CN116258357A - Heterogeneous unmanned aerial vehicle cooperative target distribution method based on polygene genetic algorithm - Google Patents
Heterogeneous unmanned aerial vehicle cooperative target distribution method based on polygene genetic algorithm Download PDFInfo
- Publication number
- CN116258357A CN116258357A CN202310549221.3A CN202310549221A CN116258357A CN 116258357 A CN116258357 A CN 116258357A CN 202310549221 A CN202310549221 A CN 202310549221A CN 116258357 A CN116258357 A CN 116258357A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- population
- target
- task
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 60
- 230000002068 genetic effect Effects 0.000 title claims abstract description 60
- 230000035772 mutation Effects 0.000 claims abstract description 51
- 210000000349 chromosome Anatomy 0.000 claims abstract description 41
- 230000003044 adaptive effect Effects 0.000 claims abstract description 36
- 230000003234 polygenic effect Effects 0.000 claims abstract description 18
- 238000002922 simulated annealing Methods 0.000 claims abstract description 18
- 230000008901 benefit Effects 0.000 claims abstract description 8
- 230000015572 biosynthetic process Effects 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 16
- 230000007246 mechanism Effects 0.000 claims description 13
- 238000000137 annealing Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 7
- 230000002079 cooperative effect Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 230000000875 corresponding effect Effects 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 5
- 108090000623 proteins and genes Proteins 0.000 claims description 5
- 230000006978 adaptation Effects 0.000 claims description 4
- 230000004083 survival effect Effects 0.000 claims description 4
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000005755 formation reaction Methods 0.000 description 22
- 238000010586 diagram Methods 0.000 description 6
- 230000008878 coupling Effects 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 4
- 238000005859 coupling reaction Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000000977 initiatory effect Effects 0.000 description 3
- 230000001276 controlling effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a heterogeneous unmanned aerial vehicle cooperative target distribution method based on a polygene genetic algorithm, which comprises the following steps: s1, determining the length of chromosomes according to the number of unmanned aerial vehicles and the number of targets, and encoding by adopting a polygenic genetic encoding method; s2, judging whether the chromosome meets the cooperative constraint of multiple unmanned aerial vehicles; if yes, adding an initial population; s3, constructing an fitness function, calculating fitness values of individuals in the initial population, and checking whether iteration times reach preset times or not; if the distribution scheme is reached, outputting the distribution scheme with the highest fitness value; otherwise, entering S4; s4, performing adaptive elite crossover operator and adaptive mutation operator operations on the common population, and performing adaptive simulated annealing operator operations on the elite population; s5, calculating individual fitness values after genetic operation, and checking whether iteration times reach preset times or not; and outputting the allocation scheme with the highest fitness value if the fitness value is reached. The invention has the advantages of improving the target distribution efficiency and the like.
Description
Technical Field
The invention mainly relates to the technical field of unmanned aerial vehicle target distribution, in particular to a heterogeneous unmanned aerial vehicle cooperative target distribution method based on a polygene genetic algorithm.
Background
The current informatization war gradually develops towards the directions of complexity, diversification and miniaturization, and the unmanned aerial vehicle plays an important role in performing military tasks on ground targets by virtue of the advantages of good concealment, strong maneuverability and low cost. Under a complex battlefield environment, certain comprehensive combat tasks such as multiple unmanned aerial vehicles observe and play cooperative tasks, and the cooperative cooperation of the functional heterogeneous intelligent agent formations is needed to be executed, so that compared with unmanned aerial vehicles with single functions, the heterogeneous unmanned aerial vehicle formations execute tasks through the cooperative cooperation, and the system has higher execution efficiency, reliability and flexibility.
In the heterogeneous unmanned aerial vehicle target allocation, the influence of various complex coupling constraints is required to be considered, in order to reduce the solving dimension of the problem, the conventional target allocation method only considers the influence of one or more constraints such as task time window constraint, weapon isomerism, target isomerism and the like, the influence of various coupling factors on allocation results is split, and the generated target allocation scheme has no reference value.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems existing in the prior art, the invention provides a heterogeneous unmanned aerial vehicle cooperative target distribution method with high target distribution efficiency based on a polygene genetic algorithm.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a heterogeneous unmanned aerial vehicle cooperative target distribution method based on a polygene genetic algorithm comprises the following steps:
s1, determining the length of a chromosome according to the number of unmanned aerial vehicles and the target number, and encoding the chromosome by adopting a polygene genetic encoding method of a fusion vacant mechanism;
s2, judging whether the chromosome generated by encoding meets the cooperative constraint of multiple unmanned aerial vehicles; if the multi-unmanned aerial vehicle cooperative constraint is met, adding an initial population until the initial population reaches a preset number in scale, and dividing the initial population into a common population and an elite population;
s3, constructing an fitness function, calculating the fitness value of each individual in the initial population, and checking whether the current iteration number reaches the preset number or not; if the preset times are reached, outputting an unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value; if the preset times are not reached, the next step is carried out;
s4, carrying out genetic operation on the initial population by a polygenic genetic algorithm; the genetic operation comprises an adaptive elite crossover operator operation and an adaptive mutation operator operation which are sequentially implemented on a common population, and an adaptive simulated annealing operator operation implemented on the elite population;
s5, calculating the fitness value of each individual in the initial population after genetic operation, and checking whether the current iteration number reaches the preset number or not; if the preset times are reached, outputting an unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value; if the preset times are not reached, the steps S4-S5 are circulated until the unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value is output.
