CN110766125A - Multi-target weapon-target allocation method based on artificial fish swarm algorithm - Google Patents
Multi-target weapon-target allocation method based on artificial fish swarm algorithm Download PDFInfo
- Publication number
- CN110766125A CN110766125A CN201810851662.8A CN201810851662A CN110766125A CN 110766125 A CN110766125 A CN 110766125A CN 201810851662 A CN201810851662 A CN 201810851662A CN 110766125 A CN110766125 A CN 110766125A
- Authority
- CN
- China
- Prior art keywords
- target
- population
- weapon
- artificial fish
- fish swarm
- 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
Images
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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A multi-target weapon-target allocation method based on an artificial fish swarm algorithm aims to overcome the defect that the solution aiming at the WTA problem in the prior art is greatly deviated from the real Pareto front edge. The method comprises the following steps: firstly, randomly generating an initial population, calculating a non-dominated solution set of the initial population, and sequencing according to the crowding distance to obtain a global optimal solution of the initial population; then, according to the clustering behavior of the artificial fish swarm algorithm, other individual fish in the fish swarm approach to the optimal solution to obtain a new swarm and a previous non-dominated solution set, and a new non-dominated solution set is calculated; and finally, performing cross variation on the clustered population to increase the population diversity, merging the clustered population with the previous non-dominated solution set again, and performing multiple iterations to obtain the final Pareto front. The method is mainly used in the field of fire fighting decision making, is closer to the real Pareto frontier compared with the prior art, has small dependence on parameters, and has great application value in multi-target weapon-target distribution.
Description
Technical Field
The invention belongs to the field of fire fighting decision making, and particularly relates to a multi-target weapon-target distribution method based on an artificial fish swarm algorithm.
Background
The problem of Weapon Target Assignment (WTA) is an important subject of firepower application, and the core problem of WTA is how to assign weapons with different killing principles and economic values to different targets to shoot to form an integrally optimized firepower striking system.
Some current methods focus on adopting a single-target planning scheme, generally, an objective function with the maximum combat effectiveness, namely the maximum damage effectiveness to an enemy target, is optimized, and then a linear programming method and intelligent algorithms such as a genetic algorithm, an ant colony algorithm, a tabu search algorithm, a particle swarm optimization method and the like are adopted for optimization solution. However, in an actual combat environment, under the condition of gradually increasing fire power consumption, the improvement on the damage efficiency of enemies is not obvious, and fire power resources are wasted, so that an objective function of 'least ammunition consumption' needs to be newly added while the traditional WTA problem pursues the maximum combat efficiency, and a single-objective optimization problem is converted into multi-objective optimization. At present, some algorithms are applied to a multi-objective optimization problem, but the algorithms have the defects of limited population diversity and large deviation from the real Pareto frontier to different degrees, the simulation effect is poor, and the practical value in future operations is not high.
Disclosure of Invention
The invention aims to provide a method for more accurately obtaining a Pareto front edge in a multi-target WTA, which can avoid the problem that the Pareto front edge is larger than the real Pareto front edge. By using the unique characteristics of the artificial fish swarm algorithm clustering behavior, dominant solutions move towards non-dominant solutions to generate new non-dominant solution sets, then the diversity of the population is improved by using the cross variation of the genetic algorithm, finally the population is combined, the non-dominant solution sets are obtained again, the final Pareto frontier is obtained through multiple iterations, and the effectiveness and the rationality of the method can be found from the final simulation effect.
The core technical content of the invention is to provide a method for more accurately obtaining Pareto frontier in multi-target WTA, firstly randomly generating an initial population, calculating a non-dominated solution set of the initial population, and obtaining a global optimal solution of the initial population by taking a crowding distance as a basis; then executing the clustering behavior of the artificial fish swarm algorithm to make other individual fish in the fish swarm randomly move towards the optimal solution, obtaining a new fish swarm, then merging the new fish swarm with the previous non-dominated solution set and calculating a new non-dominated solution set; and finally, carrying out cross variation on the fish groups to increase the population diversity, combining again to obtain a non-dominated solution set, and carrying out multiple iterations to obtain the final Pareto front.
