CN105739304B - A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online - Google Patents

A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online Download PDF

Info

Publication number
CN105739304B
CN105739304B CN201610059833.4A CN201610059833A CN105739304B CN 105739304 B CN105739304 B CN 105739304B CN 201610059833 A CN201610059833 A CN 201610059833A CN 105739304 B CN105739304 B CN 105739304B
Authority
CN
China
Prior art keywords
individual
ucav
opposition
population
current
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
Application number
CN201610059833.4A
Other languages
Chinese (zh)
Other versions
CN105739304A (en
Inventor
刘莉
温永禄
龙腾
王祝
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN201610059833.4A priority Critical patent/CN105739304B/en
Publication of CN105739304A publication Critical patent/CN105739304A/en
Application granted granted Critical
Publication of CN105739304B publication Critical patent/CN105739304B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention proposes a kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA and strikes target online distribution method, it devises first and meets strike target online each dimension numerical value of assignment problem of more UCAV and be not mutually equal the design variable of constraint, assignment problem is struck target online to standard genetic algorithm customization improvement for more UCAV, then antithetical ideas is introduced in genetic manipulation, increases the diversity of population.Method integration antithetical ideas treatment mechanism and genetic algorithm proposed by the present invention, it avoids and is absorbed in locally optimal solution in traditional algorithm solution procedure and takes too long, and the method formed has stronger global convergence ability, solves the problems, such as that prior art efficiency when the more UCAV of solution strike target distribution online is low.