Preferably, in step S1, the polygenic genetic coding method is implemented based on three layers of one module, wherein the three layers of one module include a target layer, a task execution sequence layer, an agent layer, and a task module; the target layer adopts a real number coding strategy and is composed of target sequence numbers, and the target layer is expanded according to the number of targets in an actual task; the unmanned plane layer fuses with an empty mechanism, and performs random-1 operation on part of genes according to a certain probability on the unmanned plane layer, which means that targets are not allocated to any unmanned plane, and other positions are formed by actual numbers of the unmanned plane, have a corresponding relation with the target layer and are used for matching screening of unmanned plane types and target types; the execution sequence layer carries out real number incremental coding according to the length of the chromosome and is used for defining the task execution sequence of a single unmanned aerial vehicle and analyzing the feasibility of a collaborative strategy among a plurality of unmanned aerial vehicles; the task module contains genetic information related to task types and is used for expanding according to the number of tasks in the problem model.
Preferably, in step S2, the unmanned aerial vehicle speed is set to a real number interval with a high-low threshold, i.e. in a multi-unmanned cooperative constraintCalculating the time intersection of unmanned aerial vehicle formation reaching a target point, taking the shortest time for the unmanned aerial vehicle formation to initiate cooperative action on the same target, wherein the mathematical model is shown as follows:
wherein ,representing that a plurality of unmanned aerial vehicles execute tasks +.>Time set of->Representation unmanned plane g 1 Execution task->Is (are) time of day->Representation unmanned plane g 2 Execution task->Is (are) time of day->Representation unmanned plane g 3 Execution task->Is (are) time of day->Representing +.>Execution task->Is (are) time of day->Representing unmanned plane->Execution task->Is (are) time of day->Representing cooperative pair of targets->Implement task->Temporary formation of->Time interval representing the simultaneous arrival of co-formation,/->Representation unmanned aerial vehicleReaching the target from the current position->Distance of->Representing unmanned plane->Speed interval of>Respectively representing the earliest and latest times of the unmanned aerial vehicle k to execute the task j.
Preferably, in step S3, when the fitness function is constructed, the objective function is set in a generalized manner, and four optimization indexes including task execution benefit, agent formation survival probability, task execution time consumption and distance consumption are included in the objective function.
Preferably, in step S4, the specific procedure of the adaptive elite crossover operator operation is:
the common population is subjected to self-adaptive crossover probabilitySelecting individuals crossing the elite population or performing self-crossing; wherein->;/>Representing the current iteration number of the population, +.>Representing the total iteration number of the population; crossing the selected two parent chromosomes; repeating this step for all individuals in the common population until a specified number of offspring populations are produced;
elite population individuals are not affected by crossover operations and replicate into offspring populations.
Preferably, the principle of crossing the two selected parent chromosomes comprises:
1) The third row does not participate in the interleaving operation;
2) The points of the parent chromosomes which are crossed need to be in one-to-one correspondence;
3) The number of crossing points remains random.
Preferably, in step S4, the specific procedure of the adaptive mutation operator operation is as follows:
s41, selecting a parent chromosome to generate a random number between 0 and 1, if the random number is smaller than the adaptive mutation probabilityPerforming a mutation operation on the individual; wherein the method comprises the steps of
wherein Representing the current iteration number of the population, +.>Representing the total iteration number of the population;
s42, randomly selecting a mutation operation for the individuals to be mutated.
Preferably, the adaptive mutation operator operation comprises a multi-point random mutation operation and a multi-point shift mutation operation, wherein the multi-point random mutation operation follows the following principle:
a) The number of variation points is random, namely the variation points of different individuals in the same population are different;
b) Only the unmanned plane layer participates in variation;
c) The type of the unmanned aerial vehicle generated by variation is required to be matched with the task type;
wherein the multi-point shift mutation operation follows the following two principles in addition to the principles a) and c) in the multi-point random mutation operation:
d) The point positions participating in shift mutation appear in pairs, namely the point positions of the selected mutation need to be paired pairwise before transformation;
e) The target layer and the unmanned plane layer both participate in shifting.
Preferably, in step S4, the specific process of adaptively simulating the annealing operator operation is:
s4.1, selecting 1/5 individuals from elite population to participate in simulated annealing operation by using a roulette method;
s4.2, randomly selecting from selected chromosomesPerforming multipoint mutation on the individual points, and applying local disturbance to elite individuals; wherein the method comprises the steps of
s4.3, determining whether to accept a new allocation scheme by using a designed Metropolis criterion; if the adaptation value of the new allocation scheme is higher, accepting; if the fitness value is not high, 1 random number between 0 and 1 is generated, if the random number is smaller thanA new allocation scheme is accepted, otherwise the allocation scheme is not accepted.
