Heterogeneous unmanned aerial vehicle cluster-oriented capability configuration optimization method

Abstract

The invention discloses a heterogeneous unmanned aerial vehicle cluster-oriented capability configuration optimization method. The method comprises the steps of S1, obtaining unmanned aerial vehicle cluster capability configuration elements; s2, obtaining an unmanned aerial vehicle cluster capability configuration optimization model according to the unmanned aerial vehicle cluster capability configuration elements; and S3, solving the unmanned aerial vehicle cluster capability configuration optimization model. By adopting the technical scheme of the invention, the potential of the heterogeneous unmanned aerial vehicle cluster can be brought into full play, and optimal configuration of resources can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of heterogeneous unmanned aerial vehicle clusters, and in particular relates to a capability configuration optimization method for heterogeneous unmanned aerial vehicle clusters. Background Art

[0002] UAVs are the fastest growing and most widely used aircraft in the world. With the development of artificial intelligence technology, new application methods of UAV swarms have emerged by leveraging the outstanding advantages of UAVs in low cost and high capability. UAV swarms perform tasks by coordinating multiple UAVs, which improves the capabilities of single aircraft and expands the overall mission capabilities, demonstrating higher survivability and mission capabilities than single aircraft. As the intersection of three types of characteristics: aircraft swarms, intelligent systems, and complex systems, UAV swarms are not simply the sum of single UAVs. Its characteristics are that it can distribute the loads of each link and subtask of the task to a large number of single UAVs that can communicate, and then complete the task in the form of a system. It has the characteristics of large scale, system flexibility, rich application scenarios, and configurable task types. It will play an increasingly important role in the field of intelligence. With the continuous advancement of UAV swarm technology, UAV swarms have developed from a single homogeneous platform to a multi-heterogeneous platform with complex multi-tasks in terms of platform heterogeneity. In terms of member scale, it has expanded from a few UAVs in traditional multi-aircraft formations to a UAV swarm of dozens or even hundreds of thousands. Therefore, how to fully tap the potential of drone swarms, ensure that they achieve the best results in various complex mission environments, and realize low-cost and large-scale application of drone swarms has become a topic that urgently needs to be studied.

[0003] Drone swarms have the advantage of flexible configuration adjustment when performing different tasks. Integrating and configuring drone swarm system resources is crucial to improving drone swarm capabilities, which requires strengthening macro-coordination and resource integration to promote resource integration and sharing. Through the optimized configuration of drone swarms, the problem of the mutual constraints between the diverse mission requirements of the cluster and the cost of large-scale operation can be solved. Faced with different tasks, the task objectives of the cluster are different, which can be prioritized by capability, cost, or cost-effectiveness. However, the existing drone swarm design often focuses on improving performance and pays less attention to cost-effectiveness, resulting in poor overall economic benefits of the swarm system. Faced with realistic problems such as limited drone resources and economic sustainability, it is necessary to balance costs and capabilities, effectively allocate and achieve the best configuration of the cluster to meet diverse mission requirements, reduce cluster costs, and improve the cost-effectiveness of the cluster system. Therefore, in the context of diversified applications of drone clusters, it is necessary to focus on the problem of capability and cost balance, upgrade from the "single-machine perspective" to the "cluster perspective", and establish a task-oriented capability comprehensive configuration optimization method for heterogeneous drone clusters in the top-level design. By systematically optimizing the overall resources of the drone cluster and investing in reasonable cluster configuration, we can ensure that it can meet complex and diverse mission requirements, thereby improving the execution efficiency and adaptability of drone clusters in different mission environments.