Description

A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution online Method
Technical field
It strikes target distribution method, belongs to online the present invention relates to a kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA In aircraft mission planning field.
Background technology
Towards increasingly sophisticated modern battlefield environment, battlefield task gradually develops to the situation of multiplicity and complexity, single Frame unmanned combat aircraft (unmanned combat aerial vehicle, UCAV) can not almost complete specified operation and appoint Business, more unmanned combat aircraft (UCAV) cooperations have become the inevitable choice of UCAV operational exertions.And more UCAV are hit online Target assignment problem is guarantee and the basis of more UCAV cooperations, and the purpose is to distribute target of attack for UCAV, it is adjoint Modern information technologies and the new and high technology to grow up, are one of the important contents of more UCAV mission plannings technical research, are one Typical torpedo_damaged warship (Weapon-Target Assignment, the abbreviation WTA) problem of kind.How WTA is with not With the weapon allocation of lethality and economic value to different targets is designed, to constitute the Strike system of global optimization. It is that a critical function of modern UCAV cooperations and an important auxiliary of modern automation command system are determined Plan.
The Weapon Target Assignment Problem of more weapon multiple targets is a kind of np complete problem, and research contents is primarily directed to more A to strike target, the command and control system of attacker can effectively distribute firepower, quickly eliminate target, while making attack Side's Least-cost paid and what is struck target injure maximum.WTA is a kind of optimum organization problem of resource, it is therefore an objective to be sought Preferably torpedo_damaged warship scheme, to improve fighting efficiency.In the 1980s, the Patrick A of the Massachusetts Institute of Technology Hosein and Michael Athans (1988) have done more systematic research to general WTA problems.U.S. army's national defence analysis Research institute (Institute for Defense Analysis, DA) has had been devoted to WTA problems since the 1990s Research.Ravindra is in nineteen ninety-five by genetic algorithm (GA) for solving the problems, such as WTA.
Genetic algorithm (Genetic Algorithm, GA) is taught in 1975 by the Holland of Univ Michigan-Ann Arbor USA It is proposed is the biological evolution process for the natural selection and genetic mechanisms for simulating Darwinian evolutionism to search for optimal solution Method.The effective object of genetic algorithm is population, and each of population individual corresponds to a feasible solution of required solution problem. Individual is usually also referred to as chromosome in microcosmic level, and chromosome presses certain forms (spread pattern of such as bit string or symbol) It encodes to indicate a solution.Genetic algorithm by it is all individual apply intersect, variation and selection etc. evolutional operations, make individual and The adaptive value of population is continuously improved, and achievees the purpose that tend to optimal.However genetic algorithm is a kind of random search algorithm, is generated Excellent solution generally requires the calculating time grown very much, some problems very high to requirement of real-time are difficult to meet time-constrain.Separately On the one hand, genetic algorithm is easy precocity when convergence rate is high, and algorithmic statement causes to finally obtain in locally optimal solution It is not globally optimal solution, the quality of solution is made to decline.
Antithetical ideas has long history in philosophy, set theory, politics, sociology and physics, for example, "high" and " low ", " cold " and " heat " etc..And antithetical ideas is never applied always in optimization algorithm, by 2007 years, Rahnamayan was rich Antithetical ideas is applied in differential evolution algorithm (Differential Evolution, DE) by scholar, and is demonstrated based on opposition Two aspects of quality that the improved differential evolution algorithm of thought is conciliate in convergence rate are better than DE.
More UCAV based on antithetical ideas improved adaptive GA-IAGA (Opposition-based GA, OGA) strike target online Antithetical ideas is introduced into genetic algorithm, is improved to genetic algorithm by distribution method, is struck target online point according to more UCAV The characteristics of with problem, design meet the individual UVR exposure mode of Target Assignment constraint.Intersect, after mutation operation passing through, foundation Opposition probability decides whether that foundation adapts to after opposition operation to carrying out opposition operation by the population at individual after cross and variation Degree functional value is selected, and is selected preferably population number of individual and is iterated operation as follow-on population at individual, to carry The ability of searching optimum of high population diversity and algorithm.It is beaten online in the more UCAV of solution based on antithetical ideas Revised genetic algorithum When hitting target assignment problem, there is stronger ability of searching optimum compared to common genetic algorithm, improve algorithm performance.
Invention content
It strikes target online assignment problem the purpose of the present invention is to solve more UCAV in modern weapons mission planning, needle , time-consuming the problems such as low to solution procedure optimal solution convergence, proposes a kind of to solve based on antithetical ideas Revised genetic algorithum More UCAV strike target assignment problem online.This method integrates antithetical ideas treatment mechanism and genetic algorithm, avoids traditional calculation Be absorbed in method solution procedure locally optimal solution and take it is too long, and formed method have stronger global convergence ability, alleviate Efficiency of the prior art in the more UCAV of solution strike target assignment problem online.
In order to better illustrate technical scheme of the present invention, the side for the model and use that the lower mask body introduction present invention establishes Method：
1, more UCAV strike target distribution model online
Assuming that in battlefield, we has the UCAV of m framves isomorphism or isomery, M={ M1,M2,…,Mm, the number of more UCAV is successively For 1~m, need to hit n target T={ T online1,T2,…,Tn, m ＞ n, target designation is followed successively by 1~n, if X={ x1, x2,…,xnIt is torpedo_damaged warship set, variable xiIndicate the UCAV numbers that i-th of target is distributed.