Preferably, the metapolis criterion is:
wherein A change value indicating individual fitness before and after annealing; wherein->Indicating an individual fitness value after annealing; />Indicating individual fitness values prior to annealing.
Compared with the prior art, the invention has the advantages that:
aiming at the problem that the traditional single-gene genetic algorithm is poor in solving the complex coupling problem model, the invention designs a negative real number coding strategy by adopting a multi-gene genetic coding method integrating an empty mechanism, and has lower complexity and higher coding efficiency compared with binary coding and Gray codes.
Aiming at the problem that the multi-unmanned aerial vehicle cooperative mode is too single in the prior method, the multi-unmanned aerial vehicle cooperative action method integrating the speed change adjustment mechanism is adopted, so that different unmanned aerial vehicles are allowed to freely form a queue by adjusting the speed of the unmanned aerial vehicles, and the cooperative mode of the multi-unmanned aerial vehicle formation is enriched.
Aiming at the problem that the searching direction of the traditional heuristic method is too dispersed, the invention adopts the operation of the adaptive elite crossing operator, constructs the adaptive function related to the iteration progress of the population, and ensures the population to always evolve to a better direction by controlling the scale of elite individuals participating in crossing; meanwhile, aiming at the problem of low single-point mutation operation efficiency of a genetic algorithm, the invention combines the characteristics of multi-gene coding, provides self-adaptive mutation operator operation, enriches the diversity of offspring population and ensures the optimization efficiency of the algorithm.
Aiming at the problem that the traditional heuristic algorithm is easy to sink into local optimum, the self-adaptive simulated annealing operator operation is adopted, and the Metropolis criterion conforming to the population iteration rule is designed, so that the probability that the target allocation method is sunk into the local optimum is effectively reduced, and the solving quality of the method is improved.
According to the heterogeneous unmanned aerial vehicle formation target distribution method based on the improved polygene genetic algorithm, the target is subjected to the investigation and playing cooperative military operation according to the number of the current unmanned aerial vehicles, the number of the targets and the related situation, and a task list of each unmanned aerial vehicle is automatically generated.
Drawings
FIG. 1 is a block diagram of a target allocation method according to an embodiment of the present invention.
FIG. 2 is a schematic diagram of a three-layer one-module multi-gene coding method of the invention which fuses empty mechanisms.
FIG. 3 is a schematic diagram of an adaptive elite crossing method according to an embodiment of the present invention; wherein a) is a parent chromosome; b) Is a offspring chromosome.
FIG. 4 is a schematic diagram of an adaptive mutation method according to an embodiment of the present invention; wherein a) is a multipoint random variation operation; b) Is a multi-point shift mutation operation.
Fig. 5 is a diagram showing simulation results of a case of the present invention in a specific application.
Fig. 6 is a diagram showing simulation results of the second embodiment of the present invention in a specific application.
FIG. 7 is a flow chart of an embodiment of the target allocation method of the present invention.
Detailed Description
The invention is further described below with reference to the drawings and specific examples.
As shown in fig. 1 and fig. 7, the heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm in the embodiment of the invention includes the following steps:
s1, determining the length of a chromosome according to the number of unmanned aerial vehicles and the target number, and encoding the chromosome by adopting a polygene genetic encoding method of a fusion vacant mechanism;
s2, judging whether the chromosome generated by encoding meets the cooperative constraint of multiple unmanned aerial vehicles; if the multi-unmanned aerial vehicle cooperative constraint is met, adding an initial population until the initial population reaches a preset number in scale, and dividing the initial population into a common population and an elite population;
s3, constructing an fitness function, calculating the fitness value of each individual in the initial population, and checking whether the current iteration number reaches the preset number or not; if the preset times are reached, outputting an unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value; if the preset times are not reached, the next step is carried out;
s4, carrying out genetic operation on the initial population by a polygenic genetic algorithm; the genetic operation comprises an adaptive elite crossover operator operation and an adaptive mutation operator operation which are sequentially implemented on a common population, and an adaptive simulated annealing operator operation implemented on the elite population;
s5, calculating the fitness value of each individual in the initial population after genetic operation, and checking whether the current iteration number reaches the preset number or not; if the preset times are reached, outputting an unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value; if the preset times are not reached, the steps S4-S5 are circulated until the unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value is output.
In a specific embodiment, in step S1, the polygene genetic encoding method is implemented based on three layers of one module, wherein the three layers of one module include a target layer, a task execution sequence layer, an agent layer and a task module, and the model information and the specific functions that are implemented are correspondingly included as follows:
target layer: the layer adopts a real number coding strategy and is composed of target sequence numbers, and the expansion is carried out according to the target number in an actual task;
unmanned aerial vehicle layer: an empty mechanism is designed on the layer, and a negative real number coding method is provided, namely, a part of genes are subjected to random-1 operation according to a certain probability on the layer, which means that targets are not allocated to any unmanned aerial vehicle; the rest positions are formed by actual numbers of the unmanned aerial vehicle, have corresponding relation with a target layer and are mainly used for matching screening of the unmanned aerial vehicle type and the target type;
execution sequence layer: the real number incremental coding is carried out on the layer according to the chromosome length, and the real number incremental coding is mainly used for defining the task execution sequence of a single unmanned aerial vehicle and analyzing the feasibility of a collaborative strategy among a plurality of unmanned aerial vehicles.