[0004] Configuration optimization is to maximize the utilization of resources and improve utilization efficiency, and to carry out corresponding configuration and optimization research. At present, the research on resource configuration optimization methods involves multiple fields such as drones, energy systems, electrical equipment, information services, and supply chains. Xiao Heye et al. proposed a modular drone configuration method to achieve efficient drone configuration under cost constraints. Shen Yaohan et al. proposed a configuration optimization method for drone delivery systems, which improved system service efficiency and reduced operating costs. Zhou Jianli et al. conducted capacity configuration optimization and decision-making model research on integrated energy systems. However, due to the differences in configuration optimization objects and ideas in different fields, there are large differences in configuration optimization models in various fields. Research on configuration optimization of drone clusters is mostly focused on resource scheduling, task allocation, and scale configuration. Duan Ting et al. built a dynamic task scheduling model for multi-drone clusters based on the capability framework. Qin Fan et al. proposed a scheduling method for networking and communication of drone clusters in the sea area. Wang Haomiao et al. studied the optimal task allocation strategy based on the improved alliance game algorithm under battlefield conditions. Zhang Huaiqiang et al. studied the scale configuration method of heterogeneous drone clusters. However, the configuration elements of drone clusters have expanded from simple flight and basic communication requirements to comprehensive requirements such as mission diversity, complex hardware heterogeneity, communication network configuration, and collaborative control, which poses new challenges to the capability configuration modeling of drone clusters. Existing research focuses on specific scenarios or single configuration elements, lacking more comprehensive overall research. Therefore, it is urgent to conduct research on the overall cluster configuration of drone clusters, which have complex characteristics and multiple configuration elements. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a capability configuration optimization method for heterogeneous UAV clusters, which can give full play to the potential of heterogeneous UAV clusters and achieve optimal configuration of resources.

[0006] To achieve the above object, the present invention adopts the following technical solution:

[0007] A capability configuration optimization method for heterogeneous UAV clusters, comprising:

[0008] Step S1, obtaining the configuration elements of the drone cluster capability;

[0009] Step S2: obtaining a UAV cluster capability configuration optimization model according to the UAV cluster capability configuration elements;

[0010] Step S3: Solve the optimization model of drone cluster capability configuration.

[0011] Preferably, the drone swarm capability configuration elements include: mission requirements, system resources and organizational strategies.

[0012] Preferably, the mission requirements include: endurance requirements, protection requirements, communication requirements, monitoring requirements, decision-making requirements and execution requirements.

[0013] Preferably, the system resources include: endurance load, protection load, communication load, monitoring load, decision load, and execution load.

[0014] As the preferred method, the organizational strategies include: the pilot method of drone swarm, the pigeon flock method, and the fully autonomous method of formation strategy.

[0015] Preferably, step S2 comprises:

[0016] According to the configuration elements of the UAV cluster capability, the comprehensive correlation matrix between the UAV cluster capability and the mission requirements, the correlation matrix between the configuration elements and the cluster capability, and the correlation matrix between the configuration elements and the mission requirements are obtained;

[0017] According to the comprehensive correlation matrix between UAV cluster capability and mission requirements, the correlation matrix between configuration elements and cluster capability, and the correlation matrix between configuration elements and mission requirements, the cluster configuration optimization evaluation model is obtained;

[0018] According to the cluster configuration optimization evaluation model, the UAV cluster capability configuration optimization model is obtained.

[0019] Preferably, the cluster configuration is evaluated based on the UAV cluster capability, cost and cost-effectiveness.

[0020] As a preferred method, a variable acceleration factor and a time-varying weight coefficient are introduced, and the cluster capacity configuration optimization model is solved by a particle swarm algorithm.

[0021] Based on the "ENDS-MEANS-WAYS" (EMW) analysis framework, the present invention comprehensively considers the mission objectives, system resources and organizational strategies of the drone cluster, establishes the capability requirement set and the alternative configuration element set of the drone cluster, and constructs the correlation matrix between capability and mission, load and capability. Furthermore, an evaluation model is designed, and six indicators, including endurance, protection, communication, collaborative perception, collaborative decision-making and collaborative execution, are selected to measure the capability of the drone cluster. Considering the cost and cost-effectiveness, an optimization model with high capability, low cost and high cost-effectiveness as the goal is established. The model is solved by the particle swarm algorithm to optimize the effectiveness of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0023] Figure 1 A flowchart of a method for optimizing capability configuration of heterogeneous drone clusters according to an embodiment of the present invention;

[0024] Figure 2 Schematic diagram of the capability configuration optimization method for heterogeneous UAV clusters;