The invention is realized in such a way that: a multi-target weapon-target distribution method based on artificial fish swarm algorithm comprises the following steps:
the method comprises the following steps: establishing an objective function and a constraint condition of WTA multi-objective optimization;
step two: encoding the fish population to form an initialized population X0；
Step three: calculating non-dominated solution set X of initial population after population initializationpAnd selecting the individual with the largest crowding degree distance as the global optimal solution X of the initial population according to the crowding distance sorting result1；
Step four: obtaining the global optimal solution X of the initial population1Then, artificial fish swarm clustering behavior is executed, so that other individual fishes in the fish swarm are enabled to be in a global optimal solution X1Close proximity, cluster andpost-formation of a new population Xn；
Step five: combining the clusters in the fourth step to form a new population XnAnd non-dominant solution set X of the initial population in step threepCalculating to obtain a new non-dominated solution set;
step six: and judging whether a termination condition is reached, if so, outputting an optimal weapon-target distribution scheme simulated by the artificial fish individual, and otherwise, skipping to execute the third step.
Preferably, steps S51 and S52 are added between step five and step six;
s51 is: forming a new parent population X after the clustering in the fourth stepnPerforming basic evolutionary operation of genetic algorithm, including variation and crossing, to obtain filial generation population Xn *；
S52 is: merging the child population X in step S51n *And the parent population X in step fournAnd updating the merged Pareto optimal solution set.
The invention has the beneficial effects that: a more realistic Pareto front can be obtained. The method has small dependence on parameters and high operation speed, and achieves the expected purpose. The method faces to the information war of the future change, reduces the dependence on external parameters, can start operation and make firepower distribution decision more quickly, and has certain practicability.
The invention is further described with reference to the following figures and detailed description.
Drawings
FIG. 1 is a coding structure of an artificial fish school;
FIG. 2 is a schematic diagram of the crowding distance;
FIG. 3 is a flow chart of the present invention;
FIG. 4 is a Pareto frontier comparison of the present invention with NSGA-II and MOPSO algorithms;
FIG. 5 is a running time box plot of the present invention with the NSGA-II and MOPSO algorithms.
Detailed Description
1. Multi-objective optimization model
The WTA multi-objective optimization aims to reduce the cost consumption of ammunition as much as possible when the damage efficiency reaches a certain value. Assuming that there are M weapons and N targets, one target can be attacked by multiple weapons simultaneously, the mathematical model is optimized as follows.
In the formula, ωjIs the threat level, p, of the target (j ═ 1,2, L N) jijRepresents the damage probability of the weapon i (i ═ 1,2, L M) to the target j (j ═ 1,2, L N) [ x ═ 2-ij]M×N(xij0, 1) is a decision matrix indicating whether or not a weapon i is assigned to a target j, and if a weapon i is assigned to a target j, x ij1 is ═ 1; otherwise, xij＝0。ciRepresenting the cost of consumption of the weapon i when used.
2. Artificial fish shoal encoding and decoding
Because the artificial fish swarm algorithm researches a numerical optimization problem with continuous variables and cannot be directly used for solving the nonlinear combined optimization decision problem of WTA multi-target planning, vectors on a solution space need to be encoded into a form similar to genes in a genetic algorithm. The adopted coding mode not only needs to represent the solution of the problem, but also needs to satisfy the constraint conditions set in the WTA multi-target planning model as much as possible, and therefore the coding structure shown in fig. 1 is considered and designed, and M is 5, and N is 6 as an example.
The invention comprises a population solving non-domination solution set, which mainly comprises the following contents: solving each individual in the population according to a set objective function, recording a fitness value, comparing each solution in the population with all other solutions in the population, judging whether the solution is inferior to any other solution in the population, and recording all non-inferior solutions, wherein a formed set is a non-inferior solution set of the population.