And a task module: the module mainly contains gene information related to task types, and can be expanded according to the number of tasks in the problem model. The task modules are set strictly according to the task priority order. For example, task A > task B in the timing requirement, then the encoding module of task A is pre-encoded.
Aiming at the problem that the traditional single-gene genetic algorithm is poor in solving the complex coupling problem model, the invention designs a negative real number coding strategy by adopting the multi-gene genetic coding method fused with the vacant mechanism, and compared with binary coding and Gray code, the method has lower complexity and higher coding efficiency.
In a specific embodiment, in step S2, the multi-unmanned cooperative constraint adopts a multi-agent cooperative motion strategy to set the speed of the unmanned aerial vehicle to a real number interval with a high-low threshold, namelyCalculating the time intersection of unmanned aerial vehicle formation reaching the target point, and taking the most point of the unmanned aerial vehicle formation initiating cooperative action on the same targetIn a short time, the mathematical model is shown in formula (1):
wherein ,representing that a plurality of unmanned aerial vehicles execute tasks +.>Time set of->Representation unmanned plane g 1 Execution task->Is (are) time of day->Representation unmanned plane g 2 Execution task->Is (are) time of day->Representation unmanned plane g 3 Execution task->Is (are) time of day->Representing +.>Execution task->Is (are) time of day->Representing unmanned plane->Execution task->Is (are) time of day->Representing cooperative pair of targets->Implement task->Temporary formation of->Time interval representing the simultaneous arrival of co-formation,/->Representing unmanned plane->Reaching the target from the current position->Distance of->Representing unmanned plane->Speed interval of>Respectively representing the earliest and latest times of the unmanned aerial vehicle k to execute the task j.
Aiming at the problem that the multi-unmanned aerial vehicle cooperative form is too single in the traditional method, the multi-unmanned aerial vehicle cooperative action method integrating the speed change adjusting mechanism is adopted, so that different unmanned aerial vehicles are allowed to freely form a queue by adjusting the speed of the unmanned aerial vehicles, and the cooperative form of the multi-unmanned aerial vehicle formation is enriched.
In one embodiment, the gene is generated based on a three-layer one-module multi-gene genetic coding methodAfter the chromosome, by judging whether it satisfies the synergic condition and other constraint conditions, finally taking +.>The conditional chromosomes make up the initial population and divide the initial population into a common population and an elite population. Wherein greedy strategy is used to select->Individuals form elite populations.
In a specific embodiment, in step S3, in terms of fitness function construction, a mode of optimizing only one to two indexes in the previous study is changed, the objective function is subjected to generalization setting, and four optimization indexes including task execution benefits, agent formation survival probability, task execution time consumption and distance consumption are included in the objective function. According to the emphasis requirement of the target allocation scheme, changing the weight of each index, and improving the flexibility of the target pre-allocation scheme, the model is described as follows:
wherein For the weight coefficient of each index, N R Representing the number of unmanned aerial vehicles performing class R tasks, N A The number of unmanned aerial vehicles for executing class A tasks is represented, and p represents the number of tasks; />Representing the benefits obtained after the execution of a number of tasks for a certain target,/>Representing the target value->,/>Representing the generated target pre-allocation matrix, wherein h (x) represents the number of unmanned aerial vehicles simultaneously executing a certain task; />Representing survival probability of unmanned aerial vehicle formation; />Task execution list representing a single unmanned aerial vehicle, +.>Representing threat degrees of individual targets; />Representing mileage consumption of unmanned aerial vehicle formation; />Representing unmanned plane->Mileage consumption in the task execution process; />Representing the time consumption of an agent formation to perform a task; />,/>Representation unmanned aerial vehicleThe time spent executing the task.
In a specific embodiment, in step S4, for the problem of overspreading the search direction in the past heuristic method, an adaptive elite crossover operator operation is adopted, and the specific process is as follows:
dividing the parent population into a common population and an elite population according to the previous selection operation;
according to the common populationAdaptive crossover probability in equation (3)Selecting individuals crossing the elite population or performing self-crossing;
wherein Representing adaptive crossover probability,/->Representing the current iteration number of the population, +.>Representing the total iteration number of the population;
elite population individuals are not affected by crossover operations, and replicate into offspring populations, and all individuals in a common population repeat the operations until a specified number of offspring populations are produced.
The crossing principle in which the two selected parent chromosomes cross is as follows:
1) The third row does not participate in the interleaving operation. Because the third row mainly determines the task execution sequence, the third row is ensured to be unchanged, and the offspring chromosomes can be effectively prevented from being repeatedly trapped into local deadlock due to the task sequence.
