Unmanned aerial vehicle system architecture alternative scheme selection method

Abstract

The invention relates to a method for selecting an alternative scheme of an unmanned aerial vehicle system architecture, and the method comprises the steps: 1, carrying out the architecture analysis of an unmanned aerial vehicle system, and screening out an evaluation index which is most sensitive to the architecture; 2, comparing the importance degrees of the evaluation indexes in pairs to obtain relative importance degrees, and establishing an index importance evaluation matrix; 3, calculating a maximum feature root of the index importance evaluation matrix and a corresponding feature vector, and obtaining a weighting coefficient of each evaluation index; step 4, performing pairwise comparison on the performance degrees of the evaluation indexes in each architecture alternative scheme to obtain relative numerical values, and establishing a relative numerical value matrix of each evaluation index; 5, calculating the maximum feature root of the relative numerical value matrix of each evaluation index and the corresponding feature vector, and obtaining the numerical value of each evaluation index in each architecture alternative scheme; and 6, calculating the total architecture value of each architecture alternative scheme, and selecting an optimal architecture scheme.

Description

Technical Field

[0001] The present application belongs to the technical field of unmanned aerial vehicle system architecture design, and specifically relates to a method for selecting alternative solutions for unmanned aerial vehicle system architecture. Background Art

[0002] At present, system architecture design and analysis methods include UPDM architecture method, RUP SE architecture method, Vitech's "onion model" method, Dassault's MMS method and other four methods.

[0003] UPDM (Unified Profile for DoDAF and MODAF) develops the definitions of DoDAF and MODAF in terms of architectural description. It is a perfect combination of the two in the field of modeling. The concepts of viewpoints and views used in the RUP SE architecture framework are consistent with the ISO / ITU 10746 standard reference model for open distributed processing (RM-ODP) and the ANSI / IEEE 1471-2000 standard recommended practices for software-intensive system architecture description. Vitech MBSE uses a progressive SE process called the "onion model" to allow complete temporary solutions to be obtained at increasing levels of detail in the system specification process. Dassault proposed the model-driven system engineering methodology MMS (Modeling Method for System). Based on the MMS architecture method, it further refined the model-driven system engineering to form an MMS architecture framework.

[0004] The complexity and emergence of drone systems grow in a nonlinear manner, and there are complex relationships between the design parameters of the system architecture, which leads to a large number of architectural alternatives and a large space for architectural trade-offs to be searched. Currently, there is a lack of a method for quickly selecting drone system architecture alternatives. In view of this, this application is proposed.

[0005] This application is proposed in view of the above-mentioned technical defects. Summary of the invention

[0006] The purpose of this application is to provide a method for selecting alternative architecture solutions for a drone system, so as to effectively realize the rapid selection of alternative architecture solutions for a drone system and obtain the optimal architecture solution.

[0007] The technical solution of this application is:

[0008] A method for selecting alternative solutions for an unmanned aerial vehicle system architecture, comprising:

[0009] Step 1: Analyze the architecture of the UAV system and select the evaluation indicators that are most sensitive to the architecture;

[0010] Step 2: Compare the importance of evaluation indicators in pairs, obtain the relative importance, and establish the indicator importance evaluation matrix;

[0011] Step 3: Calculate the maximum eigenvalue of the indicator importance evaluation matrix and the corresponding eigenvector to obtain the weighted coefficient of each evaluation indicator;

[0012] Step 4: Compare the performance of the evaluation indicators in each architecture alternative solution in pairs, obtain relative values, and establish a relative value matrix for each evaluation indicator;

[0013] Step 5: Calculate the maximum eigenvalue of the relative numerical matrix of each evaluation index and the corresponding eigenvector to obtain the value of each evaluation index in each architecture alternative;

[0014] Step 6: Calculate the total architecture value of each architecture alternative using the values ​​of each evaluation indicator in each architecture alternative and its weighted coefficient, and select the architecture alternative with the largest total architecture value as the optimal architecture alternative.

[0015] According to at least one embodiment of the present application, in the above-mentioned method for selecting alternative solutions for the drone system architecture, in step 1, the architecture is analyzed from the aspects of function, performance, interface, weight, security, safety, reliability and maintainability;

[0016] Screen out the evaluation indicators that are most sensitive to the architecture, including reliability, security, emergency backup, latency, scalability, weight, cost and maturity.

