CN113792985B - Multi-target distribution method for dynamic joint tasks of sensor and weapon - Google Patents
Multi-target distribution method for dynamic joint tasks of sensor and weapon Download PDFInfo
- Publication number
- CN113792985B CN113792985B CN202110967568.0A CN202110967568A CN113792985B CN 113792985 B CN113792985 B CN 113792985B CN 202110967568 A CN202110967568 A CN 202110967568A CN 113792985 B CN113792985 B CN 113792985B
- Authority
- CN
- China
- Prior art keywords
- weapon
- sensor
- combat
- target
- task
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 18
- 239000011159 matrix material Substances 0.000 claims description 21
- 239000013598 vector Substances 0.000 claims description 19
- 238000005457 optimization Methods 0.000 claims description 16
- 108090000623 proteins and genes Proteins 0.000 claims description 15
- 230000035772 mutation Effects 0.000 claims description 10
- 230000008030 elimination Effects 0.000 claims description 5
- 238000003379 elimination reaction Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000018199 S phase Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 4
- 206010064571 Gene mutation Diseases 0.000 claims description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011423 initialization method Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Aiming, Guidance, Guns With A Light Source, Armor, Camouflage, And Targets (AREA)
Abstract
The invention discloses a sensor-weapon dynamic joint task multi-target distribution method, which considers the mutual influence among different combat units and can effectively and rapidly provide a plurality of command schemes for the reconnaissance task distribution and the striking task distribution of a sensor platform and a weapon platform. Comprising the following steps: calculating targets which can be detected by the sensor and the weapon in each combat stage, and respectively obtaining initial feasibility matrixes of the weapon and the sensor; constructing an initial population according to the initial feasibility matrixes of the weapon and the sensor, performing constraint condition processing, and performing iterative search to obtain an external population; judging whether the iterative search reaches the iterative step number, and stopping and outputting the external population, namely a group of task allocation, if the iterative search reaches the iterative step number.
Description
Technical Field
The invention belongs to the technical field of intelligent control, relates to an optimization and distribution technology of combat resources, and particularly relates to a sensor-weapon dynamic joint task multi-target distribution method.
Background
The task allocation is a key link of the combat command, directly influences the progress and the win-lose of combat, and is an important military problem of competitive research of military countries. The main expression mode of future warfare is combined combat, and many weapons can have better hit rate by continuous guidance of the sensor platform, and the weapon platform and the sensor platform jointly implement combat under unified command of an combined command mechanism. Through the consensus among the networked sensors, weapons and decision makers, the command speed can be improved, the combat rhythm can be accelerated, the combat destructive performance can be increased, the survival rate can be improved, and the better combat efficiency can be realized.
In the process of allocating the combined combat task of the weapon platform and the sensor platform, the problems of limited sensors, limited weapon quantity, limited combat time, limited combat cost and the like are required to be considered. How to perform joint allocation of the sensor and weapon task through an optimization algorithm, so that the performance of the weapon sensor is effectively exerted, and the optimal combat efficiency is particularly important.
Most of the prior art only considers weapon distribution, but fails to consider combined operations of weapons and sensors. For joint distribution of sensor weapons, a certain academic study exists, but a construction method is adopted for one-time distribution of the sensor and the weapons, the coupling between the sensor weapons cannot be considered, the dynamic characteristics of actual combat cannot be considered, and when the number of nodes in the weapons and the sensor network is large, a better solution cannot be obtained.
Disclosure of Invention
The invention discloses a sensor-weapon dynamic joint task multi-target distribution method, which considers the mutual influence among different combat units and can effectively and rapidly provide a plurality of command schemes for sensor platform and weapon platform detection task distribution and striking task distribution.
The invention is realized by the following technical scheme.
A method for sensor-weapon dynamic joint task multi-objective allocation, comprising:
Calculating targets which can be detected by the sensor and the weapon in each combat stage, and respectively obtaining initial feasibility matrixes of the weapon and the sensor;
constructing an initial population according to the initial feasibility matrixes of the weapon and the sensor, performing constraint condition processing, and performing iterative search to obtain an external population;
Judging whether the iterative search reaches the iterative step number, and stopping and outputting the external population, namely a group of task allocation schemes, if the iterative search reaches the iterative step number.