More UCAV distribution that strikes target online is that the overall combat effectiveness formed into columns with entire UCAV is optimal for target, and singly The loss for the attack time, target Damage efficiency and UCAV that attack time that a UCAV strikes target, more UCAV strike target is Evaluate the leading indicator of fighting efficiency.
(1) single UCAV attack times are most short
Single UCAV attack times are most short, i.e., Target Assignment is that each UCAV distributes attack distance nearest strike mesh first Mark, then calculates the attack time needed for the target of attack from the position of current UCAV to selection.
(2) more UCAV attack times are most short
Due to only meeting the shortest index of list UCAV attack times it is possible that multi rack UCAV while distance The close situation of one target, so needing to be thought of as the most short attack time of more UCAV Target Assignments from the overall situation so that total voyage Most short, i.e., total attack time is most short.
(3) target Damage efficiency
Target Damage efficiency Maximum Index is by assessing the target value destroyed when UCAV execution tasks, to guide The optimization of Target Assignment and decision are towards making the maximized direction of fighting efficiency carry out.The index makes UCAV be intended to attack high price It is worth target.
(4) loss of UCAV
UCAV loss objectives are distributed by minimizing the cost guiding target of UCAV targets of attack towards reduction UCAV tasks The direction for injuring cost carries out.The index makes UCAV be intended to fly in safety fairway, keeps the Threat suffered by UCAV minimum.
It is most short for index with more UCAV attack times in the present invention, and assume that all UCAV are identical and each target tool There is equal value, while assuming every frame UCAV at most to attack a target.Weapon mesh is carried out under the above specific condition Mark distribution, detailed process are as follows：
(1) unfriendly target is analyzed according to the information of the information of we and UCAV detections, unfriendly target is compiled Number and threat sequercing.
(2) according to our UCAV quantity, spatial position, firepower value and the quantity of unfriendly target, spatial position, threat Value carries out cost value calculating.
(3) using the shortest UCAV- objective cross of attack time as the result of torpedo_damaged warship.
Weapon Target Assignment Problem is substantially an optimization problem, and the object function established herein is more UCAV attack times Most short, mathematical model is：
Wherein, xiIndicate the UCAV numbers that i-th of target is distributed；t(xi, i) and indicate xthiA UCAV completes i-th of mesh Time needed for mark strike.
2, antithetical ideas
Assuming that x ∈ [a, b], a, b are real number, then the reversed number of x is defined as
The definition of same principle, inverse algorithms can expand to high dimension vector.
Assuming that P=(x1,x2... ..., xD) be D dimension spaces a vector, x here1,x2,……,xD∈ R and xi∈ [ai,bi],Opposite vectorCompletely by each component in vector It is reversed number definition
Using the principle of opposite vector, inverse algorithms are defined as follows.
Assuming that P=(x1,x2... ..., xD) be D dimension spaces a vector, f () is that fitness function is used for calculating The fitness of body.According to reversely several definition,It is vectorial P=(x1,x2... ..., xD) it is reversed Vector.If the fitness of opposite vectorIt is so vectorialIt can replace P.So being adapted in order to allow More preferably individual continues to multiply degree, vectorIt need to be carried out at the same time evaluation with vectorial P.
3, it is based on antithetical ideas Revised genetic algorithum
The coding mode of genetic algorithm generally use is binary coding, however in Weapon Target Assignment Problem, two into System coding can not intuitively indicate that the matching relationship of weaponry target, the present invention are encoded using the decimal system, i.e., each dyeing Body is made of according to target tactic UCAV numbers, and each gene representation distributes to the UCAV numbers of corresponding target (in advance All UCAV are numbered).Such as a chromosome is：[4,5,3,1,2,6,8,7] indicate that the UCAV for being 4 by number is distributed It strikes target to the 1st, the UCAV that number is 7 distributes to the 8th and strikes target.
Obviously, decimally coded representation chromosome can more clearly indicate the relations of distribution of weaponry target, the present invention It is assumed that a frame UCAV can only distribute to a target, so the phenomenon that chromogene repetition occur is infeasible, thus Forbid when initialization population generate Duplication chromosome, and below intersection or mutation operator in using spy Different interleaved mode (PMX), variation mode (DM) is to avoid there is the phenomenon that chromogene repetition in new individual, in this way whole The case where individual UVR exposure is unsatisfactory for constraint is avoided the occurrence of during a iteration.
The invention is realized by the following technical scheme.
A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online, including step is such as Under：
Step 1, initialization of population strikes target the specific of the given design variable of assignment problem online according to more UCAV Constraint, i.e. each dimension numerical value of design variable are not mutually equal, and are assigned all initial population individuals one and are met design variable at random The value of particular constraints, each of initial population individual are that more UCAV strike target a feasible solution of assignment problem online.
Step 2, examine whether current iteration number meets convergence criterion, there are many convergence criterions for algorithm, such as reach most Big iterations, maximum model call number and optimal solution deviation etc..Examine whether current iteration number meets convergence criterion Specific method is：Utilize certain one or more condition in formula (4), (5) and (6), if meeting convergence criterion, current iteration Optimal solution is that current more UCAV strike target the globally optimal solution or suboptimal solution of assignment problem online, exports current optimal result, Iteration terminates；
gen≤gen_max (4)
nfe≤NFE_max (5)
Wherein, gen is current genetic algebra, and gen_max is maximum genetic algebra, and nfe is "current" model call number, NFE_max is maximum model call number, and ε is that convergence error is manually set,WithFor optimum individual under current genetic algebra Adaptive value size.
