A method and device for wargame simulation

Abstract

The application provides a kind of wargame simulation method and device, the method comprises the following steps: obtaining the deployment scheme of the force of both sides of the game;Get the action instruction of the both sides of the game, according to the action instruction, the corresponding force entity model is played, the operation result is fed back, the both sides of the game generate the next action instruction according to the game result;Record the game process and the final game result of the both sides of the game;Wherein, the force entity model is obtained by extracting the attribute of force entity in the deployment scheme of the force, based on the modeling of each functional parameter in the attribute of the force entity to the force entity.The technical scheme of the application realizes the multifunctional simulation of wargame, in the process of modeling the force entity, through the simulation of the functional attribute of the force entity, so that the simulation method and device of the application can adapt to the mainstream reinforcement learning algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, and in particular to a method and device for simulating a game of chess with pieces. BACKGROUND

[0002] Chess deduction is a combat simulation method that simulates various combat equipment, weapons and combat scenarios, and commanders execute combat operations according to the experience and rules of actual combat, achieve combat objectives, and realize combat process deduction. It can be used for optimization and evaluation of tactical strategies, as well as innovative research of strategies under the condition of new weapons and equipment, and is an effective tool for studying and understanding war.

[0003] The essence of the chess game scheme is sequential decision-making, and intelligent decision-making is an important development direction of the new generation of artificial intelligence. The basis for studying intelligent decision-making algorithms is to have a simulation method or system for chess game that is suitable for reinforcement learning algorithms. SUMMARY

[0004] The present application provides a method for simulating a game of chess with pieces, which solves the defects of the prior art, such as single function of the chess game method or system, and inability to adapt to mainstream reinforcement learning algorithms, and realizes multifunctional simulation of the chess game, which can better adapt to mainstream reinforcement learning algorithms.

[0005] The present application provides a method for simulating a game of chess with pieces, which solves the defects of the prior art, such as single function of the chess game method or system, and inability to adapt to mainstream reinforcement learning algorithms, and realizes multifunctional simulation of the chess game, which can better adapt to mainstream reinforcement learning algorithms.

[0006] Obtaining a force deployment scheme of both sides of the game;

[0007] Obtaining the action instructions of the two sides of the game, performing game operations on the corresponding force entity model according to the action instructions, feeding back the operation results, and generating the next action instructions according to the game results;

[0008] Recording the game process and the final game result of the two sides of the game;

[0009] Wherein, the force entity model is obtained by extracting the attributes of the force entity in the force deployment scheme, and modeling the force entity based on the functional parameters in the attributes of the force entity.

[0010] According to the method for simulating a game of chess with pieces provided by the present application, the functional parameters in the attributes of the force entity include the moving function, and the process of moving simulation of the force entity model includes:

[0011] Obtaining discrete direction values d' and discrete speed values v', wherein the values of d' and v' are -1, 1 or 0;

[0012] Querying the attribute database of the force entity, obtaining the direction interval [d', d' +1] of the force entity, and querying the speed interval [v', v' +1] of the force entity;min d max ] and speed interval [v min ,v max ];

[0013] According to the discrete direction value d' and the end value of the direction interval, a probability distribution of the direction of the force entity is established;

[0014] According to the discrete speed value v' and the end value of the speed interval, a probability distribution of the speed of the force entity is established;

[0015] According to the probability distribution of the direction of the force entity and the probability distribution of the speed of the force entity, the direction value and the speed value of the force entity are sampled to obtain a continuous direction value d and a continuous speed value v of the force entity;

[0016] The force entity model is controlled to move based on the continuous direction value d and the continuous speed value v.

[0017] According to the method for simulating a game of war provided by the application, the function parameter of the attribute of the force entity includes a search function, and the process of simulating the search of the force entity model includes:

[0018] The mutual distance between the force entity model and the visible enemy units within the radar range centered on the force entity model is calculated, and the enemy unit closest to the force entity model is marked as p0;

[0019] The value of all visible enemy units within the radar range centered on the force entity model is calculated according to the state and cost of the enemy units, and the enemy unit with the highest value is selected and marked as p1;

[0020] When the received instruction is 0, the force entity model outputs p0 as the search result;

[0021] When the received instruction is 1, the force entity model outputs p1 as the search result.

