CN111797966B - Multi-machine collaborative global target distribution method based on improved flock algorithm - Google Patents
Multi-machine collaborative global target distribution method based on improved flock algorithm Download PDFInfo
- Publication number
- CN111797966B CN111797966B CN202010354151.2A CN202010354151A CN111797966B CN 111797966 B CN111797966 B CN 111797966B CN 202010354151 A CN202010354151 A CN 202010354151A CN 111797966 B CN111797966 B CN 111797966B
- Authority
- CN
- China
- Prior art keywords
- target
- unmanned aerial
- algorithm
- collaborative
- target distribution
- 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.)
- Active
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 82
- 244000144992 flock Species 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000009191 jumping Effects 0.000 claims abstract description 7
- 241001494479 Pecora Species 0.000 claims description 39
- 239000000446 fuel Substances 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 10
- 230000006872 improvement Effects 0.000 claims description 6
- 230000008901 benefit Effects 0.000 abstract description 12
- 230000002068 genetic effect Effects 0.000 description 9
- 230000003993 interaction Effects 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000009304 pastoral farming Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 241000282461 Canis lupus Species 0.000 description 1
- 241001074086 Peprilus triacanthus Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000019637 foraging behavior Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multi-machine collaborative global target distribution method based on an improved flock algorithm, which comprises the following steps: step 1, determining cost benefits to ensure that the overall efficiency of the collaborative combat of multiple unmanned aerial vehicles is maximized; step 2, determining constraint conditions to ensure that the overall efficiency of the collaborative combat of the multiple unmanned aerial vehicles is maximized; step 3, determining a multi-machine collaborative global target allocation model; and 4, solving a multi-machine collaborative global target distribution model. The multi-machine collaborative global target distribution method based on the improved flock algorithm has the advantages of high stability, high optimal solution quality and high convergence speed, has the capability of jumping out of a local optimal solution, and can more effectively ensure the maximization of the overall efficiency of multi-unmanned plane collaborative combat.
Description
Technical Field
The invention relates to the technical field of information of a novel flock algorithm and multi-machine collaborative global target distribution, in particular to a multi-machine collaborative global target distribution method based on an improved flock algorithm.
Background
With the development and maturity of unmanned aerial vehicle technology and the continuous improvement of intelligent level, unmanned aerial vehicle will become the leading person of future sky and the main equipment of world armed forces, has huge combat potential in future battlefield. In modern warfare with informatization, networking and systematization against high-speed development, tasks such as information reconnaissance, battlefield striking and the like which are carried out by relying on a single unmanned aerial vehicle can not meet current task demands far, and the cooperative execution of battlefields by utilizing a plurality of unmanned aerial vehicles for a plurality of targets has become a necessary trend. Due to the fact that the operational environment complexity and the number of unmanned aerial vehicles are increased, the multi-unmanned aerial vehicle cooperative target allocation is particularly important for improving the operational efficiency, the multi-unmanned aerial vehicle automatic cooperative task completion key technology is adopted, and practicality of mutual cooperation and target reasonable allocation of unmanned aerial vehicles is determined.
In recent years, research on the problem of target allocation of multiple unmanned aerial vehicles is widely focused, and common methods include mathematical planning, negotiation-based methods and heuristic algorithms. The mathematical programming is a deterministic method for centrally solving the target allocation, the method has more specific requirements on the research object, the mathematical model needs to be changed and adjusted, and when the model scale is overlarge, the calculation amount of solving is increased exponentially. The negotiation-based method belongs to a distributed task planning method, and a contract network and an auction algorithm are commonly used. The method is suitable for task allocation and decision making under the scene with strong uncertainty, high dynamic and high real-time requirements. Compared with the prior art, the heuristic algorithm represented by the particle swarm algorithm, the genetic algorithm, the simulated annealing algorithm, the ant colony algorithm, the gray wolf algorithm and the like has low computational complexity, flexible application and easy realization, is widely used for solving the target allocation problem at present, but the traditional heuristic algorithm is difficult to find a reliable initial allocation scheme and has non-ideal convergence rate.