Pareto optimal solution set and crowd distance calculation
In the minimum multi-objective optimization problem, the Pareto optimal solution is defined as that for the variable X, if and only if no other variable X exists in the feasible region of the design variable*Satisfy f without violating the constrainti(X)≤fi(X*) At least one i is present such that fi(X)<fi(X*) If true, it is called variable X*Is a non-dominant solution, namely a Pareto optimal solution. The Pareto optimal solutions are not unique, and a plurality of Pareto optimal solutions form a Pareto optimal solution set. In the invention, each solution in the population is compared with all other solutions in the population to see whether the solution is inferior to any other solution in the population, and all non-dominant solutions are recorded, and the formed set is the Pareto optimal solution set.
Calculating crowding distance by taking the horizontal distance of two individuals (i-1) and (i +1) along two sides of each objective function of the individual i and the number d in order to estimate the density of the individuals around the individual i in the populationiThe estimated value of the sum of the M distances is referred to as a congestion distance. As shown in fig. 2, if three individuals adjacent to the same non-inferior layer are i, (i-1) and (i +1), respectively, the crowding distance of the ith individual is di＝dx+dy. The larger the value, the more dispersed the solution set, the smaller the density, and the better the diversity. Therefore, when the global optimal solution is selected, the individual with the largest crowding distance is selected as the global optimal solution.
4. Artificial herd clustering behavior
The invention comprises a clustering behavior part of artificial fish schools, which mainly comprises the following contents: and judging whether the optimal solution is in the visual field range or not through the set visual field range of the fish school, if so, moving one step to the optimal solution, and otherwise, moving one step to the center of the fish school in the visual field range.
In the basic artificial fish swarm algorithm, the state of an artificial fish individual is represented by a vector X ═ X1,x2,L xn) Indicating, current artificial fish XiIs defined as: n ═ Xj|di,j<Visum }, wherein di,j＝||Xi-XiL, Visual represents the perception range of artificial fish, i, j ═ 1,2, L n; if the perception range is set to be large, the algorithm has strong global search capability and can quickly converge, but in the later stage of convergence, the phenomenon that the artificial fish oscillates back and forth near the optimal value can occur; if the sensing range is set to be smaller, the convergence speed of the algorithm is low, and although the convergence accuracy can be improved, under the condition of a multi-peak extreme value, the algorithm is easy to fall into a local extreme value and is difficult to obtain a real optimal solution. Therefore, a linear inertia weight reduction strategy in the particle swarm optimization is introduced, and a new self-adaptive sensing range process is formed. The improved perception range is as follows.
VisualK+1＝floor(VisualK*(1-K/Max_gen)) (4)
In the formula, K is the current iteration number, and Max _ gen is the maximum iteration number. The behavior changes the perception range under different iteration times, so that the perception range is reduced along with the increase of the iteration times, and the operation precision and the operation speed of the algorithm are improved.
When the artificial fishes perform the clustering behavior, each artificial fish is gathered to the center position of the whole fish swarm, the behavior integrates the state information of the whole fish swarm, the information transmission speed is accelerated, and the convergence speed and the running speed of the algorithm are improved.
Setting the central position of the whole artificial fish school as XcThe scale of the whole artificial fish school is n, if Ycn<δYiWhere Y is the food concentration at X and δ represents the crowding factor. Then pressing formula (5) to fish school center position XcOne step forward.
By combining the multi-objective optimization problem, the invention judges whether the optimal solution is in the visual field range or not through the set visual field range of the fish school, if so, the optimal solution is moved by one step, otherwise, the optimal solution is moved by one step to the center of the fish school in the visual field range.
5. Evolution operations
Performing basic evolution operation of a genetic algorithm, mainly comprising variation and intersection, and increasing population diversity so as to more accurately obtain a Pareto frontier; and then merging the parent population and the child population, and updating the Pareto optimal solution set.
The overall algorithm flow chart is shown in fig. 3.
Assuming three weapons platforms, a total of 12 weapons platforms, a number of targets of 4, a target threat level vector of [0.15, 0.36, 0.18, 0.31], a damage probability matrix P and a ammunition cost matrix C are defined as follows:
C＝[0.62 0.63 0.69 0.80 0.72 0.90 0.96 0.680.72 0.65 0.66 0.65]
other simulation parameters are shown in table 1.
