CN116702633B - Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization - Google Patents

Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization Download PDF

Info

Publication number
CN116702633B
CN116702633B CN202310990039.1A CN202310990039A CN116702633B CN 116702633 B CN116702633 B CN 116702633B CN 202310990039 A CN202310990039 A CN 202310990039A CN 116702633 B CN116702633 B CN 116702633B
Authority
CN
China
Prior art keywords
task
reliability
objective
missile
target
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
CN202310990039.1A
Other languages
Chinese (zh)
Other versions
CN116702633A (en
Inventor
伊枭剑
余徽阳
许涛
陈俊男
王晓光
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 CN202310990039.1A priority Critical patent/CN116702633B/en
Publication of CN116702633A publication Critical patent/CN116702633A/en
Application granted granted Critical
Publication of CN116702633B publication Critical patent/CN116702633B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • 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
    • 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

Abstract

The invention discloses a heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization, which belongs to the technical field of heterogeneous warhead task planning, and comprises the following steps: taking the maximum efficiency-cost ratio of the martial arts distribution and the maximum task reliability of the martial arts distribution as optimization targets, and constructing a multi-target dynamic optimization model of the heterogeneous warhead group; carrying out iterative solution on the multi-objective dynamic optimization model based on a dynamic self-adaptive MOEA/D-AM2M method, and stopping iteration until a dynamic algorithm preset termination condition is met; in the iterative solving process, when the environment changes, starting a dynamic response mechanism in a dynamic self-adaptive MOEA/D-AM2M method to obtain a pareto front solution set of a multi-objective optimization problem in a new environment; the method not only can effectively improve the task reliability of weapon target distribution, but also can improve the efficiency of weapon target distribution, and is suitable for scenes with extremely high requirements on task reliability.