[0017] According to at least one embodiment of the present application, in the above-mentioned method for selecting alternative drone system architecture solutions, in step 2, if the two evaluation indicators are equally important, the relative importance is 1; if the former of the two evaluation indicators is slightly better than the latter, the relative importance is 3, otherwise it is 1 / 3; if the former of the two evaluation indicators is better than the latter, the relative importance is 5, otherwise it is 1 / 5; if the former of the two evaluation indicators is significantly better than the latter, the relative importance is 7, otherwise it is 1 / 7; if the former of the two evaluation indicators is particularly better than the latter, the relative importance is 9, otherwise it is 1 / 9; for other cases, refer to the middle value to construct an indicator importance evaluation matrix.

[0018] According to at least one embodiment of the present application, in the above-mentioned method for selecting alternative solutions for the drone system architecture, in step 3, Aω=γω is solved to obtain the maximum characteristic root γ of the calculation index importance evaluation matrix A, and the corresponding characteristic vector ω, and the weighted coefficient ω of each evaluation index is obtained. j , where j = 1,…,n, and n is the number of evaluation indicators.

[0019] According to at least one embodiment of the present application, in the above-mentioned method for selecting an alternative solution for the drone system architecture, step three further includes:

[0020] The weighted coefficients of each evaluation indicator are checked for consistency. If the consistency check does not meet the requirements, the elements of the indicator importance evaluation matrix A are readjusted to calculate the weighted coefficients ω of each evaluation indicator. j ;

[0021] The weighted coefficients of each evaluation indicator are checked for consistency, specifically:

[0022] Calculate the consistency index CI of the indicator importance evaluation matrix A:

[0023]

[0024] Calculate the consistency ratio CR of the indicator importance evaluation matrix A:

[0025]

[0026] Among them, RI is the random consistency index, which is 0 when n is 1; 0 when n is 2; 0 when n is 3; 0.58 when n is 4; 0.9 when n is 5; 1.12 when n is 6; 1.24 when n is 7; 1.32 when n is 8; 1.41 when n is 9; 1.45 when n is 10; 1.49 when n is 10.

[0027] If CR is less than 0.1, it is judged that the weighted coefficients of each evaluation index pass the consistency check, otherwise, it is judged that the consistency check of the weighted coefficients of each evaluation index does not meet the requirements.

[0028] According to at least one embodiment of the present application, in the above-mentioned method for selecting alternative architectures of a drone system, in step 4, if the evaluation index has the same performance in the two alternative architectures, the relative value is 1;

[0029] If the evaluation indicator performs slightly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 3, otherwise 1 / 3; if the evaluation indicator performs better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 5, otherwise 1 / 5; if the evaluation indicator performs significantly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 7, otherwise 1 / 7; if the evaluation indicator performs particularly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 9, otherwise 1 / 9; in other cases, refer to the middle value.

[0030] According to at least one embodiment of the present application, in the above-mentioned method for selecting an alternative solution for the drone system architecture, in step 5, B j θj =∈ j θ j , calculate the relative numerical matrix B of each evaluation index j The largest characteristic root ∈ j , and the corresponding eigenvector θ j , and obtain the values ​​θ of each evaluation indicator in each architecture alternative ij , where j = 1,…,m, and m is the number of architecture alternatives.

[0031] According to at least one embodiment of the present application, in the above-mentioned method for selecting an alternative solution for the drone system architecture, step five further includes:

[0032] Perform consistency check on the relative values ​​of the evaluation indicators of each architecture alternative solution. If the consistency check does not meet the requirements, readjust the relative value matrix B of each evaluation indicator. j The elements in , calculate the value of each evaluation indicator in each architecture alternative θ ij ;

[0033] The relative values ​​of the evaluation indicators of each architecture alternative are checked for consistency, specifically:

[0034] Calculate the relative numerical matrix B of each evaluation index j The consistency index CI j :

[0035]

[0036] Calculate the relative numerical matrix B of each evaluation index j The overall consistency ratio CR 总 :

[0037]

[0038] Among them, RI 1 , RI 2 , RI m is a random consistency index. When m is 1, it is 0; when m is 2, it is 0; when m is 3, it is 0.58; when m is 4, it is 0.9; when m is 5, it is 1.12; when m is 6, it is 1.24; when m is 7, it is 1.32; when m is 8, it is 1.41; when m is 9, it is 1.45; when m is 10, it is 1.49;

[0039] If CR 总 <0.1, it is judged that the relative values ​​of the evaluation indicators of each architecture alternative solution have passed the consistency check; otherwise, it is judged that the consistency check of the relative values ​​of the evaluation indicators of each architecture alternative solution does not meet the requirements.