[0025] Figure 3 Design architecture for cluster configuration elements based on EMW ideas

[0026] Figure 4 Structured description of resources for heterogeneous drone cluster systems

[0027] Figure 5 A UAV swarm formation strategy based on the pilot method

[0028] Figure 6 A UAV swarm formation strategy based on the pigeon flock method

[0029] Figure 7 A UAV swarm formation strategy based on fully autonomous method DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] Embodiment 1:

[0033] like Figure 1 As shown, an embodiment of the present invention provides a capability configuration optimization method for a heterogeneous UAV cluster, including:

[0034] Step S1, obtaining the configuration elements of the drone cluster capability;

[0035] Step S2: obtaining a UAV cluster capability configuration optimization model according to the UAV cluster capability configuration elements;

[0036] Step S3: Solve the optimization model of drone cluster capability configuration.

[0037] As an implementation method of an embodiment of the present invention, the configuration elements of the drone cluster capability include: mission requirements, system resources and organizational strategies. Mission requirements include: endurance requirements, protection requirements, communication requirements, monitoring requirements, decision requirements and execution requirements. System resources include: endurance load, protection load, communication load, monitoring load, decision load, and execution load. Organizational strategies include: drone cluster pilot method, pigeon flock method, and fully autonomous formation strategy.

[0038] Furthermore, in step S1, based on the idea of ​​the “ENDS-MEANS-WAYS” (EMW) analysis framework, the cluster configuration element architecture is designed from three aspects: task requirements, system resources, and organizational strategies. The relationship between them is described as follows: Figure 3 As shown in the figure, mission requirements (ENDS) are the driving force for the organization of actions of drone swarms, driving the utilization of system resources and the selection of organizational strategies; system resources (MEANS) are the basis for drone swarms to complete tasks, and their availability and performance directly affect the realization of mission requirements; organizational strategies (WAYS) directly affect the performance of drone swarms in actual tasks by optimizing resource utilization to meet mission requirements.

[0039] The task requirements are as follows:

[0040] When performing tasks, drone swarms can flexibly match highly adaptable solutions according to different task requirements, and have the advantages of strong task adaptability and rapid response. In order to ensure that drone swarms can stably exert their best capabilities in a diverse and complex combat environment, it is necessary to analyze typical task requirements. Let the task requirement set of drone swarms be:

[0041] E=[E1,E2,…,E i , …, E I ] T

[0042] Among them, E i represents the i-th type of task requirement; I is the number of types of task requirements, and each type of task requirement contains multiple specific subtask requirements, that is,

[0043] E i =[e i1 , e i2 ,…,e ij ,…,e iJ ]

[0044] Where J is the number of task requirements of the i-th category; e ij is the jth sub-demand in the i-th task requirement.

[0045] When performing various mission activities, drone swarms need to have a series of key capabilities to ensure the successful completion of the mission. In response to these diverse mission requirements, we focus on six core capability requirements: endurance requirements, protection requirements, communication requirements, monitoring requirements, decision-making requirements, and execution requirements, as shown in Table 1. These capabilities together constitute the basic framework for drone swarms to perform missions, ensuring their efficient operation and responsiveness in various complex environments.

[0046] Table 1

[0047]

[0048]

[0049] Therefore, the task requirements for establishing a drone swarm are:

[0050]

[0051] Among them, e ij , i=1~6 respectively represent the j-th sub-demand of the drone cluster for endurance demand, protection demand, communication demand, monitoring demand, decision-making demand, and execution demand.

[0052] The system resource analysis is as follows:

[0053] The system resources of a drone cluster can be divided into several parts, including hardware resources, software resources, human resources, and infrastructure. The focus is on the optimal configuration of hardware resources. These hardware resources are the core of achieving efficient operation of heterogeneous drone clusters under variable mission conditions. In particular, the cluster's endurance module, protection module, communication module, and dedicated function module are the focus of research, and these modules contain a variety of payload options. The payload of each module is not only unique in function, but also different in performance level, thus constituting the diversity and flexibility of heterogeneous drone clusters. Table 2 lists in detail the payloads included in the capability and cost analysis of each payload.