TABLE 1 parameter settings
From fig. 4 and fig. 5, we can find that the method can obtain a more accurate Pareto front in a short time.
Finally, it should be noted that the above examples are only intended to describe the technical solutions of the present invention and not to limit the technical methods, the present invention can be extended in application to other modifications, variations, applications and embodiments, and therefore all such modifications, variations, applications, embodiments are considered to be within the spirit and teaching scope of the present invention.
Claims (6)
1. A multi-target weapon-target distribution method based on artificial fish swarm algorithm is characterized by comprising the following steps:
the method comprises the following steps: establishing an objective function and a constraint condition of WTA multi-objective optimization;
step two: encoding the fish population to form an initialized population X0；
Step three: initialization seedComputing non-dominated solution set X of initial population after clusteringpAnd selecting the individual with the largest crowding degree distance as the global optimal solution X of the initial population according to the crowding distance sorting result1；
Step four: obtaining the global optimal solution X of the initial population1Then, artificial fish swarm clustering behavior is executed, so that other individual fishes in the fish swarm are enabled to be in a global optimal solution X1After approaching and clustering, a new population X is formedn；
Step five: combining the clusters in the fourth step to form a new population XnAnd a non-dominance solution set Xp of the initial population in the third step, and calculating to obtain a new non-dominance solution set;
step six: and judging whether a termination condition is reached, if so, outputting an optimal weapon-target distribution scheme simulated by the artificial fish individual, and otherwise, jumping to the third step.
2. The method for multi-target weapon-target assignment based on artificial fish swarm algorithm as claimed in claim 1, wherein: steps S51 and S52 are added between step five and step six;
s51 is: forming a new parent population X after the clustering in the fourth stepnPerforming basic evolutionary operation of genetic algorithm, including variation and crossing, to obtain filial generation population Xn *；
S52 is: merging the child population X in step S51n *And the parent population X in step fournAnd updating the merged Pareto optimal solution set.
3. The multi-target weapon-target allocation method based on artificial fish swarm algorithm according to claim 1 or 2, characterized in that: the WTA multiple targets in the step one comprise two targets of maximum operational efficiency and minimum ammunition consumption, and the mathematical model of the WTA multiple targets is
Constraint conditions
Wherein M is the number of weapons, N is the number of targets, ωjIs the threat level, p, of the target (j ═ 1,2, L N) jijRepresents the damage probability of the weapon i (i ═ 1,2, L M) to the target j (j ═ 1,2, L N) [ x ═ 2-ij]M×N(xij0, 1) is a decision matrix indicating whether or not a weapon i is assigned to a target j, and if a weapon i is assigned to a target j, xij1 is ═ 1; otherwise, xij＝0；ciRepresenting the cost of consumption of the weapon i when used.
4. The multi-target weapon-target allocation method based on artificial fish swarm algorithm according to claim 1 or 2, characterized in that: initializing population X in step two0The parameters include the number of the colonies, the perceived distance of the artificial fish, the crowdedness factor and the number of iterations.
5. The method for multi-target weapon-target assignment based on artificial fish swarm algorithm as claimed in claim 4, wherein: and the termination condition in the sixth step is iteration times.
