CN117850234A - Minimum computation time delay task allocation method and system for unmanned aerial vehicle formation - Google Patents
Minimum computation time delay task allocation method and system for unmanned aerial vehicle formation Download PDFInfo
- Publication number
- CN117850234A CN117850234A CN202410030986.0A CN202410030986A CN117850234A CN 117850234 A CN117850234 A CN 117850234A CN 202410030986 A CN202410030986 A CN 202410030986A CN 117850234 A CN117850234 A CN 117850234A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- population
- genetic algorithm
- distributed
- 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.)
- Pending
Links
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 76
- 230000002068 genetic effect Effects 0.000 claims abstract description 76
- 238000005457 optimization Methods 0.000 claims abstract description 32
- 238000004364 calculation method Methods 0.000 claims description 44
- 210000000349 chromosome Anatomy 0.000 claims description 38
- 230000008569 process Effects 0.000 claims description 11
- 230000035772 mutation Effects 0.000 claims description 9
- 238000004891 communication Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 4
- 230000001172 regenerating effect Effects 0.000 claims description 4
- 230000002759 chromosomal effect Effects 0.000 claims description 2
- 238000012935 Averaging Methods 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 11
- 230000006870 function Effects 0.000 description 8
- 238000011156 evaluation Methods 0.000 description 4
- 108090000623 proteins and genes Proteins 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Abstract
The invention discloses a method and a system for allocating minimum computation delay tasks of unmanned aerial vehicle formation, and relates to the technical field of unmanned aerial vehicle formation, wherein the method comprises the following steps: generating a distributed genetic algorithm parent population by the organic machine based on tasks to be executed by the organic machine/unmanned aerial vehicle formation; the method comprises the steps that an organic machine establishes a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay based on a distributed genetic algorithm parent population; the method comprises the steps that an organic machine solves a distributed genetic algorithm sub-population allocation optimization model to obtain the number of the populations to be allocated to each unmanned aerial vehicle; the unmanned aerial vehicle adopts a genetic algorithm to solve and generate a new population based on the population number to be allocated to each unmanned aerial vehicle; and after the man-machine judges that the iteration termination condition is met, sending the optimal task allocation scheme to the unmanned aerial vehicle. The invention can maximally utilize the available computing resources of each machine to realize online task allocation.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicle/unmanned aerial vehicle formation, in particular to a minimum computation time delay task allocation method and system for unmanned aerial vehicle/unmanned aerial vehicle formation.
Background
Unmanned aerial vehicle/unmanned aerial vehicle formation is a main formation form of future combat, and is characterized in that one unmanned aerial vehicle brings a plurality of isomorphic or heterogeneous unmanned aerial vehicles to execute combat tasks. According to the formation characteristics of the unmanned aerial vehicle, the task decision is usually executed by the unmanned aerial vehicle, and the unmanned aerial vehicle is responsible for executing the tasks issued by the unmanned aerial vehicle. However, in the flight process, when the tasks to be distributed are too many, the requirement of on-line planning instantaneity cannot be met because the calculation resources of the on-board computer of the organic machine are limited. The task allocation solving method based on the distributed strategy can respond to the change of the external environment well, has good real-time performance, but is not suitable for on-line task allocation of the unmanned aerial vehicle/unmanned aerial vehicle formation because the intelligent level of each unmanned aerial vehicle is not high at the present stage and the decision leading effect of the unmanned aerial vehicle in the formation is ignored by the completely distributed computing method. In the unmanned aerial vehicle/unmanned aerial vehicle formation, the unmanned aerial vehicle models are not the same, so that the calculation performance of the onboard computer is different; and in the flight process, the information processed in real time by each unmanned aerial vehicle is different, which also causes that the available computing resources of each unmanned aerial vehicle are not consistent.
Disclosure of Invention
The invention aims to provide a minimum computation time delay task allocation method and a minimum computation time delay task allocation system for formation of a man-machine-unmanned aerial vehicle, which adopt a distributed genetic algorithm to realize formation on-line task planning, reasonably allocate the computation amount of each machine task allocation computation solution and maximally utilize the available computation resources of each machine to realize on-line task allocation.