[0054] Table 2

[0055]

[0056] Characterize the system resources of the heterogeneous UAV cluster, and assume that the system resource set is:

[0057] M=[M1, M2,…,Mk,…,M K ] T

[0058] Among them, K is the total number of modules; M k represents the kth module, which can be expressed as:

[0059] Mk=[mk1,mk2,…,mk1]

[0060] Among them, m kl is the lth load in the kth module.

[0061] For a heterogeneous UAV cluster, its system resource set M can be expressed as:

[0062]

[0063] Among them, M k , k = 1, 2, 3, 4 represent the endurance module, protection module, communication module and special function module of the UAV cluster respectively; m kl represents the load l in module k;

[0064] The modules and their internal loads in the heterogeneous UAV cluster system resources can be structurally described as follows: Figure 4 shown.

[0065] Organizational strategy analysis

[0066] The organizational strategy of the drone swarm coordinates the actions of drones by optimizing system resources, and then matches the cluster mission requirements to ensure that the drone swarm can operate efficiently and complete tasks in various mission scenarios. The impact of the drone formation strategy on the optimization of the drone cluster capability configuration is mainly considered. The embodiment of the present invention mainly considers the impact of three formation strategies: pilot method, pigeon flock method, and full autonomy method on the overall capability of the drone swarm. For a drone swarm of size N, its organizational strategy and cost are shown in Table 3. The drone swarm formation strategy based on the pilot method requires the designation of a leader and the planning of the leader's tasks. Other drones as followers need to adjust their own flight status according to the leader's flight status. Its structural control is as follows: Figure 5 The UAV cluster formation strategy based on the pigeon flock method is to define the behavior of each UAV in the cluster. Each UAV is at the same level. Its structure control is as follows: Figure 6 The fully autonomous UAV swarm formation strategy is that each UAV in the swarm has sufficient autonomy to form a formation autonomously. Its structural control is as follows: Figure 7 shown.

[0067] Table 3

[0068]

[0069] Referring to the system resource characterization method of drone clusters, the organizational strategy of drone clusters is characterized:

[0070] W = [W1, W2, W3]

[0071] Among them, W1, W2, and W3 represent the three formation strategies of drone swarms: pilot method, pigeon flock method, and full autonomy method.

[0072] As an implementation of the embodiment of the present invention, step S2 includes:

[0073] Step S21: According to the UAV cluster capability configuration elements, a comprehensive correlation matrix between the UAV cluster capability and the task requirements, a correlation matrix between the configuration elements and the cluster capability, and a correlation matrix between the configuration elements and the task requirements are obtained;

[0074] Step S22, obtaining a cluster configuration optimization evaluation model according to the comprehensive correlation matrix between the UAV cluster capability and the task requirement, the correlation matrix between the configuration element and the cluster capability, and the correlation matrix between the configuration element and the task requirement;

[0075] Step S23: Obtain a UAV cluster capability configuration optimization model based on the cluster configuration optimization evaluation model.

[0076] Further, in step S21, the comprehensive correlation matrix between the UAV cluster capability and the mission requirements is as follows:

[0077] For different tasks, the weights corresponding to various requirements in the task requirement set of the drone cluster are different. For task q, the weights corresponding to various requirements in the task requirement set E are:

[0078] w E q =[w e1 q , w e2 q , w e3 q , w e4 q , w e5 q , w e6 q ] T

[0079] Among them, w ei , i=1-6 respectively represent the weights corresponding to the endurance requirement, protection requirement, communication requirement, monitoring requirement, decision-making requirement, and execution requirement of the drone cluster.

