and the minimum value of all candidate schemes, when the original index value is positioned in the optimal section, the standardization evaluation value is one, and when the original index value is higher than the upper limit of the optimal section, the standardization evaluation value is a ratio of subtracting the difference between the original index value and the upper limit of the optimal section and dividing the difference between the maximum value of all candidate schemes and the upper limit of the optimal section, and each section type index comprises the lowest environmental temperature, the highest environmental temperature, the optimal operation humidity range and the cruising speed; S2.4, adopting standardized processing of correction of an atmospheric density ratio for each altitude index, wherein each altitude index comprises a maximum take-off altitude and an elevation limit altitude, and the atmospheric density ratio is calculated by adopting an international standard atmospheric model according to the altitude of a planned operation area; And step S2.5, combining the standardized evaluation values of all indexes obtained in the steps S2.1 to S2.4 to form a standardized forward matrix.
4. A method according to claim 3, wherein said step S3 comprises the steps of: S3.1, constructing an improved probability distribution matrix, wherein the calculation mode of elements in the matrix is to add the products of the scene reference adjustment coefficients of corresponding elements in the standardized forward matrix and corresponding indexes of the elements and the mapping industry threshold factors and divide the sum of all elements in the column where the elements are located and the corresponding constraint factors; S3.2, calculating the information entropy value of each index, wherein the information entropy value is equal to the sum of the quotient of the negative one divided by the natural logarithm of the total number of the candidate schemes multiplied by the sum of the product of the improved probability value of each candidate scheme corresponding to the index and the natural logarithm of the index; s3.3, calculating an improved difference coefficient of each index, wherein the improved difference coefficient is equal to the power of a difference adjustment factor of a difference value of subtracting the entropy value of the index information; S3.4, calculating the self-adaptive entropy weight of each index, wherein the self-adaptive entropy weight is equal to the ratio of the sum of the improved difference coefficient of the index and the priori information confidence coefficient multiplied by the priori weight of the index expert divided by the sum of the improved difference coefficient of all the indexes and the confidence coefficient correction term; And S3.5, constructing a weight vector by the self-adaptive entropy weight of all indexes.
5. The method according to claim 4, wherein the step S4 specifically comprises the steps of: S4.1, multiplying each element in the standardized forward matrix by the self-adaptive entropy weight of the corresponding index of the element to obtain a weighted standardized decision matrix; s4.2, taking the maximum value in all candidate schemes as an ideal solution of the index for each index, and taking the minimum value in all candidate schemes as a negative ideal solution of the index; S4.3, calculating a group utility value of each candidate scheme, wherein the group utility value is equal to the ratio of the weight of the adaptive entropy value of the index after dividing the difference between the ideal solution and the weighted evaluation value of the scheme under each index by the difference between the ideal solution and the negative ideal solution, taking the nonlinear compression factor corresponding to the load type as the power of the index, and summing all indexes; S4.4, calculating individual regrets of each candidate scheme, wherein the individual regrets are equal to the ratio of the difference between the ideal solution and the weighted evaluation value of the scheme under each index to the weight of the adaptive entropy value of the index to be the maximum value after dividing the difference between the ideal solution and the negative ideal solution; step S4.5, calculating a compromise index of each candidate scheme, wherein the compromise index is equal to the sum of a normalized value of a group utility value of the adaptive decision coefficient multiplied by the scheme and a normalized value of an individual reglet value subtracted by the adaptive decision coefficient multiplied by the scheme, and the adaptive decision coefficient is calculated by dividing one by a sensitivity parameter with a negative plus natural constant index multiplied by the power of the difference between the group utility value average of all schemes and the individual reglet value average; and S4.6, sorting the candidate schemes according to the compromise index from small to large.
6. The method according to claim 5, wherein in the step S4.3, the nonlinear compression factor is determined according to a load type, and the value is greater than 1 when the load is a lidar, less than or equal to 1 when the load is an orthographic camera, 1.2 when the load is a oblique photographic camera, and 1.0 when the load is a multispectral sensor.
7. The method of claim 6, wherein in step S5, the acceptable advantage condition is that a difference between a compromise index of the ranked first candidate and a compromise index of the ranked second candidate is greater than or equal to a quotient of one divided by a total number of candidates minus one, and the acceptable stability condition is that the ranked first candidate is ranked first in at least one of its group utility value ranking or individual regrind value ranking.
8. A base mapping-oriented unmanned aerial vehicle type selection decision system for implementing a base mapping-oriented unmanned aerial vehicle type selection decision method as claimed in any one of claims 1 to 7, comprising: The index system construction module is used for constructing an index system consisting of a flight demand index, an unmanned aerial vehicle and load parameter index, an environment demand index and other factor indexes according to basic mapping service requirements; The data quantization standardization module is used for carrying out grading assignment on each qualitative index, respectively adopting corresponding standardization treatment on each positive index, each negative index and each interval type index to obtain a standardized evaluation value, carrying out atmospheric density correction on each height type index to obtain a standardized evaluation value, and constructing a standardized positive matrix by the standardized evaluation values of all the indexes; The weight calculation module is used for introducing scene reference constraint factors to construct an improved probability distribution matrix, calculating the information entropy value of each index, calculating the improved difference coefficient of each index through the difference adjustment factors, fusing expert priori weights of each index to obtain adaptive entropy value weights of each index, and forming weight vectors by the adaptive entropy value weights of all the indexes; The scheme ordering module is used for multiplying each element in the standardized forward matrix by the corresponding self-adaptive entropy weight to obtain a weighted standardized decision matrix, determining an ideal solution and a negative ideal solution of each index, calculating the group utility value and the individual regret value of each candidate scheme, calculating the compromise index of each candidate scheme through the self-adaptive decision coefficient, and ordering each candidate scheme from small to large according to the compromise index; and the optimal detection module is used for detecting the acceptable advantage condition and the acceptable stability condition of the first-ranked candidate solution and outputting a final type selection decision result according to the detection result.