Step 3, adaptive value that each individual of population is calculated according to fitness function, using roulette selection strategy from working as Selected in preceding population and wait for crossover operation individual, according to crossover probability, using crossover operator (PMX) treat crossover operation individual into Row crossover operation.Then, according to mutation probability, mutation operation is carried out to the individual after intersection using mutation operator (DM), is passed through Intersect, the individual after mutation genetic operation remains able to the particular constraints for meeting design variable.
Step 4, according to the random number between 0~1 randomly generated, judge whether the random number is less than pair of design Vertical probability, step 5 is transferred to if meeting；Otherwise it is transferred to step 7.Opposition probability refers to whether executing the probability of opposition operation.
Step 5, opposition operation is done to the individual in group according to probability, does the individual after opposition operation and still meets design change The particular constraints of amount.To individual in population P=(x1,x2... ..., xD) carry out opposition operation specific method be：Utilize formula (7) the opposition value of each individual dimension is calculated；The particular constraints that the individual after opposition operation still meets design variable are executed, It is not in the individual that each dimension variable value of design variable repeats.
Wherein D is the dimension of individual P, xi∈[ai,bi], wherein xiIndicate the numerical value of each element in individual, ai、biRespectively Indicate the upper and lower boundary of each element.The opposition individual of individual P isThe operation that opposes is a kind of base In the operational criterion of antithetical ideas.
Step 6, the adaptive value of newly-generated opposition individual is calculated according to fitness function, more former population at individual and corresponding Oppose individual adaptive value size, selects the big individual of adaptive value as current population.Opposition individual refers in population Individual, use opposition operation to carry out obtained individual after operation.
Step 7, continue iterative cycles using current population as population of new generation, be transferred to step 2.
The present invention has following advantageous effect：
The present invention realize more UCAV in modern weapons task grouping strike target online assignment problem solution, ensure The Optimality of understanding avoids and is absorbed in locally optimal solution in traditional algorithm solution procedure and takes too long problem, and formed Method has stronger global convergence ability.Antithetical ideas treatment mechanism and genetic algorithm are combined, formd global with processing The design method of optimization ability solves efficiency of the prior art in the more UCAV of solution strike target assignment problem online and asks Topic.
Description of the drawings
Fig. 1 is the algorithm data process flow in specific implementation mode；
Fig. 2 is current location and the relations of distribution of the UCAV and target in embodiment one；
Fig. 3 is the allocation result of the UCAV and target in embodiment two.
Specific implementation mode
Purpose in order to better illustrate the present invention and advantage, below by simulation calculation contrast test, in conjunction with table, attached Figure the present invention will be further described, and by with traditional optimization results contrast, to the present invention comprehensive performance test Card analysis.
The validity of extracting method in order to verify, is respectively adopted and (is abbreviated as based on antithetical ideas Revised genetic algorithum OGA), traditional genetic algorithm (being abbreviated as GA) customization solves more UCAV in modern weapons task grouping and strikes target online point With problem.Two examples that 8 frame UCAV attack 4 targets are selected to be illustrated.Wherein OGA and GA in testing, the rule of population Mould takes 50, and maximum iteration takes 50, crossover probability Pc=0.9, mutation probability Pm=0.05, reversed probability Po=0.9.
Embodiment one
Assuming that a flight formation there are 8 frame UCAV, 4 known targets are attacked, each UCAV can only at most attack a mesh Mark.In the case of well-known theory optimum allocation result, genetic algorithm is used respectively and is calculated based on the improved heredity of antithetical ideas The required averaging model call number of optimum allocation result is calculated in method.All UCAV and the changing coordinates of target such as table 1, Schematic diagram is as shown in Fig. 2, the speed of UCAV is set as 1.The number of unmanned vehicle is followed successively by 1~8 from left to right, the number of target It is followed successively by 1~4.Obviously, weaponry target optimum allocation result is 4 → 1,5 → 2,6 → 3,7 → 4, i.e., optimum individual is (3 45 6)。
The current location of UCAV and target in 1 embodiment one of table
Using based on the more online target assignment problem specific implementation steps of UCAV of antithetical ideas Revised genetic algorithum processing It is as follows：
Step 1, design variable dimension and its value range are determined according to UCAV and destination number, 4 is attacked by 8 frame UCAV Known target, problem dimension 4.Strike target the particular constraints of the given design variable of assignment problem online according to more UCAV, The value that all initial population individuals one meet design variable constraint at random is assigned, each of initial population individual is more UCAV One feasible solution of the online assignment problem that strikes target.Such as [1,2,3,4], [3,5,7,8] and [6,2,5,7] are exactly initial kind Individual in group.
Step 2, examine current iteration number whether meet convergence criterion, if meet convergence criterion (current optimal solution with The error of globally optimal solution is limited to 10e-6), then current iteration optimal solution is that current more UCAV strike target assignment problem online Globally optimal solution or suboptimal solution, iteration terminate.
Step 3, adaptive value that each individual of population is calculated according to fitness function, using roulette selection strategy from working as It is selected in preceding population and waits for crossover operation individual, crossover operation individual will be waited for according to crossover probability PcIt=0.9 and is handed over according to specific It pitches operator (PMX) and carries out crossover operation, then according to mutation probability Pm=0.05 and specific mutation operator (DM) to intersection after Individual carry out mutation operation, the individual by intersecting, after mutation genetic operation remain able to meet design variable it is specific about Beam.