[0080] Assuming that the drone cluster contains L capabilities, the capability set can be expressed as:

[0081] H=[H1,H2,…,H l , …, H L ]

[0082] Then the general cluster capability-task requirement correlation matrix is:

[0083]

[0084] Among them, a li represents the correlation between the lth cluster capability and the ith requirement. Considering the weight of the requirement w E q After that, the comprehensive correlation matrix of cluster capability and task requirements in task q is obtained:

[0085] M H-E q =M H-E ·w E q =[λ1 q ,λ2 q , …, λ l q , …, λ L q ] T

[0086] Among them, λ l q It represents the comprehensive correlation between the lth cluster capability and all task requirements in task q, and its expression is:

[0087]

[0088] Further, in step S21, the configuration element and cluster capability correlation matrix is ​​as follows:

[0089] The configuration elements of the drone cluster include two aspects: the system resources and organizational strategies of the drone cluster, that is, K payloads and three formation strategies. Suppose the configuration element set of the drone cluster is:

[0090] P=[P1,P2,…,P r ,…,P R ]

[0091] Among them, P r represents the rth configuration element; R is the total number of configuration elements, and R=K+3.

[0092] The capability set corresponding to the cluster configuration element is:

[0093] h=[h1,h2,…,h r ,…,h R ]

[0094] The selection vector of the capability configuration of the drone cluster is:

[0095] n=[n1,n2,…,n r ,…,n R ] T

[0096] Cluster capacity as a function of the type, performance, and number of configuration elements can be mathematically expressed as:

[0097] F l =f l (h1,…,h R ,n1,…,n R )

[0098] Among them, F l , l = 1-6, indicating the lth capability of the drone cluster; f l Represents the function corresponding to the lth ability.

[0099] Let the cluster configuration element-cluster capability association matrix be expressed as:

[0100]

[0101] Among them, b rl Represents the degree of association between the rth configuration element and the lth capability.

[0102] If f is a known analytical function and is differentiable, the absolute value of the partial derivative can be used to evaluate the relevance of each configuration element to the cluster capability. That is:

[0103]

[0104] If f is a function based on empirical data or a complex model and is not differentiable, Monte Carlo simulation, sensitivity analysis, experimental design and other methods can be used to evaluate the correlation between each configuration element and cluster capability.

[0105] Further, in step S21, the correlation matrix between the configuration elements and the task requirements is as follows:

[0106] Using the UAV cluster capability as a “bridge”, the relationship between configuration elements and mission requirements is established. That is, according to the matrix M H-E q and M P-H It can be obtained that in task q, the configuration element-requirement comprehensive correlation matrix can be expressed as:

[0107] M P-E q =[ε1 q , ε2 q , …, ε r q , …ε R q ]

[0108] Among them, ε r qis the correlation between the rth configuration element and the cluster task requirement set, indicating the degree to which the configuration element satisfies all requirements. Its expression is:

[0109]

[0110] Further, in step S22, cluster capability and cost are the core driving forces for the development of drone cluster systems, which directly affect the success of the mission and also determine the competitiveness and potential of drone cluster missions. In different tasks of drone clusters, the task objectives of the cluster are different. The pursuit of excessive capabilities may lead to high costs, while excessive focus on costs may damage capabilities. How to achieve the best balance between capabilities and costs and enhance the competitiveness of drone clusters has become the focus of research attention. Therefore, in order to develop a drone cluster capability configuration optimization model that meets different mission objectives, the embodiment of the present invention selects three types of indicators: drone cluster capability, cost, and cost-effectiveness ratio to evaluate the cluster configuration scheme.

[0111] (1) Capabilities

[0112] The six indicators of endurance E1, protection E2, communication E3, collaborative perception E4, collaborative decision-making E5, and collaborative execution E6 are selected to measure the UAV cluster capability. E q , the overall capability of the drone cluster is calculated:

[0113]

[0114] Among them, ε r q is the correlation between the rth configuration element and the cluster task requirement set; N is the size of the drone cluster; n is the selection vector of the configuration element; h is the capability set corresponding to the cluster configuration element.

[0115] (2) Cost

[0116] The cost covers the entire life cycle costs including research and development, production and use support. The cost level affects the effective use of resources and strategic execution. The main considerations are the cost of the UAV fuselage, various types of payloads and control structures.

[0117] Build the cost vector for the cluster configuration element:

[0118] C b =[c1,…,c r , …, c R ]

[0119] Among them, C r Represents the configuration element P r cost.