9. The system of claim 8, wherein the data quantization normalization module internally comprises: the qualitative index assignment unit is used for carrying out grading assignment on each qualitative index according to a preset grading scale; the interval index standardization unit is used for obtaining standardized evaluation values by adopting three-section standardization processing for each interval index; and the height index correction unit is used for obtaining a standardized evaluation value by adopting the standardized processing of the atmospheric density ratio correction for each height index.
10. The system of claim 9, wherein the weight calculation module internally comprises: The probability matrix construction unit is used for constructing an improved probability distribution matrix, wherein a scene reference adjustment coefficient and a mapping industry threshold factor are introduced; A difference coefficient calculation unit for calculating an improved difference coefficient of each index, the improved difference coefficient being equal to a difference adjustment factor power of a difference value obtained by subtracting the entropy value of the index information; and the weight synthesis unit is used for fusing the improved difference coefficient of each index with the expert priori weight to obtain the self-adaptive entropy weight.

Description

Unmanned aerial vehicle selection decision-making method and system for basic mapping Technical Field The invention relates to the technical field of unmanned aerial vehicle type selection, in particular to an unmanned aerial vehicle type selection decision method and system for basic mapping. Background In recent years, with the rapid development of unmanned aerial vehicle technology, the demand for unmanned aerial vehicles in the field of basic mapping is increasing. In the face of various unmanned aerial vehicle products with different performances on the market, how to scientifically and quantitatively select unmanned aerial vehicles suitable for self business requirements becomes an important problem facing mapping institutions. The traditional unmanned aerial vehicle type selection method mainly depends on qualitative means such as personnel experience, manufacturer investigation or sales personnel announcement, and lacks overall consideration of multidimensional indexes such as flight requirements, load performance, environmental adaptability, full life cycle economy and the like, so that the selected unmanned aerial vehicle has the problem of redundant or insufficient performance in practical application, and resource waste and repeated purchasing are caused. In the prior art, a learner applies a multi-attribute decision model such as an entropy method, a VIKOR method and the like to the field of equipment type selection. However, the traditional method has the following defects that the index standardization only processes positive indexes and negative indexes, a reasonable processing mode is lacking for interval indexes with optimal value intervals, the entropy method gives too low or even zero weight to key indexes with smaller data difference when the number of candidate schemes is small, the key indexes do not accord with the actual requirements of engineering, the VIKOR method adopts linear accumulation when the group utility value is calculated, the tolerance difference of different load types to index short plates cannot be reflected, and the nonlinear influence of air density change under special environments such as a plateau on the performance of unmanned aerial vehicle such as rising limit, endurance and the like is not included in the quantization model. Therefore, an improved unmanned aerial vehicle model selection decision method capable of comprehensively considering the section type index, the environment correction and the load sensitivity is needed, so that the scientificity and the engineering practicability of the model selection result are improved. Disclosure of Invention Aiming at the defects of the prior art, the invention provides an unmanned aerial vehicle type selection decision-making method and system for basic mapping. In order to achieve the above purpose, the invention adopts the following technical scheme: The unmanned aerial vehicle type selection decision-making method for basic mapping comprises the following steps: Step S1, constructing an index system consisting of a flight demand index, an unmanned aerial vehicle and a load parameter index, an environment demand index and other factor indexes according to basic mapping business requirements; Step S2, carrying out grading assignment on each qualitative index, respectively adopting corresponding standardization treatment on each positive index, each negative index and each interval type index to obtain a standardized evaluation value, carrying out atmospheric density correction on each height type index to obtain a standardized evaluation value, and constructing a standardized positive matrix by the standardized evaluation values of all the indexes; s3, introducing scene reference constraint factors to construct an improved probability distribution matrix, calculating information entropy values of all indexes, calculating improved difference coefficients of all indexes through difference adjustment factors, fusing expert priori weights of all indexes to obtain self-adaptive entropy value weights of all indexes, and forming weight vectors by the self-adaptive entropy value weights of all indexes; S4, multiplying each element in the standardized forward matrix by a corresponding self-adaptive entropy weight to obtain a weighted standardized decision matrix, determining an ideal solution and a negative ideal solution of each index, calculating a group utility value and an individual regressive value of each candidate scheme, calculating a compromise index of each candidate scheme through a self-adaptive decision coefficient, and sequencing the candidate schemes from small to large according to the compromise index; And S5, checking the acceptable advantage condition and the acceptable stability condition of the first-ranked candidate solution, and outputting a final type selection decision result according to the checking result. Further, in the step S1, the flight requirement indexes include minimum personnel conf