The invention has the beneficial effects that:
The invention considers the dynamic cooperation between weapon sensors and has higher combat efficiency compared with a single combat mode. Meanwhile, a heuristic method of comprehensive exploration, development and weight vector is adopted to initialize the population, constraint processing is carried out by adopting a constraint processing method of fusing greedy strategies, and sensor and weapon tasks with invalid judgment are self-adaptively eliminated, so that the solving efficiency and capacity of the algorithm are improved, the method can be applied to large-scale fight situations, damage to enemy targets is effectively improved, fight consumption is reduced, command flexibility is improved, and fight efficiency is effectively improved.
Drawings
FIG. 1 is a flow chart of a method for dynamically joint task multi-objective distribution of a sensor-weapon in accordance with the present invention;
FIG. 2 is a flow chart of one such iterative search in an embodiment.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
As shown in fig. 1, a method for dynamically and jointly distributing tasks by using a sensor and a weapon in the present embodiment specifically includes:
Step one, calculating targets which can be detected by a sensor and a weapon in each combat stage, and respectively obtaining initial feasibility matrixes of the weapon and the sensor;
In this embodiment, the initial feasibility matrix of the weapon is denoted as x= [ X mst]M×S×T, where X mst∈{0,1},xmst =1 means that weapon m can attack target t in s-phase; the initial feasibility matrix of the sensor is denoted as y= [ Y nst]N×S×T, where Y nst∈{0,1},ynst = 1 means that sensor n can detect the target t in s-phase.
Constructing an initial population according to the initial feasibility matrix of the weapon and the sensor, performing constraint condition processing, and performing iterative search to obtain an external population;
In this embodiment, the construction of the initial population is performed in the following manner:
Setting p individuals X= { X 1,x2,...xn } as an initial population, the individual length N is sx (M+N),
Where S is the total number of stages, M is the total number of weapons, and N is the total number of sensors.
The construction method of each gene locus is as follows:
First, p weight vectors λ= { w 1,w2 }, where w 1+w2 = 1,
Determining the probability of assigning tasks to the combat resources according to the weight vector lambda= { w 1,w2 }:
wi＝aw1+b
si＝cw1+d
Wherein w 1 represents the weight of the objective function 1, w 2 represents the weight of the objective function 2, wi is the probability of assigning tasks to the weapon, si is the probability of assigning tasks to the sensor, a, b, c, d are preset coefficients;
Then, for each gene position x i, calculating whether a task is required to be allocated to the combat resource according to wi and si, and when the task is required to be allocated to the combat resource, calculating a corresponding combat stage s according to the formula i/(M+N), wherein combat resources corresponding to different gene positions x i can be calculated according to the following formula:
Wherein i represents the subscript of the gene position x i, M i％(M+N) represents the x (M+N) th weapon, and N i％(M+N) -M represents the i (M+N) th sensor.
When the calculated combat resource is a weapon, according to the corresponding weapon and combat stage, according to the feasibility matrix X in the first step, a target which can be attacked by the weapon in the combat stage can be obtained, and a feasible target is randomly selected as an allocation result. When the calculated combat resource is a sensor, according to the corresponding weapon and combat stage, according to the feasibility matrix Y in the first step, a target which can be detected by the sensor in the combat stage can be obtained, and a feasible target is randomly selected as an allocation result.
The preliminary individual X is calculated in the steps, but when the individual generated by the steps fails to meet all constraint conditions and fails to combine the coupling of the weapon and the sensor, the weapon can successfully hit the target only by the guidance of the sensor for a period of time, and the weapon cannot be continuously fired, so that constraint condition processing is required for the individual; the individual is subjected to constraint processing, and when a weapon has been assigned a task at a previous stage or is not continuously directed by the sensor, the weapon is set to perform no task at that stage. Thus ultimately constructing an initial population.