[0022] According to the method for simulating a game of war provided by the application, the function parameter of the attribute of the force entity includes a missile launching function, and the process of simulating the missile launching of the force entity model includes:

[0023] The probability of hitting the enemy unit by launching a missile is calculated according to the distance between the force entity model and the enemy unit, the types of the units of both sides, and whether the enemy unit has a defensive secondary gun;

[0024] When it is judged that the probability of hitting the enemy unit is greater than a preset value, the force entity model obtains the instruction of launching a missile.

[0025] According to the wargame simulation method provided by the application, the state parameters in the attribute of the force entity include:

[0026] The state information of the force entity in the current time slice and the state information in a preset time period before the current time slice.

[0027] According to the wargame simulation method provided by the application, the feedback operation result, and the two parties of the game generate the next action instruction according to the game result, specifically including:

[0028] The battlefield after the action of the force entity model is rendered to the screen of human-computer interaction;

[0029] The battlefield situation is packaged into a group of observation data and sent to an intelligent agent, and the intelligent agent generates the next operation instruction according to the observation data.

[0030] The application also provides a wargame simulation device, including:

[0031] A scheme acquisition module is configured to acquire the force deployment scheme of the two parties of the game.

[0032] A game module is configured to acquire the action instruction of the two parties of the game, perform the game operation on the corresponding force entity model according to the action instruction, feed back the operation result, and generate the next action instruction according to the game result.

[0033] A recording module is configured to record the game process and the game result of the two parties of the game.

[0034] The force entity model is obtained by extracting the attribute of the force entity in the force deployment scheme and modeling the force entity based on each functional parameter in the attribute of the force entity.

[0035] The application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to realize the wargame simulation method of any one of the above.

[0036] The application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the wargame simulation method of any one of the above.

[0037] The application also provides a computer program product, including a computer program, and the computer program is executed by a processor to realize the wargame simulation method of any one of the above.

[0038] The application provides a wargame simulation method and device, which comprises the following steps: obtaining a force deployment scheme of two parties in a game; obtaining action instructions of the two parties in the game, performing a game operation on a corresponding force entity model according to the action instructions, feeding back an operation result, and generating a next action instruction of the two parties in the game according to the game result, wherein the force entity model is obtained by extracting attributes of a force entity in the force deployment scheme and modeling the force entity based on each functional parameter in the attributes of the force entity; and recording a game process and a final game result of the two parties in the game. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort.

[0040] Figure 1 Fig. 1 is a flowchart of a wargame simulation method provided by the present application;

[0041] Figure 2 Fig. 2 is another flowchart of the wargame simulation method provided by the present application;

[0042] Figure 3 Fig. 3 is a third flowchart of the wargame simulation method provided by the present application;

[0043] Figure 4 Fig. 4 is a structural diagram of a wargame simulation device provided by the present application;

[0044] Figure 5 Fig. 5 is a structural diagram of a wargame simulation system provided by the present application;

[0045] Figure 6 Fig. 6 is a structural diagram of an electronic device provided by the present application. DETAILED DESCRIPTION

[0046] In order to make the objects, technical solutions and advantages of the present application clearer, the following will combine the drawings in the present application to clearly and completely describe the technical solutions in the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without any creative effort belong to the protection scope of the present application.

[0047] The application is described below Figures 1-3 A method for wargame simulation is provided.

[0048] Figure 1 A flowchart of a method for wargame simulation is provided, as shown in Figure 1 The method comprises the following steps:

[0049] In step 110, the deployment scheme of the wargame parties is obtained. Specifically, the deployment scheme of the wargame parties can be input into the computer by human-computer interaction by the wargame participants. Alternatively, the deployment scheme is pre-stored in the storage device, and the pre-stored deployment scheme is directly called. In general wargame process, the wargame parties are named as red and blue parties.

[0050] In step 120, the action instruction of the wargame parties is obtained, the corresponding force entity model is operated according to the action instruction, the operation result is fed back, and the next action instruction is generated according to the wargame result. In some embodiments, the simulation wargame is a dynamic game between the red and blue parties, including human-computer wargame in which one party is human and the other party is intelligent agent, and also including mixed game between intelligent agents.

[0051] Through the human-computer interaction unit, an operation interface is provided for the human player in the simulation wargame process, so as to facilitate the human-computer mixed game, and the process of the game can also be rendered to the screen in real time. Since the wargame parties can be human or intelligent agent, when the red and blue parties are human-computer game, the battlefield after the action of the force entity model is first rendered to the screen of human-computer interaction, the next instruction operation of the force entity model is performed by observing the real-time battle record on the screen, and the battlefield situation is packaged into a set of observation data and sent to the intelligent agent. The intelligent agent generates the next operation instruction according to the observation data.

[0052] In step 130, the wargame process and the final wargame result of the wargame parties are recorded. The completed wargame is scored, and the wargame process and the final wargame result are recorded.

[0053] In the specific embodiment, when simulating a game, the simulator environment receives an action instruction of an agent or a human player at each time slice, and performs corresponding operations on the corresponding entity. Then the battlefield after the action is fed back to the human-computer interaction unit, finally rendered to the screen, and the battlefield situation is packaged into a set of observation data for the agent to output to the agent; the battlefield situation includes vector form entity information, matrix form map information and action information, and the entity information includes various attribute parameters of the platforms of the own side. The battlefield information includes the firing area, the enemy and friendly force distribution, and the reachable area of the own side. At the same time, the current instantaneous reward value is automatically calculated according to the current situation at each time slice, which is used to train the reinforcement learning agent.

[0054] In step 120, the entity model is obtained by extracting the attributes of the force entities in the force deployment scheme, and modeling the force entities based on the function parameters in the attributes of the force entities.

[0055] Specifically, the attributes of the force entities in the force deployment scheme include the type, number, various fixed and variable parameters, deployment position of the fire platforms, the type and number of various weapons and ammunition, and the platforms on which the weapons and ammunition are deployed, wherein the attributes include the moving speed and moving direction of the fire platforms.

[0056] When modeling the force entities, the state parameters in the attributes of the force entities not only include the state information of the current time slice, but also include the state information of the previous n time slices, so as to meet the Markov property in the random process. The functions of the force entities include moving, searching, launching missiles, taking off and landing aircraft (aircraft carriers), etc.

[0057] The moving action can directly receive two continuous coordinate values, or can be two discrete values, and the entity can convert the two discrete values into the steering and speed directions of the Gaussian distribution according to the angular velocity interval and the speed interval of the entity. Then the specific heading angle and speed value are obtained by sampling from the two distributions, and the coordinates of the next time slice are calculated.

[0058] Specifically, as shown in Figure 2 The function parameters in the attributes of the force entities include the moving function, and the process of moving simulation of the force entity model specifically includes the following steps:

[0059] Step 210, obtaining a discrete direction value d' and a discrete speed value v', wherein the values of d' and v' are -1, 1 or 0.

[0060] Step 220, querying the attribute database of the force entity to obtain the direction interval [d min ,d max ] and the speed interval [v min ,v max ] of the force entity. min ,d max ] and the speed interval [v min ,v max ] of the force entity.

[0061] Step 230, according to the discrete direction value d' and the end value of the direction interval, the probability distribution of the force entity direction is established. Specifically, according to the discrete direction value d' and the end value of the direction interval, the probability distribution of the force entity direction is established Wherein:

[0062]

[0063] Step 240, according to the discrete speed value v' and the end value of the speed interval, the probability distribution of the force entity speed is established. Specifically, according to the discrete speed value v' and the end value of the speed interval, the probability distribution of the force entity speed is established Wherein:

[0064]

[0065] Step 250, according to the probability distribution of the force entity direction and the probability distribution of the force entity speed, the direction value and the speed value of the force entity are sampled to obtain the continuous direction value d and the continuous speed value v of the force entity.

[0066] Step 260, based on the continuous direction value d and the continuous speed value v, the force entity model is controlled to move.

[0067] The moving action of the force entity model can directly accept two continuous coordinate values, or two discrete values. Since in the mainstream reinforcement learning algorithm, the value-based method can only output discrete action values, the random gradient policy algorithm in the policy-based method is essentially also outputting discrete values, and the Monte Carlo tree search and other planning methods also require the output action to be a discrete value. Therefore, in order to better adapt to the mainstream reinforcement learning algorithm, the force entity model should be able to accept discrete action values and output continuous coordinate values.

[0068] Specifically, after receiving the discrete action value, the force entity model converts it into a Gaussian distribution in the direction of steering and speed according to the angular velocity interval and the speed interval of the unit. Then the specific heading angle and speed value are sampled from the two distributions, and the coordinates of the next time slice are calculated.

[0069] Through the above moving simulation steps of the force entity model, the input simple discrete value can be converted into continuous speed and angular velocity, greatly compressing the action space of the intelligent decision algorithm.

