An unmanned ship cluster designated time position decision method and device based on game theory

Abstract

The application discloses a kind of unmanned ship cluster specified time position decision-making method and device based on game theory, by converting unmanned ship cluster seeking optimal confrontation position problem into non-cooperative game problem, constructs water unmanned ship cluster game model, and designs distributed nash equilibrium solution algorithm under specified time;The method can solve the nash equilibrium of non-cooperative game problem in any specified time, quickly solve the optimal confrontation position problem of unmanned ship cluster seeking, enhance the attack and defense capability of our unmanned ship, improve the efficiency of decision-making;In addition, the initial time-varying gain item can be zero by using TBG function, which avoids the problem of excessive input at the initial time. Since the design and calculation of time-varying gain θ (t) are relatively simple, the application is relatively simple in structure and easy to implement.

Description

Technical Field

[0001] This invention relates to the field of unmanned surface vessel (USV) swarm game-based adversarial technology, and in particular to a method and apparatus for making decisions on the location of a USV swarm at a specified time based on game theory. Background Technology

[0002] Game theory is a theoretical approach to studying phenomena involving struggle or competition. Applying game theory to unmanned surface vessel (USV) swarm warfare can effectively characterize the competition and decision-making evolution between combatants. In game theory, if each participant adopts an optimal strategy, and no single player can gain a higher payoff by unilaterally changing their strategy, then the current strategy combination of all participants is called a Nash equilibrium. Under Nash equilibrium, the strategy combinations of all parties form a stable state. In USV swarm warfare, analyzing Nash equilibrium can optimize positional decisions, not only enhancing the attack power of friendly USVs but also reducing the threat posed by enemy USVs.

[0003] To accelerate the convergence speed of Nash equilibrium solutions, significant progress has been made in solving Nash equilibrium problems with finite and fixed time convergence. However, the convergence time of these methods is affected by the initial state and system parameters and cannot be predicted in advance, which limits their practical applicability to some extent. To address these issues, specified-time distributed game algorithms proposed in recent years offer an effective approach.

[0004] While some research has been conducted on solving Nash equilibria with specified-time convergence, relatively few studies have applied it to unmanned surface vessel (USV) swarm adversarial games, particularly in the area of ​​USV swarm positioning. In the face of the complex and ever-changing modern battlefield, timely and efficient collaborative decision-making and adversarial strategies are crucial for seizing opportunities and gaining the initiative in war. Therefore, applying the specified-time convergence Nash equilibrium algorithm to the positioning of USV swarms has both theoretical and strategic significance. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a method and apparatus for decision-making on the location of unmanned surface vessel (USV) swarms at a specified time based on game theory. By combining the TBG function, consensus protocol and gradient method, a distributed non-cooperative game algorithm with convergence at a specified time is designed to quickly and effectively solve the problem of USV swarms seeking the optimal adversarial position and improve decision-making efficiency.

[0006] To address the aforementioned technical problems, this invention provides a game theory-based method for determining the location of unmanned surface vessel (USV) swarms at a specified time, comprising the following steps:

[0007] Step 1: Establish a scenario of adversarial unmanned surface vessel (USV) swarm confrontation between friendly and enemy forces, and randomly generate the initial positions of the USV swarm.

[0008] Step 2: Based on the initial position, communication distance, maximum range and safe range of the weapons carried by the unmanned surface vessel swarm, the problem of finding the optimal confrontation position of our unmanned surface vessel swarm is modeled as a non-cooperative game problem.

[0009] Step 3: Based on the TBG function, consensus protocol and gradient method, design a distributed Nash equilibrium solution algorithm with convergence in a specified time.

[0010] Step 4: Develop a location decision for our unmanned surface vessel cluster.

[0011] Preferably, in step 1, establishing the scenario of adversarial unmanned surface vessel (USV) swarm confrontation specifically involves: establishing an operational area Q for both USV swarms, assuming a combat line exists in the sea areas where both sides are located, and neither side can cross this combat line to enter the other's operational domain; the USV swarm is denoted as... Unmanned surface vessel a i Location for Its initial position is determined by It means that N A This indicates the number of our unmanned surface vessel (USV) swarms, where A represents our US USV swarm. This represents the x-coordinate of our i-th unmanned surface vessel. This represents the ordinate value of our i-th unmanned surface vessel; the set of enemy unmanned surface vessels can be denoted as... Define enemy unmanned surface vessel b j Location for Its initial position can be determined by It means that N B This indicates the number of enemy unmanned surface vessels (USVs), where B represents the enemy USV swarm. This represents the x-coordinate of the j-th enemy unmanned surface vessel. This represents the ordinate value of the j-th enemy unmanned surface vessel.