As shown in fig. 2, in this embodiment, the steps of one iterative search are as follows:
step 1, crossing: randomly selecting two individuals x j and x k from T, randomly selecting n continuous gene sites to cross to generate a new individual x, and performing heuristic constraint processing of a fused greedy algorithm on the x;
Step 2, mutation: carrying out single-gene mutation on each individual in the population, firstly randomly selecting a gene, calculating a corresponding combat stage and combat unit according to the position, selecting a feasibility target from the feasibility matrix, setting a probability unassigned target of pi in the mutation process, wherein pi is a preset coefficient, preferably taking 0.3-0.5, and finally carrying out constraint processing on a new individual x by adopting a heuristic method of a fused greedy algorithm;
In this embodiment, the heuristic constraint processing of the fused greedy algorithm adopts the following manner:
a) For weapon fire-turning constraint, setting the probability of adopting a greedy strategy as pg, wherein the calculation formula is pg=i/(i+i), I is the current iteration number, and I is the total iteration number; when a weapon w violates fire turning constraint, eliminating constraint conflict; and adopting a greedy strategy to respectively calculate the weighted sum of the objective functions under different conditions, and adopting a mode of maximizing the weighted sum. In specific implementation, if greedy strategy is not adopted, any mode is randomly adopted.
The elimination of constraint conflicts may take three forms: the w weapon does not perform tasks in the previous stage, the w weapon does not perform tasks in the current stage, and the w weapon does not perform tasks in both stages.
B) When the weapon and the sensor fail to meet the continuous guidance constraint, a feasible idle sensor allocation task is randomly selected in the feasibility matrix without changing the sensors to which the tasks have been allocated to meet the guidance constraint, and if no feasible idle sensor exists, the allocated weapon task is disabled.
Step 3, optimizing the individual, and eliminating invalid combat tasks with probability to obtain a new individual x;
The ineffective combat task in this embodiment includes the following two types: firstly, the target has allocated more weapons, and the weapons are not allocated to the target any more; secondly, the tracking of the target by the sensor fails to successfully guide the weapon.
Step 4, updating, for the new generated individual x, for all optimization targets j= {1,2}, updating the worst reference point z (z 1,z2), wherein z 1 represents the smallest f 1(x),z2 in the individuals of the first optimization target and the smallest f 2 (x) in the individuals of the second optimization target, and if f j(x)＜zj, setting z j＝fj (x).
Step 5, updating the neighborhood solution and updating the neighborhood range, namely generating new individuals x' corresponding to the vector lambda by crossing, mutation or optimization for all other individuals x i (corresponding to the weight vector lambda i) in the neighborhood, for all targets j= {1,2}, ifThen x i = x' is set, where i is the current algebra;
And 6, updating the external population EP according to the crowding degree relation, namely removing all individuals which are subjected to x ' from the EP according to each new individual x ', and adding x ' into the EP if no individual in the EP is subjected to x ' and the Euclidean distance between the EP and the individual x ' is not smaller than d min,dmin, wherein the minimum distance is set by the user.
And thirdly, judging whether the iterative search reaches the iterative step number, and stopping and outputting the external population if the iterative search reaches the iterative step number.
From the above steps, it can be seen that the outer population contains a plurality of individuals, each individual representing a solution for task allocation, and thus the outer population is a set of solutions, i.e. a set of task allocation methods.
Example 1:
the operational scenario aimed at by this embodiment specifically includes: the defending party deploys a plurality of fight platforms, the fight is divided into S stages, the weapon platform set is W= { W 1,w2,...wM }, the sensor platform set is S= { sen 1,sen2,...senN }, the sensor unit and the weapon unit can communicate with each other, the attacking party has T airplanes, and the aim is to break through the defending. The defender needs to perform two tasks of detecting and tracking and target hitting on a plurality of attack targets of the enemy. Wherein the sensor unit is responsible for detecting the capturing task, and the weapon unit strikes the tracked and captured enemy target after the enemy target is successfully detected and guided. The optimization objective is to maximize the elimination of the objective threat and minimize the combat resource consumption, with objective functions f 1 (x) and f 2 (x). The combination of the sensor and the weapon is characterized in that the weapon can have better hit rate only by continuous guidance of the sensor platform, the tracking of targets by multiple stages of the sensor platform is required to be ensured to guide the weapon to the targets, and the weapon unit itself needs to meet fire turning constraint, namely, after being distributed to a given target for a period of time, the current stage cannot be distributed to other targets.
For this kind of combat scenario, the method of the invention is described in detail below according to specific implementation steps:
step 1, setting an external population EP as an empty set for storing an optimal solution. And generating a group of weight vectors which are uniformly distributed in a two-dimensional space, calculating the Euclidean distance between any two weight vectors, and taking the nearest T weight vectors of each weight vector as neighborhood vectors according to the Euclidean distance, wherein the set of the T weight vectors is TV= { lambda 1,λ2...λT }.