The flock algorithm (Sheep Optimization, SO) is a novel clustered intelligent algorithm that simulates flock foraging behavior as proposed by qu-robeng and Xu Lunxiang et al in 2018. The algorithm is launched from the core of the intelligent clustering algorithm, and three strategies of global searching, local development and local optimum jumping out in the algorithm are designed by simulating three behaviors of the head sheep lead of the flock, the flock interaction and the shepherd supervision. Compared with the particle swarm algorithm, the algorithm can obtain a solution with higher quality, and has higher convergence speed and better stability. However, each sheep in the common flock algorithm is searched based on continuous space (interval), the initial position and the position updating mode are continuous functions, the variables in the multi-unmanned-plane cooperative target allocation problem are discrete, in addition, the stability of the target allocation problem solved by the algorithm is greatly influenced by grazing operation, and the basic flock algorithm is correspondingly improved according to the problems.
Aiming at the problems of unreliable optimal solution quality and non-ideal convergence speed of the traditional heuristic algorithm, the method establishes a target distribution model with multiple constraint conditions according to the characteristics of multi-machine collaborative global target distribution, and adopts an improved flock algorithm to solve.
Disclosure of Invention
Aiming at the problems, the invention provides a multi-machine collaborative global target distribution method based on an improved flock algorithm, which has the advantages of high stability, high optimal solution quality, high convergence speed, capability of jumping out of a local optimal solution and capability of more effectively guaranteeing the maximization of the overall efficiency of multi-unmanned plane collaborative combat.
Step 1: determining cost and benefit to ensure maximization of overall efficiency of multi-unmanned aerial vehicle collaborative combat
In order to ensure that the overall efficiency of the collaborative combat of multiple unmanned aerial vehicles is maximized, the fuel consumption cost of the unmanned aerial vehicles and the damage cost when the target is attacked are required to be as low as possible, and meanwhile, the income cost when the target is attacked is as high as possible. Therefore, the cost benefit is determined from three aspects of fuel consumption cost, attack damage cost and attack benefit cost, so that the overall efficiency maximization of unmanned plane cooperative warfare is ensured.
(1) Cost of fuel consumption
The unmanned aerial vehicle is required to consider the fuel consumption problem when accomplishing the task, and the fuel consumption of its flight is relevant with flight course, flight time, and the shorter the flight distance, the less fuel consumption, the available flight course size represents the fuel consumption cost. The total fuel consumption cost can be expressed as:
in the formula (1), d i (T j ,T j+1 ) Unmanned plane U i The distance from the jth target to the (j+1) th target in the allocated targets represents the unmanned plane U when j is 1 i Distance from the departure point to the first target point. When the target is pre-allocated, the specific flight range of the unmanned aerial vehicle cannot be predicted, so that the flight range of the unmanned aerial vehicle is expressed by the linear distance between the unmanned aerial vehicle and the target point.
(2) Cost of attack damage
The unmanned aerial vehicle can be influenced by threats such as enemy firepower, topography in a flight environment and obstacles when attacking the target, and the damage cost is minimized, so that the threat degree of the unmanned aerial vehicle in the task execution process is ensured to be minimum.
Suppose unmanned plane U i Attack target T i The damage probability of the unmanned aerial vehicle is h ij The attack damage cost of all unmanned aerial vehicles is:
(3) Cost of attack benefit
The attack income cost refers to the target value income which can be obtained by the unmanned aerial vehicle when the unmanned aerial vehicle attacks the target. The objective attack income cost is maximized as an objective, and in order to calculate the objective function value conveniently, the cost is estimated by adopting the objective residual value.