[0012] Preferably, in step 2, based on the initial position of the unmanned surface vessel (USV) swarm, communication distance, and the maximum and safe range of the weapons carried by the USVs, the problem of finding the optimal confrontation position for our USV swarm is modeled as a non-cooperative game problem, specifically including the following steps:

[0013] Step 21: Establish the capability for our i-th unmanned surface vessel to destroy the enemy's unmanned surface vessel. 1,i ;

[0014] Step 22: Establish the capability of our i-th unmanned surface vessel to defend against attacks from enemy unmanned surface vessels. 2,i ;

[0015] Step 23: Construct the objective function f for our i-th unmanned surface vessel. i .

[0016] Preferably, in step 21, the initial positions of the unmanned surface vessels (USVs) of both sides and the maximum firing range d of the weapons carried by the USVs are determined. max With safe range d min Define f 1,i for:

[0017]

[0018] Where i = 1, ..., N A j = 1, ..., N B N A and N B These represent the number of our side's and the enemy's unmanned surface vessels, respectively. For our unmanned surface vessel a i Location, For enemy unmanned surface vessel b j Location; For our unmanned surface vessel a i Destroy enemy unmanned surface vessel b j The probability of is expressed in the following form:

[0019]

[0020] Where, d ij For our unmanned surface vessel a i and enemy unmanned surface vessel b j The distance between them, when the function f 1,i The smaller the value, the better for our unmanned surface vessel a. i With enemy unmanned surface vessel b j The closer the distance between them is to d min The stronger our i-th unmanned surface vessel's ability to destroy enemy unmanned surface vessels, the better.

[0021] Preferably, in step 22, the collaborative and defensive capabilities of the unmanned surface vessel swarm are improved. 2,i Defined as:

[0022]

[0023] Where i = 1, ..., N A , Let d be the position of our i-th unmanned surface vessel; i,i-1 and d i,i+1 These represent unmanned surface vessel a. i With unmanned surface vessel a i-1 and unmanned surface vessel a i+1 The optimal collaboration distance; to ensure that each unmanned surface vessel (USV) can maintain effective communication with at least two other USVs and collaborate to complete complex combat missions, in the special case where i=1, i-1=N is set. A and i=N AWhen i+1 = 1, set i+1 = 1; when function f 2,i The smaller the value, the better for our unmanned surface vessel a. i They were respectively with two other unmanned surface vessels a i-1 and a i+1 The closer the distance between them is to the optimal collaboration distance d i,i-1 and d i,i+1 The stronger the defense capability of our i-th unmanned surface vessel, the better.

[0024] Preferably, in step 23, based on the aforementioned capability f of our i-th unmanned surface vessel to destroy the enemy's unmanned surface vessel... 1,i and the ability to defend against enemy unmanned surface vessel attacks. 2,i objective function f i It can be established as:

[0025] f i =α1f 1,i +α2f 2,i ,α1+α2=1

[0026] Where i = 1, ..., N A α1≥0 and α2≥0 represent the weights that our unmanned surface vessel cluster places on its destroy capability and defense capability, respectively.

[0027] Preferably, in step 3, the design of a distributed Nash equilibrium solution algorithm with convergence at a specified time, based on the TBG function, consensus protocol, and gradient method, specifically includes the following steps:

[0028] Step 31: Define the TBG function; a continuously differentiable time-varying function ξ(t) is called a TBG function and must satisfy the following conditions:

[0029]

[0030] And its derivative with respect to the time variable t satisfies:

[0031]

[0032] Among them, t f The convergence time is arbitrarily given and does not depend on any initial state; it can be set according to actual engineering needs.

[0033] Step 32, Unmanned Surface Vessel a i Update your own strategy

[0034] Step 33: Update the unmanned surface vessel a i For a j Decision estimation y ij .

[0035] Preferably, in step 32, to enhance both attack and defense capabilities, each unmanned surface vessel adjusts its own strategy based on local information to minimize the objective function value. Its gradient-based strategy update formula is as follows:

[0036]

[0037] Wherein, time-varying function k is a function of TBG; 1,i >0 controls the gain; It is an unmanned surface vessel (USV). i Estimates of the decisions made by all unmanned surface vessels except themselves.

[0038] Preferably, in step 33, since each unmanned surface vessel (USV) can only obtain decision information from its neighbors and cannot obtain decisions from non-neighbor USVs, each USV updates the decision estimates of other USVs through information exchange with its neighbors, so that the estimates converge to the true values. Based on the consensus protocol, the estimated value y... ij The specific update formula is as follows:

[0039]

[0040] Where, ω ij This indicates the communication relationship between unmanned surface vessels (USVs). If USV a i and unmanned surface vessel a j If information exchange is possible, then ω ij =1, otherwise ω ij =0.

[0041] Correspondingly, a game theory-based unmanned surface vessel (USV) swarm location decision-making device for a specified time includes: a scenario construction module, a model building module, and a location decision-making module. The scenario construction module is used to create and simulate the environment of USV swarm confrontation, setting key information such as the combat areas of both sides, the number and initial position of the USV swarm, and the maximum and safe range of its weapons, facilitating the simulation and analysis of various combat situations. The model building module is used to design the subsequent location decision-making algorithm for the specified time. Based on information such as the initial position and communication distance of the USV swarm, it models the problem of finding the optimal confrontation position for our USV swarm as a non-cooperative game problem. The location decision-making module is used to quickly determine the optimal position of each of our USVs in USV swarm confrontation, thereby enhancing the attack and defense capabilities of our USVs and ensuring real-time response to changes in the battlefield situation.

[0042] The beneficial effects of this invention are as follows: This invention transforms the problem of unmanned surface vessel (USV) swarm seeking the optimal adversarial position into a non-cooperative game problem, constructs a game model of USV swarm, and designs a distributed Nash equilibrium solution algorithm for a specified time. This method can solve the Nash equilibrium of the non-cooperative game problem at any specified time, quickly solve the problem of USV swarm seeking the optimal adversarial position, enhance the attack and defense capabilities of our USVs, and improve decision-making efficiency. In addition, the TBG function used can make the initial time-varying gain term zero, avoiding the problem of the input being too large at the beginning. Since the design and calculation of the time-varying gain θ(t) is relatively simple, this invention is structurally simple and easy to implement. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0044] Figure 2 This is a schematic diagram showing the initial positions randomly generated by both enemy and friendly unmanned surface vessels according to the present invention.

[0045] Figure 3 This is a schematic diagram of the time and location strategy of our unmanned surface vessel swarm in this invention.

[0046] Figure 4 This is a trajectory diagram of the horizontal coordinate values ​​of the position of our unmanned surface vessel based on the exponentiation-type TBG function of this invention.

[0047] Figure 5 This is a trajectory diagram of the vertical coordinate values ​​of the position of our unmanned surface vessel based on the exponentiation-type TBG function of this invention.

[0048] Figure 6 This is a schematic diagram illustrating the effectiveness of the algorithm under different parameters of the present invention.

[0049] Figure 7 This is a trajectory diagram of the horizontal coordinate value of our unmanned surface vessel's position under the shorter convergence time based on the exponential TBG function of this invention.

[0050] Figure 8 This is a trajectory diagram of the vertical coordinate value of our unmanned surface vessel's position under the shorter convergence time based on the exponential TBG function of this invention.

[0051] Figure 9 This is a trajectory diagram of the relative error under the shorter convergence time based on the exponential TBG function of this invention.

[0052] Figure 10 This is a comparison chart of the relative error trajectories of the specified time and the asymptotic convergence algorithm of this invention.

[0053] Figure 11 This is a trajectory diagram of the abscissa value of our unmanned surface vessel based on the sin-type TBG function according to the present invention.

[0054] Figure 12 This is a trajectory diagram of the longitudinal coordinate values ​​of our unmanned surface vessel based on the sin-type TBG function according to the present invention.

[0055] Figure 13 This is a relative error trajectory diagram based on the sin-type TBG function of this invention. Detailed Implementation

[0056] like Figure 1 As shown, a game theory-based method for determining the location of an unmanned surface vessel (USV) swarm at a specified time includes the following steps:

[0057] Step 1: Establish the combat zone for both sides' unmanned surface vessel (USV) clusters, randomly generate the initial positions of the USV clusters, and ensure that both sides have full knowledge of the number and initial positions of the other's USV clusters.

[0058] Step 2: Based on the initial position of the unmanned surface vessel (USV) swarm, communication distance, and the maximum and safe firing range of the weapons carried by the USVs, the problem of finding the optimal confrontation position for our USV swarm is modeled as a non-cooperative game problem, including the following steps:

[0059] Step 21: Establish the capability for our i-th unmanned surface vessel to destroy the enemy's unmanned surface vessel. 1,i ;

[0060] Step 22: Establish the capability of our i-th unmanned surface vessel to defend against attacks from enemy unmanned surface vessels. 2,i ;

[0061] Step 23: Construct the objective function f for our i-th unmanned surface vessel. i .

[0062] Step 3: Based on the TBG function, consensus protocol, and gradient method, design a distributed Nash equilibrium solution algorithm that converges within a specified time, including the following steps:

[0063] Step 31: Define the TBG function;

[0064] Step 32, Unmanned Surface Vessel a i Update your own strategy

[0065] Step 33: Update the unmanned surface vessel a i For a j Decision estimation y ij .

[0066] Step 4: Develop a location decision for our unmanned surface vessel cluster.

[0067] In this embodiment, MATLAB 2022b is used as the simulation software to simulate a game-theoretic scenario of unmanned surface vessel (USV) swarm. Based on two TBG functions, the time-location decision-making method of this invention is used to quickly solve a set of rational and effective USV swarm location strategies.

[0068] In this embodiment, MATLAB 2022b is used as the simulation software to test the influence of the algorithm parameter σ on the convergence performance of the game theory-based unmanned surface vessel cluster position decision method of the present invention at a specified time, as well as the convergence performance at a specified time. The simulation comparison between the specified time convergence and the distributed Nash equilibrium solution algorithm under asymptotic convergence of the present invention is also performed.

[0069] Assume we have N A =5 unmanned surface vessels, the enemy has N B = 4 unmanned surface vessels. The effective operational area Q of our unmanned surface vessel swarm. A = [-7,0)×[-8,8], the enemy's combat area is Q. B = (0,7]×[-8,8]. The initial positions of the unmanned surface vessel swarms of both sides are randomly generated, such as... Figure 2 As shown.

[0070] The parameters for the simulated unmanned surface vessel (USV) swarm game scenario include: the optimal cooperative distance d for our USVs is 2 unit lengths, and the maximum range d of the weapons they carry is... max The length is 15 units, and the safe range is d. min The length is 8 units. The weighting of our unmanned surface vessel swarm's emphasis on destroy capability is α1 = 0.5 and its emphasis on defense capability is α2 = 0.5.

[0071] Example 1: Based on the time-distributed Nash equilibrium solution algorithm described in step 3, the parameters are set as follows: σ = 0.000001, constant control gain k 1,i =1, specifying the convergence time t f = 1s. Based on the specified time t. f Design a TBG function with the following exponentiation form:

[0072]

[0073] Figure 3 This is a positioning strategy for our unmanned surface vessel (USV) swarm, determined by a time-distributed game theory algorithm specified in this invention, based on the initial position of the enemy USVs. For example... Figure 4 and Figure 5 As shown, the position status of our unmanned surface vessel cluster at a specified time t f The algorithm converges to Nash equilibrium in 1 second. To better demonstrate the convergence time characteristics of this invention's algorithm, the relative error is defined as... in, and The horizontal and vertical coordinates of our unmanned surface vessel represent the strategy. This is a theoretical Nash equilibrium.

[0074] To test the impact of the algorithm parameter σ on convergence performance, consider the following three different simulation parameters: a) σ = 10 -6 b) σ = 10 -4 c) σ = 3 × 10 -7 The simulation results are as follows: Figure 6 As shown in the figure, the iterative process of the relative error is illustrated. It can be seen from the figure that setting a smaller σ value can reduce the convergence error, but a smaller σ will lead to a larger time-varying gain θ(t), which will also degrade the transient performance of the algorithm. Therefore, it is necessary to set a suitable σ through multiple trials to obtain a convergence error that is acceptable in practical applications. In this embodiment, when σ = 10... -6 When the algorithm converges, its performance is better.

[0075] To further test the convergence performance of the method of the present invention at a predetermined time, t is set. f =0.1s, then the specific form of the TBG function is:

[0076]

[0077] Let σ = 10 -6 The simulation results are as follows: Figure 7 , Figure 8 and Figure 9 As shown in the figure, the position and status of the unmanned surface vessel can be seen. Can be achieved at t=t f If the convergence to Nash equilibrium is 0.1s, it can be proven that the convergence time can be arbitrarily designed by the designer according to the task requirements.

[0078] Depend on Figure 10 It is known that asymptotic convergence algorithms can guarantee that decision variables converge to Nash equilibrium, but the convergence time is not adjustable; the decision variables only converge to Nash equilibrium when the time approaches infinity. Compared with general distributed game algorithms under asymptotic convergence, this invention can quickly and effectively solve the problem of unmanned surface vessel swarms seeking optimal confrontation positions, reducing the time and resource costs of military operations and improving combat efficiency.

[0079] Example 2: To further reduce the computational load, based on the definition of the TBG function, consider another TBG function in sin form, ξ3(t), whose specific form is as follows:

[0080]

[0081] Where ρ = -π / 2. The parameter σ is set to 10. -7 The specified convergence time tf =1s, and the other parameters are the same as in Example 1.

[0082] Compared to the TBG functions ξ1(t) and ξ2(t) given in Example 1, this TBG function ξ3(t) only requires a time t. f Multiplication operations do not require t f Perform exponentiation. For example... Figure 11 , Figure 12 and Figure 13 As shown, under the action of the TBG function ξ3(t), convergence at a specified time is guaranteed, the calculation process is simplified, and the computational burden is reduced.

[0083] Based on the same inventive concept, this invention provides a game theory-based unmanned surface vessel (USV) swarm location decision-making device for a specified time, comprising: a scenario construction module, a model building module, and a location decision-making module. The scenario construction module is used to create and simulate an environment for USV swarm confrontation, setting key information such as the combat areas of both sides, the number and initial position of the USV swarm, and the maximum and safe range of its weapons, facilitating the simulation and analysis of various combat situations. The model building module is used to design the subsequent specified-time location decision-making algorithm, modeling the problem of finding the optimal confrontation position for our USV swarm as a non-cooperative game problem based on information such as the initial position and communication distance of the USV swarm. The location decision-making module is used to determine the optimal position of each of our USVs in USV swarm confrontation, making rapid location decisions based on the Nash equilibrium solution algorithm for specified time, thereby enhancing the attack and defense capabilities of our USVs and ensuring real-time response to changes in the battlefield situation.

Claims

1. A game theory-based method for determining the location of an unmanned surface vessel (USV) swarm at a specified time, characterized in that, Includes the following steps: Step 1: Establish a scenario of adversarial unmanned surface vessel (USV) swarm confrontation between friendly and enemy forces, and randomly generate the initial positions of the USV swarm; Step 2: Based on the initial position of the unmanned surface vessel (USV) swarm, communication distance, and the maximum and safe firing range of the weapons carried by the USVs, the problem of finding the optimal confrontation position for our USV swarm is modeled as a non-cooperative game problem; specifically, it includes the following steps: Step 21, Establish our side's first The ability of a single unmanned surface vessel to destroy an enemy unmanned surface vessel Based on the initial positions of both sides' unmanned surface vessels (USVs) and the maximum range of the weapons carried by the USVs. With safe range ,definition for: in, , , and These represent the number of our side's and the enemy's unmanned surface vessels, respectively. For our unmanned surface vessel Location, For enemy unmanned surface vessels Location; For our unmanned surface vessel Destroy enemy unmanned surface vessels The probability of is expressed in the following form: in, For our unmanned surface vessel and enemy unmanned surface vessels The distance between them, when the function The smaller the value, the better for our unmanned surface vessel. Enemy unmanned surface vessel The closer the distance between them At that time, our side's first The stronger a single unmanned surface vessel's ability to destroy enemy unmanned surface vessels; Step 22, Establish our side's first The ability of an unmanned surface vessel to defend against attacks from enemy unmanned surface vessels Improve the collaborative and defensive capabilities of unmanned surface vessel swarms. Defined as: in, , For our side The location of the unmanned surface vessel; and These represent unmanned surface vessels. with unmanned surface vessels and unmanned surface vessels The optimal collaboration distance; to ensure that each unmanned surface vessel (USV) can maintain effective communication with at least two other USVs to collaboratively complete complex combat missions, and in special circumstances... At that time, set ,as well as At that time, set When the function The smaller the value, the better for our unmanned surface vessel. They were respectively with two other unmanned boats and The closer the distance is to the optimal collaboration distance and At that time, our side's first The stronger the defensive capabilities of an unmanned surface vessel; Step 23, Construct our first Objective function of unmanned surface vessel Based on the above, our side's first The ability of a single unmanned surface vessel to destroy an enemy unmanned surface vessel and the ability to defend against enemy unmanned surface vessels. objective function Established as: in, , and These represent the relative importance and weighting of our unmanned surface vessel (USV) swarm towards fire-and-run capabilities and defensive capabilities, respectively. Step 3: Based on the TBG function, consensus protocol, and gradient method, design a distributed Nash equilibrium solution algorithm that converges within a specified time; specifically, it includes the following steps: Step 31: Define the TBG function; a continuously differentiable time-varying function. To be called a TBG function, the following conditions must be met: And its relationship with time variables The derivative satisfies: in, The convergence time is arbitrarily given and does not depend on any initial state; it can be set according to the actual engineering needs. Step 32, Unmanned Surface Vessel Update your own strategy ; Step 33: Update the unmanned surface vessel right Decision estimation ; Step 4: Develop a location decision for our unmanned surface vessel cluster.

2. The game theory-based method for determining the location of unmanned surface vessel swarms at a specified time as described in claim 1, characterized in that, In step 1, establishing a scenario of adversarial combat between enemy and friendly unmanned surface vessel (USV) swarms specifically involves: establishing combat zones for both sides' USV swarms. Assuming there is a combat line in the sea area where both sides are located, neither side can cross the combat line into the other's combat territory; Our unmanned surface vessel swarm is denoted as... Unmanned surface vessel Location for Its initial position is determined by It means that, among them, This indicates the number of our unmanned surface vessels in the swarm. Represents our unmanned surface vessel cluster; Indicates our side's first The x-coordinate value of the unmanned surface vessel. Indicates our side's first The ordinate value of the unmanned surface vessel; The assembly of the enemy unmanned surface vessel swarm is denoted as Define enemy unmanned surface vessels Location for Its initial position is determined by It means that, among them, Indicates the number of enemy unmanned surface vessels. Represents an enemy unmanned surface vessel swarm; Indicates the enemy's first The x-coordinate value of the unmanned surface vessel. Indicates the enemy's first The ordinate value of the unmanned surface vessel.

3. The game theory-based method for determining the location of unmanned surface vessel swarms at a specified time as described in claim 1, characterized in that, In step 32, to enhance both attack and defense capabilities, each unmanned surface vessel adjusts its strategy based on local information to minimize the objective function value. Its gradient-based strategy update formula is as follows: Wherein, time-varying function For functions related to TBG; It controls the gain; It is an unmanned surface vessel. Estimates of the decisions made by all unmanned surface vessels except themselves.

4. The game theory-based method for determining the location of unmanned surface vessel swarms at a specified time as described in claim 1, characterized in that, In step 33, since each unmanned surface vessel (USV) can only obtain decision information from its neighbors and cannot obtain decisions from non-neighbor USVs, each USV updates the decision estimates of other USVs through information exchange with its neighbors, so that the estimates converge to the true values. Based on the consensus protocol, the estimated values... The specific update formula is as follows: in, This indicates the communication relationship between unmanned surface vessels (USVs). and unmanned surface vessels To conduct information exchange, ,otherwise .

5. A decision-making device for a game theory-based unmanned surface vessel (USV) swarm location decision-making method at a specified time as described in claim 1, characterized in that, include: Scene building module, model building module, and location decision module; The scenario construction module is used to create and simulate the environment of unmanned surface vessel (USV) swarm confrontation, setting key information such as the combat areas of both sides, the number and initial position of the USV swarm, and the maximum and safe range of the weapons carried, to facilitate the simulation and analysis of various combat situations. The model building module is used for the design of subsequent time-specified position decision algorithms. Based on the initial position and communication distance information of the USV swarm, it models the problem of finding the optimal confrontation position of our USV swarm as a non-cooperative game problem. The position decision module is used to quickly determine the optimal position of each of our USVs in USV swarm confrontation, so as to enhance the attack and defense capabilities of our USVs and ensure real-time response to changes in the battlefield situation.

Images