[0120] The total cost can be expressed as:

[0121]

[0122] Among them, N represents the size of the drone cluster; Ca represents the cost of the drone body.

[0123] (3) Cost-effectiveness

[0124] Based on the evaluation of the UAV swarm mission capability, the cost-effectiveness of the UAV is evaluated by considering the cost. The cost-effectiveness is defined as the ratio of capability E to cost C, and its calculation formula is:

[0125]

[0126] Among them, ECR is the cost-effectiveness ratio; E is the capability; and C is the cost.

[0127] Furthermore, in step S23, the cluster capability configuration load includes a load in each module, such as a high endurance load, a medium endurance load or other load in the endurance module. At the same time, the drone configured with the monitoring load is configured with a high endurance load, the drone configured with the decision-making load must carry a high communication load, and the drone configured with the execution load must be configured with a high protection load. In actual use, it is necessary to consider the impact of mission requirements and cost constraints on cluster configuration, set the maximum cost to C0, and the minimum cost of each type of cluster capability to E. i0 , (i=1~6 are the endurance, protection, communication, collaborative perception, collaborative decision-making and collaborative execution capabilities of the UAV cluster respectively). Considering the configuration load, mission requirements, cost constraints and other constraints of the UAV cluster, a cluster capability configuration optimization model is established.

[0128] According to the three evaluation models of cluster capability, cost and cost-effectiveness ratio, the configuration optimization schemes of UAV cluster capability with a scale of N are established respectively.

[0129] The goal is to maximize cluster capacity:

[0130] MaxE=Maxf(N,n,h)

[0131]

[0132] Aiming to minimize cost:

[0133] MinC=N*C a +C b *n+N*C c

[0134]

[0135] The goal is to maximize the cost-effectiveness:

[0136]

[0137] As an implementation method of the present invention, in step S3, when solving the cluster capacity configuration optimization model by particle swarm algorithm, a variable acceleration factor and a time-varying weight coefficient are introduced to improve the speed and range of the particle swarm search in the early stage. The specific steps of solving are as follows:

[0138] Step 31: Initialize the particle swarm. Randomly generate the position x of T groups of particles i and speed v i , each particle represents a possible configuration scheme for the UAV swarm capability. Each dimension represents the number of a payload. For example, the first dimension is the number of high-endurance payloads n1 in the endurance module.

[0139] Step 32: Calculate the fitness function. Evaluate the configuration scheme of each particle, and calculate the fitness of each particle based on the capability E, cost C and cost-effectiveness ratio ECR function of the drone cluster.

[0140] Step 33, update local optimum and global optimum: compare the fitness value of each particle's current position with the fitness value of the previously recorded individual optimal position, and update the individual optimal position P i Among all particles, find the particle with the highest fitness value and take its position as the global optimal solution P g .

[0141] Step 34, update the particle position and velocity. Update the velocity and position of each particle based on the individual optimal and global optimal positions. The update formula using the new position and velocity of the particle is as follows:

[0142] v i (t+1)=w·v i (t)+c1·r1·(p i -x i (t))+c2·r2·(P g -x i (t))

[0143] x i (t+1)=x i (t)+v i (t+1)

[0144] Where w is the inertia weight, c1 and c2 are acceleration constants, and r1 and r2 are random numbers in the interval [0,1].

[0145]

[0146] Among them, c 1max 、c 1min 、c2max 、c 2min ,ω max ,ω min are the maximum and minimum values ​​of the acceleration factor and inertia weight, respectively, t i is the real-time iteration number, and T is the maximum iteration number.

[0147] Step 35, iteration. If the configuration solution meets the constraints, repeat steps 2 to 4 until the quality of the solution reaches a preset threshold or the maximum number of iterations is reached to meet the stop condition. If the constraints are not met, reset the particle position vector.

[0148] Step 36: Output the optimal solution. Output the global optimal drone cluster configuration solution P g , obtain the particle position vector x with the highest capability, lowest cost, or highest cost-effectiveness i .

[0149] Based on the "ENDS-MEANS-WAYS" (EMW) analysis framework theory, the research on the optimization of UAV cluster capability configuration can be generally studied from three aspects: the cluster's mission objectives (ENDS), system resources (MEANS) and organizational strategies (WAYS). Mission objectives drive the selection and allocation of system resources. The availability and performance of system resources directly affect the realization of mission requirements. The organizational method optimizes resource utilization to meet mission objectives. Specifically, it is necessary to first clarify the mission objectives of the UAV cluster, such as monitoring, transportation, rescue, etc., as well as the priority of capabilities, costs, and cost-effectiveness. Secondly, evaluate the cluster resources, including the types and numbers of UAVs, payload types, payload performance, etc., and adopt appropriate organizational strategies such as cluster formation strategies and task allocation strategies to achieve the optimal configuration of resources under limited resource conditions. Finally, guided by the mission objectives, the overall capability and adaptability of the cluster can be improved by coordinating and adjusting cluster resources and strategies, and the mission can be completed efficiently.

[0150] According to the above analysis, the embodiment of the present invention starts from the perspective of clusters and based on the "EMW" concept, studies the overall configuration optimization of drone clusters to improve the capabilities of drone clusters at low cost. Aiming at practical problems such as drone cluster resource and cost constraints, guided by task requirements, and utilizing cluster system resources and organizational strategies, a drone cluster capability configuration optimization model with high capability, low cost, and high efficiency cost ratio as the goal is established. For heterogeneous drone clusters, a high-capability and low-cost cluster capability configuration solution is formed to promote the transformation of drone design from "single machine" to "cluster", "optimal performance" to "capability and cost coordination", and "single configuration" to "overall configuration", so that drone clusters can be used on a large scale, providing technical support and scientific guidance for the development of drone clusters. This has important scientific significance and strategic value for promoting the development of drone cluster systems and improving competitiveness.

[0151] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A capability configuration optimization method for heterogeneous UAV clusters, characterized in that: include: Step S1, obtaining the configuration elements of the drone cluster capability; Step S2: Obtaining a UAV cluster capability configuration optimization model according to the UAV cluster capability configuration elements; Step S3: Solve the optimization model of drone cluster capability configuration.

2. The capability configuration optimization method for heterogeneous UAV clusters according to claim 1, characterized in that: The configuration elements of drone swarm capabilities include: mission requirements, system resources and organizational strategies.

3. The capability configuration optimization method for heterogeneous UAV clusters according to claim 2 is characterized in that: Mission requirements include: endurance requirements, protection requirements, communication requirements, monitoring requirements, decision-making requirements and execution requirements.

4. The capability configuration optimization method for heterogeneous UAV clusters according to claim 3 is characterized in that: System resources include: endurance payload, protection payload, communication payload, monitoring payload, decision payload, and execution payload.

5. The capability configuration optimization method for heterogeneous UAV clusters according to claim 4 is characterized in that: The organizational strategies include: drone swarm piloting method, pigeon flock method, and fully autonomous formation strategy.

6. The capability configuration optimization method for heterogeneous UAV clusters according to claim 5, characterized in that: Step S2 includes: According to the configuration elements of the UAV cluster capability, the comprehensive correlation matrix between the UAV cluster capability and the mission requirements, the correlation matrix between the configuration elements and the cluster capability, and the correlation matrix between the configuration elements and the mission requirements are obtained; According to the comprehensive correlation matrix between UAV cluster capability and mission requirements, the correlation matrix between configuration elements and cluster capability, and the correlation matrix between configuration elements and mission requirements, the cluster configuration optimization evaluation model is obtained; According to the cluster configuration optimization evaluation model, the UAV cluster capability configuration optimization model is obtained.

7. The capability configuration optimization method for heterogeneous UAV clusters according to claim 6 is characterized in that: Evaluate swarm configurations based on drone swarm capabilities, costs, and cost-effectiveness.

8. The capability configuration optimization method for heterogeneous UAV clusters according to claim 7, characterized in that: Variable acceleration factors and time-varying weight coefficients are introduced, and the cluster capacity configuration optimization model is solved using the particle swarm algorithm.

Images