Step 4, according to the random number between 0~1 randomly generated, judge whether the random number is less than pair of design Vertical probability, step 5 is transferred to if meeting；Otherwise it is transferred to step 7.
Step 5, opposition operation is done to the individual in group according to probability, does the individual after opposition operation and still meets design change The particular constraints of amount.
Step 6, the adaptive value that newly-generated opposition individual is calculated according to fitness function, to original seed group and newly-generated individual The larger individual identical with former population number of adaptive value, which is selected, according to adaptive value size is used as current population.
Step 7, continue iterative cycles using current population as population of new generation, be transferred to step 2.
The method of the present invention and customization GA are compared, two methods have carried out 100 solutions, system to above-mentioned model Meter the results are shown in Table shown in 2, including the statistical informations such as average value, minimum value and median of model call number in 100 solutions.
The method of the present invention and customization GA result of calculations in 2 embodiment one of table
It being compared according to result of calculation, the method for the present invention and customization GA can obtain globally optimal solution per suboptimization, still, Averaging model call number needed for the method for the present invention, which is considerably less than, customizes GA, this illustrates that the method for the present invention is better than customization GA.
Embodiment two
In the case of true, to show that the optimum allocation of weaponry target is very time-consuming by theoretical calculation, it is difficult to full The requirement of sufficient battlefield decision real-time sets maximum model calling time so in the case where theoretical optimum allocation result is unknown Number is 2000 times, uses the method for the present invention and customization GA to solve more online target assignment problems of UCAV respectively, compares the meter of the two Calculate result.Assuming that a flight formation there are 8 frame UCAV, 4 known targets are attacked, each UCAV can only at most attack a mesh Mark, UCAV and target current location such as table 3.
The current location of UCAV and target in 3 embodiment two of table
Using based on the more online target assignment problem specific implementation steps of UCAV of antithetical ideas Revised genetic algorithum processing It is as follows：
Step 1, design variable dimension and its value range are determined according to UCAV and destination number, 4 is attacked by 8 frame UCAV Known target, problem dimension 4.Strike target the particular constraints of the given design variable of assignment problem online according to more UCAV, The value that all initial population individuals one meet design variable constraint at random is assigned, each of initial population individual is more UCAV One feasible solution of the online assignment problem that strikes target.
Step 2, examine whether current iteration number meets convergence criterion, (maximum model calls if meeting convergence criterion Number is that 2000), then current iteration optimal solution is that current more UCAV strike target the globally optimal solution or secondary of assignment problem online Excellent solution, iteration terminate.
Step 3, adaptive value that each individual of population is calculated according to fitness function, using roulette selection strategy from working as It is selected in preceding population and waits for crossover operation individual, crossover operation individual will be waited for according to crossover probability and according to specific crossover operator (PMX) crossover operation is carried out, then according to mutation probability and specific mutation operator (DM) to individual into row variation after intersection Operation, the individual by intersecting, after mutation genetic operation remain able to the particular constraints for meeting design variable.
Step 4, according to the random number between 0~1 randomly generated, judge whether the random number is less than pair of design Vertical probability, step 5 is transferred to if meeting；Otherwise it is transferred to step 7.
Step 5, opposition operation is done to the individual in group according to probability, does the individual after opposition operation and still meets design change The particular constraints of amount.
Step 6, the adaptive value that newly-generated opposition individual is calculated according to fitness function, to original seed group and newly-generated individual The larger individual identical with former population number of adaptive value, which is selected, according to adaptive value size is used as current population.
Step 7, continue iterative cycles using current population as population of new generation, be transferred to step 2.
Two methods have carried out 100 solutions to above-mentioned model, and statistical result is shown in Table 4, including 100 solutions In acquire the number of current optimal solution, worst individual and its target function value in solving for 100 times, solve for 100 times in optimal Body and its target function value.During 100 times solve, the method for the present invention is identical with the current optimal solution that customization GA is acquired, such as Fig. 3 institutes Show.And the number that the method for the present invention acquires current optimal solution is 71 times, and customizes GA and only have 60 times and obtain current optimal solution.Together When, the method for the present invention obtains worst individual and is better than customization GA.
The method of the present invention and customization GA result of calculations in 4 embodiment two of table
Above-described specific descriptions have carried out further specifically the purpose, technical solution and advantageous effect of invention It is bright, it should be understood that above is only a specific embodiment of the present invention, for explaining the present invention, being not used to limit this The protection domain of invention, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all It is included within protection scope of the present invention.

Claims (3)

  1. The distribution method 1. a kind of more UCAV of the genetic algorithm based on antithetical ideas strike target online, which is characterized in that including Steps are as follows：
    Step 1, initialization of population, i.e., the particular constraints for the given design variable of distribution that strikes target online according to more UCAV, is assigned The value that all initial population individuals one meet the particular constraints of design variable at random is given, each of initial population individual is more UCAV strikes target a feasible solution in distribution method online；
    Step 2, examine whether current iteration number meets convergence criterion, if meeting convergence criterion, current iteration optimal solution The globally optimal solution or suboptimal solution for the assignment problem that strikes target online for current more UCAV, iteration terminate；
    Step 3, the adaptive value that each individual of population is calculated according to fitness function, using roulette selection strategy from current kind It is selected in group and waits for crossover operation individual, according to crossover probability, treated crossover operation individual using crossover operator PMX and intersected Operation；Then, according to mutation probability, mutation operation is carried out to the individual after intersection using mutation operator DM, by intersecting, making a variation Individual after genetic manipulation remains able to the particular constraints for meeting design variable；
    Step 4, according to the random number between 0~1 randomly generated, it is general to judge whether the random number is less than the opposition designed Rate, opposition probability refer to whether executing the probability of opposition operation, and step 5 is transferred to if meeting；Otherwise it is transferred to step 7；
    Step 5, opposition operation is done to the individual in group according to probability, opposition operation is that a kind of operation based on antithetical ideas is accurate Then, the particular constraints that the individual after opposition operation still meets design variable are done；
    Step 6, the adaptive value of newly-generated opposition individual, more former population at individual and corresponding opposition are calculated according to fitness function The adaptive value size of individual selects the big individual of adaptive value as current population, and opposition individual refers in population Body carries out the individual obtained after operation using opposition operation；
    Step 7, continue iterative cycles using current population as population of new generation, be transferred to step 2.
  2. The distribution side 2. a kind of more UCAV of genetic algorithm based on antithetical ideas according to claim 1 strike target online Method, it is characterised in that：Examining current iteration number whether to meet the specific method of convergence criterion described in step 2 is：Utilize public affairs Certain one or more condition in formula (4), (5) and (6), if meeting convergence criterion, current iteration optimal solution is current more UCAV strikes target the globally optimal solution or suboptimal solution of assignment problem online, exports current optimal result, iteration terminates；
    gen≤gen_max (4)
    nfe≤NFE_max (5)
    Wherein, gen is current genetic algebra, and gen_max is maximum genetic algebra, and nfe is "current" model call number, NFE_ Max is maximum model call number, and ε is that convergence error, f is manually setk *WithIt is adapted to for optimum individual under current genetic algebra It is worth size.
  3. The distribution side 3. a kind of more UCAV of genetic algorithm based on antithetical ideas according to claim 1 strike target online Method, it is characterised in that：To individual in population P=(x described in step 51,x2... ..., xD) carry out opposition operation specific method It is：The opposition value of each individual dimension is calculated using formula (7)；It executes the individual after opposition operation and still meets design variable Particular constraints, that is, be not in the individual that each dimension variable value of design variable repeats；
    Wherein D is the dimension of individual P, xi∈[ai,bi], wherein xiIndicate the numerical value of each element in individual, ai、biIt indicates respectively The upper and lower boundary of each element；The opposition individual of individual P is
CN201610059833.4A 2016-01-28 2016-01-28 A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online Active CN105739304B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610059833.4A CN105739304B (en) 2016-01-28 2016-01-28 A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610059833.4A CN105739304B (en) 2016-01-28 2016-01-28 A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online