[0040] According to at least one embodiment of the present application, in the above-mentioned method for selecting alternative architectures of a drone system, in step 6, the total architecture value of each alternative architecture is calculated, specifically:

[0041]

[0042] in,

[0043] Q i is the total architectural value of the i-th architectural alternative. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the performance analysis of the drone system architecture provided by the embodiment of the present application;

[0045] Figure 2 It is a schematic diagram of the analysis of the UAV system architecture interface provided by the embodiment of the present application;

[0046] Figure 3 It is a schematic diagram of the calculation relationship of the method for selecting alternative solutions for the drone system architecture provided in the embodiment of the present application;

[0047] Figure 4 It is a flow chart of a method for selecting alternative solutions for the drone system architecture provided in an embodiment of the present application.

[0048] In order to better illustrate the present embodiment, some contents of the drawings may be omitted, enlarged or reduced, which is only used for illustrative purposes and should not be construed as limiting the present application. DETAILED DESCRIPTION

[0049] In order to make the technical solution and advantages of the present application clearer, the technical solution of the present application will be described in further detail in detail and in detail with reference to the accompanying drawings. It can be understood that the specific embodiments described here are only partial embodiments of the present application, which are only used to explain the present application, not to limit the present application. It should be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, and other related parts can refer to the general design.

[0050] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of this application should be the common meanings understood by those skilled in the art in the field to which this application belongs. The term "include" used in the description of this application means that the concepts appearing before the term include the concepts listed after the term and their equivalents, without excluding other related concepts.

[0051] A method for selecting alternative architectures for unmanned aerial vehicle systems is provided. By cutting and analyzing the processes of relevant evaluation criteria and methods in system engineering manuals such as INCOSE, analysis and evaluation criteria and methods for the evaluation and selection of alternative architectures for unmanned aerial vehicle systems are formed. By designing an architecture selection scheme based on the hierarchical analysis method, the experience of architecture designers and relevant performance data in the architecture design are fully utilized to determine the parameter relationship between evaluation indicators, optimize candidate architectures, and select the preferred architecture throughout the entire life cycle of the system. This is explained in the following aspects.

[0052] a. Determine the system architecture and select space parameters

[0053] By analyzing the architecture, parameters that describe the characteristics of the architecture are selected as evaluation indicators. The architecture can be analyzed from the aspects of function, performance, interface, weight, security, safety, reliability and maintainability, and then the evaluation indicators can be screened out.

[0054] Functional analysis: The process of system functional requirements analysis needs to start with the system's state and mode analysis, determine the state and mode of system operation, and then use this to carry out the decomposition and logical analysis of the system functions, and analyze the behavioral logic of the system functions under different states and modes. It can be divided into state and mode analysis, system top-level function definition, system function decomposition, system function logic analysis, system function interface analysis, state mode function matrix and system function hierarchy diagram.

[0055] Performance analysis: First, establish a digital prototype of the UAV system and establish the relationship between the subsystems. Then, establish a hierarchical demand indicator system for UAVs and establish a mapping relationship between the demand indicator system and the subsystems. On this basis, extract the overall performance indicators from the demand indicator system and decompose the overall performance indicators to establish an overall performance indicator system, which may include flight capability indicators, flight control capability indicators, endurance capability indicators, and stability. At the same time, establish analysis models in different disciplines, which may include ANSYS models, Simulink domain models, and SymML models. On the basis of the initial overall performance indicators, perform indicator optimization analysis. If the expected indicators are met, output the performance indicators. If not, re-decompose the overall performance indicators, such as Figure 1 shown.

[0056] Interface analysis: Interface analysis is divided into external interface analysis and functional interface analysis. The interface definition should include input with information source and output with destination, and define the interface content between various systems of the drone. This interface content needs to be considered in each cycle stage of the system; data flow analysis method should be used for interface analysis. In the model-based analysis process, the analysis of object flow between functions in the system function modeling process can sort out the system function interface, and adopt the general N 2 The diagram is organized so that the components of the system are placed on the diagonal. The rest of the N×N matrix represents the interface. Arrows pointing to the system represent the input of data, and arrows pointing outward represent the output of data, such as Figure 2 shown.

[0057] Weight analysis: The system-level weight of a drone is mainly decomposed from the weight of the drone. These weight targets are usually derived from the design experience of other drones with similar purposes, and are expected to be improved during the technical design phase. As the functions are decomposed, the weight design of the entire system is iteratively improved, with a weight target for each basic function. Next, during the system architecture definition process, the weight targets of different functions will be used to evaluate the maximum acceptable weight required by the system under different system architectures.