In order to achieve the above object, the present invention provides the following solutions:
a minimum computation delay task allocation method for unmanned aerial vehicle formation comprises the following steps:
generating a distributed genetic algorithm parent population by the organic machine based on tasks to be executed by the organic machine/unmanned aerial vehicle formation;
the method comprises the steps that an organic machine establishes a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay based on a distributed genetic algorithm parent population;
the unmanned aerial vehicle solves the distributed genetic algorithm sub-population allocation optimization model to obtain the population number to be allocated to each unmanned aerial vehicle;
the unmanned aerial vehicle adopts a genetic algorithm to solve and generate a new population based on the population number to be allocated to each unmanned aerial vehicle;
and after the man-machine judges that the iteration termination condition is met, sending the optimal task allocation scheme to the unmanned aerial vehicle.
In order to achieve the above purpose, the present invention also provides the following solutions:
a unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation system, comprising:
the distributed genetic algorithm parent population generation module is used for generating a distributed genetic algorithm parent population based on tasks to be executed by the man-machine based on the man-machine/unmanned aerial vehicle formation;
the distributed genetic algorithm sub-population allocation optimization model module is used for establishing a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay based on the distributed genetic algorithm parent population by an organic machine;
the solving module is used for solving the distributed genetic algorithm sub-population allocation optimization model by the unmanned aerial vehicle to obtain the population number to be allocated to each unmanned aerial vehicle;
the new population generation module is used for solving and generating a new population by adopting a genetic algorithm based on the population number to be allocated to each unmanned aerial vehicle;
and the optimal task allocation scheme sending module is used for sending the optimal task allocation scheme to the unmanned aerial vehicle after the unmanned aerial vehicle judges that the iteration termination condition is met.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention establishes the matching relation between each calculation link and the organic and unmanned aerial vehicle based on the calculation flow of the distributed genetic algorithm, thereby realizing the reasonable distribution of each calculation link of the distributed genetic algorithm; establishing a gene coding criterion from three aspects of a task layer, a track layer and a constraint layer, and establishing a task allocation optimization target from the shortest total range and the shortest total task duration; aiming at the requirement of high real-time performance of on-line task allocation, a distributed genetic algorithm sub-population allocation model with minimum time delay as a cost is established, airborne computing resources of each machine are fully utilized, and an algorithm for solving the model is designed. Based on the invention, the unmanned aerial vehicle/unmanned aerial vehicle formation can obtain feasible solutions meeting requirements and constraints in a very short time, and the available computing resources of each machine are reasonably utilized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a minimum computation delay task allocation method for an unmanned aerial vehicle formation provided by the invention;
fig. 2 is a detailed flowchart of a method for assigning minimum computation delay tasks for unmanned aerial vehicle formation provided by the invention;
FIG. 3 is a schematic diagram of inter-machine data transfer;
FIG. 4 is a flowchart of the solution algorithm calculation;
FIG. 5 is a flowchart of genetic algorithm calculation;
fig. 6 is an iterative schematic diagram of fitness for each machine to perform genetic algorithm calculations.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a method frame for allocating minimum calculation time delay tasks of a formation of a man-machine-unmanned aerial vehicle, which adopts a distributed genetic algorithm to realize the on-line task planning of the formation, reasonably allocates the calculated amount of calculation solution of each machine task allocation, and maximally utilizes the available calculation resources of each machine to realize the rapid calculation of the on-line task allocation.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
In this embodiment, the unmanned aerial vehicle/unmanned aerial vehicle formation is composed of one unmanned aerial vehicle and three unmanned aerial vehicles, respectively, wherein the unmanned aerial vehicle is responsible for task release and decision, and the UAV 1 And UAV (unmanned aerial vehicle) 2 Can carry out reconnaissance, evaluation task, UAV 3 A striking task may be performed.
The calculation flow of the distributed genetic algorithm is as follows:
(1) Generating an initial chromosome population, wherein the total number of chromosomes is N;
(2) Splitting the chromosome population into m sub-populations, wherein m is the number of calculation units participating in calculation;
(3) Each calculation unit calculates a corresponding sub-population by adopting a genetic algorithm;
(4) After the calculation of all the calculation units is finished, summarizing the preferred chromosome individuals, judging whether the termination condition is met, and if so, finishing; otherwise, entering the step (5);
(5) Generating a new population through crossover and mutation operations, wherein the total number of chromosomes is N, and returning to the step (2).
In the formation of the unmanned aerial vehicle/the unmanned aerial vehicle, the unmanned aerial vehicle is usually used as a core command role and does not directly participate in the execution of the task, namely the unmanned aerial vehicle is a decision layer, and the unmanned aerial vehicle is an execution layer. Therefore, according to the above-mentioned distributed genetic algorithm calculation flow, the initialization and decision parts (1), (2), (4), (5) can be handed over to the unmanned aerial vehicle for calculation, and the specific calculation part (3) can be handed over to the unmanned aerial vehicle for calculation.
Based on this, as shown in fig. 1-2, the method for allocating the minimum computation delay task of the unmanned aerial vehicle-unmanned aerial vehicle formation provided by the invention comprises the following steps:
s1: the unmanned aerial vehicle generates a parent population of the distributed genetic algorithm based on tasks to be executed by the unmanned aerial vehicle/unmanned aerial vehicle formation.
In this embodiment, the unmanned aerial vehicle/unmanned aerial vehicle formation needs to execute the task of reconnaissance, striking and evaluation of 5 targets, so that the gene code of a single chromosome is divided into 5 layers, the first layer is the target number, the second layer is the task type, the third layer is the unmanned aerial vehicle number, the fourth layer is the track direction during reconnaissance and evaluation, the sign is represented by the flight direction during the hover over the target, and the fifth layer is the number of carrying missiles, so that the single chromosome is a matrix of 5 rows and 15 columns.
The generation flow of the parent population of the distributed genetic algorithm is as follows:
s11: the chromosomal gene codes.
The chromosome individual is randomly generated, and the chromosome coding is carried out from three aspects of a task decision layer, a track layer and a constraint condition layer.
Task layer: adopting [0,1] decision variables to construct a matching relation between each unmanned aerial vehicle and a task, wherein 0 represents that the unmanned aerial vehicle does not execute the task, and 1 represents that the unmanned aerial vehicle executes the task;
track layer: means the constraint of the unmanned plane position or the equivalent representation of the flight direction when executing the special task;
constraint condition layer: the unmanned aerial vehicle load constraint is indicated, what type of task can be executed by the unmanned aerial vehicle is represented, and the relationship between the unmanned aerial vehicle and different task demand loads is represented by adding a plurality of layers.
S12: judging whether the generated chromosome individual meets constraint conditions, if yes, putting the chromosome individual into a parent population, and if not, discarding, and regenerating, wherein the constraint judging conditions are whether task load constraints, execution time sequence constraints of a single task sequence { C, A, V } and range L constraints are met.
S13: and judging whether the number of chromosome individuals in the parent population reaches N, if not, returning to S11, otherwise, ending the parent population generation process.
S2: the method comprises the steps of establishing a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay by an organic machine based on a distributed genetic algorithm parent population.
As shown in FIG. 3, the method for establishing the optimization model of the distributed genetic algorithm sub-population distribution comprises the following steps:
s21: establishing unmanned plane U i Calculation time delay model for genetic algorithm solutionThe computational delay model can be expressed as:
wherein: n (N) p For the number of iterations of the p-th loop, N i For the number of individuals in the sub-population, beta is the number of CPU cycles required for single crossover mutation operation of single parent and mother generation gene individuals, f i Is the clock frequency of the unmanned aerial vehicle onboard computer.
S22: build man-machine M and unmanned plane U i Communication time delay model for transmitting sub-population informationThe communication delay model may be expressed as:
wherein: r is R i For the transmission rate of the channel during communication, W i The size of the data to be transmitted, which is positively correlated with the number of sub-populations, can be expressed as W i ＝g(N i )。
S23: according to the distributed genetic algorithm solving process, in one external circulation and unmanned aerial vehicle internal circulation calculation, one unmanned aerial vehicle calculation (genetic algorithm solving), two unmanned aerial vehicle calculations (sub-population allocation calculation and parent population generation calculation) and data transmission between the two unmanned aerial vehicles and the unmanned aerial vehicle are included, so that a sub-population allocation original model taking the minimum time delay as an optimization target can be expressed as:
s.t.N i,p not less than 0 and is a positive integer
f i ≤f i,max
Wherein: k is the external circulation number of the whole distributed genetic algorithm, m is the number of unmanned aerial vehicles, g (N) i,p ) Is unmanned plane U i The size of the data volume transmitted in the P-th cycle, which is related to the number of sub-populations that the drone needs to calculate, deltaT M,p For the calculation time delay of man-machine in the p-th external circulation, N i,p Unmanned plane U for p-th external circulation i Sub-population number f to be calculated i,max Is unmanned plane U i Is used to determine the maximum available computer clock frequency.
S24: since each external loop calculation is independent of each other and DeltaT M,p Independent of the number of sub-population assignments for each drone, therefore, during the optimization process, Δt M,p Can be regarded as an optimization variable N i Therefore, simplifying the optimization model into:
s.t.N i,p not less than 0 and is a positive integer
f i ≤f i,max
The simplified model successfully carries out dimension reduction processing on the original problem, and converts the coupling problem which needs to consider global optimization into the optimization solution of a plurality of sub-problems. This simulation had three unmanned aerial vehicles in total, so m=3, each chromosome was a 5×15 matrix, g (N i,p ) Simplified to KN i,p K is the data size occupied by a single matrix, and K is 1kB and N i,p Setting parameters manually, setting the parameters to be 100, setting beta to be 4.25X10-6 s/GHz, and setting R to be i Are all 5000kB/s, f i The frequency of the external circulation is set to be 0.5GHz, 0.8GHz and 0.6GHz respectively, N is 120, and each external circulation parameter is set to be the setting for simplifying the calculation, and the frequency of the external circulation is set to be 5.
S3: and the man-machine solves the distributed genetic algorithm sub-population allocation optimization model to obtain the population number to be allocated to each unmanned aerial vehicle.
As shown in fig. 4, the specific steps are:
s31: in the calculation of the p-th cycle, the minimum delay DeltaT of the cycle is calculated p ：
S32: based on minimum delay DeltaT p Inverse solution is carried out to obtain each unmanned plane U under average time delay i Corresponding population number N' i,p (N′ i Not necessarily a positive integer);
s33: all N' i,p Rounding down to obtain
S34: calculating the number of individuals to be distributed X p ：
S35: all N' i,p Each 1 is added to obtainTo N i,p Respectively calculate N i,p Corresponding time delay, reserving the population number of the unmanned aerial vehicle corresponding to the minimum value of the time delay, subtracting 1 from the population number of the rest unmanned aerial vehicles, and dividing the number X of the individuals to be distributed p Subtracting 1;
s36: judgment of X p If 0, the process is ended, the population number corresponding to each unmanned aerial vehicle is recorded, and if not 0, the process returns to S35.
Simulation test is carried out by using MATLAB2020b, the memory of the computing platform is 16G, the CPU frequency is 4.0GHz, and the consumption time of the whole sub-population distribution computing process is 7 multiplied by 10 -6 s, calculating to obtain delta T p = 0.0432s, the sub-population distribution results at each stage and the estimated time consumption of each machine for genetic algorithm calculation are:
s4: the unmanned aerial vehicle adopts a genetic algorithm to solve and generate a new population based on the population number to be allocated to each unmanned aerial vehicle.
The calculation flow is as shown in fig. 5:
s41: receiving sub-population data transmitted by a man-machine;
s42: establishing an fitness function based on the shortest total range of formation members and the shortest total task completion time, wherein the fitness function is as follows:
Val＝max(g i )-g i,j
wherein Val is an fitness function, g i,j For UAVs i Objective function of j-th chromosome to be calculated, g i,j Can be expressed as:
wherein, gamma is a weight coefficient, the value is 0.5, L i For UAVs i Is a flying course of the aircraft;
s43: retaining part of excellent individuals to the next generation population by adopting elite strategy, setting the population optimization probability to be 40%, and selecting part of population individuals as parent chromosomes to carry out cross operation by adopting a roulette wheel method, wherein the roulette wheel method is calculated by the following formula:
s44: performing cross operation and mutation operation on the selected individuals to generate a new population, wherein the mutation rate is set to be 0.1;
s45: if the iteration number is reached, outputting the result to the man-machine, otherwise, returning to S43.
S5: and after the man-machine judges that the iteration termination condition is met, sending the optimal task allocation scheme to the unmanned aerial vehicle.
In the implementation, 5 times of cyclic calculation between unmanned aerial vehicles are set, 100 times of iterative times are calculated in the unmanned aerial vehicle genetic algorithm, and the steps S1-S4 are executed in a cyclic mode until the termination condition is met. The final distribution results are shown in table 1, C, A, V respectively refer to the reconnaissance, striking and evaluation tasks corresponding to the tasks, the numbers represent task serial numbers, and the adaptive iterative calculation results in the calculation of each genetic algorithm are shown in fig. 6.
TABLE 1 task assignment results
| Unmanned aerial vehicle serial number | Executing a task sequence |
| UAV 1 | 4C-4V-1C-1V-2C-2V-5V-3V |
| UAV 2 | 5C-3C |
| UAV 3 | 4A-1A-2A-5A-3A |
Example two
In order to execute the corresponding method of the above embodiment to achieve the corresponding functions and technical effects, a minimum computation delay task allocation system for unmanned aerial vehicle formation is provided below.
The system comprises:
and the distributed genetic algorithm parent population generation module is used for generating a distributed genetic algorithm parent population based on tasks required to be executed by the man-machine based on the man-machine/unmanned aerial vehicle formation.
And the distributed genetic algorithm sub-population allocation optimization model module is used for establishing a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay based on the distributed genetic algorithm parent population by an organic machine.
And the solving module is used for solving the distributed genetic algorithm sub-population allocation optimization model by an organic machine to obtain the population number to be allocated to each unmanned aerial vehicle.
And the new population generation module is used for solving and generating a new population by adopting a genetic algorithm based on the population number to be allocated to each unmanned aerial vehicle.
And the optimal task allocation scheme sending module is used for sending the optimal task allocation scheme to the unmanned aerial vehicle after the unmanned aerial vehicle judges that the iteration termination condition is met.
Further, the generation module of the parent population of the distributed genetic algorithm specifically comprises:
and the chromosome individual generation unit is used for randomly generating chromosome individuals and carrying out chromosome coding from three aspects of a task decision layer, a track layer and a constraint condition layer.
The judging module is used for judging whether the generated chromosome individual meets the constraint condition, if so, putting the chromosome individual into the parent population, and if not, discarding the chromosome individual and regenerating the chromosome individual.
And the parent population generation module is used for ending the parent population generation process when the number of chromosome individuals in the parent population reaches N.
Further, the solving module specifically includes:
average time delay delta T p A calculation unit for calculating average time delay delta T of the p-th cycle based on the distributed genetic algorithm sub-population distribution optimization model p 。
An inverse solving unit for solving the phase difference based on the average time delay delta T p Inverse solution is carried out to obtain each unmanned plane U under average time delay i Corresponding population number N' i,p 。
Rounding unit for rounding all N' i,p Rounding down to obtain
A first calculation unit for calculating according to the formulaCalculating the number of individuals to be distributed X p 。
A second calculation unit for integrating allEach 1 is added to obtain N i,p Respectively calculate N i,p Corresponding time delay, reserving the population number of the unmanned aerial vehicle corresponding to the minimum value of the time delay, subtracting 1 from the population number of the rest unmanned aerial vehicles, and dividing the number X of the individuals to be distributed p Subtracting 1; when X is p And when the number is 0, obtaining the population number to be allocated to each unmanned aerial vehicle.
Further, the new population generation module specifically includes:
and the fitness function establishing unit is used for establishing a fitness function based on the shortest total range and the shortest total task duration.
And the selection unit is used for selecting excellent individuals by adopting elite strategies based on the fitness function.
And the new population generation unit is used for performing crossover operation and mutation operation on the excellent individuals to generate a new population.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In summary, the present description should not be construed as limiting the invention.
Claims (10)
1. The minimum computation time delay task allocation method for the unmanned aerial vehicle formation is characterized by comprising the following steps of:
generating a distributed genetic algorithm parent population by the organic machine based on tasks to be executed by the organic machine/unmanned aerial vehicle formation;
the method comprises the steps that an organic machine establishes a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay based on a distributed genetic algorithm parent population;
the unmanned aerial vehicle solves the distributed genetic algorithm sub-population allocation optimization model to obtain the population number to be allocated to each unmanned aerial vehicle;
the unmanned aerial vehicle adopts a genetic algorithm to solve and generate a new population based on the population number to be allocated to each unmanned aerial vehicle;
and after the man-machine judges that the iteration termination condition is met, sending the optimal task allocation scheme to the unmanned aerial vehicle.
2. The method for assigning minimum computation delay tasks for unmanned aerial vehicle formation according to claim 1, wherein the generation of the parent population of the distributed genetic algorithm by the unmanned aerial vehicle based on the tasks to be executed by the unmanned aerial vehicle formation comprises the following steps:
randomly generating chromosome individuals, and carrying out chromosome coding from three aspects of a task decision layer, a track layer and a constraint condition layer;
judging whether the generated chromosome individual meets the constraint condition, if so, putting the chromosome individual into a parent population, and if not, discarding the chromosome individual, and regenerating;
when the number of chromosomal individuals in the parent population reaches N, the parent population generation process ends.
3. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation method of claim 1, wherein the expression of the distributed genetic algorithm sub-population allocation optimization model is:
s.t.N i,p not less than 0 and is a positive integer
f i ≤f i,max
Wherein k is the external circulation number of the distributed genetic algorithm, m is the number of unmanned aerial vehicles, g (N) i,p ) Is unmanned plane U i Data size, N, transmitted in the P-th cycle p For the number of iterations of the solution of the unmanned aerial vehicle in the p-th cycle using a genetic algorithm, N i,p Unmanned plane U for p-th external circulation i The number of sub-populations to be calculated, beta is the number of CPU cycles required by single crossover mutation operation of single parent and mother generation genetic individuals, R i Is unmanned plane U i Transmission rate of channel during communication, f i Is unmanned plane U i The clock frequency of the onboard computer, N is the number of chromosome individuals in the parent population, f i,max Is unmanned plane U i Is used to determine the maximum available computer clock frequency.
4. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation time delay task allocation method of claim 3, wherein the unmanned aerial vehicle solves the distributed genetic algorithm sub-population allocation optimization model to obtain the population number to be allocated to each unmanned aerial vehicle, and specifically comprises the following steps:
calculating average time delay delta T of the p-th cycle based on the distributed genetic algorithm sub-population distribution optimization model p ；
Based on average time delay DeltaT p Inverse solution is carried out to obtain each unmanned plane U under average time delay i Corresponding population number N' i,p ；
All N' i,p Rounding down to obtain
According to the formulaCalculating the number of individuals to be distributed X p ；
All are put togetherEach 1 is added to obtain N i,p Respectively calculate N i,p Corresponding time delay, reserving the population number of the unmanned aerial vehicle corresponding to the minimum value of the time delay, subtracting 1 from the population number of the rest unmanned aerial vehicles, and dividing the number X of the individuals to be distributed p Subtracting 1;
when X is p And when the number is 0, obtaining the population number to be allocated to each unmanned aerial vehicle.
5. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation method according to claim 1, wherein the unmanned aerial vehicle adopts a genetic algorithm to solve and generate a new population based on the population number to be allocated to each unmanned aerial vehicle, and specifically comprises:
establishing an fitness function based on the shortest total range and the shortest total task duration;
selecting excellent individuals by adopting elite strategy based on the fitness function;
and performing crossover operation and mutation operation on the excellent individuals to generate a new population.
6. A minimum computation latency task allocation system for unmanned aerial vehicle-unmanned aerial vehicle formation, comprising:
the distributed genetic algorithm parent population generation module is used for generating a distributed genetic algorithm parent population based on tasks to be executed by the man-machine based on the man-machine/unmanned aerial vehicle formation;
the distributed genetic algorithm sub-population allocation optimization model module is used for establishing a distributed genetic algorithm sub-population allocation optimization model based on the minimum total time delay based on the distributed genetic algorithm parent population by an organic machine;
the solving module is used for solving the distributed genetic algorithm sub-population allocation optimization model by the unmanned aerial vehicle to obtain the population number to be allocated to each unmanned aerial vehicle;
the new population generation module is used for solving and generating a new population by adopting a genetic algorithm based on the population number to be allocated to each unmanned aerial vehicle;
and the optimal task allocation scheme sending module is used for sending the optimal task allocation scheme to the unmanned aerial vehicle after the unmanned aerial vehicle judges that the iteration termination condition is met.
7. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation system of claim 6, wherein the distributed genetic algorithm parent population generation module specifically comprises:
the chromosome individual generation unit is used for randomly generating chromosome individuals and carrying out chromosome coding from three aspects of a task decision layer, a track layer and a constraint condition layer;
the judging module is used for judging whether the generated chromosome individual meets the constraint condition, if so, putting the chromosome individual into a parent population, and if not, discarding the chromosome individual and regenerating the chromosome individual;
and the parent population generation module is used for ending the parent population generation process when the number of chromosome individuals in the parent population reaches N.
8. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation system of claim 6, wherein the expression of the distributed genetic algorithm sub-population allocation optimization model is:
s.t.N i,p not less than 0 and is a positive integer
f i ≤f i,max
Wherein k is the external circulation number of the distributed genetic algorithm, m is the number of unmanned aerial vehicles, g (N) i ) Is unmanned plane U i Data size, N, transmitted in the P-th cycle p For the number of iterations of the solution of the unmanned aerial vehicle in the p-th cycle using a genetic algorithm, N i,p Unmanned plane U for p-th external circulation i The number of sub-populations to be calculated, beta is the number of CPU cycles required by single crossover mutation operation of single parent and mother generation genetic individuals, R i Is unmanned plane U i Transmission rate of channel during communication, f i Is unmanned plane U i The clock frequency of the onboard computer, N is the number of chromosome individuals in the parent population, f i,max Is unmanned plane U i Is used to determine the maximum available computer clock frequency.
9. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation method of claim 8, wherein the solving module specifically comprises:
average time delay delta T p A calculation unit for calculating average time delay delta T of the p-th cycle based on the distributed genetic algorithm sub-population distribution optimization model p ；
An inverse solving unit for averaging time basedDelay delta T p Inverse solution is carried out to obtain each unmanned plane U under average time delay i Corresponding population number N' i,p ；
Rounding unit for rounding all N' i,p Rounding down to obtain
A first calculation unit for calculating according to the formulaCalculating the number of individuals to be distributed X p ；
A second calculation unit for integrating allEach 1 is added to obtain N i,p Respectively calculate N i,p Corresponding time delay, reserving the population number of the unmanned aerial vehicle corresponding to the minimum value of the time delay, subtracting 1 from the population number of the rest unmanned aerial vehicles, and dividing the number X of the individuals to be distributed p Subtracting 1; when X is p And when the number is 0, obtaining the population number to be allocated to each unmanned aerial vehicle.
10. The unmanned aerial vehicle-unmanned aerial vehicle formation minimum computation delay task allocation system of claim 6, wherein the new population generation module specifically comprises:
the fitness function building unit is used for building a fitness function based on the shortest total range and the shortest total task duration;
a selection unit for selecting excellent individuals by elite strategy based on the fitness function;
and the new population generation unit is used for performing crossover operation and mutation operation on the excellent individuals to generate a new population.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410030986.0A CN117850234A (en) | 2024-01-09 | 2024-01-09 | Minimum computation time delay task allocation method and system for unmanned aerial vehicle formation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410030986.0A CN117850234A (en) | 2024-01-09 | 2024-01-09 | Minimum computation time delay task allocation method and system for unmanned aerial vehicle formation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117850234A true CN117850234A (en) | 2024-04-09 |
Family
ID=90545832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410030986.0A Pending CN117850234A (en) | 2024-01-09 | 2024-01-09 | Minimum computation time delay task allocation method and system for unmanned aerial vehicle formation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117850234A (en) |
-
2024
- 2024-01-09 CN CN202410030986.0A patent/CN117850234A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112351503B (en) | Task prediction-based multi-unmanned aerial vehicle auxiliary edge computing resource allocation method | |
| CN111582691B (en) | Double-layer planning-based transportation capacity matching method for multiple transportation modes of passenger transport hub | |
| CN109682380A (en) | A kind of communication unmanned plane method for optimizing route and equipment | |
| CN107330588B (en) | Task planning method for cooperative reconnaissance of multi-base heterogeneous unmanned aerial vehicle | |
| CN113346944A (en) | Time delay minimization calculation task unloading method and system in air-space-ground integrated network | |
| CN113032904B (en) | Model construction method, task allocation method, device, equipment and medium | |
| CN112784362A (en) | Hybrid optimization method and system for unmanned aerial vehicle-assisted edge calculation | |
| CN114545975A (en) | Multi-unmanned aerial vehicle system task allocation method integrating multi-target evolution algorithm and contract network algorithm | |
| CN111553469A (en) | Wireless sensor network data fusion method, device and storage medium | |
| CN114585006B (en) | Edge computing task unloading and resource allocation method based on deep learning | |
| CN115374853A (en) | Asynchronous federal learning method and system based on T-Step polymerization algorithm | |
| CN114415735A (en) | Dynamic environment-oriented multi-unmanned aerial vehicle distributed intelligent task allocation method | |
| CN107679669A (en) | A kind of airport aircraft gate dispatching method and system based on meta-heuristic method | |
| CN112801540A (en) | Intelligent cooperative architecture design based on unmanned cluster | |
| CN113033970A (en) | AUV cluster layered distribution method for large-scale tasks | |
| CN116893861A (en) | Multi-agent cooperative dependency task unloading method based on space-ground cooperative edge calculation | |
| CN113422795B (en) | Vehicle-mounted edge task centralized scheduling and resource allocation joint optimization method based on deep reinforcement learning | |
| CN115016537A (en) | Heterogeneous unmanned aerial vehicle configuration and mission planning joint optimization method under SEDA scene | |
| CN112445617B (en) | Load strategy selection method and system based on mobile edge calculation | |
| Zhao et al. | Adaptive Swarm Intelligent Offloading Based on Digital Twin-assisted Prediction in VEC | |
| CN117850234A (en) | Minimum computation time delay task allocation method and system for unmanned aerial vehicle formation | |
| CN112750298A (en) | Truck formation dynamic resource allocation method based on SMDP and DRL | |
| Gao | Computing resource allocation strategy based on mobile edge computing in internet of vehicles environment | |
| CN117939667A (en) | Communication resource and computing resource joint allocation method in unmanned aerial vehicle auxiliary emergency communication | |
| CN116909717B (en) | Task scheduling method |
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 |