6. The method for multi-target weapon-target assignment based on artificial fish swarm algorithm as claimed in claim 4, wherein: the perceived distance of an artificial fish is constrained by the following formula:
VisualK+1＝floor(VisualK*(1-K/Max_gen))
wherein Visual is the sensing distance of the artificial fish, K is the current iteration number, and Max _ gen is the maximum iteration number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810851662.8A CN110766125A (en) | 2018-07-26 | 2018-07-26 | Multi-target weapon-target allocation method based on artificial fish swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810851662.8A CN110766125A (en) | 2018-07-26 | 2018-07-26 | Multi-target weapon-target allocation method based on artificial fish swarm algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110766125A true CN110766125A (en) | 2020-02-07 |
Family
ID=69328866
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810851662.8A Pending CN110766125A (en) | 2018-07-26 | 2018-07-26 | Multi-target weapon-target allocation method based on artificial fish swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110766125A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112004208A (en) * | 2020-06-07 | 2020-11-27 | 西南民族大学 | Artificial fish school clustering method, storage medium and device in Internet of vehicles |
| CN112070418A (en) * | 2020-09-21 | 2020-12-11 | 大连大学 | Weapon target allocation method of multi-target whale optimization algorithm |
| CN112163763A (en) * | 2020-09-25 | 2021-01-01 | 大连大学 | Weapon target allocation solving method based on improved multi-target HQPsOGA algorithm |
| US20220108186A1 (en) * | 2020-10-02 | 2022-04-07 | Francisco Daniel Filip Duarte | Niche Ranking Method |
-
2018
- 2018-07-26 CN CN201810851662.8A patent/CN110766125A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112004208A (en) * | 2020-06-07 | 2020-11-27 | 西南民族大学 | Artificial fish school clustering method, storage medium and device in Internet of vehicles |
| CN112070418A (en) * | 2020-09-21 | 2020-12-11 | 大连大学 | Weapon target allocation method of multi-target whale optimization algorithm |
| CN112163763A (en) * | 2020-09-25 | 2021-01-01 | 大连大学 | Weapon target allocation solving method based on improved multi-target HQPsOGA algorithm |
| CN112163763B (en) * | 2020-09-25 | 2023-09-01 | 大连大学 | Weapon target distribution solving method based on improved multi-target HQPSOGA algorithm |
| US20220108186A1 (en) * | 2020-10-02 | 2022-04-07 | Francisco Daniel Filip Duarte | Niche Ranking Method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108919641B (en) | Unmanned aerial vehicle flight path planning method based on improved goblet sea squirt algorithm | |
| CN110766125A (en) | Multi-target weapon-target allocation method based on artificial fish swarm algorithm | |
| CN107330902B (en) | Chaotic genetic BP neural network image segmentation method based on Arnold transformation | |
| CN111224966B (en) | Optimal defense strategy selection method based on evolutionary network game | |
| CN109670655B (en) | Multi-target particle swarm optimization scheduling method for electric power system | |
| CN111709511A (en) | Harris eagle optimization algorithm based on random unscented Sigma point variation | |
| CN115470704B (en) | Dynamic multi-objective optimization method, device, equipment and computer readable medium | |
| CN113836803A (en) | Improved wolf colony algorithm-based unmanned aerial vehicle cluster resource scheduling method | |
| CN113031650B (en) | Unmanned aerial vehicle cluster cooperative target distribution design method under uncertain environment | |
| CN116702633B (en) | Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization | |
| CN112070418A (en) | Weapon target allocation method of multi-target whale optimization algorithm | |
| Liu et al. | Dynamic multi-swarm differential learning Harris Hawks Optimizer and its application to optimal dispatch problem of cascade hydropower stations | |
| CN116341605A (en) | Grey wolf algorithm hybrid optimization method based on reverse learning strategy | |
| CN115933693A (en) | Robot path planning method based on adaptive chaotic particle swarm algorithm | |
| CN114995460A (en) | Robot path planning method based on self-adaptive improved ant colony algorithm | |
| CN111797966B (en) | Multi-machine collaborative global target distribution method based on improved flock algorithm | |
| CN117093019A (en) | Unmanned aerial vehicle power line path planning method for improving wolf algorithm | |
| CN113141272B (en) | Network security situation analysis method based on iteration optimization RBF neural network | |
| Li et al. | Improved artificial fish swarm algorithm approach to robot path planning problems | |
| Ruining et al. | Improved genetic algorithm for weapon target assignment problem | |
| CN117590757B (en) | Multi-unmanned aerial vehicle cooperative task allocation method based on Gaussian distribution sea-gull optimization algorithm | |
| Gerdelan et al. | A genetic‐fuzzy system for optimising agent steering | |
| CN111382896A (en) | WTA target optimization method of adaptive chaotic parallel clonal selection algorithm | |
| Meng et al. | An improved predator-prey particle swarm optimization algorithm for Nash equilibrium solution | |
| CN114510876B (en) | Multi-platform weapon target distribution method based on symbiotic search biological geography optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200207 |