Step 2, calculating targets which can be detected by the sensor and the weapon in each combat stage through the predicted target track, and obtaining an initial feasibility matrix of the weapon and the sensor, wherein the initial feasibility matrix of the weapon is expressed as x= [ X mst]M×S×T ], X mst∈{0,1},xmst =1 indicates that the weapon can attack the target t in the s stage, and likewise, the initial feasibility matrix of the sensor is expressed as y= [ Y nst]N×S×T ], and Y nst∈{0,1},ynst =1 indicates that the sensor can detect the target t in the s stage.
And 3, constructing an initial population by adopting a heuristic method considering weight influence. Individuals in the population are set to pi. For each individual, the following procedure was followed.
Step 3.1, the length of each individual is set to s× (m+n), where S is the number of combat stages, M is the total number of weapons, and N is the total number of sensors. With the integer t encoding, different integers t represent different enemy targets, wherein part of positions represent that weapon m is allocated to target t or unallocated task at stage s, and the other part of positions represent that sensor n is allocated to target t or unallocated task at stage s, and the calculation method is as follows:
For the gene position x i, the corresponding combat stage is i/(m+n), and the corresponding combat resource is calculated by the following formula:
wherein i represents the subscript of the genetic locus x i, M i％(M+N) represents the x (m+n) -th weapon, and N i％(M+N)-M represents the i (m+n) -th sensor.
And (3) initializing according to the combat stage and combat resources (weapons or sensors) corresponding to each gene position and the weight vector and the feasibility matrix established in the step (2).
Step 3.2, determining the probability of assigning tasks to the combat resources according to a weight vector lambda= { w 1,w2 }, wherein x 1 represents the weight of the target threat elimination target, and x 2 represents the weight of the target of minimizing combat resource consumption. For tasks that are more in need of eliminating target threats, a higher probability of task allocation is set during initial allocation. Let wi be the probability of assigning tasks to the weapon, si be the probability of assigning tasks to the sensor, preferably the task assignment probability is set according to the following formula:
wi＝aw1+b
si＝cw1+d
Where a, b, c, d are coefficients, preferably c, d is set to 1.5 times a, b takes 0.1, a is determined by dividing the total number of available allocations in the weapon feasibility table by 3-7 times the total number of targets.
And (3) after the corresponding combat resources and combat stages are calculated, randomly selecting a feasible target as an allocation result according to the feasibility matrix established in the step (2).
Step 3.3, the individual generated by steps 3.1, 3.2 fails to meet all constraint conditions, fails to consider continuous guiding constraint and weapon fire turning constraint, so constraint condition processing needs to be carried out on the generated individual, and when a certain weapon is assigned with tasks or is not continuously guided by a sensor, the weapon is set to execute no task at the stage.
The group obtains the initial population P ini after repeating the above operations.
Step 4, setting an initial reference point z ini＝{z1,z2, and calculating the fitness function value of each target of each individual in the current generation for the initial population P ini The reference point z 0 of the initial population P ini is found.
The following steps 5 to 10 are taken as an iterative process, and the current algebra is set as i:
And 5, crossing. Two individuals x j and x k are selected from T randomly, n continuous gene sites are selected randomly to be crossed to generate a new individual x, and heuristic constraint processing of a fused greedy algorithm is carried out on the x.
And 5.1, setting the probability of adopting a greedy strategy as pg for weapon fire-turning constraint, wherein the calculation formula is preferably pg=i/(i+i), I is the current iteration number, and I is the total iteration number. When a weapon w violates fire transfer constraints, there are three methods to eliminate the constraint conflict, namely w weapon does not perform the task in the previous stage, w weapon does not perform the task in the current stage and w weapon does not perform the task in both stages. When a greedy strategy is adopted, the weighted sum of the objective functions in the three cases is calculated, and a method is adopted to maximize the weighted sum. When greedy strategies are not adopted, the method is randomly adopted.
Step 5.2, when the weapon and the sensor fail to meet the continuous guidance constraint, randomly selecting a feasible idle sensor allocation task in the feasibility matrix to meet the guidance constraint without changing the sensor with the allocated task, and if the feasible idle sensor does not exist, disabling the allocated weapon task.
And 6, mutation. And (3) carrying out single-gene mutation on each individual in the population, firstly randomly selecting one gene, calculating a corresponding combat stage and combat unit according to the position, selecting a feasibility target from a feasibility matrix, setting a probability unassigned target of p in the mutation process, preferably, generally taking 0.3-0.5 for p, and finally carrying out constraint processing on a new individual x generated after mutation by adopting the constraint processing method in the step (5).
Step 7, optimizing the individual, and eliminating invalid combat tasks with probability, wherein the combat tasks comprise the following two types: one target has allocated more weapons, at which point the weapons are allocated to that target again; when the tracking of the target by the sensor fails to successfully direct the weapon. The cancellation probability Pe is set to, preferably,And setting the invalid combat task as a task failure according to the probability Pe, and obtaining a new individual x through optimization operation.
And 8, updating the reference point z. For the new individual x generated, for all task targets j= {1,2}, if the objective function f j(x)＜zj, z j＝fj (x) is set.
And 9, updating the neighborhood solution and updating the neighborhood range. The new individual x' generated by crossover, mutation or optimization for the individual corresponding to vector λ is for all other individuals x i in the neighborhood (its corresponding weight vector λ i), for all targets j= {1,2}, ifThen x i = x' is set.
And step 10, updating the external population EP according to the crowding degree relation. For each new individual x ', all individuals that are dominated by x' are removed from the EP. If no individual dominates x ' in the EP and neither Euclidean distance to individual x ' is less than d min,dmin is a self-set minimum distance, then x ' is added to the EP.
And step 11, judging whether the iteration step number is reached, and stopping and outputting the EP as an optimal allocation scheme set if the iteration step number is reached. Otherwise, returning to the step 5, and carrying out the next iteration.
According to the method, the weapon sensor feasibility matrix is used as an input variable of an optimization algorithm, and the optimization objective function is considered to maximize elimination of target threat and minimize combat resource consumption. On the basis of a multi-objective optimization algorithm based on decomposition, an initialization method is improved, an initial individual is constructed by a heuristic method taking weight vectors into consideration, the individual is optimized in an iteration process, invalid combat tasks are reduced, an adaptive greedy method is adopted for constraint condition processing, and an adaptive domain matching strategy is adopted. The improved method considers the coupling characteristic of the sensor weapon and the dynamic performance of the fight, improves the searching capability of the algorithm through optimizing the population initialization, constraint condition processing strategy and invalid individuals, avoids the situation of local optimum possibly occurring in the traditional decomposition algorithm when the fight scale is large, and can still obtain excellent solution sets under the large scale.
The foregoing is merely a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A method for multi-objective allocation of sensor-weapon dynamic joint tasks, comprising:
Calculating targets which can be detected by the sensor and the weapon in each combat stage, and respectively obtaining initial feasibility matrixes of the weapon and the sensor; the initial feasibility matrix of the weapon is denoted as x= [ X mst]M×S×T, where X mst∈{0,1},xmst = 1 means that weapon m can attack target t in s-phase; the initial feasibility matrix of the sensor is denoted as y= [ Y nst]N×S×T, where Y nst∈{0,1},ynst = 1 means that sensor n can detect target t in s-phase;
Constructing an initial population according to the initial feasibility matrixes of the weapon and the sensor, performing constraint condition processing, and performing iterative search to obtain an external population; the construction of the initial population adopts the following modes:
Setting p individuals X= { X 1,x2,…xn } as an initial population, wherein the individual length N is S× (M+N), S is the total stage number, M is the total weapon number, and N is the total sensor number;
3.1 construct p weight vectors λ= { w 1,w2 }, where w 1+w2 = 1,
3.2, Determining the probability of assigning tasks to the combat resources according to the weight vectors:
wi＝aw1+b
si＝cw1+d
Wherein w 1 represents the weight of the objective function 1, w 2 represents the weight of the objective function 2, wi is the probability of assigning tasks to the weapon, si is the probability of assigning tasks to the sensor, a, b, c, d are preset coefficients;
3.3, calculating whether a task is required to be allocated to the combat resource according to wi and si for each gene position x i, and when the task is required to be allocated to the combat resource, calculating a corresponding combat stage s according to the formula x/(M+N), wherein combat resources corresponding to different gene positions x i can be calculated according to the following formula:
Wherein i represents the subscript of gene locus x i, M i％(M+N) represents the i (m+n) -th weapon, and N i％(M+N)-M represents the i (m+n) -th sensor;
3.4, when the combat resource is calculated as a weapon, according to the weapon and the combat stage corresponding to the combat resource, obtaining targets which can be attacked by the weapon in the combat stage according to the feasibility matrix X, and randomly selecting a feasible target as an allocation result; when the calculated combat resource is a sensor, according to the corresponding weapon and combat stage, according to the feasibility matrix Y, obtaining a target which can be detected by the sensor in the combat stage, and randomly selecting a feasible target as an allocation result;
Judging whether the iterative search reaches the iterative step number, and stopping and outputting the external population, namely a group of task allocation, if the iterative search reaches the iterative step number.
2. A method of sensor-weapon dynamic joint task multi-objective allocation according to claim 1, characterized in that a weapon is set to perform no tasks at a stage before the stage, when the weapon has been allocated tasks or is not continuously directed by the sensor.
3. A method of sensor-weapon dynamic joint task multi-objective distribution according to claim 1 or 2, characterized in that one of said iterative searches is performed as follows:
step 1, crossing: randomly selecting two individuals x j and x k from T, randomly selecting n continuous gene sites to cross to generate a new individual x, and performing heuristic constraint processing of a fused greedy algorithm on the x;
Step 2, mutation: carrying out single-gene mutation on each individual in the population, firstly randomly selecting one gene, calculating a corresponding combat stage and combat unit according to the position, selecting a feasibility target from the feasibility matrix, setting a probability unassigned target of pi in the mutation process, wherein pi is a preset coefficient, and finally carrying out constraint processing on a new individual x by adopting a heuristic method of a fused greedy algorithm;
step 3, optimizing the individual, and eliminating invalid combat tasks with probability to obtain a new individual x;
Step 4, updating, for the new generated individual x, for all optimization targets j= {1,2}, updating the worst reference point z (z 1,z2) of the new generated individual x, wherein z 1 represents the smallest f 1(x),z2 in the individuals of the first optimization target and the smallest f 2 (x) in the individuals of the second optimization target, and if f j(x)<zj, setting z j＝fj (x);
Step 5, updating the neighborhood solution and updating the neighborhood range, namely generating new individuals x by crossing, mutation or optimization for individuals corresponding to the vector lambda, for all other individuals x i in the neighborhood, for all targets j= {1,2}, if Then x i = x, where i is the current algebra;
And 6, updating the external population EP according to the crowding degree relation, namely removing all the individuals which are subjected to x from the EP for each new individual x, and adding x into the EP if no individual in the EP is subjected to x and the Euclidean distance between the EP and the individual x is not smaller than the minimum distance which is set by the user and d min,dmin.
4.A method of sensor-weapon dynamic joint task multi-objective distribution according to claim 3, characterized in that said preset coefficient pi takes the range of 0.3-0.5.
5. The method for dynamically assigning multiple targets for combined tasks of a sensor-weapon according to claim 4, wherein said heuristic constraint processing of the fused greedy algorithm is performed by the following means:
a) For weapon fire-turning constraint, setting the probability of adopting a greedy strategy as pg, wherein the calculation formula is pg=i/(i+i), I is the current iteration number, and I is the total iteration number; when a weapon w violates fire turning constraint, eliminating constraint conflict; respectively calculating the weighted sum of the objective functions under different conditions by adopting a greedy strategy, and adopting a mode of maximizing the weighted sum;
b) When the weapon and the sensor fail to meet the continuous guidance constraint, a feasible idle sensor allocation task is randomly selected in the feasibility matrix without changing the sensors to which the tasks have been allocated to meet the guidance constraint, and if no feasible idle sensor exists, the allocated weapon task is disabled.
6. A method for dynamic joint task multi-objective distribution of a sensor-weapon according to claim 5, wherein said elimination of constraint conflicts can be achieved in three ways: the w weapon does not perform tasks in the previous stage, the w weapon does not perform tasks in the current stage, and the w weapon does not perform tasks in both stages.
7. A method of sensor-weapon dynamic joint task multi-objective allocation according to claim 6, wherein said ineffective combat task comprises two of: firstly, the target has allocated more weapons, and the weapons are not allocated to the target any more; secondly, the tracking of the target by the sensor fails to successfully guide the weapon.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110967568.0A CN113792985B (en) | 2021-08-23 | 2021-08-23 | Multi-target distribution method for dynamic joint tasks of sensor and weapon |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110967568.0A CN113792985B (en) | 2021-08-23 | 2021-08-23 | Multi-target distribution method for dynamic joint tasks of sensor and weapon |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113792985A CN113792985A (en) | 2021-12-14 |
| CN113792985B true CN113792985B (en) | 2024-08-09 |
Family
ID=78876224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110967568.0A Active CN113792985B (en) | 2021-08-23 | 2021-08-23 | Multi-target distribution method for dynamic joint tasks of sensor and weapon |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113792985B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114861417B (en) * | 2022-04-14 | 2024-04-19 | 中国人民解放军国防科技大学 | Multi-stage weapon target distribution method based on variable neighborhood search |
| CN117575299B (en) * | 2024-01-17 | 2024-04-02 | 南京信息工程大学 | Collaborative multitasking distribution method based on improved particle swarm algorithm |
Citations (1)
| 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 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101526893B1 (en) * | 2013-10-08 | 2015-06-09 | 국방과학연구소 | Simulation method for determining type and requirement quantity of weapons in engagement of air defense and System thereof |
| CN110597199A (en) * | 2019-09-03 | 2019-12-20 | 唐晓川 | Helicopter weapon scheduling method and system based on optimal model of shooting vector |
| CN112149959B (en) * | 2020-08-26 | 2022-10-21 | 北京理工大学 | Distributed sensor-weapon-target joint allocation method |
-
2021
- 2021-08-23 CN CN202110967568.0A patent/CN113792985B/en active Active
Patent Citations (1)
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113792985A (en) | 2021-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107832885B (en) | Ship formation fire power distribution method based on self-adaptive migration strategy BBO algorithm | |
| CN113792985B (en) | Multi-target distribution method for dynamic joint tasks of sensor and weapon | |
| CN108594645B (en) | Planning method and system for single-station multi-unmanned aerial vehicle distribution and flight route | |
| CN115328189B (en) | Multi-unmanned plane cooperative game decision-making method and system | |
| CN112926025B (en) | Method for solving weapon target distribution problem based on cross entropy genetic algorithm | |
| Leboucher et al. | Novel evolutionary game based multi-objective optimisation for dynamic weapon target assignment | |
| CN106779210A (en) | Algorithm of Firepower Allocation based on ant group algorithm | |
| CN116681223A (en) | Multi-stage combat resource collaborative allocation method based on enhanced MOEA/D | |
| CN110163502B (en) | Multi-bullet cooperative multi-stage target distribution method | |
| CN116090356B (en) | Heterogeneous warhead multi-objective task planning method based on task reliability constraint | |
| Xu et al. | MOQPSO‐D/S for Air and Missile Defense WTA Problem under Uncertainty | |
| CN108734334B (en) | Bullet and cannon combined firepower distribution method based on D number and threat priority | |
| CN108804741B (en) | D-S evidence theory cannonball combined fire power distribution method under maximum efficiency condition | |
| CN115204524B (en) | Method for generating command decision attack scheme and electronic equipment | |
| Ghanbari et al. | A survey on weapon target allocation models and applications | |
| Ruining et al. | Improved genetic algorithm for weapon target assignment problem | |
| CN115619607A (en) | Multi-stage resource attack and defense allocation method and system based on reinforcement learning | |
| CN115409351A (en) | Simplified particle swarm weapon target distribution method with priority | |
| CN114861417A (en) | Multi-stage weapon target distribution method based on variable neighborhood search | |
| Zou et al. | Solving multi-stage weapon target assignment problems by C-TAEA | |
| CN116229766B (en) | Target allocation method based on efficiency under game countermeasures | |
| CN113112079B (en) | Task allocation method based on heuristic dynamic deepening optimization algorithm | |
| CN114880857B (en) | Weapon resource multi-stage optimization distribution method based on hybrid intelligent search | |
| Mei et al. | Multi-platform Cooperative Target Assignment Method Base on Receding Horizon Control Heuristic | |
| Zhang et al. | An Improved Cuckoo Search Algorithm for Target Assignment |
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 |