Suppose unmanned plane U i Attack target T i A time pair,
The killing probability of the target is p ij Target T i Has a value of v j The attack benefit cost of all unmanned aerial vehicles is:
in actual combat, each index cannot be guaranteed to be optimal, and for a multi-objective problem, different indexes can be processed by adopting a normalization method according to the relative weights of all the objectives, so that the problem is converted into a single-objective optimization problem. The optimal objective function for multi-drone cooperative target allocation can be expressed as:
min f＝c 1 α 1 f F +c 2 α 2 f A +c 3 α 3 f V (4)
in the formula (4), c 1 ,c 2 ,c 3 The weight coefficient represents the importance degree of each optimization index, and the value range is 0,1]And satisfy c 1 +c 2 +c 3 ＝1，α 1 ,α 2 ,α 3 For the scaling factor, it is guaranteed that the cost values are in the same order.
Step 2: determining constraint conditions ensures that overall efficiency of collaborative combat of multiple unmanned aerial vehicles is maximized
The multi-unmanned aerial vehicle cooperative target allocation is a complex multi-constraint optimization problem, and in order to ensure that the overall efficiency of unmanned aerial vehicle cooperative combat is maximized, constraint conditions considered by the application include:
(1) Maximum voyage constraint
Unmanned aerial vehicle is limited by airborne fuel, and single flyable distance is limited, and unmanned aerial vehicle U is assumed i The total flying distance is L i ，Unmanned plane U i Maximum distance that can be flown, the total range of the unmanned aerial vehicle for executing the task should be smaller than the maximum distance that can be flown, namely
(2) Maximum execution capability constraint
The number of ammunition that each unmanned aerial vehicle can carry is limited by the load capacity, given that each ammunition can only attack a target once, num i Is unmanned plane U i The target number of the unmanned aerial vehicle attack is less than or equal to the maximum execution capacity, namely
(3) Target execution order constraint
When unmanned aerial vehicles attack targets cooperatively, important targets need to be attacked preferentially, and priority is required to be higher than other targets for certain specific targets. Assume target T i Priority of (2) is greater than target T j The execution order is required to satisfy:
t j ＞t i +Δt (7)
in formula (7), t i ,t j Respectively are targets T i And target T j The time of attack, Δt, is the minimum time interval in which two targets are attacked, Δt > 0.
(4) Decision variable constraints
Because the number of unmanned aerial vehicles and the number of targets have different magnitude relations, the constraints on the unmanned aerial vehicles and the targets are different when the targets are distributed. When u is more than t, each unmanned aerial vehicle attacks at least one target; when u < t, each target is attacked at least once. The constraint can be expressed as:
step 3: determining a multi-machine collaborative global goal allocation model
3.1 target Allocation combat scene
In the actual combat process, the situations of the unmanned aerial vehicle and the targets cannot be determined, and in order to enable the efficiency of the method in the target distribution process to be higher, a target distribution model is built by taking a plurality of unmanned aerial vehicles to execute cooperative attack tasks on a plurality of targets as a research background.
Suppose that there is U (u.gtoreq.1) unmanned plane U= { U on my side 1 U 2 ... U u T= { T of T (T is not less than 1) determination targets of enemy 1 T 2 ... T t Each aircraft has different flight performance and load size, and different targets have different attack values and resistance.
Because the number of unmanned aerial vehicles and targets is uncertain during actual combat, the method and the device establish a target distribution model aiming at three typical situations, specifically:
when u=t, the unmanned aerial vehicle is required to be in one-to-one correspondence with the target, the situation model is simple, and the cooperative constraint relationship is few;
when u is more than t, each unmanned aerial vehicle only distributes one target, and a situation that a plurality of unmanned aerial vehicles attack one target in a cooperative manner exists, and under the situation, the time cooperative requirement on unmanned aerial vehicle formation distributed with the same target is high;
when u < t, each target is only allocated once, and a situation exists in which one unmanned plane attacks a plurality of targets, and the allocated targets need to follow the time sequence constraint of task execution.
When the target is distributed, the corresponding relation between the unmanned aerial vehicle and the target is determined by a decision variable x ij The definition is determined as shown in a formula (9):
for different quantitative relationships, the decision variables can be expressed as:
in the formula (10), i 'represents the number of the unmanned aerial vehicle of the attack target j, and j' represents the number of the attack target of the unmanned aerial vehicle i.
