Unmanned aerial vehicle quantity allocation and spectrum resource optimization method and device, equipment and medium

Abstract

The application discloses a kind of unmanned aerial vehicle quantity allocation and spectrum resource optimization method, device, equipment and medium, it is related to unmanned aerial vehicle communication technical field, the method includes obtaining the number of fixed target distributed on the study ground area and the location of each fixed target;According to the above fixed target information and target optimization function, the optimal individual is determined using genetic tabu search algorithm;Target optimization function is the function that minimizes the total energy consumption of multi-unmanned aerial vehicle system under the restriction of the capacity of unmanned aerial vehicle battery, optimizes the number of unmanned aerial vehicles, the reconnaissance order of each unmanned aerial vehicle and spectrum use time;Genetic tabu search algorithm is the algorithm obtained by improving genetic algorithm using tabu search algorithm;Individual includes the number of unmanned aerial vehicles and the corresponding reconnaissance order of each unmanned aerial vehicle.The energy consumption of completing cooperative reconnaissance task can be further reduced by the application.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle communication, and in particular to a method and device for unmanned aerial vehicle quantity deployment and spectrum resource optimization for cooperative reconnaissance tasks, and a medium. BACKGROUND

[0002] Due to the high maneuverability, rapid deployment and flexible load carrying of unmanned aerial vehicles, they are widely used in various task scenarios, such as environmental reconnaissance, disaster search and rescue, and relay communication. With the increase in the number of fixed targets and task types, a single unmanned aerial vehicle is already unable to meet the task requirements, and multiple unmanned aerial vehicles need to cooperate with each other to meet the task requirements. Compared with a single unmanned aerial vehicle, a multiple unmanned aerial vehicle system has better task adaptability and application potential, and is more suitable for performing complex tasks. Although the multiple unmanned aerial vehicle system has great advantages, it also faces many challenges, such as the number deployment of unmanned aerial vehicles and spectrum usage conflicts. The multiple unmanned aerial vehicle system needs to determine a reasonable number deployment and form an efficient cooperative relationship. An unreasonable number deployment will increase the probability of resource waste and spectrum conflict of unmanned aerial vehicles. An unreasonable spectrum allocation may prolong the time of the task, and thus increase the energy consumption of the unmanned aerial vehicles. The number deployment, reconnaissance order and spectrum allocation of unmanned aerial vehicles are coupled with each other, and jointly determine the optimal reconnaissance scheme planning.

[0003] The cooperative reconnaissance task of deploying multiple unmanned aerial vehicles involves a task allocation problem, which is an active research field in recent years, and a large number of research results can be used for reference. Some research works study the task allocation of unmanned aerial vehicles from the perspective of energy consumption. For example, one document studies the multi-task cooperation problem of unmanned aerial vehicle groups, considers the overlapping and complementary relationship of task types, models the distributed task allocation problem of unmanned aerial vehicles as a coalition formation game problem, and proposes an energy-saving task cooperation scheme for heterogeneous multiple unmanned aerial vehicles to minimize the energy consumption of the coalition. One document studies the hierarchical unmanned aerial vehicle task scheduling problem in a self-organizing network, in which one layer of unmanned aerial vehicles performs regional coverage tasks and another layer of unmanned aerial vehicles provides edge computing services, and through block coordinate descent method and continuous convex approximation technology, the overall energy consumption of the unmanned aerial vehicles is minimized. One document studies the task joint formation method of manned aircraft and unmanned aerial vehicles, in which the demand of unmanned aerial vehicles for limited resources and unlimited resources is introduced in the task allocation of unmanned aerial vehicles, and a multi-objective optimization algorithm is proposed to reduce resource redundancy. One document studies the task allocation and route planning problem of multiple unmanned aerial vehicles from the perspective of energy efficiency, and uses a genetic algorithm to obtain a mutual preference list of unmanned aerial vehicles and tasks in path planning, and uses a matching algorithm to solve the task problem of multiple unmanned aerial vehicles. One document studies the optimization strategy of multiple unmanned aerial vehicles relay-assisted communication to improve energy efficiency.

[0004] Network optimization in multi-UAV system is the key to performance, and the existing research has done a lot of work on spectrum resource optimization. In order to avoid the communication self-interference of UAV cluster, the existing technology researches the spectrum allocation problem of task-driven UAV communication network, models the coupling relationship between task allocation and spectrum allocation as a game model, and proposes a coalition formation game to jointly optimize task selection and spectrum allocation. In view of the limitation that UAV can only join one coalition, resulting in low efficiency, the existing technology proposes an overlapping coalition formation game, which optimizes the allocation of task resources through partial cooperation of overlapping coalition members.

[0005] In the existing research on UAV task allocation, the main focus is on optimizing the energy consumption and resource benefit of the task, while ignoring the number of UAVs required to complete the task and the spectrum usage conflict problem it brings. In actual task planning, it is often necessary to calculate the number of UAVs required, and to allocate reasonable tasks and plan the supporting spectrum for each UAV. Spectrum sharing can effectively improve the spectrum efficiency and network performance of multi-UAV system, but at the same time it also puts higher requirements on the spectrum conflict resolution technology. SUMMARY

[0006] In view of this, the present application provides a UAV number allocation and spectrum resource optimization method, device, equipment and medium for cooperative reconnaissance tasks.

