Attack and defense strategy game method and system for multi-unmanned aerial vehicle target allocation

Abstract

The invention provides an attack and defense strategy game method and system for multi-unmanned aerial vehicle target allocation. Multi-unmanned aerial vehicle formation information of the first unmanned aerial vehicle formation and the second unmanned aerial vehicle formation is acquired, a strategy pair of multiple unmanned aerial vehicles is generated according to the unmanned aerial vehicle information, a strategy group set of the multi-unmanned aerial vehicle formation is generated according to the strategy pair, and the strategy pair can comprise a current unmanned aerial vehicle, an attack and defense strategy and a strategy target; the attack and defense strategies comprise an attack strategy, an avoidance strategy and an interference strategy. The probability distribution of each unmanned aerial vehicle under different attack and defense strategies is calculated, so that an income matrix is constructed, and a zero-sum game matrix model is further generated. And based on the zero-sum game matrix model, solving an attack and defense strategy game scheme of multi-unmanned aerial vehicle target allocation. According to the method, three attack and defense strategies of the unmanned aerial vehicle are fully considered, and the attack and defense strategy game scheme is solved, so that the solving quality of the target allocation problem of the unmanned aerial vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to an attack and defense strategy game method and system for multi-unmanned aerial vehicle target allocation. Background Art

[0002] With the improvement of the intelligence level and performance of drones, multiple drones can execute multiple combat strategies during the entire air combat process, which will have an important impact on the target allocation results and thus affect the combat effectiveness. During the combat process, the opposing parties need to make reasonable offensive and defensive strategies according to the real-time confrontation situation, choose to attack targets, avoid or interfere with attacks from opponents. Appropriate attack decision-making plans can reasonably allocate limited weapon resources and better carry out firepower strikes; interfering with targets can reduce the attack effectiveness of their weapons, thereby reducing the damage to one's own drones; effective avoidance based on threat assessment of targets can prevent one's own drones from being damaged. Similarly, opponents will also attack, avoid or interfere. How to set up target allocation for collaborative operations for drone formations is a major problem at present.

[0003] Most of the existing multi-UAV target allocation methods construct the problem as an optimization model, which does not consider the impact of the opponent's behavior and only optimizes the own side. The actual application effect is not ideal. In addition, in the multi-UAV target allocation problem considering the attack and defense strategy, since the specific behavior strategy of the UAV is also involved, the strategy space of the problem is significantly increased. When the number of UAVs is large, it is difficult for the existing methods to find an effective solution within a reasonable time, and sometimes even no results can be obtained. Therefore, the solution quality of the existing multi-UAV target allocation method is poor. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides an attack and defense strategy game method and system for multi-UAV target allocation, which solves the technical problem of low quality in the prior art when solving the target allocation of multiple UAVs.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] The present invention solves the technical problem by providing an attack and defense strategy game method for multi-UAV target allocation, wherein the attack and defense strategy game method is executed by a computer and comprises the following steps:

[0009] Acquire multi-UAV formation information of the first UAV formation and the second UAV formation; the multi-UAV formation information includes UAV information;

[0010] Generate strategy pairs of multiple drones according to the drone information, and generate a strategy group set of multiple drone formations according to the strategy pairs; the strategy pairs include the current drone number, the attack and defense strategy, and the strategy target; the attack and defense strategy includes the attack strategy, the avoidance strategy, and the interference strategy; the strategy group includes the strategy pairs of multiple drones in the multi-drone formation;

[0011] Obtaining the strategy probability of each attack and defense strategy of the plurality of drones respectively, and calculating the profit matrix according to the strategy probability;

[0012] Generate a zero-sum game matrix model; the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation;

[0013] The zero-sum game matrix model is used to solve the attack and defense strategy game scheme for multi-UAV target allocation.

[0014] Preferably, calculating the profit matrix according to the strategy probability includes:

[0015] Construct a strategy parameter matrix; the strategy parameter matrix includes a first strategy parameter matrix of the first UAV formation and a second strategy parameter matrix of the second UAV formation;

[0016] Calculate a first expected survival value of a first UAV formation according to the strategy parameter matrix and the strategy probability; calculate a second expected survival value of a second UAV formation according to the strategy parameter matrix and the strategy probability;

[0017] Calculate the benefit value of the first UAV formation according to the first expected survival value and the second expected survival value;

[0018] A profit matrix is ​​generated according to the profit values.

[0019] Preferably, the first strategy parameter matrix includes a first attack strategy parameter matrix, a first avoidance strategy parameter matrix and a first interference strategy parameter matrix;

[0020] The first attack strategy parameter matrix is:

[0021]

[0022] in,

[0023] a ij represents the attack parameter of drone i in the first drone formation against drone j in the second drone formation;

[0024] m represents the number of drones in the first drone formation, and n represents the number of drones in the second drone formation;

[0025] represents the attack and defense strategy adopted by drone i; They represent attack strategy, avoidance strategy and interference strategy respectively; It indicates the strategic target that drone i targets using the attack and defense strategy;

[0026] The first avoidance strategy parameter matrix is:

[0027]

[0028] in,

[0029] h i represents the avoidance parameter of UAV i to the second UAV formation;

[0030] The first interference strategy parameter matrix is:

[0031]

[0032] in,

[0033] l ij represents the interference parameter of UAV i to UAV j.

[0034] Preferably, the second strategy parameter matrix includes a second attack strategy parameter matrix, a second avoidance strategy parameter matrix and a second interference strategy parameter matrix;

[0035] The second attack strategy parameter matrix is:

[0036]

[0037] in,

[0038] b ji represents the attack parameter of drone j on drone i;

[0039] represents the attack and defense strategy adopted by UAV j, They represent attack strategy, avoidance strategy and interference strategy respectively; represents the strategic goal of UAV j;

[0040] The second avoidance strategy parameter matrix is:

[0041]

[0042] in,

[0043] h j represents the avoidance parameter of UAV j to the first UAV formation;

[0044] The second interference strategy parameter matrix is:

[0045]

[0046] in,

[0047] l ji represents the interference parameter of UAV j to UAV i.

[0048] Preferably, calculating the first expected survival value of the first UAV formation according to the strategy parameter matrix and the strategy probability includes:

[0049]

[0050] in,

[0051] u R (s R ,s B ) represents the first expected survival value of the first UAV formation;

[0052] v i represents the value of drone i in the first drone formation;

[0053] q i represents the probability of UAV i avoiding the second UAV formation, t ij represents the probability of interference from UAV i to UAV j, p ji It represents the killing probability of UAV j to UAV i.

[0054] Preferably, calculating the second expected survival value of the second UAV formation according to the strategy parameter matrix and the strategy probability includes:

[0055]

[0056] in,

[0057] u B (s R ,s B ) represents the second expected survival value of the second UAV formation;

[0058] v j represents the value of drone j in the second drone formation;

[0059] q j represents the probability of UAV j avoiding the first UAV formation, t ji represents the interference probability of UAV j to UAV i, p ij It represents the killing probability of UAV i against UAV j.

[0060] Preferably, the profit matrix includes:

[0061]

[0062] u(s R ,s B )=u R (s R ,s B )-u B (s R ,s B )

[0063] Among them, S R represents the first UAV formation strategy group set, s R ∈S R represents a strategy group of the first UAV formation, S B represents the second UAV formation strategy set, s B ∈S B represents a strategy group of the second UAV formation;

[0064] (s R ,s B ) indicates that the first UAV formation adopts strategy group s R , the second drone formation adopts the strategy group s B Strategy group combination; u(s R ,s B ) is the first UAV formation in the strategy group combination (s R ,s B ) under the profit value.

[0065] Preferably, solving the attack and defense strategy game scheme for multi-UAV target allocation according to the zero-sum game matrix model includes:

[0066] Acquire a strict game set based on the zero-sum game matrix model;

[0067] Solving a Nash equilibrium solution based on the strict game set, and selecting a first initial strategy group of the first drone formation and a second initial strategy group of the second drone formation based on the Nash equilibrium solution;

[0068] Calculate a first optimal game strategy group that maximizes the marginal return of the first drone formation based on the second initial strategy group, and calculate a second optimal game strategy group that maximizes the marginal return of the second drone formation based on the first initial strategy group;

[0069] If the first optimal game strategy group and / or the second optimal game strategy group do not exist in the strict game set, updating the strict game set based on the first optimal game strategy group and / or the second optimal game strategy group, and re-executing the step of solving the Nash equilibrium solution based on the strict game set based on the updated strict game set;

[0070] If both the first optimal game strategy group and the second optimal game strategy group exist in the strict game set, the Nash equilibrium solution is determined as the attack and defense strategy game scheme for multi-UAV target allocation.

[0071] The present invention solves the technical problem by providing an attack and defense strategy game system for multi-UAV target allocation, the system comprising:

[0072] An information acquisition module is configured to acquire multi-UAV formation information of the first UAV formation and the second UAV formation; the multi-UAV formation information includes UAV information;

[0073] a strategy generation module, configured to generate strategy pairs of multiple drones according to the drone information, and generate a strategy group set of a multi-drone formation according to the strategy pairs; the strategy pairs include a current drone number, an attack and defense strategy, and a strategy target; the attack and defense strategy includes an attack strategy, an avoidance strategy, and an interference strategy; the strategy group includes strategy pairs of multiple drones in the multi-drone formation;

[0074] A benefit matrix generation module is configured to respectively obtain the strategy probability of each attack and defense strategy of the plurality of drones, and generate a benefit matrix according to the strategy probability;

[0075] A model generation module is configured to generate a zero-sum game matrix model; the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation;

[0076] The scheme allocation module is configured to solve the attack and defense strategy game scheme for multi-UAV target allocation according to the zero-sum game matrix model.

[0077] The present invention solves its technical problem by providing a computer-readable storage medium, which stores a computer program for the attack and defense strategy game of multi-UAV target allocation, wherein the computer program enables a computer to execute the attack and defense strategy game method of multi-UAV target allocation as described above.

[0078] (III) Beneficial effects

[0079] The present invention provides an attack and defense strategy game method and system for multi-UAV target allocation. Compared with the prior art, it has the following beneficial effects:

[0080] The present invention obtains the multi-UAV formation information of the first UAV formation and the second UAV formation to generate a strategy pair of multiple UAVs according to the UAV information, and generates a strategy group set of the multi-UAV formation according to the strategy pair. The strategy pair may include the current UAV, the attack and defense strategy and the strategy target; the attack and defense strategy includes the attack strategy, the avoidance strategy and the interference strategy. By calculating the probability distribution of each UAV under different attack and defense strategies, a profit matrix is ​​constructed, and a zero-sum game matrix model is further generated. Based on the zero-sum game matrix model, the attack and defense strategy game scheme of multi-UAV target allocation is solved. The present invention fully considers the three attack and defense strategies of the UAV, solves the attack and defense strategy game scheme, and thus improves the solution quality of the UAV target allocation problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0082] Figure 1 A schematic diagram of a scenario of an attack and defense strategy game method for multi-UAV target allocation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0084] The embodiments of the present application provide an attack and defense strategy game method and system for multi-UAV target allocation, thereby solving the problem of low solution quality of multi-UAV target allocation in the prior art and improving the solution quality of UAV target allocation.

[0085] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0086] An embodiment of the present invention provides an attack and defense strategy game method for multi-UAV target allocation, which is executed by a computer and includes the following steps:

[0087] S1. Acquire multi-UAV formation information of a first UAV formation and a second UAV formation; the multi-UAV formation information includes UAV information;

[0088] S2. Generate strategy pairs of multiple drones according to the drone information, and generate a strategy group set of multiple drone formations according to the strategy pairs; the strategy pairs include the current drone number, attack and defense strategies, and strategy targets; the attack and defense strategies include attack strategies, avoidance strategies, and interference strategies; the strategy group includes strategy pairs of multiple drones in the multi-drone formation;

[0089] S3, respectively obtaining the strategy probability of each attack and defense strategy of the plurality of drones, and calculating the profit matrix according to the strategy probability;

[0090] S4, constructing a zero-sum game matrix model; the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation;

[0091] S5. Solve the attack and defense strategy game scheme for multi-UAV target allocation based on the zero-sum game matrix model.

[0092] Figure 1 This is a schematic diagram of a scenario of the attack and defense strategy game method for multi-UAV target allocation provided by an embodiment of the present invention. Each step is analyzed in detail below.

[0093] In step S1, multi-UAV formation information of the first UAV formation and the second UAV formation is obtained, wherein the multi-UAV formation information includes UAV information.

[0094] P = {R, B} represents the multi-drone formation information, characterizing the participants in the game. The two sides of the game include two multi-drone formations, which are respectively referred to as the red side and the blue side in the embodiment of the present application. R represents the red side, which is referred to as the first drone formation in the embodiment of the present application. B represents the blue side, which is referred to as the second drone formation in the embodiment of the present application.

[0095] It should be noted that each drone in the first drone formation is collectively referred to as the first drone; each drone in the second drone formation is collectively referred to as the second drone. Each drone formation can include a number of drones. The red side has m drones and the blue side has n drones.

[0096] The drone information also includes the value of each drone. The value of the R-squared drone is expressed as The value of Party B’s drone is expressed as

[0097] In step S2, strategy pairs of multiple drones are generated according to the drone information, and a strategy group set of a multi-drone formation is generated according to the strategy pairs.

[0098] Among them, the strategy pair includes the current drone number, attack and defense strategy and strategy target; the attack and defense strategy includes attack strategy, avoidance strategy and interference strategy; the strategy group includes strategy pairs of multiple drones in a multi-drone formation.

[0099] When executing a multi-drone game, each drone can adopt three strategies: attack strategy, avoidance strategy and interference strategy. At the same time, drones can use strategies to target any opponent drone. The target is called a strategic target. Drones can be equipped with fire attack weapons and electronic jamming weapons. Drones can attack a certain enemy drone, avoid attacks from a certain enemy drone, or interfere with the attacks of a certain enemy drone, but only one strategy can be selected at the same time.

[0100] Drones can use strategy pairs to play games. Strategy pairs refer to the offensive and defensive strategies and strategic goals adopted by a drone.

[0101] Taking the i-th drone of the red side as an example, its strategy pair is (p, j), which means that it adopts the attack and defense strategy p, and the strategy target is the j-th drone of the blue side. Since the red drone can adopt 3 strategies, and the strategy target can be set to any of the n drones of the blue side, each red drone can have (3×n) strategy pairs. Similarly, the blue drone can adopt 3 strategies, and the strategy target can be set to any of the m drones of the red side, so each blue drone can have (3×m) strategy pairs.

[0102] It should be noted that for a drone formation, when performing a mission, each drone will adopt a strategy pair, and the strategy pairs adopted by all drones together constitute the current strategy group of the drone formation. As the strategy pairs adopted by each drone may be different, the overall strategy group of the drone formation will also be different.

[0103] In the present application embodiment, using represents a strategy group of R party, It represents the strategy pair adopted by drone i of party R, including attack and defense strategies and strategy goals. represents the strategy adopted by drone i, They represent the three offensive and defensive strategies of attack, avoidance, and interference respectively; Indicates the strategy target of attack, avoidance, and interference of drone i of party R. For example, the strategy group formed by the attack and defense strategy of party R is [(1,2),(2,3),(3,3)], which means that drone 1 of party R attacks drone 2 of party B, drone 2 of party R avoids the attack of drone 3 of party B, and drone 3 of party R interferes with drone 3 of party B.

[0104] use It can represent a strategy group of Party B. Represents the strategy pair adopted by drone j of party B. represents the attack and defense strategy adopted by UAV j, They represent three offensive and defensive strategies: attack, evade, and interfere; Indicates the target for the drone to attack, avoid, or interfere with.

[0105] For drone formations, each drone selects a strategy pair, and the strategy pairs of all drones are counted to form a strategy group for the drone formation. Different strategy groups can be obtained by different strategy pairs selected by drones. All possible strategy group situations are counted to generate a strategy group set as a game plan that can be adopted by the drone formation.

[0106] In step S3, the strategy probability of each attack and defense strategy of the plurality of drones is obtained respectively, and a profit matrix is ​​calculated according to the strategy probability.

[0107] Each attack and defense strategy can set a strategy probability, including the kill probability corresponding to the attack strategy, the avoidance probability corresponding to the avoidance strategy, and the interference probability corresponding to the interference strategy.

[0108] The probabilities of the three strategies are expressed as follows:

[0109] (1) Kill probability

[0110] The kill probability represents the probability that one drone successfully destroys another drone. The kill probability of R party i against B party j can be expressed as The killing probability matrix of party R against party B can be expressed as where p ij represents the probability of R drone i killing B drone j. Similarly, the probability of B j killing R drone i can be expressed as The killing probability matrix of party B against party R can be expressed as

[0111] (2) Avoidance probability

[0112] The avoidance probability represents the probability that a drone successfully avoids an attack by another drone. The avoidance probability of party R i against party B’s attack can be expressed as The avoidance probability matrix of party R to party B can be expressed as Similarly, the probability of B party j avoiding R party's attack can be expressed as The avoidance probability matrix of party B to party R can be expressed as

[0113] (3) Interference probability

[0114] The interference probability represents the probability that one drone successfully interferes with the attack of another drone. The interference probability of R party i to B party j can be expressed as The interference probability matrix of party R to party B can be expressed as Similarly, the interference probability of party B j to party R i can be expressed as The interference probability matrix of party B to party R can be expressed as

[0115] Each probability can be pre-generated, and each strategy probability of the drone can be randomly generated.

[0116] Calculating the profit matrix according to the strategy probability includes the following steps:

[0117] S211, constructing a strategy parameter matrix, wherein the strategy parameter matrix includes a first strategy parameter matrix of the first UAV formation and a second strategy parameter matrix of the second UAV formation.

[0118] The first strategy parameter matrix includes a first attack strategy parameter matrix, a first avoidance strategy parameter matrix and a first interference strategy parameter matrix.

[0119] The first attack strategy parameter matrix is:

[0120]

[0121] in,

[0122] a ij represents the attack parameter of drone i in the first drone formation against drone j in the second drone formation;

[0123] m represents the number of drones in the first drone formation, and n represents the number of drones in the second drone formation;

[0124] represents the attack and defense strategy adopted by drone i; They represent attack strategy, avoidance strategy and interference strategy respectively; It indicates the strategic target that drone i targets using the attack and defense strategy.

[0125] The first avoidance strategy parameter matrix is:

[0126]

[0127] in,

[0128] h i It represents the avoidance parameter of UAV i to the second UAV formation.

[0129] The first interference strategy parameter matrix is:

[0130]

[0131] in,

[0132] l ij represents the interference parameter of UAV i to UAV j.

[0133] The second strategy parameter matrix includes a second attack strategy parameter matrix, a second avoidance strategy parameter matrix, and a second interference strategy parameter matrix.

[0134] The second attack strategy parameter matrix is:

[0135]

[0136] in,

[0137] b ji represents the attack parameter of drone j on drone i;

[0138] represents the attack and defense strategy adopted by UAV j, They represent attack strategy, avoidance strategy and interference strategy respectively; represents the policy goal of UAV j.

[0139] The second avoidance strategy parameter matrix is:

[0140]

[0141] in,

[0142] h j represents the avoidance parameter of UAV j to the first UAV formation.

[0143] The second interference strategy parameter matrix is:

[0144]

[0145] in,

[0146] l ji represents the interference parameter of UAV j to UAV i.

[0147] S212: Calculate a first expected survival value of the first UAV formation according to the strategy parameter matrix and the strategy probability. Calculate a second expected survival value of the second UAV formation according to the strategy parameter matrix and the strategy probability.

[0148] Among them, the first expected survival value is expressed as:

[0149]

[0150] in,

[0151] u R (s R ,s B ) represents the first expected survival value of the first UAV formation;

[0152] v i represents the value of drone i in the first drone formation;

[0153] q i represents the probability of UAV i avoiding the second UAV formation, t ij represents the probability of interference from UAV i to UAV j, p ji It represents the killing probability of UAV j to UAV i.

[0154]

[0155] The total expected survival value of the machine.

[0156] represents the survival probability of drone i on party R;

[0157] represents the probability that drone i fails to evade successfully and is destroyed, represents the probability that drone i fails to evade the attack;

[0158] represents the probability of drone i being destroyed;

[0159] represents the probability that drone i is not destroyed;

[0160] represents the probability that drone i is not destroyed by drone j of party B, represents the probability that drone i is destroyed by drone j, It represents the probability that drone j is not successfully interfered by drone R.

[0161] The second expected survival value is expressed as:

[0162]

[0163] in,

[0164] u B (s R ,s B ) represents the second expected survival value of the second UAV formation;

[0165] v j represents the value of drone j in the second drone formation;

[0166] q jrepresents the probability of UAV j avoiding the first UAV formation, t ji represents the interference probability of UAV j to UAV i, p ij It represents the killing probability of UAV i against UAV j.

[0167] S213. Calculate the benefit value of the first UAV formation according to the first expected survival value and the second expected survival value.

[0168] The benefit value of the first drone formation is expressed as:

[0169] u(s R ,s B )=u R (s R ,s B )-u B (s R ,s B )

[0170] in,

[0171] u(s R ,s B ) represents the profit value of the first drone formation.

[0172] S214: Generate a profit matrix according to the profit value. The profit matrix can be expressed as:

[0173]

[0174] Among them, S R represents the first UAV formation strategy group set, s R ∈S R represents a strategy group of the first UAV formation, S B represents the second UAV formation strategy set, s B ∈S B Represents a strategy group of the second drone formation.

[0175] (s R ,s B ) indicates that the first UAV formation adopts strategy group s R , the second drone formation adopts the strategy group s B Strategy group combination; u(s R ,s B ) is the first UAV formation in the strategy group combination (s R ,s B ) under the profit value.

[0176] In step S4, a zero-sum game matrix model is constructed, wherein the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation.

[0177] Considering that the strategies adopted by the two drone formations will affect each other's benefits, it is modeled as a zero-sum game matrix model G = (P, S, U). The game model includes the drone part, the strategy part and the benefit part.

[0178] The drone part P is the multi-drone formation information, the strategy part S is the multi-drone strategy information, and the benefit part U is the benefit matrix.

[0179] Among them, the strategy part S = S R ×S B , represents the strategy set of the two drone formations in the game. R represents the strategy group set of the first UAV formation, that is, the strategy group set of the red team, s R ∈S R Represents a strategy group of the red team. B represents the strategy group set of the second UAV formation, that is, the strategy group set of the blue team, s B ∈S B Indicates a strategy group of the blue team. The strategy group includes the strategy pair of each drone in the drone formation.

[0180] For the red team, since the m red drones can each have (3×n) strategy pairs, the number of strategy pairs of the red team formation is τ R is (3×n) m Similarly, the number of strategies for the blue team is τ B (3×m) n The strategy space of the two adversaries is equal to the product of the number of possible strategy pairs that the two adversaries may adopt. The size of the strategy space of the game is S|=τ R ·τ B =(3×n) m (3×m) n .

[0181] For the payoff matrix of the game, u(s R ,s B ) is the R square in the strategy group combination (s R ,s B ), the benefit value of Party B is -u(s R ,s B ).

[0182] In step S5, the attack and defense strategy game scheme for multi-UAV target allocation is solved according to the zero-sum game matrix model.

[0183] Specifically, the following steps are included:

[0184] S501. Obtain a strict game set based on the zero-sum game matrix model.

[0185] p strategy groups can be selected from the first drone formation strategy group set, and q strategy groups can be selected from the second drone formation strategy group set; wherein p and q are positive integers, and p≥2, q≥2. Specifically, in the first drone formation strategy group set S R Select two strategy groups from the second UAV formation strategy group S B Select two policy groups.

[0186] A first strict game set for the first drone formation is generated based on the p strategy groups, and a second strict game set for the second drone formation is generated based on the q strategy groups.

[0187] For example, the first strict game set may include R The two strategy groups selected in S B Two strategy groups selected from

[0188] S502, solving a Nash equilibrium solution based on the strict game set, and selecting a first initial strategy group of the first drone formation and a second initial strategy group of the second drone formation based on the Nash equilibrium solution. The steps include:

[0189] The first strict game set and the second strict game set are processed respectively based on a preset algorithm to obtain a first Nash equilibrium solution and a second Nash equilibrium solution. The first Nash equilibrium solution includes a first probability distribution of p strategy groups, and the second Nash equilibrium solution includes a second probability distribution of q strategy groups. The preset algorithm can be a probability distribution algorithm.

[0190] The policy group with the highest probability in the first probability distribution is determined as the first initial policy group, and the policy group with the highest probability in the second probability distribution is determined as the second initial policy group.

[0191] S503. Based on the second initial strategy group, solve the first optimal game strategy group that maximizes the marginal return of the first drone formation, and based on the first initial strategy group, solve the second optimal game strategy group that maximizes the marginal return of the second drone formation.

[0192] S504. If the first optimal game strategy group and / or the second optimal game strategy group do not exist in the strict game set, that is, both optimal game strategy groups do not all exist in the strict game set, then update the strict game set based on the first optimal game strategy group and / or the second optimal game strategy group.

[0193] The updating method is to add the first best response strategy group and / or the second best response strategy group which are not present in the strict game set to the strict game set.

[0194] And based on the updated strict game set, the step of solving the Nash equilibrium solution based on the strict game set is re-executed.

[0195] If both the first optimal game strategy group and the second optimal game strategy group exist in the strict game set, the Nash equilibrium solution is determined as the attack and defense strategy game scheme for multi-UAV target allocation.

[0196] The embodiment of the present invention further provides an attack and defense strategy game system for multi-UAV target allocation, characterized in that the system includes:

[0197] An information acquisition module is configured to acquire multi-UAV formation information of the first UAV formation and the second UAV formation; the multi-UAV formation information includes UAV information;

[0198] a strategy generation module, configured to generate strategy pairs of multiple drones according to the drone information, and generate a strategy group set of a multi-drone formation according to the strategy pairs; the strategy pairs include a current drone number, an attack and defense strategy, and a strategy target; the attack and defense strategy includes an attack strategy, an avoidance strategy, and an interference strategy; the strategy group includes strategy pairs of multiple drones in the multi-drone formation;

[0199] A benefit matrix generation module is configured to respectively obtain the strategy probability of each attack and defense strategy of the plurality of drones, and generate a benefit matrix according to the strategy probability;

[0200] A model generation module is configured to generate a zero-sum game matrix model; the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation;

[0201] The scheme allocation module is configured to solve the attack and defense strategy game scheme for multi-UAV target allocation according to the zero-sum game matrix model.

[0202] It is understandable that the target allocation system provided in the embodiment of the present invention corresponds to the attack and defense strategy game method. The explanation, examples, beneficial effects and other parts of the relevant contents can refer to the corresponding contents in the attack and defense strategy game method for multi-UAV target allocation, which will not be repeated here.

[0203] The embodiment of the invention further provides a computer-readable storage medium storing a computer program for attack and defense strategy game of multi-UAV target allocation, wherein the computer program enables a computer to execute the attack and defense strategy game method of multi-UAV target allocation as described above. In summary, compared with the prior art, the present invention has the following beneficial effects:

[0204] The embodiment of the present invention obtains the multi-UAV formation information of the first UAV formation and the second UAV formation to generate a strategy pair of multiple UAVs according to the UAV information, and generates a strategy group set of the multi-UAV formation according to the strategy pair. The strategy pair may include the current UAV, the attack and defense strategy and the strategy target; the attack and defense strategy includes the attack strategy, the avoidance strategy and the interference strategy. By calculating the probability distribution of each UAV under different attack and defense strategies, a profit matrix is ​​constructed, and a zero-sum game matrix model is further generated. Based on the zero-sum game matrix model, the attack and defense strategy game scheme of multi-UAV target allocation is solved. The present invention fully considers the three attack and defense strategies of the UAV, solves the attack and defense strategy game scheme, and thus improves the solution quality of the UAV target allocation problem.

[0205] It should be noted that, through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform. The above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform the method of each embodiment or some parts of the embodiment. In the specification provided here, a large number of specific details are explained. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some examples, known methods, structures and techniques are not shown in detail so as not to blur the understanding of this specification.

[0206] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0207] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An attack and defense strategy game method for multi-UAV target allocation, the attack and defense strategy game method is executed by a computer, characterized in that: The following steps are involved: Acquire multi-UAV formation information of the first UAV formation and the second UAV formation; the multi-UAV formation information includes UAV information; Generate strategy pairs of multiple drones according to the drone information, and generate a strategy group set of multiple drone formations according to the strategy pairs; the strategy pairs include the current drone number, the attack and defense strategy, and the strategy target; the attack and defense strategy includes the attack strategy, the avoidance strategy, and the interference strategy; the strategy group includes the strategy pairs of multiple drones in the multi-drone formation; Obtaining the strategy probability of each attack and defense strategy of the plurality of drones respectively, and calculating the profit matrix according to the strategy probability; Constructing a zero-sum game matrix model; the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation; The zero-sum game matrix model is used to solve the attack and defense strategy game scheme for multi-UAV target allocation.

2. The attack and defense strategy game method according to claim 1, characterized in that: Generate a profit matrix according to the strategy probability, including: Construct a strategy parameter matrix; the strategy parameter matrix includes a first strategy parameter matrix of the first UAV formation and a second strategy parameter matrix of the second UAV formation; Calculate a first expected survival value of a first UAV formation according to the strategy parameter matrix and the strategy probability; calculate a second expected survival value of a second UAV formation according to the strategy parameter matrix and the strategy probability; Calculate the benefit value of the first UAV formation according to the first expected survival value and the second expected survival value; A profit matrix is ​​generated according to the profit values.

3. The attack and defense strategy game method according to claim 2, characterized in that: The first strategy parameter matrix includes a first attack strategy parameter matrix, a first avoidance strategy parameter matrix and a first interference strategy parameter matrix; The first attack strategy parameter matrix is: in, a ij represents the attack parameter of drone i in the first drone formation against drone j in the second drone formation; m represents the number of drones in the first drone formation, and n represents the number of drones in the second drone formation; represents the attack and defense strategy adopted by drone i; They represent attack strategy, avoidance strategy and interference strategy respectively; It indicates the strategic target that drone i targets using the attack and defense strategy; The first avoidance strategy parameter matrix is: in, h i represents the avoidance parameter of UAV i to the second UAV formation; The first interference strategy parameter matrix is: in, l ij represents the interference parameter of UAV i to UAV j.

4. The attack and defense strategy game method according to claim 3, characterized in that: The second strategy parameter matrix includes a second attack strategy parameter matrix, a second avoidance strategy parameter matrix and a second interference strategy parameter matrix; The second attack strategy parameter matrix is: in, b ji represents the attack parameter of drone j on drone i; represents the attack and defense strategy adopted by drone j, They represent attack strategy, avoidance strategy and interference strategy respectively; represents the strategic goal of UAV j; The second avoidance strategy parameter matrix is: in, h j represents the avoidance parameter of UAV j to the first UAV formation; The second interference strategy parameter matrix is: in, l ji represents the interference parameter of UAV j to UAV i.

5. The attack and defense strategy game method according to claim 4, characterized in that: Calculating a first expected survival value of a first UAV formation according to the strategy parameter matrix and the strategy probability includes: in, u R (s R ,s B ) represents the first expected survival value of the first UAV formation; v i represents the value of drone i in the first drone formation; q i represents the probability of UAV i avoiding the second UAV formation, t ij represents the probability of interference from UAV i to UAV j, p ji It represents the killing probability of UAV j to UAV i.

6. The attack and defense strategy game method according to claim 5, characterized in that: Calculating a second expected survival value of a second UAV formation according to the strategy parameter matrix and the strategy probability includes: in, u B (s R ,s B ) represents the second expected survival value of the second UAV formation; v j represents the value of UAV j in the second UAV formation; q j represents the probability of UAV j avoiding the first UAV formation, t ji represents the interference probability of UAV j to UAV i, p ij It represents the killing probability of UAV i against UAV j.

7. The attack and defense strategy game method according to claim 6, characterized in that: The profit matrix includes: u(s R ,s B )=u R (s R ,s B )-u B (s R ,s B ) Among them, S R represents the first UAV formation strategy group set, s R ∈S R represents a strategy group of the first UAV formation, S B represents the second UAV formation strategy set, s B ∈S B represents a strategy group of the second UAV formation; (s R ,s B ) indicates that the first UAV formation adopts strategy group s R , the second drone formation adopts the strategy group s B Strategy group combination; u(s R ,s B ) is the first UAV formation in the strategy group combination (s R ,s B ) under the profit value.

8. The attack and defense strategy game method according to claim 1, characterized in that: The attack and defense strategy game scheme for multi-UAV target allocation is solved according to the zero-sum game matrix model, including: Acquire a strict game set based on the zero-sum game matrix model; Solving a Nash equilibrium solution based on the strict game set, and selecting a first initial strategy group of the first drone formation and a second initial strategy group of the second drone formation based on the Nash equilibrium solution; Calculate a first optimal game strategy group that maximizes the marginal return of the first drone formation based on the second initial strategy group, and calculate a second optimal game strategy group that maximizes the marginal return of the second drone formation based on the first initial strategy group; If the first optimal game strategy group and / or the second optimal game strategy group do not exist in the strict game set, updating the strict game set based on the first optimal game strategy group and / or the second optimal game strategy group, and re-executing the step of solving the Nash equilibrium solution based on the strict game set based on the updated strict game set; If both the first optimal game strategy group and the second optimal game strategy group exist in the strict game set, the Nash equilibrium solution is determined as the attack and defense strategy game scheme for multi-UAV target allocation.

9. An attack and defense strategy game system for multi-UAV target allocation, characterized in that: The system comprises: An information acquisition module is configured to acquire multi-UAV formation information of the first UAV formation and the second UAV formation; the multi-UAV formation information includes UAV information; a strategy generation module, configured to generate strategy pairs of multiple drones according to the drone information, and generate a strategy group set of a multi-drone formation according to the strategy pairs; the strategy pairs include a current drone number, an attack and defense strategy, and a strategy target; the attack and defense strategy includes an attack strategy, an avoidance strategy, and an interference strategy; the strategy group includes strategy pairs of multiple drones in the multi-drone formation; A benefit matrix generation module is configured to respectively obtain the strategy probability of each attack and defense strategy of the plurality of drones, and generate a benefit matrix according to the strategy probability; A model generation module is configured to generate a zero-sum game matrix model; the zero-sum game matrix model includes multi-UAV formation information, multi-UAV strategy information and a payoff matrix; the multi-UAV strategy information includes a strategy group set of the multi-UAV formation; The scheme allocation module is configured to solve the attack and defense strategy game scheme for multi-UAV target allocation according to the zero-sum game matrix model.

10. A computer-readable storage medium, characterized in that: It stores a computer program for attack and defense strategy game of multi-UAV target allocation, wherein the computer program enables the computer to execute the attack and defense strategy game method of multi-UAV target allocation as described in any one of claims 1 to 8.

Images