For the multi-target optimization problem, an objective function needs to be established as an optimization index to judge the advantages and disadvantages of a target distribution result, and the optimization index considered in the method comprises unmanned aerial vehicle fuel consumption cost, damage cost when a target is attacked and income cost of the target is attacked, and meanwhile constraint conditions such as unmanned aerial vehicle flight distance, flight time, load size and target execution sequence need to be met.
3.2 construction of fitness function
The constraint conditions of the multi-machine cooperative target allocation problem are numerous, and a proper mode is selected to process so as to obtain the fitness function. Constraint condition processing is carried out by adopting a penalty function method, and the corresponding penalty function is as follows:
(1) Maximum range
If the flight distance of a certain unmanned aerial vehicle in the target distribution result exceeds the maximum flight distance, penalty is applied to the unmanned aerial vehicle:
in the formula (11), L represents the flight distance of the unmanned aerial vehicle, L max And representing the maximum flight distance of the unmanned aerial vehicle, wherein l is a penalty value applied when the maximum range constraint is not met.
(2) Maximum execution capability
If a certain unmanned aerial vehicle in the target allocation result exceeds the constraint of the maximum execution capacity, penalty is applied to the unmanned aerial vehicle:
in the formula (12), num represents the target number allocated to the unmanned aerial vehicle, num represents the unmanned aerial vehicle payload, and n is a penalty value applied when the maximum execution capacity constraint is not satisfied.
Order of target execution
If a certain unmanned aerial vehicle in the target allocation result does not meet the target execution order constraint, penalty is applied to the unmanned aerial vehicle:
after a plurality of constraint conditions are processed by adopting a penalty function method, the fitness function when the flock algorithm solves the optimal target allocation scheme can be expressed as follows:
in the formula (14), f is an objective function value corresponding to a sheep, C is a penalty term, and when C is 0, the individual is feasible.
Step 4: multi-machine collaborative global target allocation model solution
4.1 flock initialization
Because the selection of the initialization mode directly affects the searching efficiency of the algorithm and the result of the allocation problem, when the discrete flock algorithm is used for solving the problem of the allocation of the cooperative targets of multiple unmanned aerial vehicles, a proper population initialization mode needs to be selected.
In the discrete flock algorithm, each sheep represents an alternative solution, and the whole flock is subjected to position updating through head sheep leading, flock interaction and shepherd monitoring behaviors so as to find the optimal solution. According to the size relation and the constraint condition between the unmanned aerial vehicle and the target, the primary flock is set in a flexible initialization mode. The dimension of the solution represented by each sheep depends on the current target allocation, assuming that the dimension of the solution is N c Then N c The value of the method is as follows:
when the number of unmanned aerial vehicles is greater than the number of targets, the solution dimension is the total number of unmanned aerial vehicles, and when the number of targets to be distributed is greater than or equal to the number of unmanned aerial vehicles, the solution dimension is the total number of targets to be distributed. Flocks are represented in a multi-dimensional array as shown in figures 2-4.
Taking u < t (u=4, t=8) as an example, assuming that the final allocation result is a solution as represented by sheep 1 in fig. 2, the corresponding decision variable matrix is as follows:
the initialization mode can meet the constraint condition of decision variables in target allocation, the advantages and disadvantages of the initial population directly influence the result after the offspring evolves, whether each scheme accords with the constraint condition is judged after the initial population is obtained in order to ensure that the schemes are effective, and if not, the scheme is initialized again.
4.2 sheep flock algorithm improvement strategy and solving step
The present application improves flock movement to a way to generate random integer updated positions. Assuming that a sheep performs a head sheep leading operation, the position updating mode of the common flock algorithm is as follows:
the improved ram flock algorithm is updated in the following way:
step function flow table 1 in equation (18) shows.
TABLE 1 Step function flow
The steps for multi-machine collaborative global target allocation using the modified flock algorithm are shown in table 2.