[0070] Further, in the search function of the force entity model, the nearest and most valuable enemy units within the radar range are searched automatically, and the commander decides whether to attack one of the two. Specifically, in the search function, considering that the number of participating fire platforms increases, the number of enemy units within the radar range of the force entity model also increases, and thus the entire action space increases exponentially. Therefore, the nearest and most valuable enemy units within the radar range are searched automatically, and the commander decides whether to attack one of the two.

[0071] Specifically, as shown in Figure 3 the function parameters in the attributes of the force entity include a search function, and the process of simulating the search of the force entity model includes the following steps:

[0072] Step 310, the mutual distance between the force entity model and the visible enemy units within the radar range centered on the force entity model is calculated, and the enemy unit closest to the force entity model is marked as p0;

[0073] Step 320, the value of all visible enemy units within the radar range centered on the force entity model is calculated according to the state (predicted armor value, predicted remaining missile quantity) and cost of the enemy units, and the enemy unit with the highest value is selected and marked as p1;

[0074] Step 330, when the received instruction is 0, the force entity model outputs p0 as the search result;

[0075] Step 340, when the received instruction is 1, the force entity model outputs p1 as the search result.

[0076] Through the above steps of simulating the search function of the force entity model, the nearest and most valuable enemy units within the radar range of the force entity model are searched automatically, which compresses the action space of the intelligent decision algorithm, and at the same time, the human expert knowledge is covered in the action space construction process.

[0077] In some embodiments, the function parameters in the attributes of the force entity include a missile launch function, and the process of simulating the missile launch of the force entity model includes:

[0078] According to the distance between the force entity model and the enemy units, the types of the two units, and whether the enemy units have a defensive secondary gun, the probability of hitting the enemy units by launching a missile is calculated;

[0079] When it is judged that the probability of hitting the enemy units is greater than a preset value, the force entity model obtains the instruction to launch a missile.

[0080] Specifically, in the missile launching function of the force entity model, the entity calculates the hit probability according to a specific scene, and specific factors considered include the distance between the two, the type between the two, and whether the target of attack has a defense secondary gun.

[0081] The technical solution of the present application, in view of the characteristics of the current mainstream intelligent game algorithm, realizes modeling of various fire platforms and weapon equipment in the simulator, further designs a force game action mechanism suitable for reinforcement learning algorithm, and returns observation values for agent training. During simulation, two interfaces for human manipulation and agent manipulation are provided, rendering functions are provided, and man-machine combat is facilitated. After the simulation is completed, the player's score can be automatically evaluated, and game data can be recorded for review and training of neural networks. This simulator provides a basic environment for the study of intelligent decision-making, military decision-making, and artificial intelligence, facilitating researchers to conduct algorithm research thereon.

[0082] The force game simulation device provided by the present application is described below, and the force game simulation device described below can be correspondingly referred to the force game simulation method described above.

[0083] As Figure 4 shown, the present application also provides a force game simulation device, comprising the following modules: a scheme acquisition module 410, a game module 430, and a recording module 440.

[0084] Specifically, the scheme acquisition module 410 is used to acquire the force deployment scheme of the two game parties.

[0085] The game module 430 is used to acquire the action instruction of the two game parties, perform game operation on the corresponding force entity model according to the action instruction, feed back the operation result, and generate the next step action instruction according to the game result. The force entity model is obtained by modeling the force entity in the force deployment scheme by the modeling module 420 based on the modeling of the force entity according to the function parameters in the attribute of the force entity.

[0086] The recording module 440 is used to record the game process and game result of the two game parties.

[0087] As Figure 5 shown, in some embodiments, the force game simulation system for the method of the present application comprises the following modules: a human-computer interaction module 510, an entity modeling module 520, a simulation game module 530, and a scoring and recording module 540.

[0088] In the entity modeling stage, the human-computer interaction module 510 is used to input the type, quantity, various fixed and variable parameters, deployment position of the fire platform of the red and blue forces, and the type and quantity of various weapons and ammunition, and the platform deployed by them; in the simulation game stage, an operation interface is provided for the human commander to facilitate human-computer hybrid game, and the game process can also be rendered to the screen in real time.

[0089] The entity modeling module 520 models each entity according to the force deployment scheme input by the human-computer interaction module. The input red and blue force deployment scheme includes the type, quantity, various fixed and variable parameters, deployment position of the fire platform, and the type and quantity of various weapons and ammunition, and the platform deployed by them. The attributes of the entity are shown in Table 1 and Table 2.

[0090] Table 1: Fire platform attributes

[0091]

[0092]

[0093] Table 2: Weapon and missile attributes

[0094]

[0095] The simulation game module 530 uses the entity model created by the entity modeling module 520 to perform dynamic game between the red and blue forces. The environment simulator at each time slice receives an action instruction from the agent or human commander, and performs corresponding operations on the corresponding entity. The battlefield situation is packaged into a set of observation data for the agent, including entity information, map information and action information. The entity information includes various attribute parameters of the platform, as shown in Table 3. The map information includes the firing area, enemy and friendly force distribution, and the reachable area of the friendly forces, as shown in Table 4. The action information stores the action samples of each unit at the current time step, and each individual action data format is shown in Table 5.

[0096] Table 3: Entity information

[0097]

[0098]

[0099] Table 4: Map information

[0100]

[0101] Table 5: Action information

[0102]

[0103]

[0104] The score record module 540 scores a game of chess completed by the simulation game module, and the score is based on the completion time, the number of destroyed enemy units, the number of friendly units damaged, the consumption of ammunition of the friendly units, etc.Then the process and the result are recorded in two formats and stored in the database.The first format can be re-input into the simulator, and the whole game process can be reviewed on the screen for the reference of the human commander.The second format processes the game plan floating point optimization and records it in the form of tf.recard as experience data to provide the neural network input of the intelligent agent so as to improve the intelligence level of the intelligent agent.

[0105] The application discloses a wargame simulation system for intelligent game, and the modeling of various fire platforms and weapon equipment is realized in the simulation system according to the characteristics of the current mainstream intelligent game algorithm, a wargame action mechanism suitable for reinforcement learning algorithm is further designed, and observation values for intelligent agent training are returned.In the simulation process, two interfaces of human operation and intelligent agent operation are provided, and a rendering function is provided to facilitate man-machine combat.After the simulation is completed, the scores of players can be automatically evaluated, and game data can be recorded for review and neural network training.The simulator provides a basic environment for the research of intelligent decision, military decision and artificial intelligence, and facilitates researchers to make algorithm research based on the basic environment.

[0106] Figure 6 An example of an entity structure diagram of an electronic device is shown in Figure 6 As shown in the figure, the electronic device can include a processor 610, a communications interface 620, a memory 630 and a communications bus 640, wherein the processor 610, the communications interface 620 and the memory 630 complete mutual communication through the communications bus 640.The processor 610 can call the logic instructions in the memory 630 to execute the wargame simulation method, and the method includes the following steps: obtaining the deployment scheme of the forces of the two parties in the game;obtaining the action instruction of the two parties in the game, performing the game operation on the corresponding force entity model according to the action instruction, feeding back the operation result, and generating the next action instruction according to the game result;recording the game process and the final game result of the two parties in the game;wherein the force entity model is obtained by modeling the force entity based on the attributes of the force entity in the deployment scheme of the force.

[0107] Further, the logic instructions in the memory 630 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or parts of the present application that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0108] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the wargame simulation method provided by the above-mentioned methods to execute the wargame simulation method, the method comprising the following steps: obtaining a force deployment scheme of a wargame party; obtaining a movement instruction of the wargame party, performing a wargame operation on a corresponding force entity model according to the movement instruction, feeding back an operation result, and generating a next movement instruction of the wargame party according to a wargame result; recording a wargame process and a final wargame result of the wargame party; wherein the force entity model is obtained by extracting attributes of a force entity in the force deployment scheme and modeling the force entity based on each functional parameter in the attributes of the force entity.

[0109] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the wargame simulation method provided by the above-mentioned methods to execute the wargame simulation method, the method comprising the following steps: obtaining a force deployment scheme of a wargame party; obtaining a movement instruction of the wargame party, performing a wargame operation on a corresponding force entity model according to the movement instruction, feeding back an operation result, and generating a next movement instruction of the wargame party according to a wargame result; recording a wargame process and a final wargame result of the wargame party; wherein the force entity model is obtained by extracting attributes of a force entity in the force deployment scheme and modeling the force entity based on each functional parameter in the attributes of the force entity.

[0110] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0111] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0112] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for simulating wargames, characterized in that, The method includes: Obtain the troop deployment plans of both sides in the game; The action instructions of the two players are obtained, and the corresponding troop entity models are used to perform the game operation according to the action instructions. The operation results are fed back, and the two players generate the next action instructions according to the game results. Record the game process and the final result between the two players; The troop entity model is obtained by extracting the attributes of troop entities in the troop deployment plan and modeling the troop entities based on the functional parameters in the attributes of the troop entities. The functional parameters in the attributes of the troop entity include movement functionality. The process of simulating movement of the troop entity model includes: Obtain discrete direction value d′ and discrete velocity value v′, where the values ​​of d′ and v′ are -1, 1 or 0; Query the attribute database of the troop entity to obtain the directional range [d] of the troop entity. min ,d max ] and speed range [v min ,v max ]; Based on the discrete direction value d′ and the end values ​​of the direction interval, a probability distribution of the direction of the troop entity is established; Based on the discrete velocity value v′ and the endpoints of the velocity interval, a probability distribution of the velocity of the troop entity is established. Based on the probability distribution of the direction of the troop entity and the probability distribution of the velocity of the troop entity, the direction value and velocity value of the troop entity are sampled to obtain the continuous direction value d and continuous velocity value v of the troop entity. The movement of the troop entity model is controlled based on the continuous direction value d and the continuous velocity value v.

2. The wargame simulation method according to claim 1, characterized in that, The functional parameters among the attributes of the troop entity include a search function. The process of performing a search simulation on the troop entity model includes: Calculate the mutual distance between the troop entity model and the visible enemy units within the radar range centered on the troop entity model, and select the enemy unit closest to the troop entity model as p0; Based on the status and cost of the enemy units, calculate the value of all visible enemy units within the radar range centered on the troop entity model, and select the enemy unit with the highest value as p1; When the received instruction is 0, the troop entity model will output p0 as the search result; When the received instruction is 1, the troop entity model will output p1 as the search result.

3. The wargame simulation method according to claim 1, characterized in that, The functional parameters of the force entity's attributes include the missile launch function. The process of simulating missile launch on the force entity model includes: The probability of a missile hitting the enemy unit is calculated based on the distance between the troop entity model and the enemy unit, the types of units on both sides, and whether the enemy unit has a defensive secondary gun. When the probability of hitting the enemy unit is greater than a preset value, the force entity model receives the command to launch a missile.

4. The wargame simulation method according to claim 1, characterized in that, The state parameters in the attributes of the military entity include: The status information of the troop entity in the current time slice and the status information in the preset time period before the current time slice.

5. The wargame simulation method according to claim 1, characterized in that, The feedback operation result, and the generation of the next action instruction by both players based on the game result, specifically include: The battlefield after the action of the troop entity model is rendered onto the human-computer interaction screen; The battlefield situation is packaged into a set of observation data and sent to the intelligent agent. The intelligent agent then generates the next operation instruction based on the observation data.

6. A simulation device for wargames, characterized in that, include: The plan acquisition module is used to acquire the troop deployment plans of both sides in the game; The game module is used to obtain the action instructions of the two players, perform game operations on the corresponding troop entity models according to the action instructions, and provide feedback on the operation results. The two players generate the next action instructions based on the game results. The recording module is used to record the game process and the game result of the two players. The troop entity model is obtained by extracting the attributes of troop entities in the troop deployment plan and modeling the troop entities based on the functional parameters in the attributes of the troop entities. The functional parameters in the attributes of the troop entity include movement functionality. The process of simulating movement of the troop entity model includes: Obtain discrete direction value d′ and discrete velocity value v′, where the values ​​of d′ and v′ are -1, 1 or 0; Query the attribute database of the troop entity to obtain the directional range [d] of the troop entity. min ,d max ] and speed range [v min ,v max ]; Based on the discrete direction value d′ and the end values ​​of the direction interval, a probability distribution of the direction of the troop entity is established; Based on the discrete velocity value v′ and the endpoints of the velocity interval, a probability distribution of the velocity of the troop entity is established. Based on the probability distribution of the direction of the troop entity and the probability distribution of the velocity of the troop entity, the direction value and velocity value of the troop entity are sampled to obtain the continuous direction value d and continuous velocity value v of the troop entity. The movement of the troop entity model is controlled based on the continuous direction value d and the continuous velocity value v.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the wargame simulation method as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the wargame simulation method as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the wargame simulation method as described in any one of claims 1 to 5.

Images