2) The crossing points of the parent chromosomes need to be in one-to-one correspondence, so that the task types of the offspring chromosomes can be matched with the chromosome types, and the effectiveness of offspring individuals is improved.
3) The number of crossing points remains random.
In a specific embodiment, in step S4, for the problem of low single-point mutation operation efficiency of the genetic algorithm, an adaptive mutation operator operation is adopted, and a multipoint random mutation operation and a multipoint shift mutation operation are introduced, which correspond to the following processes:
s41, selecting a parent chromosome to generate 1 random number between (0, 1), if the random number is smaller than the self-adaptive numberProbability of variationAnd then, carrying out mutation operation on the individual, wherein the mutation probability obtaining method is shown as a formula (4):
S42, randomly selecting a mutation operation for the individuals to be mutated.
Wherein the multipoint random variation operation mainly follows the following three principles:
a) The number of variation points is random, namely the variation points of different individuals in the same population are different;
b) Only the unmanned plane layer participates in variation;
c) The type of unmanned aerial vehicle generated by variation is required to be matched with the task type.
The multi-point shift mutation operation follows the following two principles in addition to the principles a) and c) in the multi-point random mutation operation:
d) The positions participating in shift mutation are paired, namely, the positions of the selected mutation need to be paired pairwise before transformation.
e) The target layer and the unmanned plane layer both participate in shifting.
Aiming at the problem that the searching direction of the traditional heuristic method is too dispersed, the self-adaptive elite crossing operator is adopted to operate, a self-adaptive function related to the iteration progress of the population is constructed, and the population is ensured to always evolve towards a better direction by controlling the scale of elite individuals participating in crossing; meanwhile, aiming at the problem of low single-point mutation operation efficiency of a genetic algorithm, the invention combines the characteristics of multi-gene coding, provides the self-adaptive mutation operator operation, enriches the diversity of offspring population and ensures the optimization efficiency of the algorithm.
In a specific embodiment, in step S4, aiming at the problem that the past heuristic algorithm is easy to fall into local optimization, the invention adopts an adaptive simulated annealing operator operation, which comprises the following specific processes:
s4.1, selecting 1/5 individuals from elite population to participate in simulated annealing operation by using a roulette method;
s4.2, randomly selecting from selected chromosomesPerforming multipoint mutation on individual points, and applying local disturbance to elite individual, wherein +.>The calculation method of (2) is shown in the formula (5):
s4.3, determining whether to accept a new allocation scheme by using a designed Metropolis criterion; if the adaptation value of the new allocation scheme is higher, accepting; if the fitness value is not high, 1 (0, 1) random number is generated, if the random number is smaller thanA new allocation scheme is accepted, otherwise the allocation scheme is not accepted.
Wherein the Metropolis criterion is:
wherein A change value indicating individual fitness before and after annealing; wherein->Indicating an individual fitness value after annealing; />Indicating individual fitness values prior to annealing.
Aiming at the problem that the traditional heuristic algorithm is easy to trap into local optimum, the self-adaptive simulated annealing operator is adopted to operate, and the Metropolis criterion conforming to the population iteration rule is designed, so that the probability that the target allocation method is trapped into the local optimum is effectively reduced, and the solving quality of the method is improved.
According to the heterogeneous unmanned aerial vehicle formation target distribution method based on the improved polygene genetic algorithm, the target is subjected to the reconnaissance and combat cooperative military operation according to the number of the current unmanned aerial vehicles, the number of the targets and the related situation, and a task list of each unmanned aerial vehicle is automatically generated.
In order to verify the effectiveness of the method, the improved polygenic genetic-simulated annealing algorithm (EA-GA for short) is adopted, and the problems of target allocation of the cooperative actions of observing and playing the targets of heterogeneous unmanned aerial vehicle formation are solved by the traditional genetic algorithm (GA for short), the simulated annealing algorithm (SA for short) and the standard polygenic genetic algorithm (MGA). The problem can be described as heterogeneous unmanned aerial vehicle formation consisting of a reconnaissance unmanned aerial vehicle and a batting unmanned aerial vehicle, each of which can only perform a reconnaissance task and a batting task, and action requiring formation requires that all known targets be performed once each. Based on this, two different case backgrounds were selected for comparison testing.
Case one: 5 reconnaissance unmanned aerial vehicles and 10 hitting unmanned aerial vehicles implement reconnaissance and hitting cooperative tasks on 5 targets.
In this setting, the initial properties of the drone are shown in table 1.
Table 1 unmanned aerial vehicle attribute list
The target attributes to be assigned are shown in table 2. Wherein the value of the class I target is 100, the initial threat degree is 0.3, the value of the class II target is 150, the initial threat degree is 0.3, the value of the class III target is 200, and the initial threat degree is 0.3.
Table 2 list of target attributes in case one
And respectively solving the problem model by using the five algorithms, and comparing the advantages and disadvantages of the target allocation scheme searched by the algorithm within the specified iteration times. The initial condition was set to a population size of 100 and a number of iterations of 500.