[0007] To achieve the above purpose, the present application provides the following solutions:

[0008] In a first aspect, the present application provides a UAV number allocation and spectrum resource optimization method, comprising:

[0009] Obtaining fixed target information distributed on the ground area of the study area; the fixed target information includes the number of fixed targets and the location of each fixed target;

[0010] According to the fixed target information and the target optimization function, an optimal individual is determined by using a genetic tabu search algorithm; the target optimization function is a function of optimizing the number of UAVs, the reconnaissance order of each UAV and the spectrum usage time under the limitation of UAV battery capacity, so as to minimize the total energy consumption of the multi-UAV system; the genetic tabu search algorithm is an algorithm obtained by improving the genetic algorithm using the tabu search algorithm; the individual includes the number of UAVs and the corresponding reconnaissance order of each UAV; the reconnaissance order is the order of reconnaissance of fixed targets.

[0011] In a second aspect, the present application provides a UAV number allocation and spectrum resource optimization device, comprising:

[0012] The data acquisition module acquires fixed target information distributed on the ground area of the research region; the fixed target information includes the number of fixed targets and the position of each fixed target;

[0013] The optimal individual determination module is configured to determine an optimal individual by using a genetic tabu search algorithm according to the fixed target information and a target optimization function; the target optimization function is a function of optimizing the number of unmanned aerial vehicles, the reconnaissance order of each unmanned aerial vehicle and the spectrum usage time so as to minimize the total energy consumption of the multi-unmanned aerial vehicle system under the limitation of the battery capacity of the unmanned aerial vehicle; the genetic tabu search algorithm is an algorithm obtained by improving the genetic algorithm by using the tabu search algorithm; the individual includes the number of unmanned aerial vehicles and the corresponding reconnaissance order of each unmanned aerial vehicle; and the reconnaissance order is the order of reconnaissance of the fixed targets.

[0014] In a third aspect, the present application provides an electronic device comprising a memory and a processor, wherein the memory is configured to store a computer program, and the processor is configured to execute the computer program to enable the electronic device to perform the unmanned aerial vehicle number deployment and spectrum resource optimization method according to the first aspect.

[0015] In a fourth aspect, the present application provides a computer readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the unmanned aerial vehicle number deployment and spectrum resource optimization method according to the first aspect.

[0016] According to the embodiments of the present application, the following technical effects are achieved:

[0017] The present application studies the joint optimization problem of the number of unmanned aerial vehicles required for cooperative reconnaissance tasks, the allocation of fixed targets and the allocation of spectrum usage time under the condition of multi-unmanned aerial vehicle sharing spectrum resources. The main contributions can be summarized as follows:

[0018] (1) A joint planning method for the number of unmanned aerial vehicles required, the allocation of fixed targets and the allocation of spectrum usage time is proposed. In this method, multiple unmanned aerial vehicles cooperatively complete the reconnaissance of multiple fixed targets distributed on the ground and return the reconnaissance information to the decision center. The method can adjust the number of unmanned aerial vehicles participating in reconnaissance according to the cooperative reconnaissance task, and plan the reconnaissance order and spectrum usage time for each unmanned aerial vehicle, thereby eliminating the interference of information transmission between unmanned aerial vehicles.

[0019] (2) An improved genetic algorithm is proposed, which can generate fixed target reconnaissance schemes with different numbers of unmanned aerial vehicles. Compared with the fixed gene length of individuals in the traditional genetic algorithm, individuals with different gene lengths are generated. The improved genetic algorithm represents the number of unmanned aerial vehicles by gene length, and adjusts the number of unmanned aerial vehicles according to the battery capacity constraint, so that the improved genetic algorithm can realize step-by-step optimization in the feasible solution space.

[0020] (3) Using the distribution centralized search strategy, the global optimization ability of the algorithm is enhanced. Specifically, the distributed search strategy of genetic algorithm is used to ensure that the algorithm searches in a large range in the solution space, and the centralized search strategy of tabu search is used for fine search in a small range, which constantly breaks through the local optimal solution obtained by genetic algorithm, and further reduces the energy consumption of completing the cooperative reconnaissance task. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0022] Figure 1 The system model diagram when M unmanned aerial vehicles go to K distributed ground fixed targets to perform cooperative reconnaissance tasks;

[0023] Figure 2 The total energy consumption diagram of the unmanned aerial vehicles with different spectrum allocation ratios of the present application;

[0024] Figure 3 The spectrum time adjustment example diagram with minimum energy consumption of the present application;

[0025] Figure 4 The multi-unmanned aerial vehicle cooperative reconnaissance scheme example diagram of the present application;

[0026] Figure 5 The flowchart of the unmanned aerial vehicle number allocation and spectrum resource optimization method of the present application;

[0027] Figure 6 The flowchart of the genetic tabu search algorithm of the present application;

[0028] Figure 7 The two chromosome example diagram of 3 unmanned aerial vehicles reconnaissance 10 fixed targets of the present application;

[0029] Figure 8 The crossover operator example diagram of the present application;

[0030] Figure 9 The unmanned aerial vehicle number adjustment example diagram of the present application;

[0031] Figure 10 The flowchart of the tabu search algorithm of the present application;

[0032] Figure 11 is a diagram of the algorithm planning results of the present application at different fixed target amounts, wherein (a) is a diagram of the number of UAVs and task planning results when the fixed target amount is 10, (b) is a diagram of the number of UAVs and task planning results when the fixed target amount is 14, (c) is a diagram of the UAV spectrum usage time planning results when the fixed target amount is 10, and (d) is a diagram of the UAV spectrum usage time planning results when the fixed target amount is 14;

[0033] Figure 12 Figure 12 is a diagram of the results of different algorithms of the present application for comparison;

[0034] Figure 13 is a diagram of the energy consumption and number of UAVs required when the fixed target amount changes for different algorithms of the present application, wherein (a) is a diagram of the energy consumption and fixed target amount for different algorithms, and (b) is a diagram of the number of UAVs required and fixed target amount for different algorithms. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0036] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0037] The present application is applied to the scene of UAV performing reconnaissance tasks. The UAV flies to the airspace above each fixed target from the decision center to hover for information sensing, and returns the collected sensing data to the decision center for processing. The battery capacity of the UAV is generally small and cannot support the energy consumption of long-time work. When there are many fixed targets, one UAV may not be able to complete the reconnaissance tasks of all fixed targets, and multiple UAVs are needed to cooperate in reconnaissance. Assuming that there are M UAVs to go to K fixed targets distributed on the ground to perform cooperative reconnaissance tasks, the set of UAVs and fixed targets is represented as M={1,...,M} and K={1,...,K} respectively.

[0038] The horizontal coordinate of the fixed target k is represented as q k ∈R 2×1 , and is known. The distance d k,c between the fixed target k and the decision center c can be represented as:

[0039] d k,c =||q k -qc || (1);

[0040] where q c is the horizontal coordinate of the decision center. Assuming that the rotor unmanned aerial vehicle is used to perform the cooperative reconnaissance task, it can hover in the air to obtain stable communication transmission quality. The unmanned aerial vehicle is limited by the battery capacity and must return to the decision center for charging before the energy is consumed. The energy consumption of the unmanned aerial vehicle is divided into two parts: motion energy consumption and communication energy consumption. Generally speaking, the communication energy consumption is several orders of magnitude lower than the motion energy consumption and is often ignored. The motion energy consumption of the unmanned aerial vehicle is mainly considered. When the unmanned aerial vehicle m moves at a speed v m , its power is:

[0041]

[0042] where P0 and P i represent the blade profile power and induced power in the hovering state respectively; U tip and v0 represent the rotor tip speed and average rotor speed in the hovering state respectively; d0 and s represent the fuselage drag ratio and rotor solidity respectively, and p and A represent the air density and rotor disc area respectively. The acceleration and deceleration process of the unmanned aerial vehicle is ignored. When the unmanned aerial vehicle maneuvers between fixed targets, it adopts uniform flight, and hovers when performing reconnaissance tasks. The energy consumption of the unmanned aerial vehicle at different flight speeds is not the same. The power of the unmanned aerial vehicle decreases first and then increases with the increase of the speed, and there exists a maximum endurance speed V me and a maximum endurance distance speed V mr . Under the same energy consumption, V me is the flight speed with the longest endurance time of the unmanned aerial vehicle, and V mr is the flight speed with the farthest distance of the unmanned aerial vehicle. In order to save energy expenditure, it is assumed that the unmanned aerial vehicle uses the speed V mr when flying between fixed targets. Therefore, the minimum flight energy consumption E required for the unmanned aerial vehicle to fly from the current fixed target k to the next fixed target k' is:

[0043]

[0044]

[0045] where P(V mr ) is the power of the unmanned aerial vehicle flying at a speed V mr , d k,k' is the distance between the current fixed target k and the next fixed target k', and t represents the time for the unmanned aerial vehicle to fly from the current fixed target k to the next fixed target k'.

[0046] The unmanned aerial vehicle adopts photographing reconnaissance on a fixed target, or information sensing mode of transmitting information back while reconnaissance, and the hovering time is calculated according to the information transmission time of the unmanned aerial vehicle. Benefiting from the lift gain of the unmanned aerial vehicle, the channel power gain of air-to-ground is mainly determined by the line-of-sight link, and based on the free space path loss model, the channel gain h k is represented as:

[0047]

[0048] Wherein, β0 represents the channel power gain at a reference distance of 1 meter, represents the horizontal distance between the fixed target k and the decision center c when the path loss system is 2.

[0049] When the unmanned aerial vehicle maneuvers between fixed targets, no spectrum resource is occupied, and the spectrum resource guarantee is needed only when hovering over the fixed target to transmit sensing information, a plurality of unmanned aerial vehicles share the same spectrum resource, which may cause spectrum conflict, and therefore reasonable spectrum planning is crucial. When two unmanned aerial vehicles need to use the spectrum resource at the same time, Figure 2 The total energy consumption of the unmanned aerial vehicle under different spectrum resource allocation ratios is given. Figure 2 It can be seen that the exclusive spectrum allocation is the most energy-saving way, that is, the spectrum is allocated to the unmanned aerial vehicle m1 with the shortest occupation time, and the remaining unmanned aerial vehicles hover at the speed V me waiting. After the unmanned aerial vehicle m1 completes information transmission, the spectrum is allocated to the remaining unmanned aerial vehicle with the shortest occupation time, and so on, until all the unmanned aerial vehicles complete information transmission. However, this spectrum sharing method may also cause interruption of the information being transmitted, and is not suitable for tasks such as voice and video that require continuous transmission. Therefore, the present application designs an energy consumption minimum spectrum time allocation method of the unmanned aerial vehicle based on uninterrupted information transmission of the same fixed target. Figure 3 A situation that the spectrum use time of two unmanned aerial vehicles exists conflict and the energy consumption minimum adjustment method thereof are shown. Before adjustment, the spectrum use time of the two unmanned aerial vehicles exists overlap, and interference with each other will occur within t2 time. When the unmanned aerial vehicle 1 transmits first, the waiting time of the unmanned aerial vehicle 2 is t2, and vice versa, the waiting time of the unmanned aerial vehicle 1 is t1. Since t2 < t1, the unmanned aerial vehicle 1 occupies the spectrum first to transmit information, and the unmanned aerial vehicle 2 hovers at the speed V me The method that the unmanned aerial vehicle 2 uses the spectrum after the unmanned aerial vehicle 1 transmits information is the energy consumption minimum method. The adjustment method of the remaining situation of spectrum conflict is similar.

[0050] The bandwidth of the available spectrum resource is W, the information amount sensed by the fixed target k is C k When the hovering time of the unmanned aerial vehicle m is:

[0051]

[0052] where p m denotes the transmit power of UAV m, σ 2 is the ambient noise power spectral density. The hovering energy consumption of UAV m at the fixed target k is:

[0053]

[0054] where P(0) denotes the power when the speed of UAV is 0, i.e., the power consumption when hovering.

[0055] Assume that UAV m completes the reconnaissance task of all fixed targets in the target set with the reconnaissance order π m = (π m (0),..., π m (k), π m (k'),..., π m (K m + 1)), K m denotes the number of fixed targets that UAV m reconnaissance, π m (0) and π m (K m + 1) denote the number of the decision center of UAV, π m (k) denotes the number of the current fixed target k, and π m (k') denotes the number of the next fixed target k'. Since the spectrum resource is shared, there is spectrum waiting time when the spectrum usage time conflicts. Assume that the total spectrum waiting time of UAV m is The energy consumption of UAV m due to waiting for spectrum is:

[0056]

[0057] where P(V me ) denotes the power when the speed of UAV is V me .

[0058] Therefore, the total energy consumption of UAV m is:

[0059]

[0060] where, denotes the minimum flight energy consumption of UAV m when flying to the next fixed target k' after completing the reconnaissance task of the fixed target k with the reconnaissance order π m , and denotes the hovering energy consumption of UAV m at the current fixed target k with the reconnaissance order π m .

[0061] Since the battery capacity of UAV is limited, the total energy consumption of UAV m must be less than the battery capacity ​That is

[0062]

[0063] A fixed target can only be surveyed by one UAV, and the association between fixed target k and UAV m is represented as

[0064]

[0065] UAV m follows the survey order π m , and hovers over the k m th fixed target for t m k m .

[0066]

[0067] wherein t denotes the flight time of UAV m from the current fixed target k to the next fixed target k' according to the survey order π m , and t denotes the hovering time of UAV m over the current fixed target k according to the survey order π m . denotes the total time of UAV m from the decision center to the k m th fixed target according to the survey order π m . Since the spectrum resources of the UAVs are used in an exclusive manner, the spectrum usage times of different UAVs cannot overlap, otherwise interference will occur at the receiver of the decision center. The hovering times of the UAVs cannot overlap, that is,

[0068]

[0069] wherein t k1 denotes the hovering time of UAV m1 over the k1th fixed target according to the survey order π , and 1≤k1≤K1, K1 is the number of surveyed fixed targets of UAV m1; k2 denotes the hovering time of UAV m2 over the k2th fixed target according to the survey order π

[0070] , and 1≤k2≤K2, K2 is the number of surveyed fixed targets of UAV m2.

[0070] The optimization goal is to use the least number of UAVs under the given spectrum resource conditions, and to plan the survey order and spectrum usage time of each fixed target for each UAV. Multiple UAVs cooperate to complete the survey task, which is essentially a multiple traveling salesman problem (MTSP). All UAVs start from the decision center, and each fixed target is surveyed by only one UAV. The fixed target number is represented as πm (k) denotes, π m (0) and denotes the decision center, and the number of UAVs is u m denotes. Figure 4 A multi-UAV cooperative reconnaissance scheme is shown, in which UAV 1 returns to the decision center after reconnaissance of fixed targets according to reconnaissance order π1(K1), and the fixed target numbered π2(1) is reconnoitered by UAV 2. The total distance of the UAVs is

[0071]

[0072] wherein, and are the distances from the last fixed target reconnoitered according to reconnaissance order π1(K1) to the decision center and from the decision center to the fixed target numbered π2(1), respectively, and s1denotes other distances except and If all targets of UAV 2 are assigned to UAV 1 for reconnaissance, the total distance of the UAVs is

[0073]

[0074] wherein is the distance from the last fixed target reconnoitered according to reconnaissance order π1(K1) to the fixed target numbered π2(1). It can be known from the sum of two sides of a triangle being greater than the third side that

[0075]

[0076] that is, s≥s'. Therefore, the fewer the number of UAVs participating in reconnaissance, the shorter the total distance, and the less the energy consumption.

[0077] The limited battery capacity of a UAV determines that a single UAV cannot meet large-scale reconnaissance tasks, and multiple UAVs need to be cooperatively coordinated. Reasonable planning of the reconnaissance order and spectrum usage time of multiple UAVs can improve the efficiency of task completion, reduce the spectrum waiting time of the UAVs, and thus reduce the energy consumption of the UAVs. Therefore, the optimization objective can be essentially converted into optimizing the number of UAVs, the reconnaissance order of each UAV, and the spectrum usage time of each UAV under the limitation of the battery capacity of the UAVs, so as to minimize the total energy consumption of the multi-UAV system. The mathematical model of the optimization problem is expressed as follows:

[0078]

[0079] wherein, π={π1,...,π M} denotes a set of reconnaissance orders of different UAVs, denotes a set of spectrum usage times of different UAVs, K mrepresents the number of fixed targets assigned to the unmanned aerial vehicle m. C1 represents that the total energy consumption of the unmanned aerial vehicle m must be within the battery capacity range of the unmanned aerial vehicle m; C2 represents that each fixed target must be surveyed by one unmanned aerial vehicle, C3 represents that all fixed targets are allowed to be surveyed by one unmanned aerial vehicle; C4 represents that the spectrum usage time of the unmanned aerial vehicle cannot overlap, otherwise mutual interference will occur, affecting the information backhaul. This is a mixed integer non-convex problem, the optimization variables are coupled with each other, and the solution of the optimal scheme is challenging. π1 represents the survey order of the unmanned aerial vehicle 1, π M represents the survey order of the unmanned aerial vehicle m; represents the spectrum usage time of the unmanned aerial vehicle 1, represents the spectrum usage time of the unmanned aerial vehicle m.

[0080] Embodiment one

[0081] In view of this, the present application provides a method for optimizing the number of unmanned aerial vehicles and spectrum resources, as shown in the formula (1), which comprises the following steps: Figure 5

[0082] Step 100: obtaining fixed target information distributed on the ground area; the fixed target information includes the number of fixed targets and the location of each fixed target;

[0083] Step 200: according to the fixed target information and the target optimization function, using a genetic tabu search algorithm to determine the optimal individual; the target optimization function is a function of optimizing the number of unmanned aerial vehicles, the survey order of each unmanned aerial vehicle and the spectrum usage time under the limitation of the battery capacity of the unmanned aerial vehicle, so as to minimize the total energy consumption of the multi-unmanned aerial vehicle system; the genetic tabu search algorithm is an algorithm obtained by improving the genetic algorithm using the tabu search algorithm; the individual includes the number of unmanned aerial vehicles and the corresponding survey order of each unmanned aerial vehicle; the survey order is the order of surveying the fixed target.

[0084] Wherein, the target function and the corresponding constraint condition are shown in the formula (17).

[0085] Further, step 200 specifically comprises:

[0086] Step 201: determining the initial population according to the fixed target information; the initial population includes a plurality of individuals.

[0087] Step 202: determining whether the current iteration number reaches the maximum iteration number; if yes, the individual of the current iteration number corresponding to the minimum target optimization function value is determined as the optimal individual; if not, the individual of the current iteration number corresponding to the target optimization function value less than the set threshold is reserved.

[0088] ​Step 203: first, cross-variation operation is performed on the remaining individuals to obtain updated individuals, then the number of unmanned aerial vehicles and the spectrum use time in the updated individuals are adjusted to obtain secondarily updated individuals, then the target optimization function value of each secondarily updated individual is calculated based on the target optimization function, and the individual corresponding to the next iteration is generated according to the target optimization function value, finally, the individual corresponding to the next iteration is taken as the initial solution of the tabu search algorithm for calculation to obtain a better individual, and the individual corresponding to the next iteration is updated according to the better individual.

[0089] Step 204: the current iteration number is updated to the next iteration number, the individual of the current iteration number is updated to the updated individual corresponding to the next iteration, and the target optimization function value of the updated individual of the current iteration number is calculated, and the step 202 is returned.

[0090] The genetic algorithm (Genetic algorithm, GA) has strong global exploration ability and can search widely in most areas in the solution space. The tabu search (Tabu Search, TS) has strong local search ability and can form a complementary relationship with the genetic algorithm. The improved genetic tabu search hybrid algorithm is designed, first, the cross and mutation operations of the genetic algorithm are used to generate individuals with large differences in the solution space, and the population diversity is ensured, then the local search ability of the tabu search algorithm is used to make the individual constantly break through the local optimal solution. The distributed search of the genetic algorithm and the centralized search of the tabu algorithm interact with each other, and finally converge to the approximate optimal solution, and the flow of the algorithm is as shown in Figure 6

[0091] After the population is crossed and mutated, a new unmanned aerial vehicle reconnaissance scheme is generated, but because of the randomness of cross-variation, individuals exceeding the solution space may be generated, and the two modules of unmanned aerial vehicle number adjustment and spectrum use time adjustment are used to ensure that the fixed target amount of unmanned aerial vehicle distribution does not exceed the battery capacity constraint, and the spectrum use time is separated from each other, and no interference is generated. In the later stage of evolution, the tabu search algorithm is used to increase the probability of obtaining a global optimal solution by the algorithm.

[0092] Two chromosomes are used to represent the reconnaissance scheme of the unmanned aerial vehicle, one of which is a chromosome with a length of K, representing the reconnaissance order of each fixed target, and the other is a chromosome with a length of M, representing the reconnaissance fixed target number of each unmanned aerial vehicle. Figure 7 Two technical examples of two chromosomes for 3 unmanned aerial vehicles to reconnoiter 10 fixed targets are shown, in which unmanned aerial vehicle 1 reconnoiters 3 targets in the order of 4, 2 and 5, and the remaining unmanned aerial vehicles are assigned similar tasks.

[0093] ​Crossover is a method of combining two parent information to produce offspring, which can produce better offspring individuals. But due to the uniqueness of target number and the directionality of reconnaissance order, the traditional crossover method is easy to produce the omission and repetition of reconnaissance targets. In order to solve this problem, the crossover operator is adjusted in this paper, which not only inherits the characteristics of different parents, but also ensures the correctness of the offspring. The specific crossover method is shown in Figure 8 As shown in the figure, two parent chromosomes produce offspring c1 after the traditional crossover operator, and the reconnaissance targets 1 and 2 are repeated, and 3 and 7 are omitted. Remove the repeated targets, and fill in the missing targets in the original order to produce the correct offspring.