Publications (2)

Publication Number Publication Date
CN105739304A CN105739304A (en) 2016-07-06
CN105739304B true CN105739304B (en) 2018-09-25

Family

ID=56247850

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610059833.4A Active CN105739304B (en) 2016-01-28 2016-01-28 A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online

Country Status (1)

Country Link
CN (1) CN105739304B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107330560B (en) * 2017-07-04 2020-05-12 北京理工大学 Heterogeneous aircraft multi-task cooperative allocation method considering time sequence constraint
CN108416441B (en) * 2018-05-10 2020-05-19 华中科技大学 Ship-to-shore impact firepower distribution method based on genetic algorithm
CN110096822B (en) * 2019-05-08 2020-09-25 北京理工大学 Multi-platform collaborative dynamic task allocation method under uncertain conditions
CN110377048B (en) * 2019-06-26 2023-02-03 沈阳航空航天大学 Unmanned aerial vehicle group defense deployment method based on genetic algorithm
CN114330715A (en) * 2021-12-27 2022-04-12 哈尔滨工业大学 Intelligent ammunition co-evolution task allocation method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101286071A (en) * 2008-04-24 2008-10-15 北京航空航天大学 Multiple no-manned plane three-dimensional formation reconfiguration method based on particle swarm optimization and genetic algorithm
CN103744290A (en) * 2013-12-30 2014-04-23 合肥工业大学 Hierarchical target allocation method for multiple unmanned aerial vehicle formations
CN103995539A (en) * 2014-05-15 2014-08-20 北京航空航天大学 Unmanned aerial vehicle autonomous formation evaluation index and MPC formation control method
CN104700165A (en) * 2015-03-27 2015-06-10 合肥工业大学 Multi-UAV (unmanned aerial vehicle) helicopter and warship cooperating path planning method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101286071A (en) * 2008-04-24 2008-10-15 北京航空航天大学 Multiple no-manned plane three-dimensional formation reconfiguration method based on particle swarm optimization and genetic algorithm
CN103744290A (en) * 2013-12-30 2014-04-23 合肥工业大学 Hierarchical target allocation method for multiple unmanned aerial vehicle formations
CN103995539A (en) * 2014-05-15 2014-08-20 北京航空航天大学 Unmanned aerial vehicle autonomous formation evaluation index and MPC formation control method
CN104700165A (en) * 2015-03-27 2015-06-10 合肥工业大学 Multi-UAV (unmanned aerial vehicle) helicopter and warship cooperating path planning method

Also Published As

Publication number Publication date
CN105739304A (en) 2016-07-06

Similar Documents

Publication Publication Date Title
CN105739304B (en) A kind of more UCAV based on antithetical ideas improved adaptive GA-IAGA strike target distribution method online
CN109444832B (en) Group intelligent interference decision method based on multiple interference effect values
CN106406346B (en) A kind of multiple no-manned plane collaboration rapid Cover search path planning method
CN105005820B (en) Target assignment optimizing method based on particle swarm algorithm of population explosion
CN107515618B (en) Heterogeneous unmanned aerial vehicle cooperative task allocation method considering time window
CN107330560B (en) Heterogeneous aircraft multi-task cooperative allocation method considering time sequence constraint
CN106779210B (en) Algorithm of Firepower Allocation based on ant group algorithm
CN107832885B (en) Ship formation fire power distribution method based on self-adaptive migration strategy BBO algorithm
CN110232518A (en) A kind of intimidation estimating method based on three decisions
CN108549402A (en) Unmanned aerial vehicle group method for allocating tasks based on quantum crow group hunting mechanism
CN110134007A (en) Multiple no-manned plane cooperates with target assignment method
Yu et al. Improve enhanced fireworks algorithm with differential mutation
CN112070418B (en) Weapon target distribution method of multi-target whale optimization algorithm
CN113821973B (en) Self-adaptive optimization method for multi-stage weapon target allocation
CN107909152A (en) A kind of variable differential evolution algorithm of crossover probability factor
CN114330715A (en) Intelligent ammunition co-evolution task allocation method
CN116702633B (en) Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization
CN113109770A (en) Interference resource allocation method and system
Zeng et al. Multi-objective cooperative salvo attack against group target
CN109101721A (en) Based on the multiple no-manned plane method for allocating tasks of Interval Intuitionistic Fuzzy information under uncertain environment
CN115222271A (en) Weapon target distribution method based on neural network
CN109299491B (en) Meta-model modeling method based on dynamic influence graph strategy and using method
CN109325598A (en) A kind of combination Chaos Genetic Algorithm of resourceoriented distribution
Pan et al. Radar jamming strategy allocation algorithm based on improved chaos genetic algorithm
Xu et al. MOQPSO‐D/S for Air and Missile Defense WTA Problem under Uncertainty

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant