An optimal layout planning method for multi-aircraft formation operation

Abstract

The application belongs to the fields of unmanned aerial vehicle cluster flight and layout optimization, and discloses an optimal layout planning method for multi-aircraft cluster operation, which is based on a multi-combination separation docking distribution model, randomly generates N groups of first structure parameters, performs non-dominated sorting on the N groups of first structure parameters, and marks the N groups of first structure parameters after the non-dominated sorting as an initial population; the initial population is iteratively processed through an M-time genetic algorithm model, and the initial population is updated after each iteration through the results of non-dominated sorting and crowding degree calculation; an evaluation model is constructed to evaluate the initial population, an evaluation result is output, and a structure parameter corresponding to the highest evaluation result is selected, so that the optimal structure parameter of the cluster layout is determined.

Description

Technical Field

[0001] This invention relates to the fields of unmanned aerial vehicle (UAV) swarm flight and layout optimization, and more specifically, to an optimal layout planning method for multi-aircraft swarm operations. Background Technology

[0002] In the logistics and transportation sector, drones possess enormous development potential. Given the complex road traffic networks resulting from rapid urban development, drones can improve logistics efficiency while alleviating road traffic congestion, enabling rapid and accurate delivery of various goods, and saving labor costs.

[0003] Quadcopter drones are the most common type of multi-rotor drone. They use the rotational speeds of their four rotors as control variables to control the drone's six degrees of freedom in space, making them a highly coupled, underactuated dynamic system. Besides possessing many advantages of general multi-rotor drones, quadcopter drones are also characterized by their small size, high maneuverability, and ease of control. When performing transport missions, they can flexibly and quickly navigate narrow and complex terrain to achieve precise delivery of targets. However, due to limitations in energy conversion efficiency and thrust-to-weight ratio, quadcopter drones have limited payload capacity, cannot transport large objects, and have relatively poor endurance.

[0004] In view of this, the present invention provides an optimal layout planning method for multi-aircraft tandem operation. Summary of the Invention

[0005] To overcome the problems in the prior art, this invention proposes an optimal layout planning method for multi-aircraft tandem operations.

[0006] According to one aspect of the present invention, an optimal layout planning method for multi-aircraft tandem operations is provided, characterized by comprising the following steps:

[0007] Step S1: Connect multiple UAVs into a flight platform system through kinematic pairs, construct a multi-combination separation and docking distribution model based on the flight platform system, and generate a first structural parameter based on the multi-combination separation and docking distribution model. The first structural parameter includes several parameter variables generated by the kinematic pairs.

[0008] Step S2: Randomly extract N sets of first structural parameters, perform non-dominated sorting on the N sets of first structural parameters, and mark the N sets of first structural parameters after non-dominated sorting as the initial population;

[0009] Step S3: The initial population is iteratively processed using a genetic algorithm model of M iterations, where M is the preset number of iterations. After each iteration, the initial population is updated using the results of non-dominated sorting and crowding calculation.

[0010] Step S4: Construct a fusion evaluation model to evaluate the initial population, output the evaluation results, and select the structural parameters corresponding to the highest evaluation results to determine the optimal structural parameters of the clustered layout.

[0011] In a preferred embodiment, the generation logic of the first structural parameter is as follows:

[0012] The kinematic joint design involves multiple aircraft connected by multiple docking units. Each docking unit includes a first docking unit, a second docking unit, and a third docking unit. The second and third docking units are symmetrically arranged about the plane where the first docking unit is located, forming an angle between the second and third docking units and the plane where the first docking unit is located.

[0013] In a preferred embodiment, the specific steps of the non-dominated sorting are as follows:

[0014] Step S21: Extract the i-th group of first structural parameters, where i = 1, 2, 3, ..., N; based on the i-th group of first structural parameters, obtain the corresponding first objective function and the sum of the second objective functions f. i ;

[0015] Step S22: For all N groups of first structural parameters, arbitrarily select the j-th group of first structural parameters, j = 1, 2, 3, ..., N, and j ≠ i. Based on the j-th group of first structural parameters, obtain the sum of the first objective function and the second objective function f. j ;

[0016] If there does not exist a sum of objective functions f j Better than the sum of the objective functions f i Then the objective function f i As a non-dominant individual;

[0017] Step S23: Perform steps S21-S22 once for all first structural parameters within the N groups of first structural parameters to obtain non-dominated individuals of all first structural parameters;

[0018] Step S24: Based on the values ​​calculated by the objective function of the N groups of first structural parameters, perform a non-dominated sort on the N groups of first structural parameters, and mark the first structural parameters after the non-dominated sort as the initial population.

[0019] In a preferred embodiment, the specific steps of step S3 are as follows:

[0020] Step S31: Obtain the initial population. The preset number of iterations for the initial population is M, where M > 0; define the number of iterations corresponding to the initial population as m = 1.

[0021] Step S32: The initial population is further processed by the genetic algorithm model to generate N sets of second structural parameters. The second structural parameters are the structural parameters after the N sets of first structural parameters are selected, crossed over and mutated.

[0022] Step S33: Merge the N sets of second structural parameters with the N sets of first structural parameters in the initial population to generate 2N sets of third structural parameters, wherein the third structural parameters are the sum of the second structural parameters and the first structural parameters;

[0023] Step S34: Extract the first N sets of third structural parameters from the 2N sets of third structural parameters through non-dominated sorting and crowding calculation as the first subpopulation, update the initial population according to the first subpopulation, and update the corresponding third structural parameters to the first structural parameters.

[0024] Step S35: Compare the number of iterations corresponding to the initial population with the preset number of iterations. If m < M, then m = m + 1. The initial population is iteratively processed by the genetic algorithm model. Repeat steps S31-S34.

[0025] If m ≥ M, then the iterative process ends and the updated initial population is output.

[0026] In a preferred embodiment, the processing logic of the genetic algorithm model is as follows:

[0027] Randomly select the k-th first structural parameter G from the N sets of first structural parameters for initializing the population. k and the first structural parameter G of the p-th group p And k = 1, 2, 3, ..., N; p = 1, 2, 3, ..., N, k ≠ p, N ≥ 1;

[0028] The first structural parameter G k and the first structural parameter G p Several variables are randomly selected and cross-interleaved to complete one selection and cross-interaction, forming the first structure cross-parameter.

[0029] Based on the first structural cross parameters, several variables are randomly extracted from the first structural cross parameters, and random values ​​are added to complete one selection and mutation to form the first structural mutation parameters; the first structural mutation parameters are marked as the second structural parameters.

[0030] In a preferred embodiment, the specific steps for calculating congestion are as follows:

[0031] The crowding degree of each third structural parameter q in the 2N groups of third structural parameters is D. q Defined as 0;

[0032] Perform non-dominated sorting on the 2N groups of third structural parameters to obtain the first N groups of third structural parameters after non-dominated sorting. Define the crowding degree of the third structural parameters corresponding to the two ends of the non-dominated sorting as infinite.

[0033] The crowding degree of the non-boundary third structural parameters after non-dominated sorting is calculated by formula. Based on the crowding degree, the non-boundary third structural parameters after non-dominated sorting are reordered again, and the reordered third structural parameters are updated as the first structural parameters.

[0034] In a preferred embodiment, the fusion evaluation model includes a spatial evaluation function and a hypervolume index function, and the fusion evaluation result is generated based on the fusion evaluation model.

[0035] According to another aspect of the present invention, an electronic device includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0036] The processor executes the aforementioned optimal layout planning method for multi-aircraft tandem operation by calling the computer program stored in the memory.

[0037] The technical effects and advantages of the optimal layout planning method for multi-aircraft tandem operations of the present invention are as follows:

[0038] This invention connects multiple quadcopter drones into a spatial layout flight platform using a dynamic model, effectively expanding the contact area for drone transportation. This compensates for the shortcomings of individual quadcopter drones, such as small load and underactuation, enabling the transportation of larger objects. This directly determines the load capacity and spatial throughput of the connected platform, meeting the multi-drone transportation needs in multi-mission scenarios and greatly improving the adaptability and flexible application capabilities of quadcopter drones. Attached Figure Description

[0039] Figure 1 This is a flowchart of the optimal layout planning method for multi-aircraft tandem operation according to the present invention;

[0040] Figure 2 This is a flowchart of the genetic algorithm model for multi-aircraft ensemble operation of the present invention;

[0041] Figure 3 This is a schematic diagram of the non-uniform distribution model of the docking units of the present invention;

[0042] Figure 4 This is a schematic diagram of the uniformly distributed docking unit model of the present invention;

[0043] Figure 5 This is the mechanical model of a single docking unit of the present invention.