Table 2 improved flock algorithm steps
The method has the advantages that the adaptability function value can be greatly influenced by small changes in each dimension of the solution in the target allocation problem, the stability of the algorithm is greatly influenced by reinitialization of the sheep after grazing. Assuming u=t=6, a sheep was grazed and the operation thereof is shown in fig. 5.
The flock algorithm realizes quick global exploration by simulating the flock of the first sheep, so that the flock is quickly approaching to the known global optimal solution; local development is realized through mutual movement of sheep flocks, so that the convergence speed is further increased; and judging whether to enter the local optimum and rapidly jump out of the local optimum solution by using a shepherd monitoring mechanism.
(1) Sheep head leading collar
The sheep with the best fitness function value in sheep flock refers to the behavior of each sheep moving to the sheep, and the global exploration mechanism of the corresponding algorithm updates the sheep position only when the new sheep fitness function value is better than the old sheep in order to ensure the searching performance, as shown in fig. 6.
(2) Sheep flock interaction
Local development mechanism of sheep flock interaction behavior corresponding algorithm, and each sheep in sheep flock is x i Will randomly select another random sheep x j And carrying out a flock interaction strategy. If sheep x is selected i The fitness value of (a) is superior to that of random sheep x j X is then i To far away from x j Location update of x j Toward x i And vice versa. Also, to ensure search performance, the sheep position is updated only when the new sheep fitness function value is better than the old sheep, as shown in fig. 7.
(3) Shepherd supervision
When the difference of the fitness function of the first generation of sheep and the last generation of sheep is smaller than a threshold epsilon, a shepherd supervision mechanism is introduced to jump out of local optimization. Each sheep will be grazed by the shepherd with a probability p, i.e. the sheep will be reinitialized with a probability p, as shown in fig. 8.
Compared with the traditional heuristic algorithm and the basic flock algorithm, the quality of the optimal solution obtained in the multi-machine collaborative global target distribution is more reliable, the capability of jumping out of the local optimal solution is provided, the convergence speed is faster, the stability is higher, and the overall efficiency maximization of multi-unmanned plane collaborative combat can be more effectively ensured.
Drawings
Fig. 1 is a block diagram of steps of a multi-machine collaborative global target allocation method based on an improved flock algorithm.
Fig. 2-4 are cases when population initialization is u=t (u=t=6), u < t (u=4, t=8), u > t (u= 8,t =4), respectively.
Fig. 5 shows the case where a sheep is grazed when u=t=6 in grazing operation.
Fig. 6 is a flowchart of a ram guidance algorithm.
Fig. 7 is a flowchart of a flock interaction algorithm.
Fig. 8 is a flow chart of a shepherd supervision algorithm.
Detailed Description
The system according to the invention and the method of use thereof are further illustrated by the following examples:
example 1
The invention discloses a multi-machine collaborative global target distribution method based on an improved flock algorithm, which specifically comprises the following steps:
step 1: and determining cost benefits to ensure that the overall efficiency of the collaborative combat of the multiple unmanned aerial vehicles is maximized. In order to ensure that the overall efficiency of the collaborative combat of multiple unmanned aerial vehicles is maximized, the fuel consumption cost of the unmanned aerial vehicles and the damage cost when the target is attacked are required to be as low as possible, and meanwhile, the income cost when the target is attacked is as high as possible.
Step 2: and determining constraint conditions to ensure that the overall efficiency of the collaborative combat of the multiple unmanned aerial vehicles is maximized. The multi-unmanned aerial vehicle cooperative target allocation is a complex multi-constraint optimization problem, and the constraint conditions considered by the application comprise: maximum voyage constraint, maximum execution capability constraint, target execution order constraint, and decision variable constraint are expressed as:
t j ＞t i +Δt
wherein unmanned aerial vehicle U i The total flying distance is L i ，Unmanned plane U i Maximum distance to fly, num i Is unmanned plane U i Maximum loading of t i ,t j Respectively are targets T i And target T j The time of attack, Δt, is the minimum time interval in which two targets are attacked, Δt > 0.