Description

Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization
Technical Field
The invention relates to the technical field of heterogeneous warhead task planning, in particular to a heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization.
Background
The core of heterogeneous warhead task planning is that weapons with different damage capacities and economic values are distributed to different targets under the constraint of certain conditions, and the fight resource consumption is reduced as much as possible on the premise of meeting the damage requirements of the targets, so that the weapon utilization rate is maximized, and the whole fire striking system is optimized.
In modern warfare, battlefield elements are various, the battlefield environment is changed suddenly, and in order to improve the overall battlefield efficiency and reliability of the flight weapon system and realize higher-level multi-flight weapon collaborative combat, tasks are required to be comprehensively planned, allocated and coordinated according to dynamic characteristics existing in the battlefield. Generally, for tactical purposes, this process of orchestration planning is referred to as "mission planning".
The related art does not consider the case of dynamic changes in the battlefield when solving the weapon target allocation problem. How to quickly formulate a new weapon distribution scheme after the environmental change and improve the task reliability of the weapon distribution scheme, no effective technical solution is proposed at present.
Therefore, how to quickly formulate a new weapon distribution plan after environmental change based on the situation of battlefield dynamic change and improve the task reliability of the weapon distribution plan is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
In view of the above problems, the present invention provides a method for planning task reliability of a heterogeneous warhead based on multi-objective dynamic optimization, so as to at least solve some of the technical problems mentioned in the background art.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization comprises the following steps:
s1, constructing a multi-objective dynamic optimization model of a heterogeneous warhead by taking the maximum efficiency-cost ratio of the martial arts distribution and the maximum task reliability of the martial arts distribution as optimization targets;
s2, carrying out iterative solution on the multi-objective dynamic optimization model based on a dynamic self-adaptive MOEA/D-AM2M method, and stopping iteration until a dynamic algorithm preset termination condition is met;
in the iterative solving process, when the environment changes, a dynamic response mechanism in the dynamic self-adaptive MOEA/D-AM2M method is started, and a pareto front solution set of the multi-objective optimization problem in the new environment is obtained.
Further, the multi-objective dynamic optimization model is expressed as:
wherein maximum represents the maximum target value; f (f) 1 Representing an objective function I; f (f) 2 Representing a second objective function; I. II, III, IV, V and VI each represent constraints of a multi-objective dynamic optimization model;
objective function f 1 Representing the cost effectiveness of maximizing the martial arts distribution, expressed as:
objective function two f 2 Representing the task reliability of maximizing the assignment of the martial arts, expressed as:
wherein, the liquid crystal display device comprises a liquid crystal display device,representation->Time objective->Is of value (1); />Representation->Time of day completion goal->Task reliability of (2); />Representation->Number of time wars; />Representation->Moment missile->Cost of (2); />Representation->Whether the ith missile is assigned to the target mission at the moment->Is->Integer decision variables->Indicating allocation of->Indicating no allocation; />Representation->Time zone->Task reliability of (2); />Representing a collection of regions.
Further, the constraint conditions of the multi-objective dynamic optimization model are specifically expressed as follows:
constraint I:means ensuring that the number of missile assignments does not exceed a maximum available number;
constraint II:the method ensures that each missile is allocated with at most one task;
constraint III:indicating that the assigned missile is ensured to be effectively fault-free;
constraint IV:indicating that the assigned missile cannot exceed the range of the missile;
constraint V:representing that the regional task reliability criteria are ensured to be met;
constraint VI:representing->Integer variable constraints;
wherein, the liquid crystal display device comprises a liquid crystal display device,a target at time t; />Representing the performance state of the missile at time t +.>Indicating normal performance of the missile, and indicating->Indicating that the missile performance has faults; />Missile->To task->Is a distance of (2); />Missile->Is a navigation course of (3).
Further, the calculation process of the task reliability of the multi-objective dynamic optimization model on any heterogeneous warhead objective allocation scheme comprises the following steps:
calculating the non-interception probability of the ith missile according to the number of missiles launched to the target area at the same time and the number of defensive missiles in the target area;
according to the non-interception probability, the reliability of the ith missile under the electromagnetic anti-severe environment and the task success probability of the ith missile on a target task are combined, and the task reliability of the ith missile is calculated and is recorded as a first task reliability;
calculating the task reliability of the ith missile for completing the target task in the target area according to the first task reliability and combining the number of the wars at the moment t and a 0-1 integer decision variable of whether the ith missile is distributed to the target task at the moment t, and recording the task reliability as a second task reliability;
all tasks with a preset number in the target area are completed as task success criteria; calculating the task reliability of the target area based on the task success criterion and according to the second task reliability;
and according to the task reliability of the target area, calculating the task reliability of the heterogeneous warhead by adopting the objective function II.
Further, the dynamic adaptive MOEA/D-AM2M method comprises a static adaptive multi-objective optimization algorithm, an environment detection mechanism and a dynamic response mechanism.
Further, the step S2 specifically includes:
solving the multi-objective dynamic optimization model through the static self-adaptive multi-objective optimization algorithm to obtain a pareto front solution set of the multi-objective optimization problem for a system to make decisions;
if the dynamic algorithm does not reach the preset termination condition, detecting external environment data by adopting an environment detection mechanism, and judging whether decision variables and model parameters change or not;
if the decision variable and the model parameter change, starting the dynamic response mechanism, and updating the parameters of the multi-objective dynamic optimization model in the iteration;
repeating the steps until the preset termination condition of the dynamic algorithm is met, and stopping iteration.
Further, the external environment data includes: group reliability status data, task situation information, task result situation information, and heterogeneous group situation information.
Further, if the decision variable is not changed, solving the multi-objective dynamic optimization model in the iteration through the static self-adaptive multi-objective optimization algorithm again.
Further, the updating the parameters of the multi-objective dynamic optimization model in the iteration specifically includes:
weapon increase, target increase: inheriting the history strategy, distributing targets for newly added weapons, and randomly selecting part of weapons to be distributed to new tasks;
weapon decrease, target increase: inheritance of historical policies, random selection of portions of weapons for allocation to new targets
Weapon increase, target decrease: inheriting the history strategy, and randomly distributing targets for newly added weapons;
weapon reduction, target reduction: inherit the history policy.
Further, in the heterogeneous warhead task reliability planning process, constraint conditions are processed by using the repair factors and the punishment functions.
Compared with the prior art, the invention discloses a heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization, which has the following beneficial effects:
1. the invention solves the problem of task reliability of the allocation scheme in the weapon target allocation problem based on the improved dynamic self-adaptive MOEA/D-AM2M algorithm, can effectively improve the task reliability of weapon target allocation, can improve the efficiency of weapon target allocation, and is suitable for scenes with extremely high requirements on task reliability.
2. The invention can automatically detect whether the environment changes under the battlefield of instantaneous change through the environment detection mechanism and the dynamic response mechanism, and can make an effective new scheme after the environment changes, thereby improving the reaction capability of the weapon platform.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of a heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization provided by an embodiment of the invention.
Fig. 2 is a schematic diagram of a serial model framework for task reliability according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a solution flow of a static adaptive multi-objective optimization algorithm based on improvement according to an embodiment of the present invention.
Fig. 4 is a flow chart of a dynamic response mechanism according to an embodiment of the present invention.
FIG. 5 is a schematic diagram of a weapon and target encoding provided by an embodiment of the present invention.
Fig. 6 is a schematic diagram of pareto front solution set according to an embodiment of the present invention.
Fig. 7 is a schematic diagram of comparing indexes of a multi-objective dynamic optimization algorithm and a static algorithm according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the embodiment of the invention discloses a heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization, which comprises the following steps:
s1, constructing a multi-objective dynamic optimization model of a heterogeneous warhead by taking the maximum efficiency-cost ratio of the martial arts distribution and the maximum task reliability of the martial arts distribution as optimization targets;
s2, carrying out iterative solution on the multi-objective dynamic optimization model based on a dynamic self-adaptive MOEA/D-AM2M method, and stopping iteration until a dynamic algorithm preset termination condition is met; in the iterative solving process, when the environment changes, a dynamic response mechanism in a dynamic self-adaptive MOEA/D-AM2M method is started, and a pareto front solution set of the multi-objective optimization problem in a new environment is obtained.
The following details the above steps, respectively:
in the above-mentioned step S1,
the multi-objective dynamic optimization model is expressed as:
wherein maximum represents the maximum target value; f (f) 1 Representing an objective function I; f (f) 2 Representing a second objective function; I. II, III, IV, V and VI each represent constraints of a multi-objective dynamic optimization model;
objective function f 1 Representing the cost effectiveness of maximizing the martial arts distribution, expressed as:
objective function two f 2 Representing the task reliability of maximizing the assignment of the martial arts, expressed as:
wherein, the liquid crystal display device comprises a liquid crystal display device,representation->Time objective->Is of value (1); />Representation->Time of day completion goal->Task reliability of (2); />Representation->Number of time wars; />Representation->Moment missile->Cost of (2); />Representation->Whether the ith missile is assigned to the target +.>Is->Integer decision variables->Indicating allocation of->Indicating no allocation;representation->Time zone->Task reliability of (2); />Representing a collection of regions.
The constraint conditions of the multi-objective dynamic optimization model are specifically expressed as follows:
constraint I:means ensuring that the number of missile assignments does not exceed a maximum available number;
constraint II:the method ensures that each missile is allocated with at most one task;
constraint III:indicating that the assigned missile is ensured to be effectively fault-free;
constraint IV:indicating that the assigned missile cannot exceed the range of the missile;
constraint V:representing that the regional task reliability criteria are ensured to be met;
constraint VI:representing->Integer variable constraints;
wherein, the liquid crystal display device comprises a liquid crystal display device,a target at time t; />Representing the performance state of the missile at time t +.>Indicating normal performance of the missile, and indicating->Indicating that the missile performance has faults; />Missile->To task->Is a distance of (2);missile->Is a navigation course of (3).
Based on the multi-objective dynamic optimization model and constraint conditions, bullet groups are distributed to the objectives, and a plurality of different schemes exist, and each scheme can calculate the reliability of the respective task; in the embodiment of the invention, the calculation process of the task reliability of any heterogeneous warhead target allocation scheme comprises the following steps:
according to the number of missiles launched simultaneously to the target areaAnd the number of defensive missiles in the target area SAssume that the defending missile of the target area S is about this->The single missile intercepts with equal probability, but the probability of success of interception of different missiles is +.>Different, calculate the non-intercepted probability of the ith missile +.>：
According to not being covered byInterception probabilityReliability of the i-th missile itself is combined +.>Reliability r of ith missile under electromagnetic severe environment ei (t), and the task success probability of the ith missile to the target task +.>Calculating the task reliability of the ith missile +.>Let it be the first task reliability +.>：
According to the first task reliabilityIn combination with the number of wars at time t +.>And a 0-1 integer decision variable for whether or not to assign the ith missile to the target task at time t +.>Calculating the task reliability of the ith missile for completing the jth target task in the target area S>(J is a positive integer, J is less than or equal to J), denoted as second task reliability ∈>：
Taking all tasks J with preset number in the target area S as task success criteria, thus taking the area target as a node to form a series model, as shown in FIG. 2; based on the task success criteria and according to the second task reliabilityTask reliability of calculation target area +.>：
According to the task reliability of the target area, the target function two f is adopted 2 Calculating the task reliability of the heterogeneous warhead under the self influence and countermeasure game under the task allocation scheme;
in the above-mentioned formula(s),representing simultaneous +.>The number of missiles launched; />Representing a target areaThe number of defensive missiles in (1), assuming that the air defense missiles are for ∈ ->The missiles are intercepted with equal probability, but for different missilesDifferent; />Representing the reliability of the ith missile itself; />The reliability of the ith missile under the electromagnetic severe environment is represented; />Representing the i-th missile pair target mission->Task success probability of (a); health status of the ith missileBy self reliability->And reliability of electromagnetic resistance to harsh environments +.>Co-determination, when->Not lower than the set reliability threshold +.>And->Not lower than the reliability threshold value of electromagnetic resistance under severe environment>The missile is healthy +.>Set to 1, otherwise->Set to 0; wherein (1)>、/>And->Is also a factor affecting the reliability of the task of weapon distribution.
In the step S2, the dynamic adaptive MOEA/D-AM2M method includes two parts, wherein one part is an improved static adaptive multi-objective optimization algorithm, and the other part is a new environment detection mechanism and dynamic response mechanism for coping with environmental changes;
based on this, the step S2 specifically includes:
s21, solving a multi-target dynamic optimization model through a static self-adaptive multi-target optimization algorithm to obtain a pareto front solution set of a multi-target optimization problem, namely obtaining a weapon target allocation scheme for a system to make decisions;
s22, stopping iteration and ending operation if the dynamic algorithm meets a preset termination condition; if the dynamic algorithm does not meet the preset termination condition, detecting external environment data by adopting an environment detection mechanism, and judging whether decision variables and model parameters change or not; wherein the external environment data includes: group reliability status data, task situation information, task result situation information, and heterogeneous group situation information.
S23, if the decision variable and the model parameter are not changed, solving the multi-objective dynamic optimization model in the iteration through a static self-adaptive multi-objective optimization algorithm again; if the decision variable and the model parameter change, starting a dynamic response mechanism, and updating the parameter of the multi-objective dynamic optimization model in the iteration;
s24, repeating the steps S21-S23 until the preset termination condition of the dynamic algorithm is met, and stopping iteration;
it should be noted that, the above-mentioned steps S21 to S24 are merely for convenience of description, and are not limited to specific steps.
The static adaptive multi-objective optimization algorithm, the environment detection mechanism and the dynamic response mechanism are described below.
1. Improved static adaptive multi-objective optimization algorithm:
the improved static self-adaptive multi-objective optimization algorithm is a multi-objective optimization algorithm based on decomposition, a multi-objective optimization problem is decomposed into a plurality of multi-objective optimization sub-problems, the difficulty of each sub-problem is reduced, the solving difficulty of the algorithm is reduced, solutions of the sub-problems are associated with each other, and the co-evolution can be realized;
the improved static self-adaptive multi-objective optimization algorithm combines two advanced multi-objective optimization algorithms, MOEA/D-AM2M and HLMEA; the convergence capacity of the MOEA/D-AM2M algorithm is inherited, the diversity capacity of the HLMEA algorithm is reserved, and the convergence capacity and the diversity capacity are improved; the improved static self-adaptive multi-objective optimization algorithm also adopts a new self-adaptive crossover operator, and can adaptively select a proper operator from 5 crossover operators, so that the convergence capacity of the algorithm is improved; and a local search strategy is adopted, so that the diversity capability of the algorithm is improved through local variation when the diversity of the population is reduced. The flow chart of the algorithm is shown in fig. 3, and comprises the following steps:
(1) Coding, initializing a population, and initializing various parameters;
(2) Selecting an intersection operator by adopting roulette according to the self-adaptive intersection operator probability table;
(3) Clustering each subspace population into a plurality of clusters;
(4) Individual cross variation within clusters to generate offspring if convergence and diversity balance parametersCarrying out local search on the offspring; calculating reward and punishment scores of the crossover operators according to the offspring;
(5) Adopting a non-dominant sorting method to perform environment selection;
(6) Dividing the population into subspaces according to the weight vectors, and outputting new subspace individuals and function values;
(7) If the weight is to be updated, updating the weight vector, updating the number of subspace population, and updating the reference point;
(8) Calculating parametersUpdating a reward and punishment score memory table and updating a crossover operator probability table;
(9) And (3) outputting a result if the termination condition is met, otherwise, returning to the step (2).
Because the genetic algorithm has a plurality of crossing operators aiming at binary coding operators, and more commonly has single-point crossing, multi-point crossing, shuffling crossing, agent crossing reduction and the like, the self-adaptive crossing operator in the step (2) rewards and penalizes the crossing operators through the representation of different crossing operators, designs a memory table, stores reward and penalty points, and calculates the probability of the different crossing operators being reselected according to the reward and penalty points. In the embodiment of the invention, the dominant relationship between the crossing offspring gene and the parent gene is adopted to award and punish crossing operators, and the award and punish strategies are as follows: if the parent genes are dominant, and the mother genes are assumed to dominant the father genes, the penalty score is increased by 1 if the offspring genes are dominant by the mother genes, otherwise the bonus score is increased by 1; if parents do not control each other, if the offspring genes do not control the parent genes, the reward score is added with 1, otherwise, the penalty score is added with 1. In each iteration, selecting a crossover operator as the crossover operator of the iteration according to the probability table in a roulette manner, accumulating the reward and punish scores in the iteration as the scores of the crossover operator, storing the scores into a memory table, and calculating and updating the probability table through the memory table, wherein the calculation manner is as follows:
the reward score memory list and the punishment score memory list are respectively large and smallMatrix of->，/>Wherein->For the set memory table length, +.>The number of types of crossover operators. To calculate +.> Probability of each crossover operator, in the embodiment of the invention, is firstly respectively corresponding to->And->Is>Column summation:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the first few generations->Middle->The number of promising solutions generated by the individual operators, whereas +.>Refers to an unsatisfactory solution. The probability of selecting a crossover operator is calculated as:
wherein, the liquid crystal display device comprises a liquid crystal display device,is a very small number, which has a value of 0.0001; />For preventing in the last few generations->Never select->Dividing by zero in the case of individual operators; in addition, a->Refers to allocation to->Probability of the individual operators;
according to the above process, the embodiment of the invention can calculateThe probability that each strategy is selected in the generation process; to make the sum of probabilities equal to 1, the probabilities are finally normalized as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is->Normalized probability of individual operatorsThe rate. In each iteration of the optimization algorithm, a crossover operator is selected as the crossover operator of the iteration by adopting a roulette mode through the normalized probability.
2. Environment detection mechanism and dynamic response mechanism:
weapon goal distribution environment changes are mainly divided into two categories, the first category is that the dimension of the decision variable changes, and the other category is that the dimension of the decision variable does not change. According to the multi-objective dynamic optimization algorithm framework of fig. 1, whether the environment changes or not is detected through external data, weapon reliability data, task situation information, task result situation information and bullet situation information. Decision variablesThe dimensions of (2) are determined by the number of weapons and the number of tasks; when the weapon number changes or the task number changes dynamically, the dimension of the decision variable changes accordingly, and the related parameter information can change dynamically at the same time, which is the first type of change; if neither the weapon number nor the task number changes, but only other parameters in the optimization model update dynamically, the decision variable dimension does not change, which is a second type of change. Dynamic response mechanisms are therefore also correspondingly divided into two categories;
the environment detection mechanism will decide whether the variables change or not and whether the other optimization model parameters change or not. Detecting a first type of change, namely detecting whether the number of weapons and the number of tasks are changed or not; the second type of change detection method comprises the following steps: by random sampling, whether the objective function changes before and after comparison, namely, randomly extracting part of solutions in the evolution algorithm process, calculating the function value of the solutions, comparing the function value with the previous function value of the solutions, and if the change exceeds a certain range, considering that the environment changes;
the dynamic response mechanism is used for detecting corresponding actions taken when the environment changes so as to quickly establish a weapon target allocation scheme adapting to the new environment; when the environment detection mechanism detects that the environment changes, the dynamic response mechanism firstly adopts an inheritance mechanism, namely inheritance of an allocation strategy in the optimization process; although the environment changes, the environment changes generally not very severely, and the history optimization scheme is still better than the random initialization scheme, so that the inheritance mechanism can shorten the optimization time and improve the convergence rate. For one type of environmental change, the changed decision variables need to be processed preferentially, and by carrying out random initialization on the decision space of the changed part, the superiority of a historical scheme is reserved, the diversity of the population can be improved, and the capability of finding the optimal solution is improved for the subsequent optimization process. For the change of the second-class environment, not only the historical elite information needs to be kept, but also the diversity of the population needs to be improved to explore a decision space, so that the severity of the environmental change is described according to the quantity change of non-dominant front solutions in the multi-objective optimization process, the proportion of the population needing random initialization is obtained, and the diversity of the population is improved. The flow chart is shown in fig. 4. The specific classification is 2:
the first category is that decision variables change, and the decision variables are specifically divided into 4 categories:
weapon increase, target increase: inheriting the history strategy, distributing targets for newly added weapons, and randomly selecting part of weapons to be distributed to new tasks;
weapon decrease, target increase: inheritance of historical policies, random selection of portions of weapons for allocation to new targets
Weapon increase, target decrease: inheriting the history strategy, and randomly distributing targets for newly added weapons;
weapon reduction, target reduction: inheriting the history policy;
the second type is that decision variables are unchanged, the number of weapons and missiles are unchanged, and at the moment, a certain proportion of population is randomly initialized to replace the original solution according to the change degree of the number of non-dominant front edge solutions.
In the heterogeneous warhead task reliability planning process, the constraint conditions are processed by using a repair factor and a penalty function, the infeasible scheme is corrected by using the repair factor to enable the infeasible scheme to meet the constraint conditions, and the genetic probability of the infeasible scheme is reduced by using a penalty function method;
specifically, for the constraint I, II, III, IV, the allocation scheme that does not satisfy these constraint is modified to an allocation scheme that satisfies the constraint by the repair factor; the method comprises the following steps:
for constraint condition I, randomly canceling missile allocation when constraint is not met, so that the number of the used missiles is reduced, and the constraint condition is achieved;
when the constraint condition II is not satisfied, distributing a missile to a plurality of targets, and repairing to randomly reserve one target or not distribute the target from the targets;
for constraint condition III, when the constraint of the health state is not satisfied, canceling the allocation of the missile;
for constraint condition IV, when the distance constraint is not met, randomly distributing the missile to another target meeting the distance constraint;
for constraint condition V, punishment is carried out on an allocation scheme which cannot reach the task reliability index, and the genetic probability of the scheme is reduced; the penalty function is defined as:
wherein, the liquid crystal display device comprises a liquid crystal display device,is->And an objective function.
Constraint VI is specified at encoding time asInteger encoding and therefore no additional processing is required.
The above-described aspects of the present invention will be further described with reference to the following examples.
1. The example initialization specifically comprises the following steps:
number of available warsSet to 20; number of targets found->Set to 6;
weapon type list, weapons_type_list= [1, 2, 0, …,1]One-dimensional array length isThe numerical value corresponds to the type of the numbered weapon, and the wars are classified into 3 types according to functions;
target type list, target_type_list= [1, 2, 0, …,1]One-dimensional array length isThe numerical value corresponds to the type of the numbered target, and the target type is classified into 3 types;
target_area_list= [1, 2, 0, …,1]One-dimensional array length isThe numerical value corresponds to the area where the numbered targets are located, 3 areas are set, 0 represents the target in the area 0, and the like;
setting variation probability0.015, population number->Set to 200; maximum number of iterationsSet to 100; weight vector +.>Set to 20; uniformly distributed in the two-dimensional space of the objective function, wherein each weight corresponds to one sub-problem, and the population quantity of each sub-problem is +.>Initially set to 10, so the total population number corresponds +.>；
Target valueRespectively represent->The value of each task; weapon value->Respectively represent->Cost of individual weapons; task success probability of->Is recorded as a two-dimensional matrix of weapon->For object->At->Time task success probability->The method comprises the steps of carrying out a first treatment on the surface of the Weapon course->；
Self reliability thresholdSet to 0.99, weapon self reliability +.>By random number generation and assuming that 10% does not reach the threshold +.>To simulate a real scene; reliability threshold value in severe environments such as electromagnetic countermeasure>Setting to 0.99, reliability under severe environment such as electromagnetic countermeasure>By random number generation and assuming that 10% does not reach the threshold +.>To simulate a real environment; according to->And->And its corresponding threshold value determine the performance status of the weapon +.>；
The positions of the warheads are two-dimensional coordinate matrixes; the target position is a two-dimensional coordinate matrix, and the weapon is calculatedTo the target->European distance->；/>
Reference point initialization to；
The coding mode of the solution is composed of decision variablesDetermine->Is->An integer matrix, each row representing the allocation result of a weapon to all tasks; FIG. 5 shows a coded configuration for a weapon number 10, a target number 3;
and is assumed to be atThe moment of time the environment changes, the number of available wars +.>For some reason 18; task number->Becomes 5; other model parameters also change to some extent;
2. initializing a population:
the number of random generation isIs a group of the species.
The neighbor vector is initialized. The neighbor vector of a certain vector is defined as an ownership weight vectorA set of vectors having the smallest distance to the vector. Initial population number per sub-question in the example +.>10.
3. After initialization, the model is iteratively solved by using an improved dynamic self-adaptive MOEA/D-AM2M algorithm, and the process is shown in figure 1, and specifically comprises the following steps:
firstly, adopting a static algorithm to solve a multi-objective optimization model, wherein the process is shown in figure 3, firstly, obtaining a crossover operator to be adopted in the iterative process through roulette wheel, clustering the population in the subspace of each sub-problem, dividing the population into a plurality of clusters, carrying out intra-cluster evolution, and carrying out individual in each clusterRandomly selecting a body from the same cluster>Participation in evolution; the evolution process is as follows:
cross variation: if diversity operatorNamely, the population diversity after iteration is increased, so that the focusing convergence is improved; />And->Through the operation of the crossover operator, the mutation operator generates mutation according to the mutation probability, and a new individual is generated；
If diversity operatorI.e. the population diversity after iteration is not increased, the focusing diversity is improved; at->And->After the crossover operator operation, local search is carried out to obtain a more excellent individual +.>；
Repairing: for new individualsJudging whether the constraint condition I, II, III, IV is met or not, and correcting individuals which do not meet the constraint condition through a repair operator; if the constraint condition I is not met, randomly canceling weapon allocation, so that the number of the used warheads is reduced, and the constraint condition is met; if constraint II is not satisfied, the weapon is moved from a plurality of targets allocatedRandomly reserving a target or not in the label; if the constraint condition III is not met, namely the constraint of the health state is not met, canceling the allocation of the weapon, and setting the decision variable to 0; if the constraint condition IV is not met, namely the distance constraint is not met, randomly distributing the weapon to another target meeting the distance constraint;
by new individualsUpdating the reward and penalty scores reward and penalty with the dominant relationship of their parents;
the process is repeated until each individual in the original population has completed the crossover variation, thereby generatingNew individuals; non-dominant ordering of new offspring and original population, selecting +.>Individual elite individuals participate in subsequent iterations; when the specified number of iterations is reached, the present embodiment sets +.>As an updating iteration interval, namely updating once every 5 times of iteration; the self-adaptive weight updating and subspace population quantity updating algorithm can adaptively adjust the weight and subspace population quantity according to elite individuals so as to centralize the computing power in a subspace with prospect, thereby accelerating the convergence speed and improving the algorithm capacity; updating the reference points helps solve the problem that multiple objective functions are not on an order of magnitude; />
Then updating the rewarding and punishing memory table, when the number of loop iteration reachesWhen calculating the relevant parametersAnd update the crossover operator probability table->；
Repeating the iterative process until the static algorithm termination condition is met;
4. dynamic response
Setting upThe environment changes at the moment, and when the static algorithm reaches the static algorithm termination condition and the dynamic algorithm termination condition is not met, the algorithm provided by the embodiment starts an environment detection program to detect whether the parameters in the optimization model change or not;
if the model is changed, executing a dynamic response program, initializing the population according to a dynamic response mechanism, and continuously adopting a static multi-objective optimization algorithm to solve the changed multi-objective optimization model.
And after the termination condition is reached, outputting the obtained pareto front solution and the corresponding objective function value.
The contrast static algorithm adopted in the embodiment adopts a completely random initialization population method when encountering environmental changes. At the same iteration numberAnd comparing the calculated scheme with a dynamic algorithm.
The pareto front calculated in this example is shown in fig. 6, the vertical axis is the cost efficiency, the horizontal axis is the task reliability, and each point shown in fig. 6 corresponds to a non-dominant wargroup task allocation scheme; as can be seen from fig. 6, either solution has its own advantages, either high cost efficiency or high task reliability; the numerous schemes solved by the embodiment can assist the final decision, and a decision maker can select an optimal scheme from the numerous schemes to perform task allocation of the warhead according to the actual scene.
The embodiment of the invention also adopts the indexes of the commonly used 4 kinds of test multi-objective optimization algorithms, calculates the indexes of the solving result, compares the indexes with the indexes of the solution obtained by the pure static algorithm, and can be seen as shown in figure 7And->The convergence reflected by the index verifies the superiority of the dynamic algorithm provided by the embodiment of the invention, and the index is far better than that of the static algorithm under the same iteration times; in FIG. 7, it can be seen at the same time +.>And->The diversity reflected by the index is more superior; this example demonstrates the excellent performance of the dynamic algorithm.
In summary, the heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization provided by the embodiment of the invention is respectively the efficiency-cost ratio and the warhead task reliability; wherein the cost effectiveness ratio is the ratio of weapon utility to weapon cost, reflecting the efficiency and economy of weapon distribution; the task reliability target reflects the capacity of the warhead to complete the task and is an index reflecting the reliability and quality of the combat task; the efficiency-to-cost ratio and the task reliability are a pair of conflicting goals, and increasing the efficiency-to-cost ratio of weapon distribution reduces the task reliability of weapon distribution; because two targets cannot reach the optimal simultaneously, a double-target model (namely a multi-target dynamic optimization model) is also constructed in the embodiment of the invention and is used for solving the pareto front solution set; meanwhile, a plurality of constraint conditions such as the number constraint of the warheads, the integer constraint of the decision variables 0-1, the range constraint, the state constraint and the reliability constraint are also considered.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (7)

1. A heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization is characterized by comprising the following steps:
s1, constructing a multi-objective dynamic optimization model of a heterogeneous warhead by taking the maximum efficiency-cost ratio of the martial arts distribution and the maximum task reliability of the martial arts distribution as optimization targets;
s2, carrying out iterative solution on the multi-objective dynamic optimization model based on a dynamic self-adaptive MOEA/D-AM2M method, and stopping iteration until a dynamic algorithm preset termination condition is met;
in the iterative solving process, when the environment changes, starting a dynamic response mechanism in the dynamic self-adaptive MOEA/D-AM2M method to obtain a pareto front solution set of a multi-objective optimization problem in a new environment;
in the step S1, the multi-objective dynamic optimization model is expressed as:
maxmize f 1 (t),f 2 (t),
s.t.I,II,III,IV,V,VI
wherein maximum represents the maximum target value; f (f) 1 Representing an objective function I; f (f) 2 Representing a second objective function; I. II, III, IV, V and VI each represent constraints of a multi-objective dynamic optimization model;
objective function f 1 Representing the cost effectiveness of maximizing the martial arts distribution, expressed as:
objective function two f 2 Representing the task reliability of maximizing the assignment of the martial arts, expressed as:
wherein V is j (t) represents the value of the target j at time t; r is R j (t) represents the task reliability of the completion target j at time t; m (t) represents the number of the bullet groups at the moment t; c (C) i (t) represents the cost of missile i at time t; x is x ij (t) 0-1 integer decision variable, x, indicating whether or not the ith missile was assigned to target j at time t ij (t) =1 indicates allocation, x ij (t) =0 indicates no allocation; r is R s (t) represents the task reliability of the region s at time t; u represents a region set;
the constraint conditions of the multi-objective dynamic optimization model are specifically expressed as follows:
constraint I:means ensuring that the number of missile assignments does not exceed a maximum available number;
constraint II:the method ensures that each missile is allocated with at most one task;
constraint III: s is(s) i (t) =1, indicating that the assigned missile is guaranteed to be effectively fault-free;
constraint IV: d, d ij (t)≤D i (t) indicating that the assigned missile cannot exceed the range of the missile;
constraint V: r is R s (t) 0.9 or more, representing ensuring that the regional task reliability criterion is met;
constraint VI: x is x ij (t) =0 or 1, representing an integer variable constraint of 0-1;
wherein A is t A target at time t; s is(s) i (t) represents the performance state of the missile at the moment t, s i (t) =1 indicates that the missile performance is normal, s i (t) =0 indicates a missile performance failure; d, d ij (t) represents the distance from missile i to task j at time t; d (D) i (t) represents the range of missile i at time t;
the calculation process of the task reliability of any heterogeneous warhead target allocation scheme comprises the following steps:
according to the number m (t) of missiles launched to the target area at the same time and the number n (t) of defensive missiles in the target area S, the defensive missiles in the target area S are assumed to intercept the m (t) missiles with equal probability, but the interception success probability P of different missiles is assumed Qi (t) differently, calculating the non-interception probability P of the ith missile ci (t)：
According to the probability P of non-interception ci (t) incorporating the reliability r of the ith missile itself i (t), reliability r of ith missile under electromagnetic severe environment resistance ei (t), and the task success probability p of the ith missile to the target task ij (t) calculating the mission reliability R of the ith missile i (t) denoted as first task reliability R i (t)：
R i (t)＝r i (t)r ei (t)p ij (t)P ci (t)
According to the first task reliability R i (t), in combination with the number of clusters M (t) at time t, and a 0-1 integer decision variable x of whether or not to assign the ith missile to the target task at time t ij (t) calculating the task reliability R of the ith missile for completing the jth target task in the target area S j (t), J is a positive integer, J is less than or equal to J, and is recorded as the second task reliability R j (t)：
With J arbitrary preset in the target area SAll tasks are completed as task success criteria, so that the area targets are used as nodes to form a series model; based on the task success criteria and according to the second task reliability R j (t) calculating target area task reliability R S (t)：
According to the task reliability of the target area, the target function two f is adopted 2 Calculating the task reliability of the heterogeneous warhead under the self influence and countermeasure game under the task allocation scheme;
in the above formula, m (t) represents the number of missiles launched simultaneously to the target area S; n (t) represents the number of defensive missiles in the target area S, assuming that an air-defense missile intercepts m (t) missiles with equal probability, but P for different missiles ci (t) is different; r is (r) i (t) represents the reliability of the ith missile itself; r is (r) ei (t) represents the reliability of the ith missile in an electromagnetically resistant harsh environment; p is p ij (t) represents the task success probability of the ith missile to the target task j; health s of the ith missile i (t) by its own reliability r i (t) and reliability r of electromagnetic countermeasure against severe environments ei (t) co-determining when r i (t) not lower than a set reliability threshold epsilon, and r ei (t) not lower than the threshold epsilon of reliability of electromagnetic countermeasure against severe environments e The missile is in a healthy state s i (t) is set to 1, otherwise s i (t) set to 0; wherein r is i (t)、r ei (t) and p ij (t) is also a factor affecting the reliability of the task of weapon distribution.
2. The heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization of claim 1, wherein the dynamic adaptive MOEA/D-AM2M method comprises a static adaptive multi-objective optimization algorithm, an environment detection mechanism and a dynamic response mechanism.
3. The heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization according to claim 2, wherein the step S2 specifically includes:
solving the multi-objective dynamic optimization model through the static self-adaptive multi-objective optimization algorithm to obtain a pareto front solution set of the multi-objective optimization problem for a system to make decisions;
if the dynamic algorithm does not reach the preset termination condition, detecting external environment data by adopting an environment detection mechanism, and judging whether decision variables and model parameters change or not;
if the decision variable and the model parameter change, starting the dynamic response mechanism, and updating the parameters of the multi-objective dynamic optimization model in the iteration;
repeating the steps until the preset termination condition of the dynamic algorithm is met, and stopping iteration.
4. A heterogeneous warfare task reliability planning method based on multi-objective dynamic optimization as claimed in claim 3, characterized in that the external environment data comprises: group reliability status data, task situation information, task result situation information, and heterogeneous group situation information.
5. A heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization according to claim 3, wherein if the decision variable is not changed, solving the multi-objective dynamic optimization model in the iteration through the static self-adaptive multi-objective optimization algorithm again.
6. The heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization of claim 3, wherein the updating parameters of the multi-objective dynamic optimization model specifically comprises:
weapon increase, target increase: inheriting the history strategy, distributing targets for newly added weapons, and randomly selecting part of weapons to be distributed to new tasks;
weapon decrease, target increase: inheritance of historical policies, random selection of portions of weapons for allocation to new targets
Weapon increase, target decrease: inheriting the history strategy, and randomly distributing targets for newly added weapons;
weapon reduction, target reduction: inherit the history policy.
7. The heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization of claim 1, wherein constraint conditions are processed by using repair factors and penalty functions in the heterogeneous warhead task reliability planning process.
CN202310990039.1A 2023-08-08 2023-08-08 Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization Active CN116702633B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310990039.1A CN116702633B (en) 2023-08-08 2023-08-08 Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310990039.1A CN116702633B (en) 2023-08-08 2023-08-08 Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization

Publications (2)

Publication Number Publication Date
CN116702633A CN116702633A (en) 2023-09-05
CN116702633B true CN116702633B (en) 2023-11-03

Family

ID=87843709

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310990039.1A Active CN116702633B (en) 2023-08-08 2023-08-08 Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization

Country Status (1)

Country Link
CN (1) CN116702633B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117349532A (en) * 2023-10-24 2024-01-05 烟台大学 Dynamic multi-target service combination optimization recommendation method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110096822A (en) * 2019-05-08 2019-08-06 北京理工大学 Multi-platform cooperative dynamic task allocation method under a kind of condition of uncertainty
CN112677957A (en) * 2021-01-07 2021-04-20 重庆大学 Parameter optimization method based on pareto optimality under dual-mode configuration multi-target condition
CN115222271A (en) * 2022-07-28 2022-10-21 西安电子科技大学 Weapon target distribution method based on neural network
CN116090356A (en) * 2023-03-29 2023-05-09 北京理工大学 Heterogeneous warhead multi-objective task planning method based on task reliability constraint

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
IL205526A0 (en) * 2010-05-04 2011-02-28 David Myr Automatic military real time command and control system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110096822A (en) * 2019-05-08 2019-08-06 北京理工大学 Multi-platform cooperative dynamic task allocation method under a kind of condition of uncertainty
CN112677957A (en) * 2021-01-07 2021-04-20 重庆大学 Parameter optimization method based on pareto optimality under dual-mode configuration multi-target condition
CN115222271A (en) * 2022-07-28 2022-10-21 西安电子科技大学 Weapon target distribution method based on neural network
CN116090356A (en) * 2023-03-29 2023-05-09 北京理工大学 Heterogeneous warhead multi-objective task planning method based on task reliability constraint

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Large-Scale Weapon Target Assignment Based on Improved MOEA/D Algorithm;Huiyang Yu等;《2022 4th International Conference on System Reliability and Safety Engineering (SRSE)》;全文 *
一种基于蚁群算法的准动态防空武器分配算法;张彦芳等;《火力与指挥控制》(09);全文 *
基于MOEA的微电网多目标多约束优化调度研究;李昕;《CNKI博士学位论文全文库》;全文 *
进化高维多目标优化算法研究综述;刘建昌等;《控制与决策》(05);全文 *

Also Published As

Publication number Publication date
CN116702633A (en) 2023-09-05

Similar Documents

Publication Publication Date Title
CN112861442B (en) Multi-machine collaborative air combat planning method and system based on deep reinforcement learning
CN116702633B (en) Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization
CN114594794B (en) Multi-machine collaborative task planning method considering subsystem execution capacity
CN106779210B (en) Algorithm of Firepower Allocation based on ant group algorithm
CN107479380A (en) Multi-Agent coordination control method based on evolutionary game theory
CN108594645B (en) Planning method and system for single-station multi-unmanned aerial vehicle distribution and flight route
Jiang et al. The emergence of individuality
CN108549402A (en) Unmanned aerial vehicle group method for allocating tasks based on quantum crow group hunting mechanism
CN111553469A (en) Wireless sensor network data fusion method, device and storage medium
CN112364972A (en) Unmanned fighting vehicle team fire power distribution method based on deep reinforcement learning
CN110766125A (en) Multi-target weapon-target allocation method based on artificial fish swarm algorithm
CN109670655B (en) Multi-target particle swarm optimization scheduling method for electric power system
CN116090356B (en) Heterogeneous warhead multi-objective task planning method based on task reliability constraint
CN105739304A (en) Multi-UCAV on-line striking target allocation method of opposition-based genetic algorithm(GA)
Shi et al. A multitier reinforcement learning model for a cooperative multiagent system
CN111797966B (en) Multi-machine collaborative global target distribution method based on improved flock algorithm
Liu et al. WRFMR: A multi-agent reinforcement learning method for cooperative tasks
CN112926825A (en) Multi-unmanned aerial vehicle task allocation method based on multi-target quantum shrimp swarm mechanism
Zhao et al. Shuffled frog leaping algorithm with non-dominated sorting for dynamic weapon-target assignment
CN114200960A (en) Unmanned aerial vehicle cluster search control optimization method for improving sparrow algorithm based on tabu table
CN115328189A (en) Multi-unmanned aerial vehicle cooperative game decision method and system
Ruining et al. Improved genetic algorithm for weapon target assignment problem
CN111382896B (en) WTA target optimization method of self-adaptive chaotic parallel clone selection algorithm
CN113962013A (en) Aircraft confrontation decision method and device
HuangFu et al. Adaptive grouping weapon-target assignment with field-of-view angle constraint

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