[0094] Mutation is to prevent better individuals from occupying the whole population and converging to local optimal solution too early, which increases the global search ability of the algorithm. Since the application adopts two-chromosome technology, the mutation of the number of unmanned aerial vehicles, the number of reconnaissance of each unmanned aerial vehicle and the reconnaissance target order needs to be considered respectively.

[0095] In order to explore the reconnaissance scheme with the least number of unmanned aerial vehicles as much as possible, in the mutation of the number of unmanned aerial vehicles, the two unmanned aerial vehicles with the least fixed target amount are combined to generate a scheme that reduces the task of one unmanned aerial vehicle. In the mutation of the reconnaissance number, two unmanned aerial vehicles are randomly selected, and one of them is increased by one, while the other is reduced by one. In the mutation of the reconnaissance target order, two reconnaissance targets are randomly selected, and their positions are exchanged. After mutation, a better new scheme can be generated, which can have a greater probability of being retained to the next generation in the selection operation of the genetic algorithm.

[0096] In order to ensure that the individual meets the requirements that the battery capacity of the unmanned aerial vehicle and the spectrum use time do not conflict, the new scheme generated by crossover and mutation needs to be processed for feasibility, including two parts of unmanned aerial vehicle number adjustment due to battery capacity limitation and spectrum use time adjustment due to spectrum conflict. The spectrum use time adjustment has been described in detail in Figure 2 , and the following describes the unmanned aerial vehicle number adjustment scheme.

[0097] The application increases the number of unmanned aerial vehicles to process the feasibility of the new scheme, so that the fixed target amount allocated to all unmanned aerial vehicles is within the range of its battery capacity. The specific operation is shown in Figure 9 , assuming that the energy consumption of unmanned aerial vehicle 2 exceeds its battery capacity, 1 unmanned aerial vehicle is added to share part of the task until the energy consumption of all unmanned aerial vehicles is within the range of its battery capacity.

[0098] With the increase of the evolution generation of genetic algorithm, the difference between different individuals in the population gradually decreases, which leads to the slow evolution of genetic algorithm. The present application introduces a tabu search algorithm. When the evolution generation of genetic algorithm reaches a certain proportion of the maximum generation, the partial individuals of the population are used as initial solutions, and the tabu search algorithm is used for local search to seek better solution scheme. The specific process of the tabu search algorithm is shown in Figure 10 The present application adopts the 2-opt neighborhood search operator, takes the individual as the initial solution, randomly selects two fixed targets in the fixed target chromosome and exchanges their order to search for a better reconnaissance order.

[0099] In order to verify the efficiency and effectiveness of the improved genetic tabu hybrid algorithm, the performance of the algorithm is compared with that of genetic algorithm and tabu search algorithm. The simulation also shows the algorithm performance under different task parameters. In order to make the simulation parameter setting more reasonable, the parameter settings of the unmanned aerial vehicle and the task are shown in Table 1.

[0100] Table 1 Simulation parameter setting

[0101]

[0102]

[0103] Figure 11 shows the results of unmanned aerial vehicle task planning and spectrum usage time division when the number of fixed targets is 10 and 14. As can be seen from Figure 11, the algorithm will dispatch different number of unmanned aerial vehicles according to the change of the target size, and plan reasonable reconnaissance order and spectrum usage time for each unmanned aerial vehicle to eliminate communication interference. The task planning of the unmanned aerial vehicle is not only the minimum flight distance as the target, but also the energy consumption of the battery capacity of the unmanned aerial vehicle, the target distance and the spectrum waiting time generated by the reconnaissance order, and the fixed target amount and target reconnaissance order of each unmanned aerial vehicle are reasonably arranged, and the corresponding spectrum usage time of each unmanned aerial vehicle is planned. When there are 10 reconnaissance targets, 3 unmanned aerial vehicles are needed to complete the task cooperatively, and the total energy consumption is 237.93kJ. With the increase of the number of fixed targets to 14, the energy consumption is 288.01kJ, which exceeds the energy carried by 3 unmanned aerial vehicles, so the number of unmanned aerial vehicles is increased to 4. In the "spectrum time adjustment" step of the algorithm, the spectrum conflict time of each unmanned aerial vehicle is avoided according to the criterion of minimum energy consumption, which minimizes the energy consumption of the unmanned aerial vehicle caused by spectrum waiting while eliminating interference.

[0104] In order to verify the effectiveness of the algorithm, it is compared with genetic algorithm and tabu search algorithm. Figure 12The results of the three algorithms in two scenarios of fixed target quantity of 10 and 14 targets are shown. The tabu search algorithm is prone to local optimal solution due to its poor global search ability, and the energy consumption of the obtained scheme is significantly higher than that of the genetic algorithm and the proposed algorithm after the same evolution generation. The genetic algorithm produces new population in the form of roulette, and the better individual is retained to the next generation with a larger probability. This way can expand the global search range of the algorithm, but also slows down the convergence speed of the algorithm. The proposed algorithm combines the global search of genetic algorithm and the local search of tabu search algorithm, which can overcome the defect of genetic algorithm. In the first 250 generations, the genetic algorithm is used to search for better feasible solutions in a large range, and then the genetic algorithm and the tabu search algorithm are combined to enhance the optimization ability of the algorithm. As can be seen from the figure, after 250 generations, the proposed algorithm is significantly better than the genetic algorithm, which is due to the advantage of the local search ability of the proposed algorithm, which can improve the defect of slow convergence speed of genetic algorithm and further reduce the energy consumption demand of the unmanned aerial vehicle. When the same fixed target quantity is completed, the proposed algorithm realizes the lowest energy consumption, which shows the effectiveness of the algorithm.

[0105] Figure 13 shows the energy consumption and unmanned aerial vehicle quantity demand results of different algorithms when the fixed target quantity changes. When the fixed target quantity is small, the solution space size is small, and the three algorithms can all get the optimal solution. As the fixed target quantity increases, the solution space size increases exponentially, and the gap between the algorithms also increases. The tabu search algorithm is prone to local optimal solution due to its poor global search ability, so it requires the most number of unmanned aerial vehicles and the largest energy consumption. Compared with the genetic algorithm and the tabu search algorithm, the proposed algorithm realizes the least number of unmanned aerial vehicles and energy consumption demand. Even if it has the same number of unmanned aerial vehicle demand as the genetic algorithm, it can further optimize the reconnaissance order of the unmanned aerial vehicle through the local search of the tabu search, and reduce the energy consumption. At the same time, the larger the fixed target quantity, the more the number of unmanned aerial vehicles required, and the more complex the solution space. The proposed algorithm effectively combines global search and local search, so the advantage is more obvious.

[0106] The unmanned aerial vehicle and spectrum resource joint planning problem of reconnaissance task is studied to solve the least number of unmanned aerial vehicles, target reconnaissance sequence and spectrum use time. The genetic algorithm is improved to realize step-by-step optimization of the reconnaissance scheme of different numbers of unmanned aerial vehicles. The coupling relationship between the optimized variables and the energy consumption is used to convert the problem into energy consumption optimization of the unmanned aerial vehicle, and a genetic tabu hybrid algorithm is designed. First, the global optimization ability of the genetic algorithm is used to generate widely distributed candidate solutions, and then the centralized search characteristics of the tabu search algorithm are used for fine local search to obtain the optimal solution. The proposed algorithm can automatically adjust the number of reconnaissance unmanned aerial vehicles according to the change of the fixed target quantity, and plan the target reconnaissance sequence and spectrum use time for each unmanned aerial vehicle, combining the advantages of genetic algorithm and tabu search algorithm, and improving the defect that the convergence speed of genetic algorithm will slow down. The simulation results prove the effectiveness of the proposed algorithm, which can reduce the number of unmanned aerial vehicles required and energy consumption. In addition, the more the fixed target quantity, the more obvious the advantage of the proposed algorithm, which can support the use of unmanned aerial vehicle clusters.

[0107] Embodiment two

[0108] The unmanned aerial vehicle number allocation and spectrum resource optimization device provided by the embodiment of the application comprises:

[0109] The data acquisition module acquires fixed target information distributed on the ground area under study; the fixed target information comprises the number of fixed targets and the position of each fixed target.

[0110] The optimal individual determination module is used for determining an optimal individual by using a genetic tabu search algorithm according to the fixed target information and a target optimization function; the target optimization function is a function for optimizing the number of unmanned aerial vehicles, the reconnaissance sequence of each unmanned aerial vehicle and the spectrum use time under the limitation of the battery capacity of the unmanned aerial vehicle, so that the total energy consumption of the multi-unmanned aerial vehicle system is minimized; the genetic tabu search algorithm is an algorithm obtained by improving the genetic algorithm by using the tabu search algorithm; the individual comprises the number of unmanned aerial vehicles and the corresponding reconnaissance sequence of each unmanned aerial vehicle; and the reconnaissance sequence is the sequence of reconnaissance of the fixed target.

[0111] Embodiment three

[0112] The electronic device comprises a memory for storing a computer program and a processor for running the computer program to enable the electronic device to execute the unmanned aerial vehicle number allocation and spectrum resource optimization method of embodiment one.

[0113] Optionally, the electronic device can be a server.

[0114] In addition, embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the UAV number allocation and spectrum resource optimization method of Embodiment 1.

[0115] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0116] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for allocating the number of unmanned aerial vehicles (UAVs) and optimizing spectrum resources, characterized in that, The method comprises the following steps: acquiring fixed target information distributed on a research ground area; the fixed target information comprises the number of fixed targets and the position of each fixed target; determining an optimal individual according to the fixed target information and a target optimization function by using a genetic tabu search algorithm; the target optimization function is a function of optimizing the number of unmanned aerial vehicles, the reconnaissance sequence of each unmanned aerial vehicle and the spectrum use time of each unmanned aerial vehicle under the limitation of the battery capacity of the unmanned aerial vehicle, so as to minimize the total energy consumption of the multi-unmanned aerial vehicle system; the genetic tabu search algorithm is an algorithm obtained by improving a genetic algorithm by using a tabu search algorithm; the individual comprises the number of unmanned aerial vehicles and the corresponding reconnaissance sequence of each unmanned aerial vehicle; the reconnaissance sequence is the sequence of reconnaissance of fixed targets; The total energy consumption of the drone m is: wherein π m = (π m (0),..., π m (k), π m (k'),..., π m (K m +1)) represents the reconnaissance order of the UAV m, π m (0) and π m (K m +1) represent the UAV decision center, π m (k) represents the number of the current fixed target k, π m (k') represents the label of the next fixed target k', and K m represents the number of fixed targets that the UAV m reconnaissance. represents the minimum flight energy consumption of the UAV m when flying from the current fixed target k to the next fixed target k' in the reconnaissance order π m represents the minimum flight energy consumption of the UAV m when flying from the current fixed target k to the next fixed target k' in the reconnaissance order π represents the minimum flight energy consumption of the UAV m when flying from the current fixed target k to the next fixed target k' in the reconnaissance order π m represents the hovering energy consumption of the UAV m over the current fixed target k represents the energy consumption of the UAV m when waiting for the spectrum 2.The method of claim 1, wherein, the target optimization function and the corresponding constraint condition are as follows: wherein M represents the number of UAVs, π = {π1,..., πM} represents the set of reconnaissance orders of different UAVs, M} represents the set of reconnaissance orders of different UAVs, represents the set of spectrum usage time of different UAVs, m represents the mth UAV, E m represents the total energy consumption of the UAV m; π1represents the reconnaissance order of the UAV 1, π M represents the reconnaissance order of the UAV M; represents the spectrum usage time of the UAV 1, represents the spectrum usage time of the UAV M; C1 represents a constraint that the total energy consumption of the drone m must be within the battery capacity of the drone m ; C2 represents the constraint that each fixed target must be surveyed by one UAV; ω m,k denotes that the fixed target k is assigned to the UAV m; C3 represents a constraint condition of allowing all fixed targets to be reconnoitered by one unmanned aerial vehicle; K represents a constraint condition of the number of fixed targets; C4 represents a constraint condition that the spectrum use time of the UAV cannot overlap; represents the reconnaissance order of the UAV m1 hovering time over the k1th fixed target, 1≤k1≤K1, K1 being the number of reconnaissance fixed targets of the UAV m1; represents the reconnaissance order of the UAV m2 hovering time over the k2th fixed target, 1≤k2≤K2, K2 being the number of reconnaissance fixed targets of the UAV m2. 3.The method of claim 1, wherein, The energy consumption generated when the UAV m waits for the spectrum is: wherein P(V me ) represents the power of the UAV m at the speed V me , represents the total spectrum latency of the UAV m, V me represents the flight speed at which the UAV has the longest endurance time. 4.The method of claim 1, wherein, The minimum flight energy consumption of the unmanned aerial vehicle m is: P(V mr ) represents the power of the unmanned aerial vehicle m at the speed V mr , P(V m ) represents the flight time of the unmanned aerial vehicle m from the current fixed target k to the next fixed target k' in the reconnaissance order π mr V mr max represents the maximum endurance speed.

5. The method of claim 1, wherein, The energy consumption of the UAV m hovering is: wherein P(0) represents the power of the UAV m when hovering, represents the hovering time of the UAV m over the current fixed target k. m the hovering time over the current fixed target k.

6. The method of claim 1, wherein, the step of determining the optimal individual according to the fixed target information and the target optimization function by using the genetic tabu search algorithm comprises the following steps: determining an initial population according to the fixed target information; the initial population comprises a plurality of individuals; judging whether the current iteration number reaches a maximum iteration number; if yes, determining the individual of the current iteration number corresponding to the minimum target optimization function value as the optimal individual; if no, reserving the individual of the current iteration number corresponding to the target optimization function value less than a set threshold value; performing a crossover and mutation operation on the reserved individual to obtain an updated individual; adjusting the number of unmanned aerial vehicles and the spectrum use time in the updated individual to obtain a second updated individual; calculating the target optimization function value of each second updated individual based on the target optimization function, and generating the individual corresponding to the next iteration according to the target optimization function value; taking the individual corresponding to the next iteration as an initial solution of the tabu search algorithm to calculate a better individual, and updating the individual corresponding to the next iteration according to the better individual; updating the current iteration number to the next iteration number, updating the individual of the current iteration number to the updated individual corresponding to the next iteration, and calculating the target optimization function value of the updated individual of the current iteration number, and returning to the step of judging whether the current iteration number reaches the maximum iteration number.

7. An apparatus for unmanned aerial vehicle number assignment and spectrum resource optimization, characterized in that, The method comprises the following steps: a data acquisition module acquires fixed target information distributed on a research ground area; the fixed target information comprises the number of fixed targets and the position of each fixed target; an optimal individual determination module is configured to determine an optimal individual according to the fixed target information and a target optimization function by using a genetic tabu search algorithm; the target optimization function is a function of optimizing the number of unmanned aerial vehicles, the reconnaissance sequence of each unmanned aerial vehicle and the spectrum use time of each unmanned aerial vehicle under the limitation of the battery capacity of the unmanned aerial vehicle, so as to minimize the total energy consumption of the multi-unmanned aerial vehicle system; the genetic tabu search algorithm is an algorithm obtained by improving a genetic algorithm by using a tabu search algorithm; the individual comprises the number of unmanned aerial vehicles and the corresponding reconnaissance sequence of each unmanned aerial vehicle; the reconnaissance sequence is the sequence of reconnaissance of fixed targets; The total energy consumption of the drone m is: wherein π m = (π m (0),..., π m (k), π m (k'),..., π m (K m +1)) represents the reconnaissance order of the UAV m, π m (0) and π m (K m +1) represent the UAV decision center, π m (k) represents the number of the current fixed target k, π m (k') represents the label of the next fixed target k', and K m represents the number of fixed targets that the UAV m has reconnoitered. represents the minimum flight energy consumption of the UAV m when flying from the current fixed target k to the next fixed target k' in the reconnaissance order π m represents the minimum flight energy consumption of the UAV m when flying from the current fixed target k to the next fixed target k' in the reconnaissance order π represents the minimum flight energy consumption of the UAV m when flying from the current fixed target k to the next fixed target k' in the reconnaissance order π m represents the hovering energy consumption of the UAV m in the airspace above the current fixed target k represents the energy consumption of the UAV m when waiting for the spectrum 8. An electronic device, comprising: An electronic device comprising a memory for storing a computer program and a processor for running the computer program to cause the electronic device to perform the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, A computer program product comprising a computer program stored on a computer readable medium, the computer program being executable by a processor to implement the method of any one of claims 1-6.

Images