[0058] Supportability analysis: Supportability analysis includes scheduled and unscheduled maintenance requirements and links to safety-related functions. When defining the "overhaul interval", the fault detection rate and fault isolation rate need to be considered. It is also necessary to define the support requirements for the signals and connections of external test equipment.

[0059] Safety analysis: Safety analysis includes the minimum performance restrictions for functional availability, i.e., functional continuity, and integrity, i.e., correctness of behavior; functional failure conditions are obtained by performing functional hazard assessment (FHA) or AISS security analysis, and the functional failure conditions are classified to obtain safety requirements.

[0060] Reliability analysis: System-level reliability analysis mainly comes from functional availability and operational availability indicators; these values ​​will continue to be refined as the functions are decomposed, and each function will correspond to a value; in the system architecture definition process, the system reliability value will be used to evaluate whether different system architectures meet the required functional availability.

[0061] Maintainability analysis: Based on the virtual analysis of maintainability of digital prototypes, a virtual environment is established that includes product digital prototypes, maintenance personnel, maintenance tools, maintenance equipment, and maintenance process information. In this environment, the maintenance of products and related processes are simulated, including visibility during the maintenance process, accessibility of parts, and disassembly and assembly of parts, and qualitative and quantitative analysis of maintainability based on digital prototypes is carried out.

[0062] b. Architecture selection and structural model establishment.

[0063] According to the functions, performance, interfaces, weight, security, safety, reliability and maintainability, the system architecture is analyzed, and the evaluation indicators that are most sensitive to the architecture in terms of functions, performance, interfaces, weight, security, safety, reliability and maintainability are selected. Based on these evaluation indicators, the architecture selection structure model is established, which is divided into the architecture target layer, the architecture total value layer and the evaluation indicator layer, such as Figure 3 As shown, based on the architecture, the structural model is selected and the indicator set parameters are determined. i The weight function Q i Defined as the total architecture value function, where i = 1, ..., n, n is the number of architecture alternatives, corresponding to the total architecture value layer, which will affect Q i The jth evaluation indicator It is defined as a multidimensional value indicator, that is, the evaluation indicator of the architecture, where j = 1,…,m, m is the number of evaluation indicators, corresponding to the evaluation indicator layer.

[0064] c. Determine the evaluation method.

[0065] Evaluate architectural alternatives based on the analytic hierarchy process.

[0066] Determine the indicator set formula: Use the linear weighting method to combine the selected evaluation indicators, as shown in formula (1), where is the jth evaluation index of the i-th architecture, where i = 1, ..., n, n is the number of architecture alternatives, j = 1, ..., m, m is the number of evaluation indicators, ω j For evaluation indicators The weighting coefficient of .

[0067]

[0068] Determine the weighting coefficient ω of the evaluation index:

[0069] Evaluation matrix establishment: According to the selected evaluation indicators, the evaluation indicators are compared in pairs. Based on the criteria in Table 1, the score ratios between the indicators are determined according to their relative importance and filled in the corresponding matrix, as shown in Table 2;

[0070] Table 1 Index evaluation scale:

[0071] Table 2 Indicator importance evaluation matrix: <![CDATA[a 1 ]]> <![CDATA[a 2 ]]> <![CDATA[a 3 ]]> …… <![CDATA[a m ]]> <![CDATA[a 1 ]]> <![CDATA[P 11 ]]> <![CDATA[P 12 ]]> <![CDATA[P 13 ]]> …… <![CDATA[P 1m ]]> <![CDATA[a 2 ]]> <![CDATA[P 21 > <![CDATA[P 22 ]]> <![CDATA[P 23 ]]> …… <![CDATA[P 2m ]]> <![CDATA[a 3 ]]> <![CDATA[P 31 ]]> <![CDATA[P 32 ]]> <![CDATA[P 33 ]]> …… <![CDATA[P 3m ]]> …… …… …… …… …… …… <![CDATA[a m ]]> <![CDATA[P m1 ]]> <![CDATA[P m2 ]]> <![CDATA[P m3 ]]> <![CDATA[P mm ]]>

[0072] Assign weighted coefficients: Since various indicators have different degrees of influence on the system solution, the importance of each of the above evaluation indicators needs to be designed when selecting the architecture alternatives. The importance is expressed by the weight value. According to Table 2, the indicator importance evaluation matrix A is established, and its maximum eigenvalue γ and the eigenvector ω corresponding to γ ​​are calculated, as shown in formula (2). The eigenvector ω is the weighted coefficient of each corresponding evaluation indicator.

[0073] Aω=γω…………(2)

[0074] Weighted coefficient consistency check: The consistency index CI is calculated according to formula (3), γ is the maximum characteristic root, n is the order of the indicator importance evaluation matrix A, and the random consistency index RI is introduced. The index is shown in Table 3. Combining CI and RI, according to formula (4), the consistency ratio CR is calculated. If CR < 0.1, it is considered that the indicator importance evaluation matrix A meets the consistency check. If CR ≥ 0.1, it is necessary to adjust the elements in the indicator importance evaluation matrix A, that is, the ratio of the importance of different indicators, so that CR < 0.1.

[0075]

[0076] Table 3 Randomness consistency index RI: n 1 2 3 4 5 6 7 8 9 10 RI 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.49

[0077]

[0078] Determine the evaluation index value a for each solution:

[0079] For each evaluation indicator a j , where j = 1, ..., m, m is the number of evaluation indicators, analyze the alternative architectures, and determine the alternative solution X for each architecture i Evaluation Metrics j The values ​​are as follows:

[0080] Establish an indicator value matrix: select an evaluation indicator a j , where j = 1, ..., m, m is the number of evaluation indicators, and any indicator a is selected from all architecture alternatives j The degree of performance in the architecture, based on the criteria in Table 1, determines its ratio P in all architectures and establishes the architecture indicator matrix B j , where j = 1, ..., m, m is the number of evaluation indicators, and is filled in as shown in Table 4. For each evaluation indicator a j , construct the corresponding indicator matrix B j Finally, the indicator matrices corresponding to all indicators form a matrix set, which contains m architecture indicator matrices B j , where j = 1,…,m, where m is the number of evaluation indicators.

[0081] Table 4 Architecture indicator matrix B j :

[0082] Determine the evaluation value: Establish the architecture indicator matrix B according to Table 4 j , where j = 1, ..., m, m is the number of evaluation indicators, and the corresponding maximum eigenvalue ∈ j , and ∈ j The corresponding eigenvector θ j , as shown in formula (5), the eigenvector θ j That is the corresponding evaluation index a j The score among all architecture alternatives is obtained by calculating the corresponding maximum characteristic root ∈ of all architecture indicator matrices according to formula (5): j , and ∈ j The corresponding eigenvector θ j , you can get the scores of all evaluation indicators in all architecture alternatives;

[0083] B j θ j =∈ j θ j …………(5)

[0084] Solution consistency check: Based on the architecture indicator matrix for a certain indicator established in Table 4, the architecture indicator matrix B is calculated based on formula (3). j , where j = 1,…,m, m is the number of evaluation indicators, and the corresponding CI j , the consistency check of the scheme is shown in formula (6). If CR 总 <0.1, the solution is considered to meet the consistency check. If CR 总 ≥0.1, then you need to adjust the matrix B j The elements in , that is, the importance ratio of a certain indicator in different architecture schemes, and recalculate the corresponding maximum characteristic root ∈ j , and ∈ j The corresponding eigenvector θ j , recalculate CR 总 , which eventually makes CR 总 <0.1.

[0085]

[0086] d. Scheme selection: based on the calculated evaluation index weighting coefficient ω j , where ω 1 ,ω 2 ,…,ω m , m is the number of evaluation indicators, and evaluation indicator aj The score θ among all architecture alternatives nj , where θ n1 ,θ n2 ··,θ nm , n is the number of architecture alternatives, j = 1,…, m, m is the number of evaluation indicators, and the final score of each solution is shown in Table 5. According to Table 5, the total value Q of the architecture solution i , the one with the highest score is the best solution.

[0087] Table 5: Total value function of architecture solution:

[0088] In a specific example, the UAV system design has three architectural alternatives. Option 1 is a two-pole integration solution based on 1394B and FC high-speed bus, Option 2 is a three-level integration solution based on 1394B and FC high-speed bus, and Option 3 is a three-level architecture solution based on 1394B, GJB289A and FC bus.

[0089] Based on the method for selecting alternative architecture schemes for the drone system disclosed in the above embodiment, the alternative architecture scheme is selected and implemented as follows.

[0090] Step 1: Perform architecture analysis on the UAV system and select the evaluation indicators that are most sensitive to the architecture.

[0091] The architecture is analyzed in terms of functionality, performance, interfaces, weight, supportability, safety, reliability and maintainability.

[0092] The most sensitive evaluation indicators for architecture are selected, including reliability, security, emergency backup, latency, scalability, weight, cost and maturity, with corresponding indicator parameters a j (j=1,…,8)as shown in Table 6.

[0093] Table 6 Evaluation index parameter comparison table:

[0094] Step 2: Compare the importance of evaluation indicators pairwise, obtain relative importance, and establish indicator importance evaluation matrix A.

[0095] If two evaluation indicators are equally important, the relative importance is 1;

[0096] If the former of the two evaluation indicators is slightly better than the latter, the relative importance is 3, otherwise it is 1 / 3; if the former of the two evaluation indicators is better than the latter, the relative importance is 5, otherwise it is 1 / 5; if the former of the two evaluation indicators is obviously better than the latter, the relative importance is 7, otherwise it is 1 / 7; if the former of the two evaluation indicators is particularly better than the latter, the relative importance is 9, otherwise it is 1 / 9; in other cases, refer to the middle value.

[0097] Establish the indicator importance evaluation matrix A, as shown in Table 7.

[0098] Table 7 System evaluation index parameter table:

[0099] Step 3: Calculate the maximum eigenvalue γ of the indicator importance evaluation matrix A and the corresponding eigenvector ω to obtain the weighted coefficient ω of each evaluation indicator. j (j=1,…,8).

[0100] From Aω=γω, we can solve the maximum characteristic root γ=8.283 of the indicator importance evaluation matrix A, and the weighted coefficient ω of each evaluation indicator is j As shown in Table 8 below.

[0101] Table 8 Evaluation index parameter comparison table:

[0102] The weighted coefficients of each evaluation indicator are checked for consistency. If the consistency check does not meet the requirements, the elements of the indicator importance evaluation matrix A are readjusted to calculate the weighted coefficients ω of each evaluation indicator. j (j=1,…,8).

[0103] Calculate the consistency index CI of the indicator importance evaluation matrix A:

[0104] Among them, n is the order of the indicator importance evaluation matrix A, that is, the number of evaluation indicators, which is 8, and the consistency index CI of the indicator importance evaluation matrix A is 0.04.

[0105] Calculate the consistency ratio CR of the indicator importance evaluation matrix A:

[0106] Among them, RI is the random consistency index, when n is 1, it is 0; when n is 2, it is 0; when n is 3, it is 0.58; when n is 4, it is 0.9; when n is 5, it is 1.12; when n is 6, it is 1.24; when n is 7, it is 1.32; when n is 8, it is 1.41; when n is 9, it is 1.45; when n is 10, it is 1.49. The consistency ratio CR of the indicator importance evaluation matrix A is 0.028, which is much less than 0.1. The weighting coefficient ω of each evaluation indicator is j Pass the consistency check. If it is greater than 1, the consistency check fails.

[0107] Step 4: Compare the performance of the evaluation indicators in each architecture alternative solution, obtain relative values, and establish the relative value matrix B of each evaluation indicator. j (j=1,…,8).

[0108] If the evaluation indicator performs to the same degree in the two architecture alternatives, the relative value is 1;

[0109] If the evaluation indicator performs slightly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 3, otherwise 1 / 3; if the evaluation indicator performs better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 5, otherwise 1 / 5; if the evaluation indicator performs significantly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 7, otherwise 1 / 7; if the evaluation indicator performs particularly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 9, otherwise 1 / 9; in other cases, refer to the middle value.

[0110] A comparative analysis of the three schemes was conducted for the performance level of each evaluation indicator, as shown in Table 9.

[0111] Table 9 Scheme comparison table:

[0112] Establish the relative value matrix B of the weight evaluation index 6 , as shown in Table 10.

[0113] Table 10 Weight framework indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 2 2 Solution 2 0.5 1 1 Solution 3 0.5 1 1

[0114] Establish the relative value matrix B of cost evaluation indicators 7 , as shown in Table 11.

[0115] Table 11 Cost structure indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 4 3 Solution 2 0.25 1 0.5 Solution 3 0.3 2 1

[0116] Establish the relative numerical matrix B of scalability evaluation indicators 5 , as shown in Table 12.

[0117] Table 12 Scalability architecture indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 2 4 Solution 2 0.5 1 2 Solution 3 0.25 0.5 1

[0118] Establish the relative numerical matrix B of delay evaluation index 4 , as shown in Table 13.

[0119] Table 13 Latency architecture indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 1 4 Solution 2 1 1 4 Solution 3 0.25 0.25 1

[0120] Establish the relative numerical matrix B of reliability evaluation index 1 , as shown in Table 14.

[0121] Table 14 Reliability architecture indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 0.5 0.5 Solution 2 2 1 0.5 Solution 3 2 2 1

[0122] Establish the relative numerical matrix B of safety evaluation index 2 , as shown in Table 15.

[0123] Table 15 Security architecture indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 0.5 0.5 Solution 2 2 1 0.5 Solution 3 2 2 1

[0124] Establish the relative numerical matrix B of emergency backup evaluation indicators 3 , as shown in Table 16.

[0125] Table 16 Emergency backup architecture indicator matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 0.5 0.5 Solution 2 2 1 0.5 Solution 3 2 2 1

[0126] Establish the relative value matrix B of maturity assessment indicators 8 , as shown in Table 17.

[0127] Table 17 Maturity Architecture Indicator Matrix: Solution 1 Solution 2 Solution 3 Solution 1 1 0.5 0.25 Solution 2 2 1 0.25 Solution 3 4 4 1

[0128] Step 5: Calculate the relative numerical matrix B of each evaluation index j (j=1,…,8) the largest characteristic root ∈ j , and the corresponding eigenvector θ j , and obtain the values ​​θ of each evaluation indicator in each architecture alternative ij (j=1,…,8,i=1,2,3).

[0129] B j θ j =∈ j θj .

[0130] Perform consistency check on the relative values ​​of the evaluation indicators of each architecture alternative solution. If the consistency check does not meet the requirements, readjust the relative value matrix B of each evaluation indicator. j (j=1,…,8), calculate the value θ of each evaluation indicator in each architecture alternative ij (j=1,…,8,j=1,2,3).

[0131] Calculate the relative numerical matrix B of each evaluation index j The consistency index CI of (j=1,…,8) j :

[0132] Among them, m is the relative numerical matrix B of the evaluation index j The order of , that is, the number of architectural alternatives, is 3.

[0133] Calculate the relative numerical matrix B of each evaluation index j The overall consistency ratio CR 总 :

[0134] Among them, RI 1 , RI 2 , RI m is a random consistency index. When m is 1, it is 0; when m is 2, it is 0; when m is 3, it is 0.58; when m is 4, it is 0.9; when m is 5, it is 1.12; when m is 6, it is 1.24; when m is 7, it is 1.32; when m is 8, it is 1.41; when m is 9, it is 1.45; when m is 10, it is 1.49.

[0135] Get the relative numerical matrix B of each evaluation index j The overall consistency ratio CR 总 It is 0.026, which is much smaller than 0.1. The relative values ​​of the evaluation indicators of each architecture alternative solution pass the consistency check. If it is greater than 1, the consistency check does not meet the requirements.

[0136] Step 6: Calculate the total architectural value Q of each architectural alternative i (i=1,2,3), the architecture alternative with the largest total architecture value is selected as the best architecture alternative.

[0137]

[0138] Architecture Alternative 1: Q 1 =1.2814, Architecture Alternative 2: Q 2 =1.1736, Architecture Alternative 3: Q3 =1.4939, as shown in Table 18.

[0139] Table 18: Total value function of architecture solution:

[0140] Architecture alternative three is selected as the best architecture solution.

[0141] The method for selecting alternative architecture schemes for a drone system disclosed in the above-mentioned embodiment determines the evaluation criteria and evaluation method for selecting the architecture scheme, determines the key decision points, and solves the difficult problems of complex relationships between drone system design parameters, large search space for architecture alternatives, and high complexity. By applying engineering practices in actual models, the criteria and methods are verified and solidified, relevant standards / specifications are formed, and finally a complete technical system is obtained.

[0142] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. Those skilled in the art should understand that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A method for selecting alternative solutions for an unmanned aerial vehicle system architecture, characterized in that: include: Step 1: Analyze the architecture of the UAV system and select the evaluation indicators that are most sensitive to the architecture; Step 2: Compare the importance of evaluation indicators in pairs, obtain the relative importance, and establish the indicator importance evaluation matrix; Step 3: Calculate the maximum eigenvalue of the indicator importance evaluation matrix and the corresponding eigenvector to obtain the weighted coefficient of each evaluation indicator; Step 4: Compare the performance of the evaluation indicators in each architecture alternative solution in pairs, obtain relative values, and establish a relative value matrix for each evaluation indicator; Step 5: Calculate the maximum eigenvalue of the relative numerical matrix of each evaluation index and the corresponding eigenvector to obtain the value of each evaluation index in each architecture alternative; Step 6: Calculate the total architecture value of each architecture alternative using the values ​​of each evaluation indicator in each architecture alternative and its weighted coefficient, and select the architecture alternative with the largest total architecture value as the optimal architecture alternative.

2. The method for selecting alternative solutions for the UAV system architecture according to claim 1, characterized in that: In step 1, the architecture is analyzed in terms of functionality, performance, interfaces, weight, supportability, safety, reliability, and maintainability; Screen out the evaluation indicators that are most sensitive to the architecture, including reliability, security, emergency backup, latency, scalability, weight, cost and maturity.

3. The method for selecting alternative solutions for the UAV system architecture according to claim 2, characterized in that: In step 2, if the two evaluation indicators are equally important, the relative importance is 1; if the former is slightly better than the latter, the relative importance is 3, otherwise 1 / 3; if the former is better than the latter, the relative importance is 5, otherwise 1 / 5; if the former is significantly better than the latter, the relative importance is 7, otherwise 1 / 7; if the former is particularly better than the latter, the relative importance is 9, otherwise 1 / 9; in other cases, refer to the middle value to construct the indicator importance evaluation matrix.

4. The method for selecting alternative solutions for the UAV system architecture according to claim 3, characterized in that: In step 3, Aω=γω, the maximum characteristic root γ of the evaluation matrix A of the calculation index importance and the corresponding characteristic vector ω are solved to obtain the weighted coefficient ω of each evaluation index. j , where j = 1,…,n, and n is the number of evaluation indicators.

5. The method for selecting alternative solutions for the UAV system architecture according to claim 4, characterized in that: Step three also includes: The weighted coefficients of each evaluation indicator are checked for consistency. If the consistency check does not meet the requirements, the elements of the indicator importance evaluation matrix A are readjusted to calculate the weighted coefficients ω of each evaluation indicator. j ; The weighted coefficients of each evaluation indicator are checked for consistency, specifically: Calculate the consistency index CI of the indicator importance evaluation matrix A: Calculate the consistency ratio CR of the indicator importance evaluation matrix A: Among them, RI is the random consistency index, which is 0 when n is 1; 0 when n is 2; 0 when n is 3; 0.58 when n is 4; 0.9 when n is 5; 1.12 when n is 6; 1.24 when n is 7; 1.32 when n is 8; 1.41 when n is 9; 1.45 when n is 10; 1.49 when n is 10. If CR is less than 0.1, it is judged that the weighted coefficients of each evaluation index pass the consistency check, otherwise, it is judged that the consistency check of the weighted coefficients of each evaluation index does not meet the requirements.

6. The method for selecting alternative solutions for the UAV system architecture according to claim 5, characterized in that: In step 4, if the evaluation indicator performs at the same level in the two architecture alternatives, the relative value is 1; If the evaluation indicator performs slightly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 3, otherwise 1 / 3; if the evaluation indicator performs better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 5, otherwise 1 / 5; if the evaluation indicator performs significantly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 7, otherwise 1 / 7; if the evaluation indicator performs particularly better in the front-end architecture alternative than in the back-end architecture alternative, the relative value is 9, otherwise 1 / 9; in other cases, refer to the middle value.

7. The method for selecting alternative solutions for the UAV system architecture according to claim 6, characterized in that: In step five, Calculate the relative numerical matrix B of each evaluation index j The largest characteristic root ∈ j , and the corresponding eigenvector Get the values ​​of each evaluation indicator in each architecture alternative Wherein, j=1,…,m, and m is the number of architecture alternatives.

8. The method for selecting alternative solutions for the UAV system architecture according to claim 7, characterized in that: Step five also includes: Perform consistency check on the relative values ​​of the evaluation indicators of each architecture alternative solution. If the consistency check does not meet the requirements, readjust the relative value matrix B of each evaluation indicator. j , calculate the values ​​of each evaluation indicator in each architecture alternative The relative values ​​of the evaluation indicators of each architecture alternative are checked for consistency, specifically: Calculate the relative numerical matrix B of each evaluation index j The consistency index CI j : Calculate the relative numerical matrix B of each evaluation index j The overall consistency ratio CR 总 : Among them, RI1, RI2, RI m is a random consistency index. When m is 1, it is 0; when m is 2, it is 0; when m is 3, it is 0.58; when m is 4, it is 0.9; when m is 5, it is 1.12; when m is 6, it is 1.24; when m is 7, it is 1.32; when m is 8, it is 1.41; when m is 9, it is 1.45; when m is 10, it is 1.49; If CR 总 <0.1, it is judged that the relative values ​​of the evaluation indicators of each architecture alternative solution have passed the consistency check; otherwise, it is judged that the consistency check of the relative values ​​of the evaluation indicators of each architecture alternative solution does not meet the requirements.

9. The method for selecting alternative solutions for the UAV system architecture according to claim 8, characterized in that: In step 6, the total architectural value of each architectural alternative is calculated, specifically: in, Q i is the total architectural value of the i-th architectural alternative.

Images