Step 3: determining a multi-machine cooperative global target distribution model, determining a target distribution combat scene and constructing an adaptability function, wherein the adaptability function when the flock algorithm solves an optimal target distribution scheme can be expressed as follows:
wherein f is the objective function value corresponding to a sheep, C is a punishment term, and when C is 0, the individual is feasible.
Step 4: solving a multi-machine collaborative global target distribution model, selecting a proper population initialization mode, and determining an improvement strategy and an improvement step of a flock algorithm.
Preferably, a target distribution model with multiple constraint conditions is established according to the characteristics of multi-machine cooperative global target distribution, an improved flock algorithm is adopted for solving, the optimal solution is obtained in the multi-machine cooperative global target distribution, the quality is reliable, the capability of jumping out of the local optimal solution is achieved, the convergence speed is high, the stability is high, and the overall efficiency maximization of multi-unmanned plane cooperative combat can be effectively guaranteed.
Example 2
In order to verify the effectiveness of the flock algorithm for the problem of multi-machine collaborative global target allocation, MATLAB simulation experiments were performed and performance comparison was performed with a Genetic Algorithm (GA).
Step one, algorithm initialization
Assuming that 8 fight unmanned aerial vehicles are available for calling before fight, 8 target points to be attacked, initial parameter information classification of the unmanned aerial vehicles and the target points is shown in tables 3 and 4.
Table 3 initial information table for unmanned aerial vehicle
TABLE 4 initial information Table of target points
The probability of damage after the target attack is shown in table 5 assuming that the probability of killing each target by the unmanned aerial vehicle is the same. The algorithm parameters are set as follows: population size np=50, maximum number of iterations 100, threshold epsilon=10 -8 The reset probability p=0.2.
TABLE 5 damage probability Table after an unmanned plane attacks a target
Because the number of unmanned aerial vehicles and the target number are uncertain in actual combat, the simulation experiment is carried out on three different application scenes, and each scene is set as follows:
scene one: u=t, unmanned plane U 1 U 6 Attack target point T 1 T 6 ；
Scene II: u > t, unmanned plane U 1 U 8 Attack target point T 1 T 4 ；
Scene III: u is less than t, unmanned plane U 1 U 4 Attack target point T 1 T 8 。
Step two, algorithm simulation
The application simulates distribution problems in different scenes, and specific distribution results are shown in tables 6, 7 and 8:
TABLE 6 scene one assignment results
TABLE 7 scene allocation results
Table 8 scenario three assignment results
Tables 6-8 show that the problem of multi-machine collaborative global target allocation under different quantitative relationships and multiple constraint conditions can be solved by utilizing the flock algorithm, and reasonable results can be obtained. Because the global target allocation problem has extremely high requirements on the quality, convergence speed and stability of the solution, the performance of the improved flock algorithm and the original genetic algorithm is compared.
Step three, performance analysis
In order to verify the searching efficiency of the flock algorithm for solving the problem of multi-machine collaborative global target allocation, the application compares the performance of the improved flock algorithm with that of a genetic algorithm, wherein the same parameter of the two algorithms is set to be the same value, and the crossover probability P in the genetic algorithm c Probability of variation P =0.9 m =0.1. Primary population after population initializationThe average fitness function values are shown in table 9.
TABLE 9 average fitness function values for initial population in each scenario
In order to avoid the influence of accidental factors of a single experiment, 30 simulation tests are respectively carried out on each scene, and the results are recorded and the average value is calculated. The cost function value of the primary population in different scenes is stable: scene one was about 33.7, scene two was about 29.8, and scene three was about 31.6. Table 9 shows that the cost function value of the primary population for 30 simulation experiments has no effect on the performance of the algorithm. The average cost function value after the flock algorithm is improved for the first time to update the population is better than that of the genetic algorithm: the ISO is 16.57 and the GA is 33.61 in the scene, the ISO is 16.14 and the GA is 29.88 in the scene II, the ISO is 20.29 and the GA is 31.52 in the scene III, and the improved flock algorithm has the capability of quick global search. The above shows that the improved flock algorithm has more ideal convergence speed, better final solution quality and better stability compared with the genetic algorithm, and the improved flock algorithm has fewer parameters compared with the genetic algorithm.
Analysis results show that the convergence rate, the solution quality and the stability of the algorithm are obviously superior to those of the genetic algorithm, the algorithm parameters are fewer, the algorithm can be rapidly converged in fewer iterations, the algorithm is stabilized to be near the optimal solution, the capability of jumping out of local optimal is achieved, and the problem of multi-machine collaborative global target distribution can be better solved.
The above examples of the present invention are intended to be illustrative only and not limiting of the embodiments of the present invention. Any modification or partial replacement by a person of ordinary skill in the art without departing from the spirit and scope of the present invention is intended to be encompassed within the scope of the claims of the present invention.
Claims (2)
1. The multi-machine collaborative global target distribution method based on the improved flock algorithm is characterized by comprising the following specific steps of:
step 1: determining cost and income to ensure that the overall efficiency of the collaborative combat of multiple unmanned aerial vehicles is maximized; in order to ensure that the overall efficiency of the collaborative combat of multiple unmanned aerial vehicles is maximized, the fuel consumption cost of the unmanned aerial vehicles and the damage cost when the target is attacked are required to be as low as possible, and meanwhile, the income cost when the target is attacked is as high as possible;
step 2: determining constraint conditions to ensure that the overall efficiency of the collaborative combat of multiple unmanned aerial vehicles is maximized; the multi-unmanned aerial vehicle cooperative target allocation is a complex multi-constraint optimization problem, and the constraint conditions considered include: maximum voyage constraint, maximum execution capability constraint, target execution order constraint, and decision variable constraint are expressed as:
t j ＞t i +Δt
wherein unmanned aerial vehicle U i The total flying distance is L i ，Unmanned plane U i Maximum distance to fly, num i Is unmanned plane U i Maximum loading of t i ，t j Respectively are targets T i And target T j The attack time, delta t is the minimum time interval between two targets, delta t > 0;
step 3: determining a multi-machine cooperative global target distribution model, determining a target distribution combat scene and constructing an adaptability function, wherein the adaptability function when the flock algorithm solves an optimal target distribution scheme can be expressed as follows:
fitness＝f+C
wherein f is a target function value corresponding to a sheep, C is a punishment item, and when C is 0, the individual is feasible;
step 4: solving a multi-machine collaborative global target distribution model, selecting a proper population initialization mode, and determining an improvement strategy and an improvement step of a flock algorithm.
2. The multi-machine collaborative global target distribution method based on the improved flock algorithm is characterized in that a target distribution model with multiple constraint conditions is established according to the characteristics of multi-machine collaborative global target distribution, the improved flock algorithm is adopted for solving, the optimal solution is obtained in the multi-machine collaborative global target distribution, the quality is reliable, the capability of jumping out of the local optimal solution is achieved, the convergence speed is high, the stability is high, and the overall efficiency maximization of multi-unmanned plane collaborative combat can be effectively ensured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010354151.2A CN111797966B (en) | 2020-04-28 | 2020-04-28 | Multi-machine collaborative global target distribution method based on improved flock algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010354151.2A CN111797966B (en) | 2020-04-28 | 2020-04-28 | Multi-machine collaborative global target distribution method based on improved flock algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111797966A CN111797966A (en) | 2020-10-20 |
| CN111797966B true CN111797966B (en) | 2024-04-05 |
Family
ID=72806701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010354151.2A Active CN111797966B (en) | 2020-04-28 | 2020-04-28 | Multi-machine collaborative global target distribution method based on improved flock algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111797966B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190041B (en) * | 2021-04-30 | 2022-05-10 | 哈尔滨工业大学 | Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology |
| CN114429290A (en) * | 2022-01-19 | 2022-05-03 | 中国航空工业集团公司西安航空计算技术研究所 | Cooperative combat scene-oriented combat unit scheduling method |
| CN115129082B (en) * | 2022-07-13 | 2023-09-29 | 安徽大学 | Perception detection unmanned aerial vehicle cluster cooperative control method for pollution source search |
| CN118469246B (en) * | 2024-07-09 | 2024-10-01 | 中国人民解放军海军航空大学 | Collaborative optimization method for offshore aircraft task allocation and ammunition configuration |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007080584A2 (en) * | 2006-01-11 | 2007-07-19 | Carmel-Haifa University Economic Corp. Ltd. | Uav decision and control system |
| CN110766254A (en) * | 2018-07-27 | 2020-02-07 | 深圳市白麓嵩天科技有限责任公司 | Multi-unmanned aerial vehicle cooperative task allocation method based on improved genetic algorithm |
-
2020
- 2020-04-28 CN CN202010354151.2A patent/CN111797966B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007080584A2 (en) * | 2006-01-11 | 2007-07-19 | Carmel-Haifa University Economic Corp. Ltd. | Uav decision and control system |
| CN110766254A (en) * | 2018-07-27 | 2020-02-07 | 深圳市白麓嵩天科技有限责任公司 | Multi-unmanned aerial vehicle cooperative task allocation method based on improved genetic algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 文化算法在有人/无人机协同作战目标分配中的应用;王勋;姚佩阳;孙昱;李锴;荣庆;;空军工程大学学报(自然科学版);20160425(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111797966A (en) | 2020-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111797966B (en) | Multi-machine collaborative global target distribution method based on improved flock algorithm | |
| CN112861442B (en) | Multi-machine collaborative air combat planning method and system based on deep reinforcement learning | |
| CN108680063B (en) | A kind of decision-making technique for extensive unmanned plane cluster dynamic confrontation | |
| CN113741508B (en) | Unmanned aerial vehicle task allocation method based on improved wolf pack algorithm | |
| CN110083971B (en) | Self-explosion unmanned aerial vehicle cluster combat force distribution method based on combat deduction | |
| CN111240353B (en) | Unmanned aerial vehicle collaborative air combat decision method based on genetic fuzzy tree | |
| CN110889625A (en) | Task planning method for swarm unmanned aerial vehicle system | |
| CN113485456B (en) | Unmanned aerial vehicle group distributed online self-adaptive task planning method | |
| CN105005820B (en) | Target assignment optimizing method based on particle swarm algorithm of population explosion | |
| CN113031650B (en) | Unmanned aerial vehicle cluster cooperative target distribution design method under uncertain environment | |
| CN112733251B (en) | Collaborative flight path planning method for multiple unmanned aerial vehicles | |
| CN114330115B (en) | Neural network air combat maneuver decision-making method based on particle swarm search | |
| CN112734278A (en) | C2 organization resource dynamic scheduling method for time-sensitive target striking | |
| CN116225049A (en) | Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm | |
| Qingwen et al. | Cooperative jamming resource allocation of UAV swarm based on multi-objective DPSO | |
| CN113887919A (en) | Hybrid-discrete particle swarm algorithm-based multi-unmanned aerial vehicle cooperative task allocation method and system | |
| CN117170405A (en) | Unmanned aerial vehicle task allocation method based on multi-target particle swarm | |
| CN115047907B (en) | Air isomorphic formation command method based on multi-agent PPO algorithm | |
| Wang et al. | An efficient clonal selection algorithm to solve dynamicweapon-target assignment game model in UAV cooperative aerial combat | |
| CN115567353A (en) | Radar networking system-oriented interference multi-beam scheduling and interference power joint optimization method | |
| Zhao et al. | Shuffled frog leaping algorithm with non-dominated sorting for dynamic weapon-target assignment | |
| Gaowei et al. | Using multi-layer coding genetic algorithm to solve time-critical task assignment of heterogeneous UAV teaming | |
| Wang et al. | Task decision-making for UAV swarms based on robustness evaluation | |
| Li et al. | Improved genetic algorithm for multi-agent task allocation with time windows | |
| Li et al. | Research on stealthy UAV path planning based on improved genetic algorithm |
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 |