[0044] In the diagram: 1. First docking unit; 2. Second docking unit; 3. Third docking unit; 4. First ball joint; 5. First link; 6. Second link; 7. Second ball joint; 8. Flight platform. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] This invention primarily utilizes multiple quadcopter drones (UAVs) to form a flight platform, effectively expanding the contact area for UAV transport. This overcomes the limitations of individual quadcopter drones, such as small payload and underactuation, enabling the transport of heavier objects. The choice of UAV aggregation type and the spatial layout of the drones are crucial, directly determining the payload capacity and maneuverability of the aggregation platform. By designing multi-UAV combination / separation docking units, multiple quadcopter drones are connected into an aggregation platform. Optimization algorithms are used to solve for suitable geometric parameters and spatial layout of the aggregation platform, allowing for the design of various multi-UAV aggregation forms to meet the multi-UAV transport needs in multi-mission scenarios. The multi-UAV aggregation structure can significantly improve the adaptability and flexibility of quadcopter drones, possessing significant practical engineering value.

[0047] Example 1

[0048] Please see Figure 1-5 As shown in this embodiment, an optimal layout planning method for multi-aircraft tandem operations includes the following steps:

[0049] Step S1: Connect multiple UAVs into a flight platform system through kinematic pairs, construct a multi-combination separation and docking distribution model based on the flight platform system, and generate a first structural parameter based on the multi-combination separation and docking distribution model. The first structural parameter includes several parameter variables generated by the kinematic pairs.

[0050] It's important to note that connecting multiple drones together via kinematic pairs to form a unified flight platform system effectively expands the contact area for drone transport, thus compensating for the limitations of a single quadcopter drone in terms of small payload and underactuation, enabling the transport of heavier objects.

[0051] In practical applications: multiple drones refer to three or more drones; all drones are linked together by kinematic joints. Similarly, the more drones there are, the higher the complexity of the corresponding flight platform system. Therefore, considering complexity and cost, choosing to link three drones together is a reasonable starting point, and then expanding according to needs.

[0052] The generation logic for the first structural parameter is as follows:

[0053] Based on the multi-vehicle ensemble operation, a multi-combination separation and docking distribution model is constructed. Several parameter variables between the multiple vehicles and the flight platform 8 are labeled as the first structural parameters, such as... Figure 3 As shown, in the operation of three UAVs in a coordinated manner, considering factors such as flight stability and force balance of the coordinated system, although the three UAVs are not uniformly distributed, at least two UAVs should be symmetrical about the plane containing another docking / separation unit. That is, the kinematic joint coordination design involves multiple aircraft connected by multiple docking units. Specifically, the docking units include a first docking unit 1, a second docking unit 2, and a third docking unit 3. The second docking unit 2 and the third docking unit 3 form an angle β about the plane containing the first docking unit 1. Since the parameter variables of the first docking unit 1 and the second docking unit 2 / the third docking unit 3 are independent of each other, the parameter variables include α. 11 α 12 , θ1, l, x1, F1, α 21 α 22 , θ2, l', x2, F2 and β, where: α 11 α 12 The parameter variables corresponding to θ1, l, x1, and F1 are the parameters corresponding to the first docking unit 1, α 21 α 22 The parameter variables corresponding to θ2, l', x2, and F2 are the parameters corresponding to the second docking unit 2 / the third docking unit 3.

[0054] For example, the mechanical models of the first docking unit 1 and the second docking unit 2 / third docking unit 3 are still as follows Figure 5 As shown, there are six adjustable variables.

[0055] The formula for the constraint conditions corresponding to the multi-combination separation docking distribution model is as follows:

[0056] 4F1cos(α 11 +α 12 -θ1)+2×4F2cos(α 21 +α 22 -θ2)=3G r +3G l1 +3G l2 +G p +G (1)

[0057] 4F1sin(α 11 +α 12 -θ1)=2×4F2sin(α 21 +α 22 -θ2)cosβ (2)

[0058]

[0059] Among them: G r G is the unloaded takeoff weight of the drone. l1 For the weight of the first link 5, G l2 For the weight of the second link 6, G p G represents the gravity of flight platform 8, G represents the gravity of the load, and the gravity of the ball joint is negligible.

[0060] α 11 The angle of the first ball joint 4 corresponding to the first docking unit 1 determines the direction of the pull force of the four rotors of the multi-rotor UAV; α 12 The angle of the second ball joint 7 corresponding to the first docking unit 1 determines the direction of the second link 6; θ1 is the angle of the first link 5 relative to the second link 6 corresponding to the first docking unit 1, determining the degree of bending of the long link; l1 is the position of the pin corresponding to the first docking unit 1 in the inner groove of the first link 5, determining the length of the long link; x1 is the position of the second ball joint 7 corresponding to the first docking unit 1 from the center of the flight platform 8, determining the position of the force point of the flight platform 8; F1 is the lift of the single rotor of the quadcopter UAV corresponding to the first docking unit 1, specifically as follows... Figure 5 As shown;

[0061] Similarly: α 21 The angle of the first ball joint 4 corresponding to the second docking unit 2 / third docking unit 3 determines the direction of the pull force of the four rotors of the multi-rotor UAV; α 22 The angle of the second ball joint 7 corresponding to the second docking unit 2 / third docking unit 3 determines the direction of the second link 6; θ2 is the angle of the first link 5 relative to the second link 6 corresponding to the second docking unit 2 / third docking unit 3, determining the degree of bending of the long link; l2 is the position of the pin corresponding to the second docking unit 2 / third docking unit 3 in the inner groove of the first link 5, determining the length of the long link; x2 is the position of the second ball joint of the second docking unit 2 / third docking unit 3 at the center of the flight platform 8, determining the position of the force point of the flight platform 8; F2 is the lift of the single rotor of the quadcopter UAV corresponding to the second docking unit 2 / third docking unit 3.

[0062] The formula for obtaining the objective function from the multi-combination separation docking distribution model is as follows:

[0063] f1 = 4F1·cos(α)11 +α 12 -θ1)+2×4F2cos(α 21 +α 22 -θ2)-3.668(4)

[0064]

[0065] w s =2sinβ·(l2sinα) 22 +(l1-l')sin(α 22 -θ2)+x2) (6)

[0066]

[0067] f2=(w' s +560)·(h′ s +150) (8)

[0068] Where: f1 is the first objective function, L is the dynamic loss function of the ensemble; w s h is the horizontal width of the shaded area. s f1 represents the vertical height of the shaded area; f2 represents the second objective function.

[0069] For a specific example, if β = 60°, then the decision space of the multi-combination separation-docking uniform distribution model formed by the three sets of docking units, due to the symmetrical distribution, should have the same decision variables for the three sets of docking units. Therefore, the decision variables of this model only have six variables: F, α1, α2, θ, l, and x. Figure 4 As shown; since the three quadcopter drones are evenly distributed, the torque balance and the horizontal force balance are satisfied, and no additional constraints need to be set. Therefore, in order to obtain the maximum load of the assembly, the lift F of the single rotor of the quadcopter drone can be directly taken as the maximum value of 0.625; the multi-combination separation docking uniform distribution model is simplified to five dimensions, namely five variables: α1, α2, θ, l, and x.

[0070] The specific formula is as follows:

[0071] The only equation that the model variables need to satisfy is shown in equation (9), which is the force balance equation of the vertical direction of the ensemble structure. The variables in the equation are defined as in 3.3.2. The basic requirement for the ensemble to take off normally is that the load mass is greater than 0.

[0072]

[0073] To ensure that the quadcopter drones in the assembly do not interfere with each other, the distance between two adjacent docking / separation units should be greater than the minimum horizontal clearance distance of a single drone, i.e.:

[0074]

[0075] The objective functions of this optimization problem are mainly the maximum load f1 of the ensemble structure and the vertical through area f2 of the ensemble structure. The maximum load of the ensemble structure can be solved by the force balance equation (9) of the ensemble in the vertical direction. Substituting the known values ​​into the equation, we can obtain the expression (11) for the first objective function f1 of the mathematical model.

[0076] f1=7.5cos(α1+α2-θ)-3.668 (11)

[0077] The power loss of the ensemble is related to the ratio of the total lift of the three quadcopter UAV propellers to the load of the ensemble. Since the horizontal force balance and torque balance do not need to be considered when the three combination / separation docking units are evenly distributed, the maximum load can be represented by the maximum power of each UAV. Therefore, the power loss L of the ensemble at this time can be represented by equation (12).

[0078]

[0079] The vertical area swept by the assembly / disassembly docking unit is represented by the corresponding horizontal width w. s As shown in equation (13).

[0080] w s =(l2sinα2+(l1-l)sin(α2-θ)+x)·cos120° (13)

[0081] The vertical height h corresponding to this area s As shown in equation (14).

[0082] h s =l2cosα2+(l1-l)cos(α2-θ) (14)

[0083] Based solely on the vertical area of ​​the entire ensemble structure, the horizontal distance plus the minimum width of the quadcopter drone, and the vertical distance plus the height of the quadcopter drone, we obtain the expression for the second objective function f2 of the mathematical model (15).

[0084] f2=(w s +560)·(h s +150) (15)

[0085] Similarly, regarding the non-uniform distribution modeling of multiple combination / separation docking units, specifically, the uniform distribution analyzed above theoretically offers higher space utilization. Considering factors such as flight stability and force balance of the ensemble, although the three quadcopter UAVs are not uniformly distributed, at least two UAVs should be symmetrical about the plane containing the other combination / separation docking unit, such as... Figure 4 As shown.

[0086] Step S2: Randomly extract N sets of first structural parameters, perform non-dominated sorting on the N sets of first structural parameters, and mark the N sets of first structural parameters after non-dominated sorting as the initial population;

[0087] For example: N=100, where the N sets of first structural parameters are specific structural parameters extracted in advance by technicians from the multi-aircraft tandem operation based on the actual situation, and how multiple UAVs in multi-mission scenarios can quickly adapt to changing scenario requirements.

[0088] The specific steps of non-dominated sorting are as follows:

[0089] Step S21: Extract the i-th group of first structural parameters, where i = 1, 2, 3, ..., N; based on the i-th group of first structural parameters, obtain the corresponding first objective function and the sum of the second objective functions f. i ;

[0090] Step S22: For all N groups of first structural parameters, arbitrarily select the j-th group of first structural parameters, j = 1, 2, 3, ..., N, and j ≠ i. Based on the j-th group of first structural parameters, obtain the sum of the first objective function and the second objective function f. j ;

[0091] If there does not exist a sum of objective functions f j Better than the sum of the objective functions f i Then the objective function f i As a non-dominant individual;

[0092] Step S23: Perform steps S21-S22 once for all first structural parameters within the N groups of first structural parameters to obtain non-dominated individuals of all first structural parameters;

[0093] Step S24: Based on the values ​​calculated by the objective function of the N groups of first structural parameters, perform a non-dominated sort on the N groups of first structural parameters, and mark the first structural parameters after the non-dominated sort as the initial population.

[0094] Step S3: The initial population is iteratively processed using a genetic algorithm model of M iterations, where M is the preset number of iterations. After each iteration, the initial population is updated using the results of non-dominated sorting and crowding calculation.

[0095] The specific operation steps of step S3 are as follows: Figure 2 As shown:

[0096] Step S31: Obtain the initial population. The preset number of iterations for the initial population is M, where M > 0; define the number of iterations corresponding to the initial population as m = 1.

[0097] Step S32: The initial population is further processed by the genetic algorithm model to generate N sets of second structural parameters. The second structural parameters are the structural parameters after the N sets of first structural parameters are selected, crossed over and mutated.

[0098] Step S33: Merge the N sets of second structural parameters with the N sets of first structural parameters in the initial population to generate 2N sets of third structural parameters, wherein the third structural parameters are the sum of the second structural parameters and the first structural parameters;

[0099] Step S34: Extract the first N sets of third structural parameters from the 2N sets of third structural parameters through non-dominated sorting and crowding calculation as the first subpopulation, update the initial population according to the first subpopulation, and update the corresponding third structural parameters to the first structural parameters.

[0100] Step S35: Compare the number of iterations corresponding to the initial population with the preset number of iterations. If m < M, then m = m + 1. The initial population is iteratively processed by the genetic algorithm model. Repeat steps S31-S34.

[0101] If m ≥ M, then the iterative process ends and the updated initial population is output.

[0102] The processing logic of the genetic algorithm model is as follows:

[0103] Randomly select the k-th first structural parameter G from the N sets of first structural parameters for initializing the population. k and the first structural parameter G of the p-th group p And k = 1, 2, 3, ..., N; p = 1, 2, 3, ..., N, k ≠ p, N ≥ 1;

[0104] First structural parameter G k The parameters and variables are represented as follows:

[0105] in: and β k These are the parameter variables in the first structure parameter;

[0106] Similarly: the first structural parameter G p The parameters and variables are represented as follows:

[0107]

[0108] The first structural parameter G k and the first structural parameter G p Several variables are randomly selected and cross-interleaved to complete one selection and cross-interaction, forming the first structure cross-parameter.

[0109] For example, selection and crossover: that is, randomly selecting the first structural parameter G with number k from N groups of first structural parameters. k Select several parameter variables (e.g., select...) ), and the first structural parameter G with number p p Select several parameter variables (e.g., select...) They are interchanged to complete one selection and crossover.

[0110] Based on the first structural cross parameters, several variables are randomly extracted from the first structural cross parameters, and random values ​​are added to complete one selection and mutation to form the first structural mutation parameters; the first structural mutation parameters are marked as the second structural parameters.

[0111] For example, selection and mutation: randomly select the first structural crossover parameter, extract several variables from it, add random values ​​to achieve mutation; after the selection, crossover and mutation operations, the second structural parameter can be obtained.

[0112] The crowding degree is calculated for the third structural parameters of the 2N groups after non-dominated sorting. The specific steps for crowding degree calculation are as follows:

[0113] The crowding degree of each third structural parameter q in the 2N groups of third structural parameters is D. q The congestion level is D. q Defined as 0; q = 1, 2, 3, ..., 2N;

[0114] Perform a non-dominated sort on the 2N groups of third structural parameters to obtain the first N groups of third structural parameters after the non-dominated sort. Define the crowding degree of the third structural parameters corresponding to the two boundary points after the non-dominated sort as infinite, i.e., 1. q =∞, N q =∞;

[0115] The crowding degree is calculated using a formula for the third structural parameter of the non-boundary structure after non-dominated sorting. The specific formula is as follows:

[0116] Where: D q This represents the crowding degree of the q-th third structural parameter; This represents the z-th objective function value corresponding to the (q+1)-th third structural parameter; Let f1 represent the z-th objective function value corresponding to the (q-1)-th third structural parameter, where z = 1 or z = 2; f2 represent the first objective function value and the second objective function value, respectively.

[0117] Based on the non-dominant relationship and crowding degree, a suitable third structural parameter is selected to form a new parent population; finally, a new offspring population is generated through the basic operations of the genetic algorithm, and so on, until the program termination condition is met. The corresponding algorithm flowchart is as follows. Figure 2 As shown.

[0118] Step S4: Construct a fusion evaluation model to evaluate the initial population, output the evaluation results, and select the structural parameters corresponding to the highest evaluation results to determine the optimal structural parameters of the clustered layout.

[0119] The fusion evaluation model includes a spatial evaluation function and a hypervolume index function. The spatial evaluation function is as follows:

[0120]

[0121] Where: PF represents the known true Pareto front, d i It refers to the Euclidean distance between two continuous vectors on the non-dominated boundary of the solution set. It is the average value of each Euclidean distance. The Spacing evaluation index only measures the diversity of multi-objective optimization algorithms. The smaller the value, the more uniform the solution set. It is more suitable for solving two-dimensional objective space problems, but its effect is not ideal in the case of a large number of objectives.

[0122] The hypervolume index function HV(S) is:

[0123]

[0124] In the formula, VOL(·) represents the Lebesgue measure; the larger the HV(S) value, the closer the S value is to the entire Pareto front, indicating that the overall performance of the optimization algorithm is better;

[0125] Based on this, the result d of the fusion evaluation generated by the fusion evaluation model is:

[0126] d=α·Spacing+(1-α)HV(S) (19)

[0127] The weights are dynamically adjusted by the scaling factor α, and d is the result of the fusion evaluation. The smaller the d value, the better the optimization effect.

[0128] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0129] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired or wireless network. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0130] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0131] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0132] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0133] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0134] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0135] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0136] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An optimal layout planning method for multi-aircraft tandem operations, characterized in that, Includes the following steps: Step S1: Connect multiple UAVs into a flight platform system through kinematic pairs, construct a multi-combination separation and docking distribution model based on the flight platform system, and generate a first structural parameter based on the multi-combination separation and docking distribution model. The first structural parameter includes several parameter variables generated by the kinematic pairs. The generation logic for the first structural parameter is as follows: The kinematic joint design involves multiple aircraft connected by multiple docking units. Each docking unit includes a first docking unit, a second docking unit, and a third docking unit. The second and third docking units are symmetrically arranged about the plane where the first docking unit is located, forming an angle between the second and third docking units and the plane where the first docking unit is located. Step S2: Random extraction The first structural parameter of the group, for The first structural parameter of the group is sorted non-dominated, and the sorted non-dominated parameters are then sorted. The first structural parameter of the group is marked as the initial population; the specific steps of non-dominated sorting are as follows: Step S21: Extract the first Group 1 structural parameters, where Based on the first The first structural parameter of the group obtains the sum of the corresponding first objective function and second objective function. ; Step S22: For all Arbitrarily select the first structural parameter from the group Group first structural parameter, ,and Based on the first The first structural parameter of the group obtains the sum of the corresponding first objective function and second objective function. ; If there is no sum of objective functions Better than the sum of the objective functions Then the objective function As a non-dominant individual; Step S23: For All first structural parameters within the group are processed once using the above steps S21-S22, resulting in non-dominated individuals of all first structural parameters. Step S24: According to The value of the objective function calculated for the first structural parameter of the group is... Perform a non-dominated sort on the first structural parameter of the group, and mark the first structural parameter after the non-dominated sort is completed as the initial population; Step S3: Initial population passes The secondary genetic algorithm model is used for iterative processing, where The initial population is updated after each iteration using the results of non-dominated sorting and crowding calculation. Step S4: Construct a fusion evaluation model to evaluate the initial population, output the evaluation results, and select the structural parameters corresponding to the highest evaluation results to determine the optimal structural parameters of the clustered layout.

2. The optimal layout planning method for multi-aircraft tandem operation according to claim 1, characterized in that, The specific steps for step S3 are as follows: Step S31: Obtain the initial population. The preset number of iterations for the initial population is... , Define the number of iterations corresponding to the initial population. ; Step S32: The initial population is further processed using a genetic algorithm model to generate... The second structural parameter of the group is: The first structural parameter of the group is the structural parameter after selection, crossover, and mutation operations; Step S33: ... Group 2 structural parameters and the initial population The first structural parameters of the group are merged to generate A third structural parameter, wherein the third structural parameter is the sum of the second structural parameter and the first structural parameter; Step S34: The third structural parameter of group A is extracted after calculation using non-dominated ordering and crowding degree. The third structural parameter of the group is the first subpopulation. The initial population is updated based on the first subpopulation, and the corresponding third structural parameter is updated to the first structural parameter. Step S35: Compare and analyze the number of iterations corresponding to the initial population with the preset number of iterations. If ,but The initial population is iteratively processed using a genetic algorithm model, repeating steps S31-S34. like If the iteration process ends, the updated initial population is output.

3. The optimal layout planning method for multi-aircraft tandem operation according to claim 2, characterized in that, The processing logic of the genetic algorithm model is as follows: From initializing the population Randomly select the first structural parameter from the group. Group first structural parameters and the Group first structural parameters ,and ; , , ; The first structural parameter and first structural parameters Several variables are randomly selected and cross-interleaved to complete one selection and cross-interaction, forming the first structure cross-parameter. Based on the first structural cross parameters, several variables are randomly extracted from the first structural cross parameters, and random values ​​are added to complete one selection and mutation to form the first structural mutation parameters; the first structural mutation parameters are marked as the second structural parameters.

4. The optimal layout planning method for multi-aircraft tandem operation according to claim 3, characterized in that, The specific steps for calculating congestion are as follows: Each third structural parameter in the group of third structural parameters The congestion level is Defined as 0; right The third structural parameter of the group is used for non-dominated sorting to obtain the first group after non-dominated sorting. The third structural parameter of the group defines the crowding degree of the two boundary points corresponding to the third structural parameter after non-dominated sorting as infinite; The crowding degree of the non-boundary third structural parameters after non-dominated sorting is calculated by formula. Based on the crowding degree, the non-boundary third structural parameters after non-dominated sorting are reordered again, and the reordered third structural parameters are updated as the first structural parameters.

5. The optimal layout planning method for multi-aircraft tandem operation according to claim 4, characterized in that, The fusion evaluation model includes a spatial evaluation function and a hypervolume index function, and the fusion evaluation results are generated based on the fusion evaluation model.

6. An electronic device, characterized in that, include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the optimal layout planning method for multi-aircraft tandem operation as described in any one of claims 1 to 5 by calling the computer program stored in the memory.

Images