The steps for processing the target distribution problem by adopting the improved polygenic genetic-simulated annealing algorithm provided by the invention are as follows:
determining the length of the chromosome according to the number of unmanned aerial vehicles and the target number, and encoding the chromosome according to a three-layer one-module encoding method of a fusion vacant mechanism in fig. 2;
judging whether the chromosome generated by the codes meets the cooperative constraint of multiple unmanned aerial vehicles, if so, adding an initial population until the initial population reaches 100;
checking whether the current iteration number reaches 500, if so, ending the algorithm, and outputting a target allocation scheme with the highest fitness value; if the maximum iteration times are not reached, calculating the fitness value of each individual in the current population, and selecting 1/5 of the individuals to form elite population to complete the selection operation;
sequentially implementing an adaptive elite crossover operator operation and an adaptive mutation operator operation on a common population, and implementing an adaptive simulated annealing operator operation on the elite population;
judging whether the individuals subjected to the genetic operation meet complex cooperative constraints, and if so, adding a offspring population; if not, continuing to implement genetic operation of the corresponding operator in the previous step.
The EA-GA method is compared with a standard genetic algorithm (GA algorithm), a simulated annealing algorithm (SA algorithm), a polygenic genetic algorithm (MGA algorithm) and an adaptive genetic algorithm (AGA algorithm) to solve the model of the first case 50 times, and the average fitness value obtained 50 times is compared, wherein the statistical result is shown in figure 5.
According to comparison of calculation results, the target distribution scheme obtained by searching by the EA-GA method is obviously higher than other four algorithms, which proves that the method is better than the other four comparison algorithms.
Case two: 5 reconnaissance unmanned aerial vehicles, 10 hit unmanned aerial vehicles and implement reconnaissance cooperative tasks to 10 targets.
In this setting, the initial properties of the drone are shown in table 1. The target properties are shown in table 3:
table 3 list of target attributes in case two
The steps for processing the target allocation problem by adopting the improved polygenic genetic-simulated annealing algorithm of the invention are as follows:
determining the length of the chromosome according to the number of unmanned aerial vehicles and the target number, and encoding the chromosome according to a three-layer one-module encoding method of a fusion vacant mechanism in fig. 2;
judging whether the chromosome generated by the codes meets the cooperative constraint of multiple unmanned aerial vehicles, if so, adding an initial population until the initial population reaches 100;
checking whether the current iteration number reaches 500, if so, ending the algorithm, and outputting a target allocation scheme with the highest fitness value; if the maximum iteration times are not reached, calculating the fitness value of each individual in the current population, and selecting 1/5 of the individuals to form elite population to complete the selection operation;
sequentially implementing an adaptive elite crossover operator operation and an adaptive mutation operator operation on a common population, and implementing an adaptive simulated annealing operator operation on the elite population;
judging whether the individuals subjected to the genetic operation meet complex cooperative constraints or not; if yes, adding a offspring population, and if not, returning to the previous step to continue to implement the genetic operation of the corresponding operator.
The EA-GA method, the GA algorithm, the SA algorithm, the MGA algorithm and the AGA algorithm are used for solving the model of the case two for 50 times, and the average fitness value obtained for 50 times is compared, so that the statistical result is shown in figure 6.
According to comparison of calculation results, the score of the target distribution scheme obtained by searching by the method is obviously higher than that of other four algorithms, which shows that the method is superior to the other four comparison algorithms.
The constraint considered by the improved polygene genetic algorithm is more complex, genetic operation can be performed in a self-adaptive manner in the population optimization process, the search space of the algorithm is further enlarged, and the target distribution efficiency is greatly improved.
As used in this disclosure and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. The terms "first," "second," and the like, as used in this disclosure, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. Likewise, the word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the invention without departing from the principles thereof are intended to be within the scope of the invention as set forth in the following claims.
Claims (10)
1. The heterogeneous unmanned aerial vehicle cooperative target distribution method based on the polygene genetic algorithm is characterized by comprising the following steps of:
s1, determining the length of a chromosome according to the number of unmanned aerial vehicles and the target number, and encoding the chromosome by adopting a polygene genetic encoding method of a fusion vacant mechanism;
s2, judging whether the chromosome generated by encoding meets the cooperative constraint of multiple unmanned aerial vehicles; if the multi-unmanned aerial vehicle cooperative constraint is met, adding an initial population until the initial population reaches a preset number in scale, and dividing the initial population into a common population and an elite population;
s3, constructing an fitness function, calculating the fitness value of each individual in the initial population, and checking whether the current iteration number reaches the preset number or not; if the preset times are reached, outputting an unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value; if the preset times are not reached, the next step is carried out;
s4, carrying out genetic operation on the initial population by a polygenic genetic algorithm; the genetic operation comprises an adaptive elite crossover operator operation and an adaptive mutation operator operation which are sequentially implemented on a common population, and an adaptive simulated annealing operator operation implemented on the elite population;
s5, calculating the fitness value of each individual in the initial population after genetic operation, and checking whether the current iteration number reaches the preset number or not; if the preset times are reached, outputting an unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value; if the preset times are not reached, the steps S4-S5 are circulated until the unmanned aerial vehicle cooperative target allocation scheme with the highest fitness value is output.
2. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygene genetic algorithm according to claim 1, wherein in step S1, the polygene genetic encoding method is implemented based on three layers of one module, wherein the three layers of one module include a target layer, a task execution sequence layer, an agent layer and a task module; the target layer adopts a real number coding strategy and is composed of target sequence numbers, and the target layer is expanded according to the number of targets in an actual task; the unmanned plane layer fuses with an empty mechanism, and performs random-1 operation on part of genes according to a certain probability on the unmanned plane layer, which means that targets are not allocated to any unmanned plane, and other positions are formed by actual numbers of the unmanned plane, have a corresponding relation with the target layer and are used for matching screening of unmanned plane types and target types; the execution sequence layer carries out real number incremental coding according to the length of the chromosome and is used for defining the task execution sequence of a single unmanned aerial vehicle and analyzing the feasibility of a collaborative strategy among a plurality of unmanned aerial vehicles; the task module contains genetic information related to task types and is used for expanding according to the number of tasks in the problem model.
3. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm according to claim 1 or 2, wherein in step S2, the unmanned aerial vehicle speed is set to a real number interval with a high and low threshold, namely, in the multi-unmanned aerial vehicle cooperative constraintCalculating the time intersection of unmanned aerial vehicle formation reaching a target point, taking the shortest time for the unmanned aerial vehicle formation to initiate cooperative action on the same target, wherein the mathematical model is shown as follows:
wherein ,representing that a plurality of unmanned aerial vehicles execute tasks +.>Time set of->Representation unmanned plane g 1 Execution task->Is used for the time of day (c),representation unmanned plane g 2 Execution task->Is (are) time of day->Representation unmanned plane g 3 Execution task->Is (are) time of day->Representing +.>Execution task->Is (are) time of day->Representing unmanned plane->Execution task->Is (are) time of day->Representing cooperative pair of targets->Implement task->Temporary formation of->Time interval representing the simultaneous arrival of co-formation,/->Representing unmanned plane->Reaching the target from the current position->Distance of->Representing unmanned plane->Speed interval of>Respectively representing the earliest and latest times of the unmanned aerial vehicle k to execute the task j.
4. The heterogeneous unmanned aerial vehicle collaborative target allocation method based on the polygene genetic algorithm according to claim 1 or 2, wherein in step S3, when the fitness function is constructed, the objective function is subjected to generalization setting, and four optimization indexes including task execution benefit, agent formation survival probability, task execution time consumption and distance consumption are included in the objective function.
5. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm according to claim 1 or 2, wherein in step S4, the specific process of the adaptive elite crossover operator operation is as follows:
the common population is subjected to self-adaptive crossover probabilitySelecting individuals crossing or self-acting with elite populationsCrossing; wherein the method comprises the steps of
Representing the current iteration number of the population, +.>Representing the total iteration number of the population; crossing the selected two parent chromosomes; repeating this step for all individuals in the common population until a specified number of offspring populations are produced;
elite population individuals are not affected by crossover operations and replicate into offspring populations.
6. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm according to claim 5, wherein the principle of crossing the two selected parent chromosomes comprises:
1) The third row does not participate in the interleaving operation;
2) The points of the parent chromosomes which are crossed need to be in one-to-one correspondence;
3) The number of crossing points remains random.
7. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm according to claim 2, wherein in step S4, the specific process of the adaptive mutation operator operation is as follows:
s41, selecting a parent chromosome to generate a random number between 0 and 1, if the random number is smaller than the adaptive mutation probabilityPerforming a mutation operation on the individual; wherein the method comprises the steps of
wherein Representing the current iteration number of the population, +.>Representing the total iteration number of the population;
s42, randomly selecting a mutation operation for the individuals to be mutated.
8. The heterogeneous unmanned aerial vehicle collaborative target allocation method based on a polygenic genetic algorithm according to claim 7, wherein the adaptive mutation operator operation comprises a multipoint random mutation operation and a multipoint shift mutation operation, wherein the multipoint random mutation operation follows the following principles:
a) The number of variation points is random, namely the variation points of different individuals in the same population are different;
b) Only the unmanned plane layer participates in variation;
c) The type of the unmanned aerial vehicle generated by variation is required to be matched with the task type;
wherein the multi-point shift mutation operation follows the following two principles in addition to the principles a) and c) in the multi-point random mutation operation:
d) The point positions participating in shift mutation appear in pairs, namely the point positions of the selected mutation need to be paired pairwise before transformation;
e) The target layer and the unmanned plane layer both participate in shifting.
9. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm according to claim 7, wherein in the step S4, the specific process of the adaptive simulated annealing operator operation is as follows:
s4.1, selecting 1/5 individuals from elite population to participate in simulated annealing operation by using a roulette method;
s4.2, randomly selecting from selected chromosomesPerforming multipoint mutation on the individual points, and applying local disturbance to elite individuals; wherein the method comprises the steps of
s4.3, determining whether to accept a new allocation scheme by using a designed Metropolis criterion; if the adaptation value of the new allocation scheme is higher, accepting; if the fitness value is not high, 1 random number between 0 and 1 is generated, if the random number is smaller thanA new allocation scheme is accepted, otherwise the allocation scheme is not accepted.
10. The heterogeneous unmanned aerial vehicle cooperative target allocation method based on the polygenic genetic algorithm according to claim 9, wherein the metapolis criterion is as follows:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310549221.3A CN116258357B (en) | 2023-05-16 | 2023-05-16 | Heterogeneous unmanned aerial vehicle cooperative target distribution method based on polygene genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310549221.3A CN116258357B (en) | 2023-05-16 | 2023-05-16 | Heterogeneous unmanned aerial vehicle cooperative target distribution method based on polygene genetic algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116258357A true CN116258357A (en) | 2023-06-13 |
CN116258357B CN116258357B (en) | 2023-07-25 |
Family
ID=86686573
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310549221.3A Active CN116258357B (en) | 2023-05-16 | 2023-05-16 | Heterogeneous unmanned aerial vehicle cooperative target distribution method based on polygene genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116258357B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116975655A (en) * | 2023-08-29 | 2023-10-31 | 天栋智能科技(天津)有限公司 | Parameter generation method, signal compression and reconstruction method, system, equipment and medium |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113885555A (en) * | 2021-09-14 | 2022-01-04 | 安徽送变电工程有限公司 | Multi-machine task allocation method and system for power transmission line dense channel routing inspection |
CN114417735B (en) * | 2022-03-08 | 2023-04-18 | 大连理工大学 | Multi-unmanned aerial vehicle cooperative task planning method in cross-regional combined combat |
CN116050086A (en) * | 2022-12-15 | 2023-05-02 | 中国航空工业集团公司西安航空计算技术研究所 | Multi-machine air combat task allocation method based on genetic algorithm |
-
2023
- 2023-05-16 CN CN202310549221.3A patent/CN116258357B/en active Active
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116975655A (en) * | 2023-08-29 | 2023-10-31 | 天栋智能科技(天津)有限公司 | Parameter generation method, signal compression and reconstruction method, system, equipment and medium |
CN116975655B (en) * | 2023-08-29 | 2024-04-05 | 天栋智能科技(天津)有限公司 | Parameter generation method, signal compression and reconstruction method, system, equipment and medium |
Also Published As
Publication number | Publication date |
---|---|
CN116258357B (en) | 2023-07-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107515618B (en) | Heterogeneous unmanned aerial vehicle cooperative task allocation method considering time window | |
CN106990792B (en) | Multi-unmanned aerial vehicle collaborative time sequence coupling task allocation method based on hybrid gravity search algorithm | |
CN109542103A (en) | A kind of robot welding paths planning method based on fireworks particle swarm algorithm | |
CN112329348A (en) | Intelligent decision-making method for military countermeasure game under incomplete information condition | |
CN116258357B (en) | Heterogeneous unmanned aerial vehicle cooperative target distribution method based on polygene genetic algorithm | |
CN107330560A (en) | A kind of multitask coordinated distribution method of isomery aircraft for considering temporal constraint | |
Fogel | An overview of evolutionary programming | |
CN112013829A (en) | Multi-UAV/UGV (unmanned aerial vehicle/user generated Unit) cooperative long-term operation path planning method based on multi-objective optimization | |
CN104661917B (en) | Airborne vehicle managing device, airborne vehicle and airborne vehicle management method | |
CN111121784B (en) | Unmanned reconnaissance aircraft route planning method | |
CN109242290B (en) | Automatic generation method for unmanned aerial vehicle group action scheme | |
CN114417735A (en) | Multi-unmanned aerial vehicle cooperative task planning method in cross-regional combined combat | |
CN108596339A (en) | A kind of multi-objective Optimization Genetic Algorithm of structure population and subproblem | |
CN117077981B (en) | Method and device for distributing stand by fusing neighborhood search variation and differential evolution | |
CN112561032B (en) | Multi-agent reinforcement learning method and system based on population training | |
Jing et al. | Cooperative task assignment for heterogeneous multi-UAVs based on differential evolution algorithm | |
CN112070418A (en) | Weapon target allocation method of multi-target whale optimization algorithm | |
CN116050086A (en) | Multi-machine air combat task allocation method based on genetic algorithm | |
CN113487031A (en) | Multi-unmanned aerial vehicle task allocation method based on improved simulated annealing fusion genetic algorithm | |
CN113919425B (en) | Autonomous aerial target allocation method and system | |
Purmonen | Predicting game level difficulty using deep neural networks | |
CN115034557A (en) | Agile satellite emergency task planning method | |
Mishra et al. | Multi-objective genetic algorithm: A comprehensive survey | |
CN108960497B (en) | Military scheme optimization method based on rule mining | |
CN116596287B (en) | Task driving decision-making method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |