CN116225049A - Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm - Google Patents
Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm Download PDFInfo
- Publication number
- CN116225049A CN116225049A CN202211651892.2A CN202211651892A CN116225049A CN 116225049 A CN116225049 A CN 116225049A CN 202211651892 A CN202211651892 A CN 202211651892A CN 116225049 A CN116225049 A CN 116225049A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- attack
- wolf
- enemy
- 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.)
- Withdrawn
Links
- 230000007123 defense Effects 0.000 title claims abstract description 75
- 241000282461 Canis lupus Species 0.000 claims abstract description 85
- 230000008901 benefit Effects 0.000 claims abstract description 52
- 230000000694 effects Effects 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 40
- 238000000034 method Methods 0.000 claims description 32
- 241000282421 Canidae Species 0.000 claims description 21
- 230000006399 behavior Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 8
- 230000009471 action Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 7
- 238000003491 array Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 6
- 238000010079 rubber tapping Methods 0.000 claims description 6
- 230000004083 survival effect Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000035899 viability Effects 0.000 claims description 3
- 230000002265 prevention Effects 0.000 claims description 2
- 238000004088 simulation Methods 0.000 description 15
- 238000011161 development Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
- G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
-
- 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)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of intelligent decision making of multiple unmanned aerial vehicles, and provides a multi-unmanned aerial vehicle wolf crowd collaborative combat attack and defense decision making algorithm. Calculating satisfaction factors of satisfaction decisions based on comprehensive advantage values, wherein the comprehensive advantage values are comprehensively obtained according to situation information of both sides of a battlefield and the performance of a fighter and target intention; and finally, calculating the income and the cost of the attack and defense decision scheme through the satisfaction factor to obtain the attack and defense decision scheme of the unmanned aerial vehicle and the enemy target. The multi-unmanned-plane-wolf-crowd cooperative combat attack and defense decision algorithm provided by the invention can play a wonder effect in a specific combat scene. The wolf group combat emphasizes the logic of role allocation and task execution, and is more suitable for cluster combat tasks with complex situations.
Description
Technical Field
The invention relates to the technical field of intelligent decision making of multiple unmanned aerial vehicles, in particular to a multi-unmanned aerial vehicle wolf crowd collaborative combat attack and defense decision making algorithm.
Background
With the development of artificial intelligence, in particular the development of intelligent unmanned aerial vehicle technology, unmanned aerial vehicle cluster systems are more and more complex and intelligent. In number, unmanned aerial vehicle clusters have evolved from simple multi-machine collaboration to hundreds of thousands of unmanned aerial vehicle collaboration. Military, unmanned aerial vehicle cluster system is from simple execution reconnaissance task to accomplish the integrative task of reconnaissance under the complex condition. In the situation of unmanned aerial vehicle formation, unmanned aerial vehicle clusters are developed from a simple interaction coordination mode of long-range planes and plane planes to autonomous coordination of self-organizing formation according to task development changes. In the aspect of human-computer interaction, unmanned plane clusters are developing towards human-computer intelligent interaction, but the current research is still in the theoretical research stage. The prior theoretical research is, for example, an unmanned cluster defense combat scheme optimizing method based on intelligent algorithm [ J ]. The military project, 2022,43 (6): 1415-1425 ], an unmanned plane cluster countermeasure multi-coupling task intelligent decision method [ J ]. The aerospace project, 2021 ] "
However, in the actual air combat application, the intelligent decision algorithm is often influenced by various uncertain factors, so that the real-time optimization is difficult to achieve, and the traditional intelligent optimization algorithm cannot meet the requirement of high air combat instantaneity.
Disclosure of Invention
In order to effectively solve the problem of complex and changeable environments facing battlefields, various complex tasks can be completed by utilizing unmanned aerial vehicle cooperative combat modes, and the unmanned aerial vehicle has excellent coordination, intelligence and autonomy, so that the clustered unmanned aerial vehicle has battlefield cognition capability and can realize complete self-organizing combat, and the invention provides a multi-unmanned aerial vehicle wolf group cooperative combat attack and defense decision algorithm. According to the invention, through simulating the behavior of the natural wolf group mutually cooperating hunting object, a novel meta-heuristic algorithm is abstracted, the algorithm is based on the natural law of 'strong survival', firstly, the head wolf unmanned aerial vehicle is fast, flexible and stealth, and the like, so that effective situation information of enemy is obtained, and tactical decision is made according to the effective situation information; and secondly, commanding other unmanned aerial vehicles (group wolves) to travel on a hiking way when a proper fighter plane is involved, quickly obtaining victory and then immediately and collectively withdrawing to wait for the next combat command of the head wolf unmanned aerial vehicle. The mode is flexible and can play a wonder effect in a specific combat scene.
The technical scheme of the invention is as follows: a multi-unmanned plane wolf group collaborative combat attack and defense decision algorithm, which calculates satisfaction factors of satisfaction decisions based on comprehensive advantage values, wherein the comprehensive advantage values are comprehensively obtained according to situation information of both sides of a battlefield, performance of a fighter and target intention; finally, calculating the income and the cost of the attack and defense decision scheme through the satisfaction factor to obtain the attack and defense decision scheme of the unmanned aerial vehicle and the enemy target;
based on the attack and defense decision-making problem of M unmanned aerial vehicles on N enemy targets, a multi-unmanned fighter plane collaborative multi-target attack and defense decision-making model is built for determining which unmanned aerial vehicle specifically attacks which enemy target and which unmanned aerial vehicle adopts a defense strategy; the target function in the multi-unmanned fighter plane collaborative multi-target attack and defense decision model is obtained according to the air combat situation dominant function, the unmanned plane performance dominant function and the target intention dominant;
the air combat situation dominance function is:
S A =k 1 P α +k 2 P d +k 3 P e
wherein P is α ,P d ,P e Respectively representing enemy target angle, enemy target distance and enemy target energy advantage, k 1 ,k 2 ,k 3 Respectively the weight coefficients;
because the types of unmanned aerial vehicles of the two parties of the fight are fewer, the performances are similar, the attack and defense decision often only considers the situation of the air combat, but the performances of the unmanned aerial vehicles are not considered, and the difference of the air combat capabilities of the two parties of the fight is more and more obvious along with the development of the technological level of different countries, so that the effect of the performances of the unmanned aerial vehicles in the air combat is more and more outstanding. The unmanned aerial vehicle performance dominance function is:
in the formula, cap i And Cap j Unmanned aerial vehicle performance, i.e., unmanned aerial vehicle performance and enemy target, respectively, is related to unmanned aerial vehicle maneuverability, striking capacity, detection capacity, maneuvering capacity, viability, voyage capacity, and electronic countermeasures;
the modern unmanned air combat is an information game process in a complex environment, a party can read battlefield information first, and the future actions of the enemy can be accurately and rapidly predicted, so that the first opportunity can be mastered in the air combat. The more dangerous the intention of the enemy machine to the me machine, the less the intention advantage of the me machine to the enemy machine. Target intention advantage takes target intention advantage value S according to intention classification of enemy plane on me plane I ;
Establishing the following objective function;
wherein x is ij ={0,1},x ij =1 indicates that the enemy unmanned aerial vehicle j is assigned to the my unmanned aerial vehicleA machine i; j (J) 1max The comprehensive advantage value is the embodiment of the air combat comprehensive advantage of the enemy unmanned aerial vehicle j by the unmanned aerial vehicle i; b (B) ij And C ij The income and the cost obtained by attacking the enemy unmanned plane j for the unmanned plane i are respectively obtained; j (J) 1max Based on the situation advantage of the air combat, the aim is to save the unmanned aerial vehicle on the my side to the maximum extent; j (J) 2max The method takes the income and the cost of the attack result as the core, and aims to ensure the effect generated by each attack; constraint conditions (1) ensure that all enemy targets are attacked; constraint condition (2) ensures that the force distribution is balanced, and the maximum number of unmanned aerial vehicles j which are allowed to cooperatively attack a specific target enemy is D j The method comprises the steps of carrying out a first treatment on the surface of the Constraint condition (3) is unmanned aerial vehicle ammunition constraint, and the maximum number of enemy targets which can be attacked simultaneously by the unmanned aerial vehicle i is smaller than the carrying capacity E of the enemy targets i ;
Based on the established multi-unmanned fighter plane cooperative multi-target attack and defense decision model, combining a wolf algorithm to perform multi-unmanned fighter plane cooperative multi-target attack and defense decision; the wolf swarm algorithm simulates hunting behaviors and hunting rules of the wolf swarm, and divides the wolf swarm into head wolves, exploring wolves and slamming wolves; setting a high-value unmanned aerial vehicle or a man-made unmanned aerial vehicle as a head wolf unmanned aerial vehicle, obtaining effective situation information of an enemy, making tactical decisions according to the effective situation information, and commanding other unmanned aerial vehicles; the unmanned aerial vehicle with the detection performance and the maneuvering performance is a wolf detection unmanned aerial vehicle, and the survival ability is set to be strong; the unmanned aerial vehicle with strong striking capability and maneuverability is a wolf unmanned aerial vehicle;
the method comprises the following specific steps:
(1) The position of the unmanned aerial vehicle i in the optimization process updates the intelligent behavior corresponding to the character classification, and the head wolf generation rule of 'winner is king' and the wolf group update mechanism of 'winner survival' are interacted to generate;
(2) A satisfaction factor; establishing a satisfaction decision based on the satisfaction factor; the satisfaction factor is set for eliminating the distribution strategy with infeasibility and low income, so that the algorithm searching efficiency is improved; optimizing satisfaction factors of satisfaction decisions by using a wolf algorithm;
on the premise of ensuring that the unmanned aerial vehicle can obtain comprehensive advantages, making a satisfactory decision; the formula (1) and the formula (2) are taken as the purposes of the wolf's group algorithmA standard function; with the position X= { X of the unmanned aerial vehicle i i1 ,x i2 ,···,x ij ,···,x m }(1≤x ij N) is less than or equal to the decision scheme of attack and defense; x is x ij =k represents an attribute value k of the target unmanned aerial vehicle j of the attack of the my unmanned aerial vehicle i, specifically whether to allocate the enemy unmanned aerial vehicle j to the my unmanned aerial vehicle for attack decision; calculating a corresponding objective function value of the enemy unmanned aerial vehicle perceived by the unmanned aerial vehicle;
according to the characteristics of the multi-unmanned fighter plane cooperative multi-target attack and defense decision-making problem, a position update formula of a wolf algorithm is defined, and the formula is as follows
The above representation randomly generates Step two-dimensional arrays (x ij ,x ik ) Where j, k=1, 2, …, L and j+.k, X is arranged in the order of the two-dimensional arrays i Exchanging the numerical value of the corresponding bit number code;
(3) In the walk behavior, the wolf-detecting unmanned aerial vehicle performs reconnaissance in h directions, namely randomly executing h times Θ (X) i ,stap a ) Wherein step is a The method comprises the steps of wandering a wolf unmanned aerial vehicle, recording perceived enemy unmanned aerial vehicle each time, and calculating a target function value;
(4) In the calling behavior, the beaten wolf unmanned aerial vehicle calls by the head wolf unmanned aerial vehicle with a step of running b The position of the head wolf unmanned aerial vehicle is closed, namely the position X of the head wolf unmanned aerial vehicle i Execution once Θ (X) i ,stap b ) The method comprises the steps of carrying out a first treatment on the surface of the The process is that the head wolf unmanned aerial vehicle guides the group of the wolf unmanned aerial vehicle while the individual characteristics of the wolf unmanned aerial vehicle are reserved;
(5) In the tapping behavior, the my unmanned aerial vehicle involved in the tapping executes Θ (X i ,stap c ) The method comprises the steps that (1) a wolf detecting unmanned aerial vehicle and a raging wolf unmanned aerial vehicle conduct a girth attack under the command of a head wolf unmanned aerial vehicle; the attack behavior can be understood as that wolves perform small-range group motion around excellent prey, perform fine search on excellent solution domain, increase population diversity and avoid calculationThe method is early maturing.
(6) In the iteration process, the unmanned aerial vehicle group continuously performs the walk, call and attack actions until the optimization accuracy requirement or the maximum iteration number k is reached max The position of the head wolf unmanned aerial vehicle is output, namely, under the constraint condition (1), the attack and defense decision scheme X for ensuring the survivability of the unmanned aerial vehicle by considering the unmanned aerial vehicle air combat situation function S Calculating the benefit B under the attack and defense decision scheme S Cost C and sum S The satisfaction factor is calculated by the following formula;
(7) After the satisfaction factor of the comprehensive advantages is obtained, a final multi-unmanned aerial vehicle cooperative multi-target attack and defense decision scheme is obtained based on a satisfaction decision; under constraint condition (2), the balance between the total attack income and the total loss cost is achieved, and in order to meet the requirement of unmanned air combat instantaneity, the balance may not be optimal, but tactical requirements can be met, and the attack task is completed.
The set of individual satisfaction for satisfaction of decision theory is defined as
Σ α ={W s (u)≥αW r (u)|u∈U}
Wherein U is a comprehensive dominance value; w (W) s (u) and W r (u) an acceptance decision function and a rejection decision function defined in a decision space, W s (u) and W r (u) are respectively
Wherein B is z And C z The method is characterized in that benefits and costs corresponding to the attack and defense decision scheme are obtained; v (V) jmax And V imax Maximum price of enemy unmanned aerial vehicle and my unmanned aerial vehicle respectivelyThe value quantity is used for normalization processing; f is penalty factor;
wherein, gamma is E (0, + -infinity) is a penalty adjustment factor, m j The number of unmanned aerial vehicles for simultaneously attacking the enemy target j; m is m j Exceeding threshold D j When the unmanned aerial vehicle is too concentrated, f can be rapidly reduced, and punishment is carried out on the behavior of the unmanned aerial vehicle for attacking a certain target.
(8) And according to the acceptance decision function and the rejection decision function, realizing the cooperative multi-objective attack and defense decision of the multi-unmanned aerial vehicle, and obtaining a final decision scheme.
The intention advantage is specifically valued as follows, and 0.1 is taken when the enemy unmanned aerial vehicle is intended to attack; when the enemy unmanned aerial vehicle is intended to be a false attack, the intended advantage is 0.3; when the enemy unmanned aerial vehicle is intended to be an electronic attack, the intended advantage is 0.4; when the enemy drone is intended to monitor, the intended advantage is 0.4; when the enemy unmanned aerial vehicle is intended to be reconnaissance, the intended advantage is 0.5; when the enemy unmanned aerial vehicle is intended to be burst prevention, the intended advantage is 0.7; when the enemy unmanned aerial vehicle intends to defend, the intention advantage is 0.9; when the enemy drone intent is unknown, the intent advantage is 0.
The high value is that the load capacity is strong, the cost is high, the flying height is high, the flying speed is high or the maneuverability is strong.
The invention has the beneficial effects that: the invention provides a multi-unmanned plane wolf group collaborative combat attack and defense decision algorithm which can play a wonder effect in a specific combat scene. The wolf group combat emphasizes the logic of role allocation and task execution, and is more suitable for cluster combat tasks with complex situations. Under the condition of taking priority on air combat advantages, a satisfaction degree factor of satisfaction decision is sought through a wolf algorithm, so that the efficiency of satisfaction decision is improved while the efficiency is ensured, and an effective balance is achieved between overall attack income and overall loss cost.
Drawings
Fig. 1 is a schematic diagram of a multi-unmanned aerial vehicle wolf crowd cooperative combat attack and defense decision-making problem.
Fig. 2 is a flowchart of a multi-unmanned aerial vehicle wolf-crowd cooperative combat attack and defense decision algorithm in the invention.
Fig. 3 is a schematic diagram of the wolf's algorithm according to the present invention.
Fig. 4 is a flowchart of unmanned aerial vehicle formation collaborative planning decisions.
FIG. 5 (a) is an air combat situation diagram of a simulation scenario;
fig. 5 (b) is a simulation scenario two-air combat situation diagram.
Detailed Description
The following describes the embodiments of the present invention further with reference to the drawings and technical schemes.
The invention is described in detail below through a multi-unmanned fighter plane collaborative multi-objective attack and defense decision model and an algorithm processing flow:
referring to fig. 1,2 and 3, the following is specific:
(1) The multi-unmanned fighter plane collaborative multi-target attack and defense decision model is as follows:
(1.1) the multi-unmanned fighter plane cooperated with the multi-target attack and defense decision-making problem, consider the attack and defense decision-making of M-frame my unmanned aerial vehicle to N enemy targets, and the principle of the multi-unmanned fighter plane cooperated with the multi-target attack and defense decision-making is shown in figure 1.
And (1.2) when the enemy unmanned aerial vehicle enters the attack range of the unmanned aerial vehicle, carrying out attack and defense decision on the unmanned aerial vehicle to form an unmanned aerial vehicle attack and defense decision scheme, wherein the decision target is to determine which unmanned aerial vehicle specifically attacks which target and which unmanned aerial vehicle adopts a defense strategy.
(1.3) in the cooperative multi-target attack and defense decision process of the multi-unmanned fighter plane, mainly considering 3 aspects of air combat situation, unmanned plane performance and target intention.
(1.4) air combat situation. Various situation factors in the unmanned air combat environment, such as changes of angles, distances, heights, tracks and the like of targets, can often directly influence the results of attack and defense decisions, and the air combat situation dominance function is designed as follows
S A =k 1 P α +k 2 P d +k 3 P e
Wherein P is α ,P d ,P e Respectively representing enemy target angle, enemy target distance and enemy target energy advantage, k 1 ,k 2 ,k 3 Is the corresponding weight coefficient.
(1.5) unmanned aerial vehicle performance. In the early stage of unmanned air combat decision, as the types of unmanned aerial vehicles of the two parties of the fight are fewer and have similar performances, the attack and defense decision often only considers the situation of the air combat without considering the performance of the unmanned aerial vehicle, but with the development of the technological level of different countries, the difference of the air combat capability of the two parties of the fight is more and more obvious, and the effect of the performance of the unmanned aerial vehicle in the air combat is also more and more outstanding. The unmanned aerial vehicle performance dominance function is designed as:
in the formula, cap i And Cap j Unmanned aerial vehicle performance, i.e., unmanned aerial vehicle performance and enemy target, respectively, relates to unmanned aerial vehicle maneuverability, striking capacity, detection capacity, maneuvering capacity, viability, voyage capacity, and electronic countermeasures.
(1.6) target intention. The modern unmanned air combat is an information game process in a complex environment, a party can read battlefield information first, and the future actions of the enemy can be accurately and rapidly predicted, so that the first opportunity can be mastered in the air combat. The more dangerous the intention of the enemy machine to the me machine, the less the intention advantage of the me machine to the enemy machine.
(2) The multi-unmanned fighter plane cooperative multi-target attack and defense decision algorithm flow is as follows:
(2.1) wolf's algorithm. The wolf swarm algorithm simulates hunting behavior and hunting rules of the wolf swarm, and divides the wolf swarm into a head wolf, a detecting wolf and a slash wolf, and the essence of the wolf swarm algorithm is that the artificial wolf is continuously updated according to the self and other wolves, so as to approach to the optimal position.
(2.2) optimization procedure. The new position of artificial wolves is the result of 3 intelligent actions of wolves wandering away by the exploring wolves, head wolves summoning and slaving the slaving wolves, and the interaction of the head wolves 'winner' rule of production and the wolf update mechanism 'survivors'. The principle of which is shown in figure 3;
(2.3) a satisfaction factor. The satisfaction decision eliminates the distribution strategy with infeasibility and low profit by setting the satisfaction factor, thereby improving the algorithm searching efficiency. This section uses a wolf algorithm to optimize the satisfaction factor for the satisfaction decision. In order to determine the satisfaction factor, a satisfaction decision is made on the premise of ensuring that the unmanned aerial vehicle can obtain the air combat advantage. With the position X= { X of the artificial wolf i i1 ,x i2 ,···,x ij ,···,x m }(1≤x ij N) is less than or equal to the decision scheme of attack and defense, x ij The value k of the attribute of the target unmanned aerial vehicle j of the unmanned aerial vehicle j, specifically whether to assign the enemy unmanned aerial vehicle j to the unmanned aerial vehicle for attack decision. The artificial wolf perceives the corresponding objective function value of the enemy unmanned aerial vehicle, and defines the position updating formula of the wolf algorithm according to the actual characteristics of the multi-unmanned fighter aircraft cooperative multi-objective attack and defense decision-making problem, as follows
The above representation randomly generates Step two-dimensional arrays (x ij ,x ik ) Where j, k=1, 2, …, L and j+.k, and arranging X in the order of the two-dimensional arrays i The values of the corresponding bit number codes are exchanged.
(2.4) in the wander behavior, the wolf-penetrating unmanned aerial vehicle heuristically scouts in h directions, i.e., randomly executes h times Θ (X i ,stap a ) Wherein step is a The method comprises the steps of walking the wolf unmanned aerial vehicle, recording the perceived enemy unmanned aerial vehicle each time, and calculating the objective function value. This process can be understood as the wolf's heuristically searching for hunting.
(2.5) in the action of calling, the first wolf of the Sum prayer is quickly called with a larger step size tap b The position of the head wolves is closed, namely the position X of the high wolves unmanned plane i Execution once Θ (X) i ,stap b ) The method comprises the steps of carrying out a first treatment on the surface of the The process can be understood as that the excellent individuals of the wolves, namely the head wolves, are unmanned while the characteristics of the individuals of the wolves are maintainedThe machine guides the group of the wolf unmanned aerial vehicle.
(2.6) in the tapping behavior, the My unmanned aerial vehicle participating in the tapping executes Θ (X) i ,stap c ) The method is characterized in that the wolf detecting unmanned aerial vehicle and the raging wolf unmanned aerial vehicle conduct the attack behavior under the command of the head wolf unmanned aerial vehicle. The attack behavior can be understood as that wolves perform small-range group motions around excellent prey, perform fine search on excellent solution domains, increase population diversity and avoid premature algorithms.
(2.7) in the iterative process, the unmanned aerial vehicle group continuously performs the walk, call and attack actions until reaching the optimization precision requirement or the maximum iterative times k max Outputting the position of the head wolf, namely, under the first objective function, giving priority to the optimal decision scheme X of the air combat dominance function of the unmanned aerial vehicle to ensure the survivability of the unmanned aerial vehicle S Calculating the benefit B under the attack and defense decision scheme S Cost C and sum S Then the satisfaction factor may be calculated by
(2.8) obtaining a final multi-unmanned aerial vehicle cooperative multi-target attack and defense decision scheme based on a satisfaction decision after obtaining a satisfaction factor of comprehensive advantages; under the second objective function, an effective balance is achieved between the total attack income and the total loss cost, and in order to meet the requirement of unmanned air combat instantaneity, the balance may not be optimal, but tactical requirements can be met, and the attack task is completed.
The individual satisfaction set of satisfaction decision theory is defined as
Σ α ={W s (u)≥αW r (u)|u∈U}
Wherein W is s (u) and W r (u) an acceptance decision function and a rejection decision function defined in a decision space, W s (u) and W r (u) are respectively designed as
Wherein B is z And C z V for the corresponding benefits and costs of the attack and defense decision scheme jmax And V imax The maximum value of the enemy and the unmanned aerial vehicle are respectively used for normalization processing, f is a punishment factor, punishs the excessively concentrated distribution result, and is designed as
Wherein, gamma is E (0, + -infinity) is a penalty adjustment factor, m j For the number of unmanned aerial vehicles simultaneously attacking enemy target j, when m j Exceeding threshold D j When the unmanned aerial vehicle is too concentrated, f can be rapidly reduced, and punishment is carried out on the behavior of the unmanned aerial vehicle for attacking a certain target;
and (2.9) realizing multi-unmanned aerial vehicle cooperative multi-objective attack and defense decision according to the acceptance decision function and the rejection decision function, and obtaining a final decision scheme.
In order to verify the effectiveness of the multi-unmanned aerial vehicle wolf-crowd collaborative combat attack and defense decision algorithm, numerical simulation research is carried out on the multi-unmanned aerial vehicle wolf-crowd collaborative combat attack and defense decision algorithm.
Considering that 4 unmanned aerial vehicles attack 6 enemy targets, the air combat situations of the simulation scene 1 and the simulation scene 2 are respectively shown in fig. 5 (a) and fig. 5 (b);
firstly, a satisfaction factor of a satisfaction decision is calculated based on a dominance function, and comprehensive dominance values of the simulation scene 1 and the simulation scene 2 on enemy fighter plane are shown in the table 1 and the table 2 respectively according to situation information of enemy and me parties in a battlefield, performance of the fighter plane and target intention.
Table 1 Emulation scenario 1 comprehensive dominance value of My fighter against enemy fighter
T1 | T2 | T3 | | T5 | T6 | ||
1 | 0.3092 | 0.2864 | 0.2516 | 0.3722 | 0.2642 | 0.2625 | |
2 | 0.5609 | 0.7534 | 0.3754 | 0.6648 | 0.4906 | 0.4068 | |
3 | 0.3057 | 0.3943 | 0.2529 | 0.3781 | 0.2705 | 0.2526 | |
4 | 0.5531 | 0.7700 | 0.3760 | 0.6330 | 0.4880 | 0.4231 |
Table 2 Emulation scenario 2 comprehensive dominance value of My fighter against enemy fighter
And setting the wolf group scale as 50, and setting the maximum iteration number of the algorithm as kmax=50, wherein the decision scheme of the attack and defense of the unmanned aerial vehicle and the enemy target is shown in the table.
Table 3 attack and defense decision scheme based on comprehensive dominance values
For simulation scenario 1, the benefits and costs under this attack and defense decision scheme can be calculated as follows: b (B) S1 =2.6312,C S1 = 1.9389. For simulation scenario 2, the benefits and costs under this attack and defense decision scheme can be calculated as follows: b (B) S2 =2.5787,C S2 =1.9250
The final attack and defense decision scheme is shown in the table after satisfactory decision
TABLE 4 final attack and defense decision scheme
Under the final attack and defense decision scheme, the income and the cost of the simulation scene 1 are respectively B 1 =2.5882,C 1 = 1.8850; the benefits and costs of simulation scenario 2 are B 2 =2.6177,C 2 =1.8106。
Simulation results show that in the simulation scene 1, the unmanned plane U1 attacks the targets T1, T3, U2 attacks T6, T4, and U3 and U4 attack T2 and T5 respectively; in the simulation scene 2, the attacks T1, T2 and T3 of the unmanned aerial vehicle U2 and the attacks T5, T6 and T3 and T4 are distributed to the U4, and the U1 is unfavorable in the environment, selects a defending strategy to save own strength, and does not accord with the actual situation in the attack simulation result of the enemy aircraft.
Claims (3)
1. A multi-unmanned plane wolf-crowd cooperative combat attack and defense decision algorithm is characterized in that the multi-unmanned plane wolf-crowd cooperative combat attack and defense decision algorithm calculates satisfaction factors of satisfactory decisions based on comprehensive advantage values, and the comprehensive advantage values are comprehensively obtained according to situation information of both sides of a battlefield and the performance of a fighter and target intention; finally, calculating the income and the cost of the attack and defense decision scheme through the satisfaction factor to obtain the attack and defense decision scheme of the unmanned aerial vehicle and the enemy target;
based on the attack and defense decision-making problem of M unmanned aerial vehicles on N enemy targets, a multi-unmanned fighter plane collaborative multi-target attack and defense decision-making model is built for determining which unmanned aerial vehicle specifically attacks which enemy target and which unmanned aerial vehicle adopts a defense strategy; the target function in the multi-unmanned fighter plane collaborative multi-target attack and defense decision model is obtained according to the air combat situation dominant function, the unmanned plane performance dominant function and the target intention dominant;
the air combat situation dominance function is:
S A =k 1 P α +k 2 P d +k 3 P e
wherein P is α ,P d ,P e Respectively representing enemy target angle, enemy target distance and enemy target energy advantage, k 1 ,k 2 ,k 3 Respectively the weight coefficients;
the unmanned aerial vehicle performance dominance function is:
in the formula, cap i And Cap j Unmanned aerial vehicle performance, i.e., unmanned aerial vehicle performance and enemy target, respectively, is related to unmanned aerial vehicle maneuverability, striking capacity, detection capacity, maneuvering capacity, viability, voyage capacity, and electronic countermeasures;
target intention advantage takes target intention advantage value S according to intention classification of enemy plane on me plane I ;
Establishing the following objective function;
wherein x is ij ={0,1},x ij =1 denotes allocation of enemy drone j to my drone i; j (J) 1max The comprehensive advantage value is the embodiment of the air combat comprehensive advantage of the enemy unmanned aerial vehicle j by the unmanned aerial vehicle i; b (B) ij And C ij The income and the cost obtained by attacking the enemy unmanned plane j for the unmanned plane i are respectively obtained; j (J) 1max Based on the situation advantage of the air combat, the aim is to save the unmanned aerial vehicle on the my side to the maximum extent; j (J) 2max The method takes the income and the cost of the attack result as the core, and aims to ensure the effect generated by each attack; constraint conditions (1) ensure that all enemy targets are attacked; constraint condition (2) ensures that the force distribution is balanced, and the maximum number of unmanned aerial vehicles j which are allowed to cooperatively attack a specific target enemy is D j The method comprises the steps of carrying out a first treatment on the surface of the Constraint condition (3) is unmanned aerial vehicle ammunition constraint, and the maximum number of enemy targets which can be attacked simultaneously by the unmanned aerial vehicle i is smaller than the carrying capacity E of the enemy targets i ;
Based on the established multi-unmanned fighter plane cooperative multi-target attack and defense decision model, combining a wolf algorithm to perform multi-unmanned fighter plane cooperative multi-target attack and defense decision; the wolf swarm algorithm simulates hunting behaviors and hunting rules of the wolf swarm, and divides the wolf swarm into head wolves, exploring wolves and slamming wolves; setting a high-value unmanned aerial vehicle or a man-made unmanned aerial vehicle as a head wolf unmanned aerial vehicle, obtaining effective situation information of an enemy, making tactical decisions according to the effective situation information, and commanding other unmanned aerial vehicles; the unmanned aerial vehicle with the detection performance and the maneuvering performance is a wolf detection unmanned aerial vehicle, and the survival ability is set to be strong; the unmanned aerial vehicle with strong striking capability and maneuverability is a wolf unmanned aerial vehicle;
the method comprises the following specific steps:
(1) The position of the unmanned aerial vehicle i in the optimization process updates the intelligent behavior corresponding to the character classification, and the head wolf generation rule of 'winner is king' and the wolf group update mechanism of 'winner survival' are interacted to generate;
(2) A satisfaction factor; establishing a satisfaction decision based on the satisfaction factor; the satisfaction factor is set for eliminating the distribution strategy with infeasibility and low income, so that the algorithm searching efficiency is improved; optimizing satisfaction factors of satisfaction decisions by using a wolf algorithm;
on the premise of ensuring that the unmanned aerial vehicle can obtain comprehensive advantages, making a satisfactory decision; taking the formula (1) and the formula (2) as objective functions of the wolf's group algorithm; with the position X= { X of the unmanned aerial vehicle i i1 ,x i2 ,···,x ij ,···,x m }(1≤x ij N) is less than or equal to the decision scheme of attack and defense; x is x ij =k represents an attribute value k of the target unmanned aerial vehicle j of the attack of the my unmanned aerial vehicle i, specifically whether to allocate the enemy unmanned aerial vehicle j to the my unmanned aerial vehicle for attack decision; calculating a corresponding objective function value of the enemy unmanned aerial vehicle perceived by the unmanned aerial vehicle;
according to the characteristics of the multi-unmanned fighter plane cooperative multi-target attack and defense decision-making problem, a position update formula of a wolf algorithm is defined, and the formula is as follows
The above representation randomly generates Step two-dimensional arrays (x ij ,x ik ) Where j, k=1, 2, …, L and j+.k, X is arranged in the order of the two-dimensional arrays i Exchanging the numerical value of the corresponding bit number code;
(3) In the walk behavior, the wolf-detecting unmanned aerial vehicle performs reconnaissance in h directions, namely randomly executing h times Θ (X) i ,stap a ) Wherein step is a The method comprises the steps of wandering a wolf unmanned aerial vehicle, recording perceived enemy unmanned aerial vehicle each time, and calculating a target function value;
(4) In the calling behavior, the beaten wolf unmanned aerial vehicle calls by the head wolf unmanned aerial vehicle with a step of running b The position of the head wolf unmanned aerial vehicle is closed, namely the position X of the head wolf unmanned aerial vehicle i Execution once Θ (X) i ,stap b ) The method comprises the steps of carrying out a first treatment on the surface of the The process is that the head wolf unmanned aerial vehicle guides the group of the wolf unmanned aerial vehicle while the individual characteristics of the wolf unmanned aerial vehicle are reserved;
(5) In the tapping behavior, the my unmanned aerial vehicle involved in the tapping executes Θ (X i ,stap c ) The method comprises the steps that (1) a wolf detecting unmanned aerial vehicle and a raging wolf unmanned aerial vehicle conduct a girth attack under the command of a head wolf unmanned aerial vehicle;
(6) In the iteration process, the unmanned aerial vehicle group continuously performs the walk, call and attack actions until the optimization accuracy requirement or the maximum iteration number k is reached max The position of the head wolf unmanned aerial vehicle is output, namely, under the constraint condition (1), the attack and defense decision scheme X for ensuring the survivability of the unmanned aerial vehicle by considering the unmanned aerial vehicle air combat situation function S Calculating the benefit B under the attack and defense decision scheme S Cost C and sum S The satisfaction factor is calculated by the following formula;
(7) After the satisfaction factor of the comprehensive advantages is obtained, a final multi-unmanned aerial vehicle cooperative multi-target attack and defense decision scheme is obtained based on a satisfaction decision; under constraint (2), a balance is achieved between overall attack gain and overall loss cost, and an individual satisfaction set for satisfying the decision theory is defined as
Σ α ={W s (u)≥αW r (u)|u∈U}
Wherein U is a comprehensive dominance value; w (W) s (u) and W r (u) an acceptance decision function and a rejection decision function defined in a decision space, W s (u) and W r (u) are respectively
Wherein B is z And C z The method is characterized in that benefits and costs corresponding to the attack and defense decision scheme are obtained; v (V) jmax And V imax The maximum value amounts of the enemy unmanned aerial vehicle and the my unmanned aerial vehicle are respectively used for normalization processing; f is penalty factor;
wherein, gamma is E (0, + -infinity) is a penalty adjustment factor, m j The number of unmanned aerial vehicles for simultaneously attacking the enemy target j; (8) And according to the acceptance decision function and the rejection decision function, realizing the cooperative multi-objective attack and defense decision of the multi-unmanned aerial vehicle, and obtaining a final decision scheme.
2. The multi-unmanned aerial vehicle wolf crowd cooperative combat attack and defense decision algorithm according to claim 1, wherein the intention dominance is specifically valued as follows, and 0.1 is taken when an enemy unmanned aerial vehicle is intended to attack; when the enemy unmanned aerial vehicle is intended to be a false attack, the intended advantage is 0.3; when the enemy unmanned aerial vehicle is intended to be an electronic attack, the intended advantage is 0.4; when the enemy drone is intended to monitor, the intended advantage is 0.4; when the enemy unmanned aerial vehicle is intended to be reconnaissance, the intended advantage is 0.5; when the enemy unmanned aerial vehicle is intended to be burst prevention, the intended advantage is 0.7;
when the enemy unmanned aerial vehicle intends to defend, the intention advantage is 0.9; when the enemy drone intent is unknown, the intent advantage is 0.
3. The multi-unmanned aerial vehicle wolf-crowd cooperative combat attack and defense decision algorithm according to claim 1 or 2, wherein the high value is advantageous in terms of high load capacity, high cost, high flying height, high flying speed or high maneuverability.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211651892.2A CN116225049A (en) | 2022-12-21 | 2022-12-21 | Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211651892.2A CN116225049A (en) | 2022-12-21 | 2022-12-21 | Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116225049A true CN116225049A (en) | 2023-06-06 |
Family
ID=86579449
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211651892.2A Withdrawn CN116225049A (en) | 2022-12-21 | 2022-12-21 | Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116225049A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116596287A (en) * | 2023-07-18 | 2023-08-15 | 中国电子科技集团公司第二十九研究所 | Task driving decision-making method and system |
CN116860395A (en) * | 2023-07-13 | 2023-10-10 | 四川大学 | Distributed layered contract net firepower distribution method |
CN117313972A (en) * | 2023-09-28 | 2023-12-29 | 烟台大学 | Attack method, system and device for unmanned ship cluster and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112284188A (en) * | 2020-09-25 | 2021-01-29 | 南京航空航天大学 | Unmanned aerial vehicle attack and defense decision method based on self-adaptive step discrete wolf cluster algorithm |
WO2021174765A1 (en) * | 2020-03-03 | 2021-09-10 | 中国科学院自动化研究所 | Control system based on multi-unmanned-aerial-vehicle collaborative game confrontation |
CN113625739A (en) * | 2021-08-25 | 2021-11-09 | 中国航空工业集团公司沈阳飞机设计研究所 | Expert system optimization method based on heuristic maneuver selection algorithm |
CN113741508A (en) * | 2021-06-29 | 2021-12-03 | 南京航空航天大学 | Improved wolf colony algorithm-based unmanned aerial vehicle task allocation method |
CN114442662A (en) * | 2022-01-05 | 2022-05-06 | 南京航空航天大学 | Improved wolf colony optimization algorithm-based unmanned aerial vehicle cluster cooperative ground strike method |
CN114995500A (en) * | 2022-06-10 | 2022-09-02 | 中国航空工业集团公司沈阳飞机设计研究所 | Multi-machine formation cooperative target allocation method |
-
2022
- 2022-12-21 CN CN202211651892.2A patent/CN116225049A/en not_active Withdrawn
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021174765A1 (en) * | 2020-03-03 | 2021-09-10 | 中国科学院自动化研究所 | Control system based on multi-unmanned-aerial-vehicle collaborative game confrontation |
CN112284188A (en) * | 2020-09-25 | 2021-01-29 | 南京航空航天大学 | Unmanned aerial vehicle attack and defense decision method based on self-adaptive step discrete wolf cluster algorithm |
CN113741508A (en) * | 2021-06-29 | 2021-12-03 | 南京航空航天大学 | Improved wolf colony algorithm-based unmanned aerial vehicle task allocation method |
CN113625739A (en) * | 2021-08-25 | 2021-11-09 | 中国航空工业集团公司沈阳飞机设计研究所 | Expert system optimization method based on heuristic maneuver selection algorithm |
CN114442662A (en) * | 2022-01-05 | 2022-05-06 | 南京航空航天大学 | Improved wolf colony optimization algorithm-based unmanned aerial vehicle cluster cooperative ground strike method |
CN114995500A (en) * | 2022-06-10 | 2022-09-02 | 中国航空工业集团公司沈阳飞机设计研究所 | Multi-machine formation cooperative target allocation method |
Non-Patent Citations (1)
Title |
---|
周同乐 等: "基于狼群算法的多无人机协同多目标攻防满意决策方法", 《指挥与控制学报》, vol. 6, no. 3, pages 251 - 256 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116860395A (en) * | 2023-07-13 | 2023-10-10 | 四川大学 | Distributed layered contract net firepower distribution method |
CN116860395B (en) * | 2023-07-13 | 2024-06-11 | 四川大学 | Distributed layered contract net firepower distribution method |
CN116596287A (en) * | 2023-07-18 | 2023-08-15 | 中国电子科技集团公司第二十九研究所 | Task driving decision-making method and system |
CN116596287B (en) * | 2023-07-18 | 2023-10-03 | 中国电子科技集团公司第二十九研究所 | Task driving decision-making method and system |
CN117313972A (en) * | 2023-09-28 | 2023-12-29 | 烟台大学 | Attack method, system and device for unmanned ship cluster and storage medium |
CN117313972B (en) * | 2023-09-28 | 2024-04-12 | 烟台大学 | Attack method, system and device for unmanned ship cluster and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112861442B (en) | Multi-machine collaborative air combat planning method and system based on deep reinforcement learning | |
CN108680063B (en) | A kind of decision-making technique for extensive unmanned plane cluster dynamic confrontation | |
CN116225049A (en) | Multi-unmanned plane wolf-crowd collaborative combat attack and defense decision algorithm | |
CN111240353B (en) | Unmanned aerial vehicle collaborative air combat decision method based on genetic fuzzy tree | |
CN112269396B (en) | Unmanned aerial vehicle cluster cooperative confrontation control method for eagle pigeon-imitated intelligent game | |
CN110083971B (en) | Self-explosion unmanned aerial vehicle cluster combat force distribution method based on combat deduction | |
CN111797966B (en) | Multi-machine collaborative global target distribution method based on improved flock algorithm | |
CN114330115B (en) | Neural network air combat maneuver decision-making method based on particle swarm search | |
CN114460959A (en) | Unmanned aerial vehicle group cooperative autonomous decision-making method and device based on multi-body game | |
CN113962012B (en) | Unmanned aerial vehicle countermeasure strategy optimization method and device | |
CN113159266B (en) | Air combat maneuver decision method based on sparrow searching neural network | |
CN108459616A (en) | Unmanned aerial vehicle group collaboration covering Route planner based on artificial bee colony algorithm | |
CN113052289A (en) | Unmanned ship cluster striking position selection method based on game theory | |
CN112284188A (en) | Unmanned aerial vehicle attack and defense decision method based on self-adaptive step discrete wolf cluster algorithm | |
CN113625767A (en) | Fixed-wing unmanned aerial vehicle cluster collaborative path planning method based on preferred pheromone gray wolf algorithm | |
CN112305913A (en) | Multi-UUV collaborative dynamic maneuver decision method based on intuitive fuzzy game | |
Liu et al. | Multi‐UUV Cooperative Dynamic Maneuver Decision‐Making Algorithm Using Intuitionistic Fuzzy Game Theory | |
CN113887919A (en) | Hybrid-discrete particle swarm algorithm-based multi-unmanned aerial vehicle cooperative task allocation method and system | |
Xianyong et al. | Research on maneuvering decision algorithm based on improved deep deterministic policy gradient | |
Wang et al. | An efficient clonal selection algorithm to solve dynamicweapon-target assignment game model in UAV cooperative aerial combat | |
CN117291254A (en) | Agent task allocation training method based on imitation learning and safety reinforcement learning | |
CN114167899B (en) | Unmanned plane bee colony collaborative countermeasure decision-making method and system | |
Özpala et al. | Multiple UAV engagement decision by game theory | |
Gao et al. | A Sequential Decision Algorithm of Reinforcement Learning for Composite Action Space | |
Sun et al. | Task allocation in multi-AUV dynamic game based on interval ranking under uncertain information |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230606 |